Engineering of the synthetic metabolic pathway for ... - IS MUNI

MASARYK UNIVERSITY

FACULTY OF SCIENCE

LOSCHMIDT LABORATORIES

DEPARTMENT OF EXPERIMENTAL BIOLOGY

Engineering of the synthetic metabolic

pathway for biodegradation of

environmental pollutant

Doctoral dissertation

Pavel Dvořák

Supervisors:

Prof. Mgr. Jiří Damborský, Dr.

Doc. RNDr. Zbyněk Prokop, Ph.D.

BRNO 2014

Poděkování

Na tomto místě bych chtěl poděkovat svému školiteli Jiřímu Damborskému za

profesionální vedení, schopnost a odhodlání dělat kvalitní, nepodlézavou vědu, ochotu

setkávat se, diskutovat a radit, která není ani mezi školiteli postgraduálních studentů zcela

automatická, a konečně za jeho víru ve zdárné konce, která mě dovedla až k sepsání této

práce.

Velmi děkuji i Zbyňku Prokopovi, mému školiteli specialistovi, za jeho optimismus a cenné

rady a všem svým současným i minulým kolegům, kteří přispívali a přispívají k tomu, že

Loschmidtovy laboratoře jsou nejenom špičkovým vědeckým týmem, ale i příjemným

místem pro práci, kvůli kterému se člověk každé pondělní ráno rád přiměje vstát z postele.

Největší dík ale patří mé ženě Monice za její lásku, přátelství, toleranci a schopnost vracet

mě z badatelských výšin zpátky nohama na zem.

Bibliographic entry

Author: Mgr. Pavel Dvořák

Loschmidt Laboratories

Department of Experimental Biology

Faculty of Science

Masaryk University

Title of dissertation: Engineering of the synthetic metabolic pathway for

biodegradation of environmental pollutant

Study Programme: Biology

Field of study: Molecular and Cellular Biology

Supervisor: Prof. Mgr. Jiří Damborský, Dr.

Supervisor-specialist: doc. RNDr. Zbyněk Prokop, PhD.

Year of defence: 2014

Keywords: biocatalysis; biodegradation; kinetic modelling; metabolic

engineering; in vitro multi-enzyme reaction; synthetic

biology; 1,2,3-trichloropropane

Bibliografický záznam

Autor: Mgr. Pavel Dvořák

Loschmidtovy laboratoře

Ústav experimentální biologie

Přírodovědecká fakulta

Masarykova univerzita

Název disertace: Inženýrství syntetické metabolické dráhy pro biodegradaci

environmentálního polutantu

Studijní program: Biologie

Studijní obor: Molekulární a buněčná biologie

Školitel: Prof. Mgr. Jiří Damborský, Dr.

Školitel specialista: doc. RNDr. Zbyněk Prokop, PhD.

Rok obhajoby: 2014

Klíčová slova: biokatalýza; biodegradace; kinetické modelování;

metabolické inženýrství; in vitro multi-enzymová reakce;

syntetická biologie; 1,2,3-trichlorpropan

Scio me nihil scire.

(I know that I know nothing)

Socrates

Success consists of going from failure to failure without loss of enthusiasm.

Winston Churchill

© Pavel Dvořák, Masaryk University 2014

CONTENT

MOTIVATION 1

ABSTRACT 2

ABSTRAKT 4

INTRODUCTION 7

1. Emerging technologies for engineering of metabolic pathways 7

1.1 Metabolic engineering 9

1.1.1 Strategies and tools 10

1.2 Synthetic biology 17


1.3 Protein engineering 21


1.4 Perspectives 24

2. Engineering of biodegradation pathways 25

2.1 Strategies and tools 27

2.2 Perspectives 32

3. Engineering of the synthetic metabolic pathway for

biodegradation of 1,2,3 trichloropropane 34

3.1 Halogenated hydrocarbons and 1,2,3-trichloropropane 34

3.2 Overview of 1,2,3-trichloropropane pathway engineering 38

CONTRIBUTION TO THE RESULTS 47

CHAPTER 1

In vitro assembly and immobilization of the synthetic pathway for biodegradation of toxic recalcitrant pollutant 1,2,3- trichloropropane 49

SUPPLEMENTARY TABLES AND FIGURES 69

CHAPTER 2

Maximizing the efficiency of in vitro multi-enzyme process by

stoichiometry optimization 79


CHAPTER 3

Computer-assisted engineering of the synthetic pathway for

biodegradation of 1,2,3-trichloropropane in heterologous host E. coli 99


CHAPTER 4

Assembly of the synthetic pathway for biodegradation of

1,2,3-trichloropropane in Pseudomonas putida KT2440 CF1 131


SUMMARY 147

REFERENCES 148

CURRICULUM VITAE 160

LIST OF PUBLICATIONS 162

LIST OF CONTRIBUTIONS AT CONFERENCES AND SYMPOSIA 163

Motivation

- 1 -

MOTIVATION

Nature possess a great potential to cope itself with many problems arising from

continuously increasing human activity on the Earth. Among others, this potential is

hidden in astonishing variability of metabolic pathways of living organisms. For example,

it is not completely rare phenomenon that bacterium can adapt its metabolic traits for

new substrate and break down a toxic polluting compound. However, in certain cases, the

evolution is not fast enough or ends in a deadlock. Such challenges can be possibly solved

by state-of-the-art tools of recently established scientific disciplines including metabolic

engineering, protein engineering and synthetic biology. Still, the huge complexity of

dynamic processes in living cell represents a major bottleneck for any attempts to

rationally engineer natural or synthetic metabolic pathways for the purpose of

biodegradation of environmental pollutants or biosynthesis of value added chemicals. The

complexity can be significantly reduced by reconstructing selected multi-enzyme

reactions in vitro or by assembling orthogonal metabolic modules within an organism.

Such model studies help us to understand the dynamic behavior of metabolic pathways

and the obtained knowledge can be step-by-step utilized for rational engineering of living

systems.

The objectives of the Ph.D. project and this Thesis:

1. Introduction to the fields of metabolic engineering and synthetic biology with special

attention devoted to the knowledge-based engineering of biodegradation pathways;

prologue to the selected model system - synthetic metabolic pathway for

biodegradation of anthropogenic pollutant 1,2,3-trichloropropane.

2. In vitro reconstruction and immobilization of the model pathway. Development of the

biodegradation process based on immobilized enzymes.

3. Detailed kinetic characterization of employed enzymes, development and validation of

kinetic model for the pathway in vitro.

4. Optimization of the pathway in vitro by employment of suitable engineered enzymes

and balancing of biocatalysts' stoichiometry using kinetic modelling.

5. Application of obtained knowledge and synthetic biology tools for rational engineering

of the pathway in heterologous host Escherichia coli and dissection of the pathway

bottlenecks in vivo.

6. Selection of suitable microbial host (chassis) for biodegradation of

1,2,3-trichloropropane and further evolution of the synthetic pathway.

Abstract

- 2 -

ABSTRACT

This Thesis describes the application of synthetic biology and metabolic engineering

approaches for rational redesign of metabolic pathway for biodegradation of important

environmental pollutant 1,2,3-trichloropropane (TCP). The emerging technologies for

engineering of biosynthetic and biodegradation pathways are described in the

Introduction of the Thesis and prologue to the origins of synthetic TCP route is provided.

The pathway consisting of three enzymes – haloalkane dehalogenase, haloalcohol

dehalogenase and epoxide hydrolase - from two different microorganisms can convert

toxic TCP into harmless product glycerol but suffers from several important bottlenecks.

The four studies described in the Results section of the Thesis aim at rational dissection of

these handicaps, optimization of pathway performance and construction of biocatalyst

utilizable for TCP removal from the contaminated sites.

Chapter 1 of the Results section describes reconstruction of the model pathway in in

vitro conditions and developing of a novel biotechnology for TCP transformation based on

immobilized enzymes. The efficiency of the pathway was enhanced by employment of

engineered haloalkane dehalogenase with improved activity toward TCP and the route

was immobilized in the form of purified enzymes or cell-free extracts. The performance of

the three-enzyme system was tested in batch and continuous operations. The study

provides the first available report on the use of an immobilized synthetic metabolic

pathway employing engineered enzyme for the biotransformation of the toxic industrial

waste into desirable commodity chemical.

Further improvement of the reaction efficiency can be achieved by tuning enzymes'

stoichiometry. Development of the workflow for maximizing the efficiency of in vitro

multi-enzyme process by stoichiometry optimization is addressed in Chapter 2. In this

study, we employed kinetic modelling to maximize the efficiency of a three-enzyme

system based on in vitro assembled TCP pathway. Mathematical modelling and one-pot

multi-enzyme laboratory experiments provided detailed insight into pathway dynamics,

enabled the selection of suitable engineered enzyme and afforded high yield of the final

product glycerol, while minimizing biocatalyst loadings. The study highlights the potential

of kinetic modelling for industrial biocatalysis and presents a broadly applicable strategy

for optimizing multi-enzyme processes.

Chapter 3 describes application of previously developed kinetic model for

computer-assisted engineering of TCP pathway in heterologous host Escherichia coli. We

assembled TCP route in the laboratory strain E. coli BL21 (DE3), and used it as an

orthogonal biological system for thorough investigation of pathway bottlenecks in vivo.

Variants of the pathway employing wild-type or engineered haloalkane dehalogenase

were designed using modified mathematical model. The E. coli recombinants with

optimized and non-optimized stoichiometry of pathway enzymes were constructed and

characterized in terms of their viability in presence of TCP and degradation efficiency. The

validated model was used to quantitatively describe the kinetic limitations of currently

available enzyme variants, and to predict improvements required for further pathway

Abstract

- 3 -

optimization. The study highlights the potential of rational engineering of microorganisms

for the degradation of toxic anthropogenic compounds.

Last chapter of the Results section, Chapter 4, is focused on transfer of TCP pathway

from laboratory strain E. coli BL21 (DE3) into robust heterologous host Pseudomonas

putida KT2440, equipped with repertoire of metabolic and physiological functions that

make it less susceptible to stress accompanying degradation of some problematic

compounds including chlorinated organic pollutants. We decided to implant the TCP

pathway into this host to verify its suitability for TCP biodegradation and further tuning of

the synthetic route by in vivo evolution. Synthetic operon encoding three enzymes from

TCP pathway was assembled and introduced into P. putida chromosome. Obtained

constructs will be further characterized and compared with their E. coli counterparts. The

concluding section of the chapter discusses future directions in engineering of the

synthetic metabolic pathway for biodegradation of TCP.

In summary, the Thesis presents an innovative concept for rational engineering of a

synthetic metabolic pathway for degradation of toxic recalcitrant compound based on

detailed understanding of studied system and exploitation of in vitro, in silico and in vivo

tools of synthetic biology and metabolic engineering.

Abstrakt

- 4 -

ABSTRAKT

Tato práce popisuje využití metod syntetické biologie a metabolického inženýrství k

racionálnímu redesignu metabolické dráhy pro biodegradaci důležitého

environmentálního polutantu 1,2,3-trichlorpropanu (TCP). V úvodu práce jsou čtenáři

přiblíženy nové technologie a přístupy pro inženýrství biosyntetických a biodegradačních

drah společně s původem a počátky studia syntetické TCP dráhy. Dráha, která se skládá ze

tří enzymů – halogenalkandehalogenasy, haloalkoholdehalogenasy a epoxidhydrolasy –

pocházejících ze dvou různých mikroorganismů, dokáže přeměnit toxický TCP na

neškodný produkt glycerol. Efektivita této přeměny je však negativně ovlivněna několika

významnými limitacemi dráhy. Čtyři studie, které jsou shrnuty ve výsledkové části této

práce, se snaží tato omezení popsat, racionálně optimalizovat fungování dráhy a připravit

biokatalyzátor, který by mohl být využit k dekontaminaci míst znečištěných TCP.

Kapitola 1 výsledkové části popisuje rekonstrukci modelové dráhy v podmínkách in

vitro a vývoj nové biotechnologie pro transformaci TCP založené na imobilizovaných

enzymech. Efektivita dráhy byla zvýšena zapojením halogenalkandehalogenasy s uměle

zvýšenou aktivitou s TCP a dráha byla imobilizována v podobě purifikovaných enzymů či

bezbuněčných extraktů. Výkon tříenzymového systému byl testován v třepaných lahvích i

v průtokovém bioreaktoru. Tato studie je jedinou doposud publikovanou prací popisující

využití imobilizované syntetické metabolické dráhy s enzymem upraveným proteinovým

inženýrstvím pro biotransformaci toxické průmyslové odpadní látky na žádanou

komoditní chemikálii.

Dalšího zvýšení efektivity reakce může být dosaženo úpravou stechiometrie enzymů

v dráze. Kapitola 2 popisuje vývoj pracovního postupu pro maximalizaci efektivity in vitro

multienzymového procesu pomocí optimalizace stechiometrie. V této studii jsme využili

kinetické modelování pro zlepšení efektivity tříenzymového systému založeného na in

vitro rekonstruované TCP dráze. Matematické modelování podpořené laboratorními

experimenty poskytlo detailní vhled do dynamického fungování dráhy, umožnilo výběr

vhodného enzymu upraveného proteinovým inženýrstvím a přispělo tak k získání

maximálního možného výtěžku finálního produktu glycerolu při využití minimálního

potřebného množství enzymů. Studie zdůrazňuje potenciál kinetického modelování pro

průmyslovou biokatalýzu a nabízí široce využitelnou strategii pro optimalizaci

multienzymových procesů.

Kapitola 3 se věnuje aplikaci dříve popsaného kinetického modelu v inženýrství TCP

dráhy v heterologním hostiteli, bakterii Escherichia coli. TCP dráha byla sestavena v

laboratorním kmeni E. coli BL21 (DE3) a tento ortogonální systém byl využit k důkladné

analýze limitací dráhy v podmínkách in vivo. Za využití modifikovaného matematického

modelu byly navrženy varianty dráhy s divokým typem halogenalkandehalogenasy i s

mutanty připravenými proteinovým inženýrstvím. Byly zkonstruovány rekombinantní

bakterie E. coli s optimalizovanou a neoptimalizovanou stechiometrií enzymů dráhy.

Konstrukty byly experimentálně charakterizovány z pohledu jejich schopnosti degradovat

TCP a přežít v jeho přítomnosti. Ověřený model byl použit ke kvantitativnímu popisu

Abstrakt

- 5 -

kinetických limitací v současnosti dostupných variant enzymů dráhy a k predikci dalších

kroků nutných ke zlepšení limitních parametrů. Tato studie zdůrazňuje potenciál

racionálního inženýrství mikroorganismů pro degradaci toxických látek antropogenního

původu.

Poslední kapitola výsledkové části, Kapitola 4, je zaměřena na popis přenosu TCP

dráhy z laboratorního kmene E. coli BL21 (DE3) do robustního hostitele, bakterie

Pseudomonas putida KT2440. Tato bakterie je přirozeně vybavena repertoárem

metabolických a fyziologických funkcí, které ji dělají méně náchylnou ke stresu

doprovázejícímu degradaci některých problematických látek včetně chlorovaných

organických polutantů. Rozhodli jsme se přenést TCP dráhu do tohoto hostitele, abychom

ověřili jeho použitelnost pro účely biodegradace TCP a dalšího vývoje dráhy pomocí in

vivo evoluce. Studie popisuje přípravu syntetického operonu kódujícího enzymy TCP

dráhy, jeho vnesení do chromosomu P. putida a základní charakterizaci vzniklých

konstruktů. Získané rekombinantní bakterie budou dále charakterizovány a srovnány s

konstrukty E. coli. V závěru kapitoly autor diskutuje možné budoucí směry dalšího

inženýrství syntetické metabolické dráhy pro biodegradaci TCP.

Tato práce tak prezentuje inovativní koncept racionálního inženýrství syntetické

metabolické dráhy pro degradaci toxického environmentálního polutantu, založený na

detailní znalosti studovaného biologického systému a využití in vitro, in silico a in vivo

nástrojů syntetické biologie a metabolického inženýrství.

Introduction

- 7 -

INTRODUCTION

1. Emerging technologies for engineering of metabolic

pathways

Metabolism is a network of life-sustaining biochemical reactions in living cells

catalyzed by enzymes. These reactions can be divided into two principal categories: (i)

catabolic, that break down organic matter and produce energy through cellular

respiration, and (ii) anabolic, that use energy to synthesize biomolecules. Series of

catabolic or anabolic enzymatic reactions are realized via metabolic pathways. The

metabolic pathways are closely interconnected with cell signalling pathways and, taking

into account the central dogma of molecular biology, also with information-storing

molecules DNA and RNA. Mutual dependences of these basic cell components and

principles of their orchestration for the purpose of cell maintenance, reproduction or

adaptation to the new conditions are still not fully understood [1]. The dynamic

complexity of a metabolism is encoded already in the word itself that is derived from

Greek μεταβολή [metabolē], which means a change. This complexity shelters yet

unexplored fortune and endless space of possible solutions for many problems of

mankind to come in 21st century, but is also an evident obstacle for our attempts to

domesticate microbes, animals or plants for biosynthesis of value-added chemicals or for

biodegradation of polluting compounds.

In contrast to their purely chemical counterparts, biocatalytic reactions can proceed

in moderate conditions, but with high rate and specificity. The humans have taken

advantage of the biochemical trails of living organisms in classical biotechnology for

thousands of years without the knowledge of molecular basis of cellular processes. Yeasts,

bacteria and fungi have been used in agriculture, food production and medicine. Recently,

the modern biotechnology adopted the principles of sustainable living and our effort to

decode and exploit the metabolic processes and the life itself has significantly increased.

In the last decades of 20th century, biologists and biochemists focused on study of complex

natural systems, e.g., bacterial cell, and tried to dissect them into pieces and understand

the structure and function of basic parts like nucleic acids and proteins [2–4]. At the same

time, they learned how to use some of the discovered biological components, e.g.,

restriction enzymes, DNA ligases or DNA polymerases, as tools for system manipulation

[5,6]. These initial discoveries in 1950s, 60s, and 70s gave rise to the recombinant DNA

technology and molecular biology.

Since that time, we have been moving from analytical age to a synthetic age. Now,

variety of parts and tools is available and the possibilities expanded together with

emerging fields of metabolic engineering, synthetic biology and protein engineering [7,8].

The toolkits of these three scientific disciplines are being combined to utilize and tailor

biocatalytic reactions both in vivo and in vitro (Figure 1). The increasing computational

Introduction

- 8 -

power and invention of chemical DNA synthesis allow the biologists and chemists to adopt

engineering principles and construct some biological parts like DNA molecules or

enzymes also de novo [9,10]. The recent corporate interest in metabolic engineering and

synthetic biology was highlighted by their placing among the TOP 10 emerging

technologies for 2012 selected by the World Economic Forum. These disciplines are also

embedded in the European Union call appeal for the Knowledge Based Bio-Economy

towards 2020. The detailed discussion on all three subjects would be behind the scope of

this text. Nevertheless, the fundamentals described in the following paragraphs will

provide sufficient introduction into the matter and explanation of important terms used in

the Results section of the Thesis.

Figure 1. Utilization of metabolic engineering, protein engineering and synthetic biology for tailoring metabolic pathways in vivo and in vitro. The toolkits and the levels of operation of three emerging technologies partially overlap. Synthetic biology targets predominantly nucleic acid level and manipulates the DNA and RNA parts in order to design controllable biological systems. Protein engineering aims to optimize stability and catalytic properties of crucial pathway enzymes. Metabolic engineering balances expression of pathway enzymes (blue, red and green circle), minimizes accumulation of intermediates (M1, M2), improves uptake of substrates (S) and maximizes formation of products (P) in the target pathway.

Introduction

- 9 -

1.1 Metabolic engineering

Metabolic engineering (ME) was firstly defined as a new scientific discipline by James

E. Bailey in 1991 as "...the improvement of cellular activities by manipulations of

enzymatic, transport, and regulatory functions of the cell with the use of recombinant

DNA technology." [11]. The first international conference on ME was held in 1996 in

Danvers, Massachusetts, USA, and in 1998 the journal Metabolic Engineering was

established. The field has evolved primarily to allow alternative biological production of

drugs, fuels, and biomaterials or biodegradation of anthropogenic pollutants by

engineered organisms. The effort is driven by the shortage of fossil reserves and

accumulation of toxic man-made chemicals in the environment. The ME has a great

potential for growth in following decades. Up to 20% of global chemistry market

(estimated at 2,292 billion US$) could by covered by biotechnological products by 2020

[12].

The ME is now understood as a practice of optimizing genetic and regulatory

processes within cells to: (i) improve the yield and productivity of native products

synthesized by organisms, (ii) extend the range of substrates or improve the uptake of

substrate, or (iii) establish production of products that are new to the host cell [7,13,14].

These goals can be achieved either by engineering of natural metabolic pathways present

in the host cell or synthetic routes assembled from enzymes (genes) originating from

different organisms. Metabolic engineers usually model these biocatalytic reactions using

mathematical apparatus, calculate a yield of useful products, and determine the

constraints for their production. Computational and experimental tools are applied to

overcome these constraints and establish cost effective process. Maximum yield or uptake

of desired substance must be balanced with the natural survival needs of the host cell.

Bacteria are widely accepted host organisms in ME. The most explored model

organism Escherichia coli is also the most frequently used host for molecular cloning and

heterologous expression of recombinant enzymes in ME [15]. Its major advantages are

sequenced genome (4.6 Mbp) and wide range of tools developed for its genetic

manipulation, rapid growth, fast doubling time (20-30 min) and high cell densities

achieved in simple synthetic media. E. coli BL21(DE3) (E. coli B F– dcm ompT hsdS(rB–,

mB–) gal λ(DE3)) is widely used strain for overexpression of recombinant proteins and

proved its utility also in many ME studies. The host is a lysogen of λDE3 and contains the

T7 RNA polymerase gene, under the control of the lacUV5 promoter, integrated into the

chromosome. IPTG is used to induce the expression of recombinant proteins cloned into

vectors downstream of a T7 RNA promoter and transformed into the BL21 (DE3) cells.

Other bacterial hosts utilized in academic or industrial laboratories are for example

Bacillus subtilis (production of food enzymes), Pseudomonas putida (biodegradation of

organic pollutants), Streptomyces (production of antibiotics) or cyanobacteria (biofuel

production). In certain cases, properties of procaryotic cells are not sufficient for

expression of target recombinant gene, e.g., due to the missing posttranslational

Introduction

- 10 -

modifications, and eucaryotic hosts like Saccharomyces cerevisiae (production of

bioethanol or terpenoids), fungi (production of antibiotics), mammalian or insect cells

(production of biopharmaceuticals) are employed. The selection of the suitable host

organism is the crucial step at the beginning of any ME project aimed at in vivo pathway

engineering. The engineering of microbial hosts will be discussed in the Introduction and

the Results parts of the Thesis.

1.1.1 Strategies and tools

Microorganisms have been employed for human benefit for thousands of years and

genetically manipulated for decades. Traditionally, species that naturally produced a

desired molecule were identified and then improved through classical strain engineering

on the basis of random mutagenesis induced by chemical (e.g., nitrosoguanidine) or

physical (e.g., UV irradiation) mutagens, application of mutator strains or adaptive

laboratory evolution in chemostat and subsequent screening for the best microbial

factories [7]. More recently, random gene knock-outs and overexpression through

transposon mutagenesis or whole-genome shuffling of selected parental strains with

beneficial properties were employed in a hunt for phenotypic improvement [16,17]. This

has been a very efficient strategy and has resulted in low-cost production of many

different chemicals including penicillin, citric acid, glutamate or lysine, to name a few

[18,19]. However, demanding screening and resulting variants with many undefined

mutations are disadvantageous characteristics of the classical approach.

The attempts to tune production strains rationally appeared with the advent of ME

and development of more sophisticated genetic engineering tools. The complexity of

solved problems made the field become highly interdisciplinary joining together

molecular biology, bioinformatics, microbiology, biochemistry, mathematics or system

biology and many omics disciplines, e.g., metabolomics, proteomics, genomics and

transcriptomics. Typical ME project now consists of three basic steps including: (i)

pathway selection and design, (ii) selection of suitable pathway enzymes, and (iii)

pathway optimization (Figure 2) [20]. Numerous computational and experimental tools

can be utilized to carry out these three steps. Thorough repertoires are listed in some

recent excellent reviews [20–23] and the following paragraphs will provide the overview

of only the most common tools.

Computational tools. The databases of anabolic and catabolic pathways like MetaCyc

or University of Minnesota Biocatalysis/Biodegradation Database can be used for

identification of pathway that can break down or produce a compound of interest. A more

advanced tool, Biochemical Network Integrated Computational Explorer (BNICE), predicts

also unknown pathways that are potentially chemically feasible, taking into account the

starting compound and/or product, the requested length of the pathway and the broader

reaction rules of the Enzyme Commission classification system [24]. Softwares like

CellDesigner or Cytoscape can visualize the defined reaction or metabolic-signalling

Introduction

- 11 -

interaction networks [25,26]. The databases and servers archiving the information about

enzymes and their structural or kinetic properties, e.g., BRENDA, BioCyc, KEGG, Protein

Data Bank, ExPASy, and databases providing access to the physical, chemical and

structural data of possible metabolites, e.g., ChemSpider or NCBI PubChem, are often

employed during the second step of the ME project to collect all available information

about the basic building blocks of selected metabolic pathway. The pathway of choice can

be assembled and engineered either in an appropriate natural or heterologous host

organism or in vitro, out of the context of living cell.

Figure 2. Workflow and tools for design and engineering of pathways in microbial chemical factories. Adopted from [20] and modified.

Pathway optimization is usually the most time- and resource- demanding part of the

process. The product titres reaching grams per litre of cell culture for pharmaceuticals and

hundreds of grams per litre for commodity chemicals are required to make the developed

biotechnology economically feasible [27]. The minimal-knowledge-based optimization of

the selected metabolic route requires the information at least on topology of the reaction

network (cell compartment), concentrations of metabolites, and the carbon flux through

the pathway. Flux is the reaction rate (unit mol.dm-3.s-1) connecting two metabolites

(A→B) or rate at which material is processed through the whole pathway [28]. Fluxes are

calculated in a steady state, thus do not describe the dynamics of enzymatic reactions.

Fluxes define the minimal information needed for description of metabolism and cell

physiology along with intracellular metabolite concentrations that are determined using

metabolomics and experimental analytical techniques encompassing gas chromatography,

Introduction

- 12 -

liquid chromatography, or capillary electrophoresis in combination with mass

spectrometry [29].

Mathematical models of metabolic networks play a central role in ME and pathway

optimization. Models are often deposited in standardized Systems Biology Markup

Language (SBML) in order to allow their sharing among diverse users. Modelling helps to

analyze the selected pathway (e.g., estimate thermodynamic feasibility) and identify

reactions that need to be modified to improve its performance. Flux Balance Analysis

(FBA) and Metabolic Flux Analysis (MFA) are two major techniques used for calculation

and modelling of intracellular fluxes [28,30]. FBA is a theoretical concept and represents a

direct application of linear programming to biological systems. It uses the stoichiometric

coefficients for each reaction in the system as the set of constraints for the optimization.

MFA determines intracellular fluxes from measurable rates of metabolite uptake and

secretion in growth medium. Substrates labelled with isotope 13C are used in chemostat

grown cultures to trace fluxes through a network. Eventually, the Cobra 2.0 toolbox for

Matlab, OptFlux or OptKnock platforms can implement constraint-based flux calculations

and suggest strategies based on knock-out or overexpression of target genes to optimize

metabolite production without compromising the cell growth [31–33]. Although, FBA and

MFA can deal with genome-scale metabolic models and do not require information on

enzyme kinetics, their output gives just steady-state approximation of dynamic reality in

living cell.

Kinetic modelling employed in the chapters 2 and 3 of the Thesis' Results section,

provides a good alternative to static flux calculations when reliable kinetic parameters of

pathway enzymes and some other input data, e.g., specific surface area of the cell,

permeability coefficient for substrate or approximate enzyme amount in the cell, are

available [15,34]. Kinetic constants can be obtained from enzyme databases or

determined experimentally with purified enzymes or cell-free extracts. One should keep in

mind that most of the kinetic constants deposited in databases are not standardized and

measured under in vivo-like conditions, which may be a limiting factor when applying

such data for kinetic modelling in in vivo systems [35,36]. Computational tools like

COPASI, Scientist, or E-Cell are eventually used to assemble the kinetic model of the

pathway in a form of kinetic and differential equations and simulate reaction time courses

for various conditions [37,38]. Once are the bottleneck reaction steps identified either by

kinetic modelling or flux analysis, experimental techniques are applied in order to target

corresponding gene(s) or regulatory mechanisms.

Experimental tools. Experimental tools for pathway optimizing are applied hand in

hand with the theoretical tools. They are mostly based on recombinant DNA technology

and their employment often generates transgenic organisms. Traditionally, engineering

input is conducted on the level of gene expression that affects the quantity of protein

molecules (Figure 3). Pathway optimizing often requires combination of gene deletions,

which may prevent the leakage of carbon to the unwanted by-products formed in

competing metabolic pathways, and balancing expression of several enzymes, preventing

Introduction

- 13 -

accumulation of intermediates potentially toxic to the host cell and streams the carbon

matter directly to the final product [39,40]. Expression of a target gene can be completely

nullified by knock-out techniques. Chromosomal gene disruption by phage λ Red

recombinase assisted homologous recombination is widely used procedure especially for

E. coli and other Gram-negative bacteria [41,42]. Engineering cuts at the level of gene

transcription and translation are utilized for fine-tuning of expression levels of pathway

components. Up and down regulation of gene expression can be achieved by: (i)

engineering synthetic promoter libraries, e.g., by random or targeted mutagenesis of

relevant promoter regions, (ii) engineering stability of mRNA molecules through tunable

intergenic regions with mRNA secondary structures or RNase cleavage sites, (iii) applying

RNA interference, or (iv) engineering ribosome binding sites [43–46]. Pathway

performance can be tuned also at the protein level through improved spatial organization

of enzymes and substrate channeling [47,48].

Assembly of new synthetic pathway in a heterologous host often requires parallel

overexpression of several genes. One popular approach, applied also in the Chapter 4 of

this Thesis' Results section, is to use the combinations of Novagen Duet plasmid vectors

(Merck Millipore, Germany) that carry compatible replication origins and independent

antibiotic markers. This allows effective propagation and maintenance of four plasmids

and simultaneous co-expression of up to eight genes in a single E. coli BL21(DE3) cell [49].

Duet vectors proved to be useful in ME studies focused on production of biodiesel,

heparosan, or pinocembrin [50–52]. Recently, Xu and co-workers modified Duet vectors

by introducing restriction sites for isocaudamer pairs AvrII, NheI, SpeI, and XbaI, allowing

rapid modular assembly of a number of regulatory elements (promoters, operators,

ribosome binding sites, and terminators) and multi-gene pathways [53]. Resulting

ePathBrick vectors provided a platform for optimizing pathway gene configuration and

gene order which was demonstrated by constructing 54 variants of flavonoid pathway

with three genes either in monocistronic, pseudo-operon or operon configuration within

one week. SEVA (Standard European Vector Architecture) plasmids with standardized

architecture (Figure 4) and nomenclature represent alternative modular vector system

tailored to allow cloning or expression of heterologous genes in P. putida and other Gram-

negative bacteria [54]. Various antibiotic markers, origins of replications, and cargos can

be combined in a single plasmid to fulfil the user's needs and rich collection of diverse

constructs is already available. The plasmids for the purpose of cloning, heterologous

expression, recombination or transposon mutagenesis were traditionally introduced into

the host by triparental mating, i.e. form of bacterial conjugation where a conjugative

plasmid present in one bacterial strain assists the transfer of a mobilizable plasmid

present in a second bacterial strain into a third bacterial strain [55]. Recently, faster and

convenient methods such as heat shock transformation or electroporation are frequently

used [56].

Introduction

- 14 -

Figure 3. Biomolecular targets on the genomic, transcriptional, translational and protein level (A) that can be engineered to adjust the amount and spatial organization of enzymes and improve functioning of metabolic pathways (B). Transcription machinery, enzyme promoters, ribosome binding sites (RBS), and translational machinery can be modified to adjust the expression of enzymes. Enzymes can be assembled using the protein scaffolds to optimize their spatial organization in a pathway and promote substrate channelling. Genome editing can be used to modulate host metabolism to improve flux through the target pathway. Schematic representation of two metabolic pathway in figure 3B shows the effect of increasing concentration of the second and third enzymes in the pathway on the titter of the product. Enzyme fluxes are represented by the size of the grey arrows and metabolite concentrations are represented by the size of the coloured circles. Adopted from [57] and modified.

Metabolic engineers have to keep in mind that the introduction of recombinant

plasmids and massive expression of heterologous genes can affect the fitness of the host

organism either by additional metabolic load or depletion of essential cofactors in redox

reactions [20,58]. Bypassing such troubles is possible by applying lower-copy-number

plasmids with weaker constitutive promoters and enzyme-mediated cofactor recycling

through overexpression of NAD+ kinase, transhydrogenases or dehydrogenases [59].

Additionally, expression of heterologous genes directly from host chromosome can be

beneficial for higher stability of desired genotype/phenotype and better host viability

[60]. Homologous recombination and site-specific or random transposition-based

techniques were developed to enable chromosomal insertion of single genes, gene clusters

Introduction

- 15 -

or whole synthetic operons [61–63]. Configuring pathway genes in the synthetic operon

further reduces metabolic burden. Expression of individual genes in an operon can be

tuned, e.g., by changing a gene order or multiplying the whole metabolic cassette in host

chromosome through increasing selection pressure in cell culture [64,65].

Figure 4. Overall organization of Standard European Vector Architecture (SEVA) plasmids. SEVA vectors are formed by three variable modules: a cargo (blue), a replication origin (green) and an antibiotic marker (magenta). Enzymes used to change the functional modules are shown with the same colour code. Modules are separated by three fixed regions (shared by all vectors) including the transcriptional terminators T0 and T1 and the origin oriT for broad host range transfer. Adopted from [54].

Design strategies. Since 1990's significant effort and funding have been directed

toward rational engineering of biotechnological processes. The process for production of

1,3-propanediol from glucose using a metabolically engineered E. coli K12 developed by

DuPont represents a textbook example of this approach [66]. 1,3-propanediol is a

commodity chemical used world-wide for the manufacturing of fabrics and range of

plastic-based materials. Achieving economically feasible yield of 135 g of 1,3-propanediol

per liter of cell culture required introduction and overexpression of six genes from three

different microorganisms, deletion of glycerol kinase and glycerol dehydrogenase genes

from competing pathway, and directed change in glucose uptake system of E. coli. The fact

that 575 scientists from DuPont and Genencor worked on this project continuously for 15

years illustrates the cost of viable biotechnology development and partially explains the

lack of similarly successful stories based on rational ME approach [15,67]. Thus far,

rational ME has generated interesting and sophisticated, but in many cases unproductive

demonstrations, due to the fact that effects of directed in vivo perturbations are still

Introduction

- 16 -

mostly unforeseen. Therefore, the field currently branches into more rapid strategies that

reduce the problematic complexity.

Multivariate modular metabolic engineering (MMME) is one of the solutions for

semi-rational design of cell factories. MMME approach divides multi-gene pathway of

choice into several metabolic modules based upon pathway bottleneck compounds and

enzyme turnover [68]. Expression of the modules is balanced via altering control elements

at the level of transcription (e.g,. promoter strength, gene copy number), translation (e.g.,

ribosome binding site), or enzyme properties (e.g., activity, inhibition). Eventually,

optimally balanced pathway is searched in reduced combinatorial space. The initial

implementation of this strategy enabled gram per liter production of taxadiene, precursor

of anticancer drug taxol, originally isolated from Taxus brevifolia Pacific yew tree, in E. coli

[69]. Treatment of a single patient would require harvesting dozens of full-grown trees or

complicated chemical synthesis of the compound consisting of up to 50 steps. Thus,

biosynthetic production provided reasonable alternative. The synthetic taxadiene

pathway was divided into upstream (8 genes) and downstream (2 genes) modules, which

were expressed separately from pairs of five selected plasmids with diverse copy numbers

and 3 different promoters. The screening of 32 constructed pathway variants resulted in

minimized accumulation of toxic intermediate and 15,000-fold improvement of taxadiene

production. The MMME strategy in combination with above mentioned Novagen Duet

vectors or ePathBrick vectors was successfully applied also for increasing flavonoids

production in E. coli or for achieving 8.6 gram per liter production of fatty acids in the

same host, respectively [52,70].

Despite the potential of utilizing microbial cell factories for biosynthesis or

biodegradation, possibly providing economic solutions to global challenges, the field of

ME still lacks a standards and universally applicable principles for strain optimization.

Nevertheless, the aim to engineer living systems on the rational basis remains the main

stream in the community. This desire made ME adopt strategies from another emerging

and closely related field of synthetic biology that has been recently established primarily

to understand the fundamentals of life and implement the knowledge for living system

manipulation.

Introduction

- 17 -

1.2 Synthetic biology

The origins of synthetic biology (SB) can be traced back to the completion of the

sequencing of human genome in 2001. Postgenomic era is characterized by

high-throughput sequencing and chemical synthesis of DNA (Figure 5). These

technological developments were followed by emergence of the companies that were able

not only read, but also write DNA and offer it as a commercial cost-effective product [14].

The age of SB was launched by constructing the first genetic counter and toggle switch in

E. coli. as well, because these innovations showed for the first time that artificial genetic

elements can be used for control of processes inside a microbe [71,72]. The competition

International Genetically Engineered Machine (iGEM) officially started in 2005 and since

that time has played an important role in the development of SB and dissemination of its

ideas among undergraduate and high school students [73]. In May 2010, an initial ascent

of SB culminated as a team of scientists led by Craig J. Venter synthesized Mycoplasma

mycoides genome de novo and introduced it into M. capricolum recipient cell [9].

The principal idea behind the SB is that any biological system can be regarded as a

combination of individual functional elements – not unlike those found in man-made

devices – and these parts can be combined in novel configurations to modify existing

properties or to create a new ones [74]. The overall aim of SB is to simplify biological

engineering by applying principles and designs adopted from electronic and computer

engineering in order to produce predictable and robust systems (e.g., genetic control

systems, metabolic pathways, chromosomes and cells) with novel functionalities that do

not exist in Nature [75]. This should be achieved through the assembly of

well-characterized, standardized, and reusable components.

Figure 5. Plot of price per base of DNA sequencing and synthesis. Adopted from www.synthesis.cc and modified.

Introduction

- 18 -


SB has been confronted with the need to precisely define its practice and to develop a

clear and generally applicable strategies from its very beginning. Synthetic biologists took

their lesson from the first decade of ME and realized that this was the only way how to

warrant and implement engineering principles into such a stochastic field as biology. The

awareness of biological complexity resulted in implementing Design-Build-Test-Analyze

(DBTA) cycle that was actually proposed for optimizing microbial production systems

already in 1991 by J.E. Bailey in his early description and vision of ME [11,75]. This

strategy can be linked to designing and building many variants of certain device or system

(e.g., genetic circuit, metabolic pathway, or whole microbial factory), selecting or

screening for variants with desired properties (e.g., the best response or the highest yield

of the product) and discarding the unproductive designs. The knowledge that is gained

from the initial rounds of DBTA cycle can be implemented in mathematical algorithms

used to improve future computer-aided designs. The more sophisticated software tools

for building biology are and the better characterized biological components we have, the

more requirement for the DBTA cycle will diminish and biological engineering will move

closer to the electronic and physical engineering disciplines [75].

There are several main areas of interest in SB that currently contribute to DBTA cycle

and the global collection of parts, devices, and basic knowledge of biological systems:

high-throughput DNA sequencing and synthesis [76]

standardization and registry of novel biological parts and devices [77]

mathematical modelling of biological systems and networks [78]

design of orthogonal systems [79]

minimal genomes and minimal cells/chasses [80]

synthetic genomes and whole-genome editing tools [81,82]

biosensors, synthetic genetic circuits and switches [83]

cell-free systems [67]

cell to cell communication [84]

de novo protein design [85]

Thorough description of all these topics is out of scope of this text and curious

readers are referred to the cited literature. The following paragraphs discuss the topics

relevant for experimental part of this Thesis.

DNA synthesis and parts standardization. Fast and affordable DNA synthesis

facilitated expansion of SB and represents the headstone for most of the research projects

including those focused on assembly of whole synthetic genomes [9,81]. Nowadays,

mostly microarray-based oligo synthesis is used to provide substrate (usually 5-50 oligos)

Introduction

- 19 -

for constructing larger synthetic fragments (usually 200-3000 bp) [76]. Scarless methods

including popular Gibson assembly, ligase cycling reaction or yeast recombination are

employed for combining sequence-verified gene-length fragments in even bigger

complexes [86,87]. Gene synthesis combined with parallel optimization of codons is very

powerful for improving expression of target genes in specific heterologous host [88].

Levels of proteins overexpressed in E. coli from codon-optimized genes can reach up to

30-40% of the total cellular protein. DNA synthesis also allows preparation and

modulation of genetic parts (promoters, ribosome binding sites, genes, terminators, etc.)

with verified sequences. Characterized parts that fulfil the defined standards in SB (e.g.,

BioBrick standard) are stored in parts registries and can be used for assembly of synthetic

genetic devices [77]. Precise definition of part property, e.g., of promoter strength or

plasmid copy number, for specific host organism and in specific conditions is crucial for

reliable input of mathematical modelling and successful design. This proved to be true

also in the Chapter 3 of the Thesis's Results section.

Mathematical modelling of biological systems and networks. The bottom-up

approach of SB and integration of DBTA cycle require close integration of experimental

and computational techniques. Similarly to ME, mathematical modelling is irreplaceable

for system design and optimization. The main tasks for SB computational tools include

collection and storage of literature data and datasheets from parts characterization,

implementation of parts in more complex modules, or model-based numerical simulations

of synthetic gene circuits and metabolic pathways [89]. Autodesk Cyborg

(www.autodeskresearch.com/projects/cyborg) and ClothoCAD (www.clothocad.org)

represent two ambitious projects that try to develop user-friendly modular software

platforms for SB data management and computer-aided design. Despite of the enormous

progress in computing capacity in recent years, the prediction accuracy of many

theoretical tools that deal with complex in vivo systems and stochastic biological

processes is still unsatisfactory. SB therefore develops several parallel strategies to

compensate for the complexity of a living cell.

Design of orthogonal systems. Orthogonal, or in other words parallel and

independent, systems are one of the key foundations in SB, making biology more

amenable to computational design. The ultimate goal for synthetic biologists is to develop

artificial genetic codes (e.g., quadruplet instead of triplet) and orthogonal transcription-

translation machineries that would allow assembly of novel metabolic pathways from

enzymes containing unnatural amino acids [79,90]. Such pathways could posses yet

unknown functions and synthesize variety of valuable compounds. Importantly,

orthogonal pathways or other modules should not interact with the natural counterparts

in the host cell. This allows better predictability of the component behaviour and

preceding of inhibitory cross-talk upon introduction of the pathway into an existing

metabolic network [91]. Also in the Chapter 3 of this Thesis, orthogonality of the studied

metabolic route in E. coli was achieved due to the fact that the reactions catalyzed by the

pathway enzymes were not naturally occurring in the heterologous host.

Introduction

- 20 -

Minimal genomes and minimal cells. SB aims to construct microbial chasses

(minimal host cells) either through lego-like de novo synthesis of viable cell or

systematical deletion of non-essential genes in the genomes of natural hosts. The latter

approach seems to be far more feasible at the moment. For example, transposon

mutagenesis-based procedure is used to produce The Keio collection of all viable E. coli

single-gene knock-outs to facilitate a systematic investigation of the regulation and

metabolism of E. coli [92]. The similar study in M. genitalium revealed that over 100 out of

482 genes were not essential for cell survival [80]. Although the function of many

essential genes is yet unknown the aim of the project is to create "M. laboratorium" chassis

with as few genes as possible. Similar attempts focus mostly on answering fundamental

questions of biology. But discarting non-essential functions in the widely used hosts like E.

coli, S. cerevisiae or P. putida can be very useful also for application purposes. It might

result in reduced complexity of the system and better efficiency of the biosynthetic or

biodegradation processes.

Engineering of cell-free systems. The major principle of cell-free synthetic biology is

that purified biomolecules or components in crude cell extracts, which can be better

monitored and modelled, substitute intact cells for constructing complex biomolecular

systems [67,93]. This approach overcomes current limitations of in vivo ME and SB, e.g.,

prospective toxicity of metabolites, interferences of multiple cell components, instability

of evolved genotypes, GMO restrictions, and provides high precision and freedom of

design. Cell-free systems can be applied for production of proteins (antibodies, vaccines,

biocatalysts) and diverse metabolites or construction of minimal cells from bottom up. In

vitro engineering is expanding also in the field of biocatalysis [29]. Metabolic pathways

reconstructed from purified enzymes or cell-free extracts allow verification of biocatalyst

functioning, determining kinetic parameters and evaluation of the network by kinetic

modelling [94-96]. In vitro multi-enzyme systems are intensively studied for their

potential in production of pharmaceuticals or biofuels. For example, Bujara and colleagues

utilized cell-free extract with endogenous glycolysis enzymes from genetically modified E.

coli for 10-step multi-enzyme synthesis of unnatural monosaccharides based on building

block dihydroxyacetone phosphate [97]. Del Campo and co-workers employed cell-free

system encompassing thirteen purified enzymes for conversion of abundant

monosacharide xylose into H2 approaching 100% of the theoretical yield [98]. Possible

disadvantages of in vitro metabolic networks are: (i) suboptimal efficiency due to the

lower enzyme concentration than in extremely dense cell cytoplasm and (ii) limited

recycling of the system [67]. These drawbacks can be solved by enzyme immobilization

and improved spatial organization via synthetic protein or DNA scaffolds that diminish

diffusion of pathway intermediates [48]. Alternatively, biocatalysts can be precipitated

and covalently interconnected in cross-linked enzyme aggregate (CLEA) particles [99].

This method was successfully applied for synthesis of toxic nucleotide analogues by

immobilized five-enzyme synthetic pathway with included ATP regeneration [100].

Similar strategy was adopted also for in vitro biodegradation of toxic environmental

pollutant described in the Chapter 1 of the Thesis.

Introduction

- 21 -

Altogether, above mentioned SB tools and approaches help closing the enormous gap

between the cellular complexity and human desire for engineering. There is no doubt that

SB has revolutionized the field of ME. Amazing example of ME and SB consonance is a ten-

year project based on co-operation of University of California with company Amyris Inc.

resulting in commercial production of semi-synthetic artemisinine [75,101]. Artemisinine

is an important antimalarian drug precursor originally derived from plant Artemisia

annua. The goal of the project was to establish economically viable microbial production

of the precursor and substitute the fluctuating supply of plant-derived drug. The number

of design and experimental steps, e.g., balancing expression of 15 codon-optimized genes

from A. annua, selection of suitable S. cerevisiae chassis, chromosomal insertions, cell free

assays or gene knock-outs, stood between the initial idea and final production of 25 g per

litre of artemisinic acid converted to the artemisinine in the last chemical step.

Importantly, this case represents a precedent that supplied valuable information in DBTA

cycle and possibly simplified the development of similar processes for biotechnological

production of some other important pharmaceuticals like anticancer paclitaxel, analgesic

morphine, anti-HIV prostatin or antibiotic bleomycin [75]. ME and SB experts would

surely agree that another emerging discipline, protein engineering, will significantly

contribute to the success of these future enterprises.

1.3 Protein engineering

Basic building blocks of metabolic pathways are enzymes - powerful catalysts that

are able to increase reaction rates by up to seventeen orders of magnitude. Due to this

capacity, isolated enzymes are now widely utilized in molecular biology, food industry,

medicine, biodegradation or production of pharmaceuticals. However, only limited

number of enzymes naturally catalyzes the reactions that biotechnologists require and

which occur under conditions that are industrially acceptable and economically feasible.

Many efforts are currently devoted to the modification of enzyme properties to meet the

demands of bio-based industry. A scientific discipline dealing with designing and

constructing more powerful biocatalysts is called protein engineering (PE). Fine-tuning

enzyme characteristics through PE is desirable also for developing highly efficient systems

for ME. Specifically in strain engineering, enzyme variants with evolved kinetic properties

can replace traditional suboptimal approaches, based on overexpression of slow

biocatalysts, that can compromise system productivity and host viability through the

increased metabolic burden [8,58].


Three distinct strategies were developed in last two decades to allow the

construction and identification of mutant enzymes possessing desirable properties such

as increased activity, stability, modified specificity and selectivity or reduced

Introduction

- 22 -

substrate/product inhibition (Figure 6) [102,103]. The earliest approach was rational

design, which has been beneficial mostly in studying catalytic mechanism of enzymes and

recently also in knowledge-based modifications of enzymes' properties. Rational design

emphasizes the understanding of protein structure and amino acids interactions at the

beginning of the process. It exploits various computational tools and computer modeling

techniques, e.g., molecular docking, homology modelling or molecular dynamics

simulations, and site-directed mutagenesis for generation of targeted mutation(s)

resulting in single or several modified variants of enzyme [104]. State-of-the-art variant of

this strategy is de novo protein design which requires detailed understanding of the

desired catalytic meachanism and works with Rosetta software package [85,105].

Directed evolution as a newer approach has proven to be a powerful tool for the

improvement of enzymes [106,107]. Contribution of directed evolution is substantial

especially in particular cases, when neither the three-dimensional structure or homology

model nor the catalytic mechanism of the enzyme is known. Through the library

construction methods, based mostly on PCR protocol, directed evolution introduces

random mutations alongside the whole gene or recombines genes and generates huge

quantities of mutated sequences. Error-prone PCR or DNA-shuffling became widely used

methods of directed evolution [108,109]. Their randomness reminds the natural evolution

and can result in unexpected protein variants with improved functions. However, the

search for desired biocatalysts in huge mutant libraries is very demanding. The right

choice of high-throughput screening or selection method, usually based on colorimetric

assays, growth assays or fluorescence activated cell sorting (FACS), is crucial for

successful outcome of the directed evolution project.

Recent advances in enzyme engineering have used a combination of random

methods of directed evolution with elements of rational enzyme modification to

successfully by-pass certain limitations of both approaches. This focused directed

evolution, semi-rational approach or data-driven PE, targets multiple specific residues or

certain protein regions selected on the basis of prior structural and functional knowledge

[111,112]. The efficient choice of mutations likely to affect enzyme function is conducted

both experimentally and, on a much greater scale, computationally, using new powerful

algorithms and statistical tools, e.g. 3DM database, HotSpot Wizzard, FOLDX or ROSETTA

[104,113]. Focused directed evolution thus results in a smaller “smart“ libraries that are

easier to search through and more likely to yield positive results [114]. Particularly,

oligonucleotide-directed saturation mutagenesis proved to be useful for construction of

focused libraries [115].

Introduction

- 23 -

Figure 6. Comparison of the main protein engineering strategies. Adopted from [110] and modified.

Introduction of PE into the ME protocols is evident especially during the last few

years. Both rational design and directed evolution have successfully been used to engineer

diverse enzyme properties affecting the flux and thus the performance of the metabolic

pathways. Zhang and co-workers rationally engineered a promiscuous 2-keto-isovalerate

decarboxylase and 2-isopropylmalate synthase for preferred substrate specificity towards

a non-natural substrate [116]. Combination of mutated enzymes expanded E. coli

metabolism to produce unnatural alcohols as potential biofuels. In another recent report,

two enzymes from the diterpenoid biosynthetic pathway geranylgeranyl-diphosphate

synthase and levopimaradiene synthase were modified for levopimaradiene production in

E. coli [117]. Combination of rational approach based on homology modelling and

phylogenetic analysis in case of the first enzyme and random mutagenesis through error-

prone PCR of the second enzyme resulted in 2600-fold increase of levopimaradiene

biosynthesis.

Introduction

- 24 -

1.4 Perspectives

Synthesis of ME, SB, and PE will for sure be continuously intensified in following

years and will result in more powerful approaches for rapid improvement of metabolic

pathways and strain optimization. More experience and more information deposited in

DBTA cycle will allow better predictions and knowledge-based designs. SB devices like

biosensors and genetic switches will allow dynamic control of metabolic and signalling

pathways in engineered organisms [118]. It is expected that microfluidics, FACS and other

high-throughput techniques will revolutionize screening for new biocatalysts and

pathways or parts and strain characterization [119]. Besides that, phenomenon of natural

evolution and adaptation will draw more attention for possible fine tuning of rationally

designed microbial cell factories [120]. Importantly, progress in biological engineering

must go hand in hand with establishing regulatory frameworks and institutions that will

communicate the outcomes of the scientific projects to public and take care of biosafety,

ethics issues, and socio-economic aspects of conducted research.

Although the development in all three described areas has been exciting, our

understanding of cell biology remains unsatisfactory. This is obvious from limited

complexity and fitness of anthropogenic constructs [120]. ME and SB succeeded in

engineering microbes for biosynthesis of valuable chemicals in defined laboratory

conditions, but design of cells that will safely function in complex natural environments is

still a challenge. This is related especially to the engineering of metabolic pathways and

whole cells for biodegradation of toxic anthropogenic compounds and environmental

pollutants. Applications of the above described tools and approaches for this purpose will

be discussed in more detail in following chapters.

Introduction

- 25 -

2. Engineering of biodegradation pathways

The most of recent ME and SB applications are focused on production of

pharmaceuticals, commodity chemicals or biopolymers. The general interest in

biosynthesis is driven by unavoidable exhaustion of fossil resources and corresponding

economic aspects. Removal of environmental pollution caused by extensive activities of

consumer society represents another serious topic that deserves attention of biological

engineering. Recent pollution of soils, ground and surface waters constitutes a major

threat to the public health, not just in developing, but also in industrial countries including

EU states, China, and USA. In the Assessment of the European Environment Agency,

published in 2007, the EU estimated that there are approximately 250,000 known

contaminated sites in Europe that need to be cleaned up. Potentially polluting activities

occure in nearly 3,000,000 sites and the number of sites requiring remediation will

increase by 50% by 2025. Beside the medical and environmental consequences, this

situation signs considerable potential for growth of eco-industry focused on pollutant

removal and clean-up technologies.

Majority of the contaminants affecting the soil and water in Europe are organic

compounds such as mineral oil hydrocarbons, polyaromatic hydrocarbons, benzene

derivatives, and halogenated hydrocarbons (Figure 7). Many of organic polluting

compounds used in agriculture (e.g., pesticides dichlordifenyltrichlorethan, atrazine,

pentachlorophenol), industry (e.g., solvents as dichloroethane or dielectric fluids as

polychlorinated biphenyls) or military (e.g., explosives as trinitrotoluene) are of

anthropogenic origin and are called xenobiotics. Xenobiotics are the chemicals that have a

structure or a substituent on their structure that is not found in natural compounds. Some

of these compounds are considered as recalcitrant, in other words are not bioavailable or

not biologically degraded in the environment. Nevertheless, most of them are more or less

susceptible to biodegradation, i.e., biologically catalyzed reduction in their complexity

[121]. In case of organic compounds, biodegradation can sometimes lead to the complete

conversion of original compound to the inorganic products, i.e. mineralization.

Although plants and, to a lesser extent, animals may cause a number of changes in

structure of polluting chemicals, the major organisms causing the biological

transformations in soils, waters, and other environments are the microorganisms carrying

catabolic pathways. The biotransformations of organic pollutants catalyzed by microbial

metabolism are usually much faster than an abiotic degradation and depending on

microorganism can occur either in aerobic or anaerobic conditions. These factors together

with many advantageous properties mentioned in previous chapters made microbes and

especially bacteria favourable mean for bioremediation technologies, i.e., technologies

focused on removal of pollutants from contaminated sites in the environment either by

natural attenuation (little or no human action), biostimulation (addition of nutrients or

electron donors/acceptors to promote the growth of certain microbe) or bioaugmentation

(addition of natural or engineered microorganisms with the desired catalytic capabilities)

Introduction

- 26 -

[122]. Certain criteria formulated already 40 years ago by Dr. Martin Alexander must be

met for biodegradation to take place in the environment and bioremediation technology

to be seriously considered as a practical mean for treatment (Box 1) [121]. Some of the

early in situ or ex situ biotechnological processes really fulfilled these prerequisites and

resulted in successful local or world-wide applications of natural degraders in

anthropogenic operations. These include large-scale wastewater denitrification, removal

of uranium or application of 1,2-dichloroethane degrading bacterium for groundwater

treatment [123–125].

The uprise of recombinant DNA technology and ME in 1990s allowed transformation

of bioremediation from an empirical practice into an engineering science. The utlimate

goal of the new field was to engineer whole microbes, their biodegradation pathways and

corresponding enzymes toward directed in situ mineralization of desired pollutants. Such

"superbugs" were expected to provide economically feasible and environmentally friendly

alternative to costly conventional technologies for pollutants removal [126–128].

Furthermore, engineered bacteria could help solving the problem of persistent organic

pollutants (POPs) as dichlordifenyltrichlorethan (DDT), polychlorinated biphenyls (PCBs)

or dioxins that are resistant to natural biodegradation. The progress in design of

microorganisms and their metabolic pathways for the purpose of biodegradation and

bioremediation of organic pollutants achieved by emerging engineering technologies is

summarized in the following paragraphs.

Figure 7. Environmental contaminants affecting soil and groundwater in Europe (in %). Included are the data from Austria, Belgium, Czech Republic, Finland, Macedonia, Greece, Hungary, Italy, Luxembourg, Slovakia, Spain, and Sweden. Adopted from http://www.eea.europa.eu and modified.

Introduction

- 27 -

Box 1. The important criteria for practical feasibility of bioremediation technology formulated

by Martin Alexander [121].

1. A microorganism must exist that have the needed catalytic activity.

2. That microorganism must have the capacity to transform the compound at

reasonable rate and bring its concentration to the level that meets regulatory

standards.

3. The microorganism must not generate products that are toxic at the concentrations

likely to be achieved during the remediation.

4. The site must not contain concentrations of chemicals that are inhibitory to the

biodegrading species.

5. The target compound must be available to the microorganism.

6. Conditions at the site or in a bioreactor must be made conducive to the microbial

growth or activity.

7. The cost of the technology must be lower or, at worst, no more expensive than other

available technologies that can destroy the chemical.

2.1 Strategies and tools

Adaptive laboratory evolution and uprise of rational strain engineering. Early

attempts in biodegradation were limited to the isolation of microbes from contaminated

sites and studying natural degradation pathways in culturable bacteria [128]. Adaptive

laboratory evolution of bacteria carrying desirable trails was applied to speed up the rate

of pollutant's removal. This was achieved mostly by prolonged exposure of microbes to

the target chemical in chemostat or shaking flasks. The early experiment of Kellogg and

co-workers conducted in 1981 is an illustrative example of how the laboratory in vivo

evolution approach can be used to isolate microorganisms capable of utilizing persistent

compounds such as 2,4,5-trichlorophenoxyacetic acid, defoliating Agent Orange abused in

Vietnam War [129]. The same strategy was used also in more recent studies. For example,

faster degradation of widely used pesticide atrazine by Pseudomonas sp. ADP was

observed as a consequence of tandem duplications of atzB gene encoding the second

enzyme from atrazine pathway after 320 generations-long cultivation with pesticide as a

sole nitrogen source [130]. Similarly, amplification of tfdA gene encoding the first enzyme

in the pathway, 2,4-D dioxygenase, resulted in faster degradation of

2,4-dichlorophenoxyacetic acid, used as a sole carbon source during 1,000 generations-

long cultivation of Ralstonia sp. TFD41 [131]. These studies highlight the effect of tandem

gene duplication in the evolution of efficient catabolic pathways.

Very interesting strategy, which takes an advantage of adaptive laboratory evolution,

is genome shuffling. Genome shuffling has proven useful for engineering of multitrait

phenotypes that would be difficult to engineer rationally. Improvement of degradation of

anthropogenic pesticide pentachlorophenol (PCP) by Sphingobium chlorophenolicum

ATCC 39723 was addressed by this method, since the authors expected requirement of

Introduction

- 28 -

multiple mutations for the adaptations reducing the toxic effects of substrate and its

metabolites [132]. Three successive rounds of protoplast fusion and genome

recombinations of parents with improved phenotypes alternating with selection for

improved growth in the presence of PCP resulted in substantial improvements in both the

rate of PCP degradation and the concentration of PCP that could be tolerated. Analysis of

several improved strains proved that various combinations of mutations contributed to

the improved phenotypes. However, the mechanism of the adaptation remained

uncertain.

A patent and publication in Science by Chakrabarty and co-workers in 1981

represented a starting point of rational engineering for biodegradation and

bioremediation [129]. The work described preparation of recombinant P. putida strains

capable of breaking down crude oil by so called plasmid-assisted molecular breeding, i.e.,

dissemination of novel catabolic capabilities through directed bacterial conjugation and

plasmid transfer. Conjugation techniques allowing interspecies horizontal gene transfer

were utilized in many follow-uping studies [133]. Also Pseudomonads used by

Chakrabarty and co-workers remained popular hosts and model organisms for

biodegradation and bioremediation. The reason is that Psudomonads, encompassing

Gram-negative, aerobic, gamma-proteobacterial species, often possess remarkable

capacity to endure both endogenous and exogenous stresses [134]. Many Pseudomonads

are naturally resistant to high levels of solvents, probably due to the efficient efflux

pumps. Several environmental Pseudomonas strains were shown to mineralize aromatic

hydrocarbons and halogenated and nitro-organic compounds. In some P. putida strains

such as CSV86, aromatic compounds are even preffered substrates over glucose [135].

Namely P. putida KT2440 is generally regarded as safe organism, thus it is a convenient

model for laboratory experiments and suitable host for use in biotechnological processes.

Recruitment of new enzymes into existing catabolic pathways. The late 1980s and

early 1990s represented the golden era of biodegradation research with numerous

engineering attempts following the example of Chakrabarty and co-workers. The

persistence of many xenobiotics was in that time attributed mainly to the absence of

complete degradative pathways in a single organism [136,137]. Thus, the recruitment of

known complementary enzyme sequences by conjugative transfer of genes and so called

"patchwork assembly" of several existing natural pathways in a suitable host was believed

to generate a functioning synthetic routes allowing complete mineralization of target

persistent compounds such as PCBs, chlorinated benzoates or toluenes. Single genes or

catabolic gene clusters were heterologously expressed either from plasmids or

deliberately introduced into the chromosome of the host strain in a form of catabolic

cassette [133]. The system based on Tn5 minitransposon mediated delivery proved to be

very useful in studies aiming at stable expression of recruited enzymes from chromosome

of Gram-negative bacteria. The portfolio of plasmid delivery vectors utilizing Tn5-based

technology for random chromosomal insertions has expanded during the years [63,138].

Recently, also a Tn7-based broad-range bacterial cloning and expression vectors were

Introduction

- 29 -

prepared to allow site-specific implantation of metabolic modules at single attTn7 site of

various Gram-negative hosts [62].

Described tools and strategy were utilized for assembly of hybrid pathways in

various studies [139–144]. To name a few, Timmis and co-workers combined enzymes

from five different catabolic pathways of three distinct soil bacteria into functional route

for the degradation and mineralization of methylphenols and methylbenzoates [140].

Interestingly, the hybrid pathway was fully regulated and expressed only in response to

the presence of the substrate. In another work of the same group, a catabolic pathway for

alkylbenzoates carried on TOL plasmid of Pseudomonas was successfully restructured for

processing of the new substrate 4-ethylbenzoate [141]. The authors integrated the genes

from mutated bacteria selected for desired regulation of the pathway. More recently,

Tn5-based minitransposon system was utilized for transfer of three catabolic clusters

with eleven genes from three distinct bacteria to Cupriavidus necator H850 [143].

Engineered strain was the first designer bacterium to show aerobic growth on a wide

range of PCBs including two commercial pesticides Aroclor 1221 and Aroclor 1232.

Despite some success of "patchwork strategy" and engineering of "superbugs" with

extended substrate scope, this rather naive approach resulted also in many

disappointments. Almost textbook example is represented by engineered Pseudomonas

strains that failed to grow on 2-chlorotoluene as the only carbon source in well defined

laboratory conditions even though they possesed all genetic components presumed to be

necessary for substrate mineralization [145]. Similar failures can be explained by the fact

that the experimenters in the early and also in some recent studies underestimated or did

not have sufficient insight into important factors such as: (i) expression levels of

introduced heterologous enzymes and possible kinetic bottlenecks of assembled synthetic

routes, (ii) leakage of the side-products, (iii) thermodynamic feasibility of hybrid catabolic

network, (iv) complexity of regulatory networks, (v) or stress response and overall

physiology of the cell upon introduction of new metabolic modules and exposure to the

toxic substrates or intermediates [126–128]. Following paragraphs discuss selected

strategies that take listed factors into account and allow knowledge-based design of

strains with desired biodegradative capacities.

Engineering expression of enzymes in metabolic pathways. Tuning of individual

enzymes expression in natural or synthetic biodegradation pathways can help to

overcome or moderate the effect of unbalanced enzyme kinetics. One of the first studies

that dealt with the expression levels and activities of enzymes in the biodegradation

pathway was that of Mattozzi and co-workers [146]. The authors successfully designed a

P. putida strain that fully mineralized paraoxon, the worldwide known synthetic

phosphotriester insecticide, and used it as a sole source of carbon and phosphorus. In the

assembled catabolic pathway, one of the initial reaction intermediates diethyl phosphate

was degraded by serial action of three enzymes encoded by genes on synthetic operon.

The authors optimized the gene order in the operon and corresponding expression levels

according to the activities of individual enzymes measured in the cell lysates. In that way,

Introduction

- 30 -

they achieved faster paraoxon hydrolysis and utilization of diethyl phosphate. Several

studies reported the improvement of expression of dszB encoding 2-hydroxybiphenyl-2-

sulfinate sulfinolyase, the last rate limiting enzyme from the dibenzothiophene

biodesulfurization pathway. The overexpression of DszB in host cells and the performance

of the whole three-enzyme pathway was improved both by mutating the 5' untranslated

region of dszB [147] and by rearranging the gene order and removing the overlapping

structure in dsz operon [148]. The later work proved again that the positioning of gene in

the operon is an important determinant for its expression level. This phenomenon was

well described and explained in the recent study of Lim and co-wokers [64]. The

expression levels of pathway enzymes in host cell were addressed also in the work

described in the Chapters 3 and 4 of this Thesis.

Computer-aided design. Systems biology contributed significantly to the

rationalizing of the biodegradation field during the last few years. At the time of writing of

this Thesis NCBI public database lists 4647 sequences of bacterial genomes

(http://www.ncbi.nlm.nih.gov/genomes) from which many encode enzymes with yet

unknown biodegradative capacities. These huge pile of read genetic information together

with new powerful algorithms and computational tools that help scientists with mining of

relevant data allowed considerable progress in our understanding of dynamic interactions

that occur inside and in between the cells and the environments. Genome-scale metabolic

models and their interconnection with transcriptomic, proteomic, metabolomic, and flux

analyses under various growth and stress conditions completely changed our view on

popular biodegradative bacteria, such as P. putida [149,150]. Available genomic data

enabled also new insights into evolution of microbial metabolism toward biodegradation

of important xenobiotics [151].

Genome-based metabolic modelling can predict suitable targets for knock-out or

overexpression within the genome of a host organism. Thus, the desired task can be

sometimes achieved even without the need to incorporate heterologous genes [152]. One

of the modelling platforms utilizing genome-scale models, OptKnock, was found useful in

designing Geobacter sulfurreducens for higher respiration rates and electron transfer

[153]. Evolved strain could possibly enhance conversion of organic compounds to the

electricity in microbial fuel cells. The structure-activity relationship tool EcoliTox can

assess toxicity of selected substrates and metabolites throughout the metabolome of E.

coli [154]. Obtained information can be used for fine-tuning pathway expression in this

host.

Rich source of information on 219 biodegradation pathways is University of

Minnesota Biocatalysis/Biodegradation Database (UMBBD; http://eawag-bbd.ethz.ch).

Importantly, UMBBD can not only search through the existing metabolic routes, but allows

also predicting novel biotransformations leading from target xenobiotic substrate to the

unharmful molecule that can be metabolized by central catabolic pathways of a host cell.

The UMBBD Pathway Prediction System can highlight reactions that are likely to occur in

aerobic environment, but says nothing about the thermodynamic or kinetic feasibility of

Introduction

- 31 -

proposed pathways. Missing information on designer pathway thermodynamics can be

provided by the computational tool BNICE introduced in chapter 1.1.1 [24]. In theoretical

study of Finley and co-workers, metabolic flux analysis and genome-scale model of P.

putida implemented in BNICE were used to predict set of feasible pathways for

mineralization of 1,2,4-trichlorobenzene [155]. Although such theoretical predictions

should be experimentally verified, this pioneering work shows the possible way for those

who wish to utilize increasing power of computational modelling for design of novel

promissing catabolic routes. Unfortunately, majority of computational tools developed

nowadays for strain design are applied for improvement of the product yields rather then

biodegradation of recalcitrant pollutants [22].

SB approaches. Also SB, aiming at computer-aided designs and lego-like assemblies

of parts from diverse living organisms, has the potential to change the field of

biodegradation. SB approaches and devices such as toggle switches and genetic regulatory

circuits could be used for engineering bacterial consortia that would fulfil the demanding

biodegradation tasks better than individual microbes [156]. It is probable that

biodegradation in complex changeable environments will perform more reliably when

diverse metabolic and regulatory modules will be spread among members of consortium.

However, thus far the pioneering tests of consortial biodegradative capacities occured

only in defined laboratory conditions. For example, co-culture of engineered Escherichia

coli SD2 and Pseudomonas putida KT2440 was successfully employed for mineralizing

insecticide parathion in shaked flasks and in biofilm cultivation [157].

Development of novel regulatory systems for protein expression, cell-to-cell

communication and whole-cell biosensing that would allow parallel degradation and

seeking for toxic or explosive compounds in contaminated soils and waters represents

another challenging tasks for synthetic biologists [84,158,159]. SB-based strategies can be

also used to accelerate connection of target substrate with corresponding catalytic

activities. Aso and co-workers transplanted specialized membrane structures called

superchannels, allowing direct uptake of macromolecules from the environment into the

cytoplasm of Sphingomonas sp. A1, in the cell envelop of dioxin-degrading S. wittichii RW1

and achieve increased bioremediation capacity of this strain [160]. Opposite approach,

display of catabolic enzymes or whole pathways on the cell surface can increase efficiency

of substrate decomposition and diminish toxic effects of reactive metabolites on the

cytoplasmic structures [161].

Alternatively, the problem of metabolite toxicity could be addressed by engineering

biodegradative systems in vitro. This may encompass utilization of immobilized purified

pathway enzymes or crude cell extracs. Biodegradation field has not yet fully explored

benefits of this strategy. Predominantly, individual immobilized enzymes or whole-cell

biocatalysts are addopted in biotechnological processes aimed at contaminants removal

[162–164]. Rather exceptional study of Geuke and co-workers showed potential

application of in vitro engineering approch for biodegradation of technical

hexachlorocyclohexane (HCH) consisting of five isomers [165]. The authors used model

Introduction

- 32 -

system of two purified recombinant enzymes, dehydrochlorinase LinA and haloalkane

dehalogenase LinB, that are known to initiate biotransformation of prohibited insecticide

γ-HCH. They separately incubated HCH isomers with various ratios of LinA and LinB and

determined metabolic profiles of occuring sequential biotransformations. The analyses of

these profiles helped to judge the environmental fate of HCH isomers and proved that the

original HCH degradation pathway is optimized for γ-HCH, but not for other isomers. In

vitro approach proved to be useful for developing promissing biotechnology for

transformation of toxic anthropogenic pollutant into desirable commodity chemical

described in Chapter 1 of this Thesis.

Combination of ME and PE approaches for biodegradation. Significant

improvement of the biodegradation route performance can be achieved through

engineering activity or selectivity of the limiting enzymes. Activities of many enzymes that

act on anthropogenic compounds are derived from promiscuous activites that are

generally inefficient [166]. For example, Rhodococcus erythropolis N-acylhomoserine

lactone acylase hydrolyzes various lactones with values of kcat/Km up to 1.7x106 M-1.s-1 and

also unnatural substrate insecticide paraoxon, but with a significantly lower kcat/Km of 0.5

M-1.s-1 [167]. Promiscuous activities often result from substrate ambiguity, i.e., the ability

to bind and convert a molecule that resembles the normal substrate.

Numerous reports describe application of common PE methods such as epPCR, DNA

shuffling, site-directed or saturation mutagensis for engineering activity or selectivity of

individual catabolic enzymes [128,168]. However, the studies that report successful

implantation of constructed mutants in the context of the whole biodegradative pathway

and receiving host organims are not so common. Iwariky and co-workers successfully

applied Alcaligenes sp. KF711 harboring engineered monooxygenase P450CAM for

dehalogenation of pentachloroethane to trichloroethane under unoxic conditions. The

product was further degraded by hybrid dioxygenase in the presence of oxygen [169]. In

another study, promiscuous toluene orthomonooxygenase and epoxide hydrolase were

engineered using DNA-shuffling and saturation mutagenesis, respectively. Subsequent

employment of modified genes in recombinant E. coli with synthetic pathway allowed

faster aerobic degradation of chlorinated ethenes with lesser accummulation of

stress-inducing intermediates [170,171].

2.2 Perspectives

Numerous examples presented in the previous chapter clearly show that engineering

of microorganisms towards novel or improved biodegradative capacities remains a

challenge. The field of biodegradation still lacks success stories similar to those known

from the field of biosynthesis. Unfortunatelly, the high risk connected with engineering of

microbes for pollution removal and legislative barriers hindering their introduction to

practical applications let many scientific teams to change the topics of their research.

Introduction

- 33 -

Accordingly, application of ME and SB to biodegradation problems are rather rare in

current literature. This unsatisfactory situation can be attributed mainly to: (i) complexity

of biological systems and (ii) controversy on application of GMO in the field studies.

It is no doubt that the environmental concerns and regulatory constraints have

limited not just in situ applications of GMOs for bioremediation, but also the quality of

fundamental research and the overall progress in the field. The experiments of Lawrence

Wackett and his team on field-scale remediation of atrazine-contaminated soil with killed

recombinant E. coli expressing atrazine chlorohydrolase most probably represents the

only well documented case of GMO utilization for in situ bioremediation [172]. The

relaxation of restrictive policy is desirable because only the field-scale experiments can

lead to better understanding of designer organism behavior in real environments, where

both biotic and abiotic factors form the complex fluctuating bioremediation space [173].

Considering potential future in situ applications, some intelectual effort has already been

invested in development of technologies that would ensure the biosecurity and fate of

released GMOs. To avoid endangering the biodiversity and uncontrolled spreading of

foreign genes or recombinant microbes, the constructed degraders could be tracked

through bioluminiscence, fluorescence or watermarking of synthetic DNA or completely

discarded after performing the desired task by inducible suicidal systems [174–176].

Employment of immobile transgenic plants and their symbiotic bacteria for

phytobioremediation or application of GM microbes only within the boundaries of

well-defined decontamination zones or bioreactors could also reduce the risk of

unintenional dissemination [177]. Concerning the first limiting point, the only way towards microorganisms

performing well in a complex natural environment leads through deeper understanding of

fundamental principles of cell biology. The more intensive exploitation of the emerging

technologies introduced in this text should help with such a task. There is lots of space for

employment and further high-throughput data mining, functional identification and

experimental characterization of biological components under standardized conditions

[120]. Desirable is also wider utilizing of flux analysis and kinetic modelling [178],

inspecting biodegradation processes at a single cell, a single pathway, and a single

molecule levels [179], studying the formation and behavior of microbial consortia [156] or

development of more precise and reliable experimental tools for balancing of metabolic

networks and reduction of diverse stresses. Better synergy of PE and ME through

employment and evaluating of new enzyme variants in biodegradation pathways would

also be beneficial. The last but not least, the idea of DBTA cycle should be adopted by

biodegradation community, same as design of orthogonal and in vitro systems that might

help to reduce both problems of biological complexity and GMO controversy [67,75].

Enzymes with altered activity and enantioselectivity and some of the mentioned

novel ME and SB approaches have been recently used for engineering and understanding

of the synthetic metabolic pathway for biodegradation of anthropogenic pollutant

1,2,3-trichloropropane. The background and history of the pathway design, assembly, and

engineering is described in the following part of the Introduction.

Introduction

- 34 -

3. Engineering of the synthetic metabolic pathway for

biodegradation of 1,2,3-trichloropropane

The attempts to find natural degraders or engineer new microbes capable of efficient

removal of man-made halogenated hydrocarbons from the environment have continued

through the decades since 1980s until today. The engineering of the synthetic pathway for

aerobic utilization of 1,2,3-trichloropropane has been initiated already in 1989 and thus,

with its 25-years-long history, represents one of the most systematic efforts of that kind

reported so far. Before we come to the more detailed description of the pathway

engineering, the rationale on such endeavour will be provided in the following chapter.

3.1 Halogenated hydrocarbons and 1,2,3-trichloropropane

Many anthropogenic compounds mentioned in chapter 2.1 were halogenated

hydrocarbons. These chemicals had been considered as bearers of progress at the

beginning of their synthetic production in the first half of the last century. They were used

as insecticides (e.g., DDT, HCH), herbicides (e.g., atrazine), soil fumigants (e.g.,

1,2-dibromoethane) and plant growth regulators (e.g., halogenated benzoic acids).

Halogenated hydrocarbons protected millions of people from malaria and other wasting

diseases and enabled the Green Revolution in agriculture between 1940s and 1960s.

Manufacturing of reactive halogenated intermediates, such as epichlorohydrin or vinyl

chloride, supported a boom of chemical industry and introduced plastics into our lives.

However, the environmental-fate and toxicity studies triggered by the evidence provided

in famous book of Rachel Carson Silent Spring published in 1962, indicated that many of

these compounds can act as a persistent pollutants in the environment and can be harmful

to the ecosystems and human health [180]. Halogens became “not-to-trust elements“,

although it is now obvious that synthetic halogenated hydrocarbons have much more

structural analogues in the living nature than was ever supposed [181].

1,2,3-trichloropropane. Among other halogenated hydrocarbons, chlorinated

alkanes including 1,2-dichloroethane, 1,2-dibromoethane or 1,2,3-trichloropropane (TCP),

extensively used in agriculture or industry, represent common groundwater contaminants

in many developing and industrial countries [182]. Particularly TCP constitutes a hard nut

for remediation technologists due to its physicochemical properties, toxicity to living

organisms and environmental recalcitrance. TCP is colorless, volatile liquid with a sharp,

sweet odor, manufactured by major chemical companies such as Dow Chemical Company

or Shell. Recent report of Dow estimates worldwide annual production of TCP to be below

50,000 metric tons [183]. TCP is formed predominantly as a by-product during synthesis

of other important chemicals such as epichlorohydrin or cross-linking agent

hexafluoropropylene. It used to be present in soil fumigant 1,3-dichloropropene known

under trade name D-D and occured in past also in customer products, such as certain

Introduction

- 35 -

paint removers and cleaning agents. Mainly due to the improper waste management and

its incorporation in D-D, TCP became a widespread contaminant of soils, ground and

underground waters. Contamination was proved in several states in EU [184,185], USA

[186] and Asia [187]. In 2010, the US Environmental Protection Agency listed TCP as an

emerging contaminant. The EU Chemicals Agency considers TCP as a chemical of very high

concern because of its carcinogenic and toxic effects. TCP is anticipated human carcinogen

based on evidence of carcinogenicity in studies with experimental animals [188]. TCP

caused DNA damage, including formation of DNA adducts, and cell proliferation in rats

and mice and gene mutations or various chromosomal aberations were observed upon

exposure in bacteria, yeast, and mammalian cells in vitro [189,190]. Accidental ingestion

or inhalation of contaminated air resulted in acute intoxication and damage of liver in

humans [191]. Concentrations ranging from 0.1 to 74 μg.L-1 have been detected in

drinking water in Hawaii and California, two countries that continuously monitor TCP in

water sources [192]. State of Hawaii established maximal TCP level in a drinking water of

0.6 μg.L-1 (cca. 4.1 nM).

Abiotic transformations of TCP. The half-life of abiotic hydrolysis of TCP under

environmental conditions (25°C, pH=7) was estimated to be in the order of hundreds of

years [193]. Persistent nature and physicochemical properties of TCP listed in Table 1,

such as high density or limited water solubility (cca. 1.5 g.L-1), make a remediation of the

contaminated sites a difficult task [194]. Methods for in situ treatment of TCP-polluted

water, including vacuum extraction and subsequent chemical oxidation with ozone or

hydrogen peroxide, application of activated carbon or reduction with zero-valent zinc; are

under investigation [195]. However, pump and treat technologies and abiotic

transformations are expensive and relatively inefficient [194]. In vivo or in vitro

biotransformation, especially when coupled with the recycling of the biocatalyst, could

provide economical and efficient alternative for TCP removal.

Table 1. Basic physicochemical properties of metabolites from TCP pathway.a

chemical SMILES

molecular

weight density

boiling

point

melting

point logPa

(g.mol

-1) (g.ml

-1) (°C) (°C)

TCP C(C(CCl)Cl)Cl 147.4 1.39 156 -14 2.29

DCP C(C(CCl)Cl)O 129.0 1.36 182 NA 0.78

ECH C1C(O1)CCl 92.5 1.18 115-117 -57 0.35

CPD C(C(CCl)O)O 110.5 1.32 213 -40 -0.71

GDL C1C(O1)CO 74.1 1.12 61-62 -54 -0.93

GLY C(C(CO)O)O 92.1 1.25 182 18 -1.93

[a] Source of information: ChemSpider database http://www.chemspider.com/

[b] Calculated using ALOGPS 2.1 program [196].

Abbreviations: TCP, 1,2,3-trichloropropane; DCP, 2,3-dichloropropane-1-ol; ECH, epichlorohydrin;

CPD, 3-chloropropane-1,2-diol; GDL, glycidol; GLY, glycerol; NA, data not available.

Introduction

- 36 -

Biotic transformations of TCP. Biodegradation of TCP was observed both under

anaerobic and aerobic conditions. Recently, several Gram-negative bacterial strains from

genus Dehalogenimonas were isolated from contaminated groundwater and showed to

reductively dehalogenate TCP under anoxic conditions [197,198]. The major drawback of

this biotransformation is that it results in potentionally toxic products such as allyl

chloride or allyl alcohol and the mechanism of reductive dechlorination has not yet been

elucidated. In situ reductive dechlorination of contaminated groundwater can be

performed by direct injecting hydrogen-releasing compounds or hydrogen, which serves

as an electron donor [195]. However, this time-demanding strategy was applied only with

TCP concentrations below 1 mg.L-1. Anaerobic transformations might be more suitable for

in situ bioremediation than aerobic processes, due to the difficulty of supplying

homogeneous oxygen to the underground. Nevertheless, aerobic biodegradation

pathways are often studied, because they may lead to the complete mineralization of

target substrate and support growth of bacteria, which is highly desirable [128]. Aerobic

conversion of TCP was observed only in cometabolic mode, catalyzed by the methane

monooxygenase from the resting cells of methanotropic bacterium Methylosinus

trichosporium OB3b [199]. Calculations performed by Dolfing and Janssen proved that

complete mineralization of TCP by bacterial oxidative metabolism through lesser

chlorinated propanes and HCl formation is thermodynamically feasible [200]. Regrettably,

no organism capable of aerobic utilization of TCP as a source of carbon and energy for

growth has been isolated from polluted sites or enrichment cultures thus far. This can be

attributed to the: (i) relatively short presence of TCP in the environment and

corresponding lack of efficient dehalogenating enzymes and (ii) low adaptability potential

of bacteria toward this substrate [182]. Direct toxic effect of TCP or its reaction

intermediates or indirect incidence of reactive oxygen species (ROS), that can be formed

during aerobic biodegradation of chlorinated pollutants, may represent significant

handicap for evolution [201,202]. Due to the absence of natural degrader, TCP, together

with some other halogenated aliphatic hydrocarbons such as 1,2-dichloropropane or

trichloroethylene, was listed among extremely difficult targets for aerobic biodegradation

(Figure 8).

Introduction

- 37 -

Figure 8. Degradability of some aliphatic and aromatic compounds including those with

halogenated substituent(s). Adopted from [182].

Toxicity and oxidative stress. TCP toxicity mode of action on bacterial cells is not

fully understood. As can be deduced from studies on eukaryotes, TCP and its presumed

reaction intermediates, such as epoxides frequently produced by oxidative conversion of

chlorinated aliphatics, can form DNA adducts and induce genetic damage [200]. They can

also react with other biomacromolecules and cellular components, e.g., cell membranes.

Reactivity of TCP towards membrane structures could be promoted by high

hydrophobicity of the compound (Table 1). Saturation of unsaturated fatty acids by

halogenation or lipid peroxidation triggered by epoxides could cause change in membrane

fluidity, resulting in collapse of the electron transport chain and electrochemical potential

that maintains viability [203]. Electron transport chains are major sites of premature

electron leakage to oxygen. Disruption of respiratory chain could promote formation of

ROS, such as superoxide and hydrogen peroxide, and oxidative stress through further

damage of DNA and oxidation of lipids and proteins. More research is needed to validate

these hypothetical modes of action and will be an objective of future research in

Loschmidt Laboratories.

Introduction

- 38 -

It is evident that rapid enzymatic dehalogenation, forming harmless product

molecule acceptable in natural metabolic network of host organism, represents a critical

step for successful biodegradation of TCP. Selection of enzymes suitable for such task and

design of aerobic catabolic pathway for TCP mineralization are discussed in the following

chapter.

3.2 Overview of TCP pathway engineering

In 1989, van den Wijngaard and co-workers isolated from freshwater sediment

Gram-negative bacterial strain AD1 growing on epichlorohydrin (ECH) as the carbon and

energy source [204]. The bacterium, known recently as Agrobacterium radiobacter AD1,

was able of aerobic utilization of dihalogenated alcohols, such as 1,3-dichloropropan-1-ol,

via ECH, 3-chloropropane-1,2-diol (CPD), glycidol (GDL). Resulting reaction product

glycerol (GLY) enters the glycolysis upon phosphorylation and oxidation. Enzymes

expected to catalyze the reactions leading to GLY were epoxide hydrolase and haloalcohol

dehalogenase. The authors proposed TCP mineralization by addition of a single

dehalogenation step to the metabolic pathway of AD1.

Hydrolysis of a carbon-halogen bond in aliphatic halogenated alkanes and alkenes is

carried out by a spectrum of well characterized haloalkane dehalogenases (HLDs),

enzymes belonging to the α/β-hydrolase fold superfamily of proteins [205,206]. HLDs

have been reported to catalyze the first dehalogenation step in aerobic biodegradation

pathways of important halogenated pollutants, such as 1,2-dichloroethane,

1,2-dibromoethane and HCH [207]. Analysis of HLDs isolated during the last two decades

indicate that dehalogenation is the main function of these enzymes. Thus, mentioned

activities are probably not promiscuous and were rather evolved from activities toward

natural counterparts of halogenated xenobiotics [182]. First HLD capable of TCP

dehalogenation was characterized by Yokota and co-workers in 1987 (Table 2) [208]. The

authors described the enzyme from Corynebacterium sp. strain m15-3. (later renamed as

Rhodococcus sp. m15-3). Dehalogenase was later assigned DhaA and its sequence was

described by Kulakova and co-workers who isolated the gene from 1-chlorobutane

degrader Rhodococcus rhodochrous NCIMB13064 [209]. The enzyme converted prochiral

TCP into both enantiomers of 2,3-dichloropropan-1-ol (DCP) in almost equimolar ratio.

Importantly, dhaA gene was found to be quite widespread, occuring in several

geographically distinct Rhodococcus strains but never in bacteria possessing enzymes for

haloalcohol utilization [210]. Therefore, complementation and further evolution of these

functions within a single organism was unlikely, which may explain the absence of natural

TCP degrader. The attempts to obtain a recombinant bacterial biocatalyst by means of ME

and PE followed.

Introduction

- 39 -

ME and PE strategies applied to TCP project. The first recombinant strain capable

of TCP utilization was constructed in 1999 through heterologous expression of wild type

DhaA in A. radiobacter AD1 and assembly of the synthetic metabolic pathway (Figure 9)

[211]. Bosma and co-workers placed dehalogenase gene under the control of strong

constitutive promoter and introduced it into AD1 on a broad host range plasmid using

triparental mating. The resulting construct showed growth on 1,2,3-tribromopropane and

1,2-dibromo-3-chloropropane as a sole carbon sources in minimal medium, but not on

TCP. Only 0.7 mM (0.14 mmol ≌ 21 mg) of TCP was utilized by 200 mL cell culture of

starting OD450 of 0.07 within 25-day cultivation. This poor conversion of TCP was

attributed to the low activity of wild-type DhaA with the anthropogenic substrate TCP

(kcat/Km=40 s-1.M-1). The catalytic efficiency was more than two orders of magnitude lower

than that of dehalogenase DhlA from Xanthobacter autothropicus GJ10, the natural

degrader utilized in a groundwater purification plant treating xenobiotic pollutant

1,2-dichloroethane [125].

Bosma and co-workers published follow-up work describing improvement of DhaA

activity with TCP by PE in 2002. Although a crystal structure of DhaA has already been

available [212], they applied directed evolution, which was more frequently used for the

purpose of enzyme optimization in that time. One round of DNA shuffling and subsequent

epPCR resulted in variant DhaAM2 carying two substitutions (C176Y, Y273F) and 8-times

higher catalytic efficiency toward TCP [213]. Molecular modelling suggested that

introduced mutations minimized the occurence of unproductive binding mode of TCP in

the active site of DhaA. 200 mL culture of recombinant AD1 of starting OD450 of 0.14

constitutively expressing dhaAM2 gene under the strong dhla promoter on plasmid

utilized 3.6 mM (0.72 mmol ≌ 106 mg) of TCP within 10 days. Significant, but not

sustainable increase of optical density of the culture (final OD450 of 0.45) was observed.

This demontration represents a significant step towards viable TCP degrader, however,

the designed strain suffered from two limitations: (i) the dehalogenation of TCP by

DhaAM2 was not sufficiently fast to ensure rapid conversion of substrate that is toxic to

the cells already at the contration of 1 mM, (ii) cumulation of (S)-DCP caused by

enantioselectivity of haloalcohol dehalogenase.

Introduction

- 40 -

Table 2. Basic properties of the enzymes from TCP pathway.

enzyme

haloalkane

dehalogenase

DhaA

EC 3.8.1.5

haloalcohol

dehalogenase

HheC

EC 4.5.1.-

epoxide

hydrolase

EchA

EC 3.3.2.3

organism

Rhodococcus

rhodochrous

NCIMB13064

Agrobacterium

radiobacter AD1

Agrobacterium

radiobacter AD1

superfamily α/β-hydrolases

NAD(P)-binding

Rossmann-fold

domains

α/β-hydrolases

GenBank ID AF060871.1 AF397296.1 Y12804.1

residue

count/monomer 293 255 294

quarternary

structure

(PDB ID)

no

(1CQW)

homotetramer

(1ZMT)

no

(1EHY)

reaction

mechanism

nucleophilic

substitution

nucleophilic

substitution

nucleophilic

substitution

catalytic

residues

Glu130, His272,

Asp106

Ser132, Tyr145,

Arg149

Asp107, His275,

Asp246

substrate scope

broad specificity,

halogenated aliphatic

and cyclic compounds

broad specificity,

aliphatic and aromatic

vicinal haloalcohols

broad specificity,

1,2-epoxyalkanes,

styrene oxide

MW

(kDa)/monomer

pI/monomer

33.2

4.95

28.0

5.29

34.0

5.18

temp. optimum

(°C) 37 50 50

pH optimum 8.0-9.5 8.0-9.0 8.0-9.0

references [207,208] [214] [216]

Abbreviations: PDB, Protein Data Bank; MW, molecular weight; pI, isoelectric point.

Introduction

- 41 -

Accumulation of epoxides was not observed, but 10-times lower specific activity of

EchA with GDL than with ECH indicated a possible problem [214]. The flux of carbon and

energy through the synthetic pathway possesing these bottlenecks was obviously not

sufficient to provide enough GLY for sustainable growth, resulting in a failure of the

attempts to degrade TCP in a continuous packed-bed bioreactor inoculated with the

recombinant strain.

Cl

Cl Cl

Cl

Cl OHO

Cl

OH

OH

Cl

OH

O

OH

OH OH

1,2,3-trichloropropane (R,S)-2,3-dichloropropane-1-ol DCP

(R,S)-epichlorohydrin ECH

(R,S)-glycidol GDL

(R,S)-3-chloropropane-1,2-diol CPD

glycerol GLY

glycolysis

DhaA HheC

HheCEchA

EchA

TCP

glycerol kinase and glycerol-3-phosphate dehydrogenase

+H2O/-HCl -HCl

-HCl

+H2O

+H2O

Figure 9. Scheme of the synthetic metabolic pathway for aerobic degradation of 1,2,3-trichloropropane to the glycerol. At the turn of millenium, several studies appeared that were focused on deeper

characterization of two enzymes that form synthetic TCP pathway together with

haloalkane dehalogenase DhaA: haloalcohol dehalogenase HheC and epoxide hydrolase

EchA (Figure 9 and Table 2) [214–218]. Both enzymes were also engineered towards

improved activity, enantioselectivity, stability or broader substrate specificity, using

either rational design or directed evolution methods [219–223]. However, none of the

studies aimed at application of HheC or EchA in TCP biodegradation and no mutants with

properties optimized for this purpose were obtained. In contrast, DhaA and its activity

with difficult substrate TCP drew considerable attention. Using computational simulations

and quantum mechanical calculations, Banas and co-workers pointed out the significance

of access tunnels, connecting the bulk solvent with the burried active site of DhaA, for the

catalysis of TCP [224]. This finding was fundamental for further enhancement of DhaA

activity. In 2009, Pavlova and co-workers published a new mutant, DhaA31, with 32-times

improved turnover number and 26-times higher catalytic efficiency (kcat = 1.26 s-1; kcat/Km

= 1,050 s-1.M-1), when compared with wild type DhaA [225]. Applying computer-assisted

Introduction

- 42 -

design and site-directed and saturation mutageneses, three new substitutions (I135F,

V245F, L246I) were introduced into the main and slot tunnels of template DhaAM2.

Computational modelling and kinetic analyses revealed that new bulky residues protected

the active site from excess of water molecules, which otherwise disfavours nucleophilic

substitution. The enantioselectivity of DhaA31 remained unaltered.

Van Leeuwen and co-workers targeted the enantioselectivity of DhaA31 in order to

obtain the variants converting TCP predominantly into either (R)- or (S)-DCP [226]. Five

rounds of focused directed evolution employing saturation mutagenesis with restricted

codon sets provided mutant DhaA r5-90R with thirteen new amino acid substitutions and

improved (R) selectivity (e.e. = 90%) in comparison with DhaA31 (e.e. = 13%).

Unfortunately, mutagenesis affected the activity with TCP which dropped to the level of

DhaA (kcat/Km = 25 s-1.M-1). New DhaA variant could be potentially utilized in TCP pathway

to complement the function of (R)- selective HheC, because no haloalcohol dehalogenase

capable of efficient conversion of (S)-DCP has been isolated thus far.

The reseach group of Dick B. Janssen provided also alternative solution for

unmatching selectivity of DhaA and HheC. Arif and co-workers published a report on

isolation and characterization of novel P. putida strain MC4 that could utilize both

enantiomers of DCP as a sole carbon and energy source for growth [227]. The growth was

allowed by the lack of enantioselectivity of the initial oxidative

dehalogenase/dehydrogenase DppA that converted DCP to toxic 2-chloroacrolein and

2-chloroacrylic acid. Further degradation of 2-chloroacrylic acid in MC4 has not been

studied, but the pathway is supposed to continue through hydrolytic dechlorination to

yield pyruvate or reduction and further dehalogenation which could yield lactate. The

authors wanted to take advantage of the properties of the new strain for TCP

biodegradation and recently introduced dhaA31 gene with constitutive dhlA promoter into

the chromosome of MC4 using the transposon delivery system [228]. The resulting

recombinant bacterium P. putida MC4-5222 utilized 3.1 mM TCP in minimal medium

within 25 days of fed-batch cultivation. (volume of culture was not reported). The growth

of culture with slow growth rate (μ = 0.0012–0.008 h-1) resembling the one reported for

engineered strain AD1 with dhaAM2 expressed from plasmid was observed. Moreover, the

immobilized cells of MC4-5222 were tested for biodegradation of TCP in a laboratory scale

packed-bed reactor. The efficiency of TCP removal, supplied continuously at the

concentrations 0.30–0.33 mM, increased from 87 – 97% during 48 days of reactor

operation. The experiment proved good stability of chromosomal insertion and the

potential of cells to survive prolonged time interval in presence of TCP supplied as the

sole carbon source.

The engineered P. putida MC4-5222 represents a rare example of genetically

modified bacterium which showed growth on an important recalcitrant environmental

pollutant. Still, the recombinant cells were sensitive to the concentrations of TCP as low as

1 mM and further improvement of strain characteristics is desirable for a full scale

industrial application [228]. Adaptive laboratory evolution in the presence of chemical

mutagen ethyl methanesulfonate and TCP at the upper limit of tolerance were not

Introduction

- 43 -

successful. The authors attributed the limited fitness to the possible side reactions and

corresponding accumulation of reactive by-products, that might occur in metabolic

network of the strain MC4-5222 upon growth on TCP. However, these constraints are

difficult to trace back because pathway that converts DCP to the metabolites utilizable in

central metabolic routes is not well understood. The original TCP pathway encompassing

DhaA, HheC, and EchA, offers following advantages over new route from P. putida: (i) all

enzymes and metabolites of the pathway are known and characterized (Tables 1 and 2),

(ii) the enzymes do not need cofactors and were shown to be active under in vitro

conditions, and (iii) the pathway yields well defined utilizable product. These

characteristics proved to be useful in rational pathway engineering reported in the

Results part of this Thesis.

In this Thesis, the synthetic TCP pathway composed of DhaA, HheC and EchA

enzymes is used as a model system for knowledge-based engineering of an organism

degrading xenobiotic pollutant. A unique DBTA cycle encompassing following five steps

was used (Figure 10): (1) study of assembled TCP pathway with wild-type or engineered

enzymes both in soluble and immobilized forms in vitro, (2) determination of kinetic

parameters of individual enzymes and construction of a kinetic model describing the five-

step conversion of TCP to GLY, (3) application of the kinetic model for study of a dynamic

behaviour of in vitro three-enzyme system, (4) utilization of validated model for rational

design of the orthogonal pathway in laboratory strain E. coli, and (5) transfer of the most

promissing pathway variants into P. putida chassis. The data and new knowledge

generated within this project should provide deeper insight into this interesting model

system, should allow rational dissection of pathway bottlenecks and will hopefully lead to

the development of better biocatalyst for TCP biodegradation. Presented concepts and

tools should be applicable to engineering organisms degrading also other recalciltrant

Introduction

- 44 -

Figure 10. A knowledge-based strategy applied for rational engineering of a synthetic metabolic pathway for biodegradation of 1,2,3-trichloropropane. A) Metabolic pathway is initially reconstructed in vitro in a form of free or immobilized purified enzymes and its functioning is verified by measuring metabolite concentrations using available analytical techniques. Kinetic parameters of individual enzymes with their target substrates and reaction mechanisms are determined. B) Kinetic parameters and equations are utilized for development of a kinetic model. Model is experimentally validated with purified enzymes and used for dynamic in silico simulations of pathway performance under selected constraints. Productivity of the pathway is maximized through optimizing enzyme stoichiometry and incorporating available engineered enzymes. C) Synthetic pathway is assembled in a modular way as an orthogonal biological system in suitable heterologous host. Established mathematical model supplemented with relevant biological parameters is used to calculate all possible pathway variants within in vivo constraints. Only the best performing variants are constructed and experimentally characterized.

- 45 -

- 46 -

Contribution to the results

- 47 -

CONTRIBUTION TO THE RESULTS

The Results section of the Thesis consists of four chapters. Chapters 1 – 3 are based

on three original peer-reviewed articles. Chapter 4 is summarising data and results

under preparation for publication.

1. Dvorak, P., Bidmanova, S., Prokop, Z., and Damborsky, J. (2014) Immobilized

synthetic pathway for biodegradation of toxic recalcitrant pollutant

1,2,3-trichloropropane. Environ. Sci. Technol. 48, 6859-6866.

2. Dvorak, P., Kurumbang, N. P., Bendl, J., Brezovsky, J., Prokop, Z., and Damborsky,

J. (2014) Maximizing the efficiency of multi-enzyme process by stoichiometry

optimization. ChemBioChem DOI: 10.1002/cbic.201402265

3. Kurumbang, N. P.*, Dvorak, P.*, Bendl, J., Brezovsky, J., Prokop, Z., and

Damborsky, J. (2014) Computer-assisted engineering of the synthetic pathway for

biodegradation of a toxic persistent pollutant. ACS Synth. Biol. 3, 172-181.

(*shared first authors)

4. Dvorak, P., Nikel, P., de Lorenzo, V., Prokop, Z., Damborsky, J. (2014) Assembly of

the synthetic pathway for biodegradation of 1,2,3-trichloropropane in

Pseudomonas putida KT2440 CF1. (in preparation)

Author's contribution to the articles:

1. design of experiments, gene cloning, preparation of soluble and immobilized

enzymes and their characterization, development of analytical methods, assembly

of packed bed reactor, batch and continuous multi-enzyme reactions,

interpretation of data, writing of the manuscript

2. design of in vitro experiments, preparation of purified enzymes, determination of

steady-state kinetic parameters of enzymes, enatioselectivity measurements,

determination of inhibition constants, production of (S)-DCP by preparative

biocatalysis, batch multi-enzyme reactions, interpretation of data, writing of the

manuscript

3. design of in vitro and part of in vivo experiments, quantification of enzyme

content in cells by SDS-PAGE analysis and activity measurements, determination

of accumulation of TCP and its metabolites in host cells, degradation of TCP in

buffer by pre-induced resting cells, recovery of cells in LB medium, interpretation

of data, writing of the manuscript together with N. P. Kurumbang

4. design of experiments, construction of synthetic operon, cloning and

transformations, characterization of recombinant bacteria bearing synthetic

operon on plasmid and in chromosome, writing of the manuscript

- 48 -

- 49 -

CHAPTER 1

In vitro assembly and immobilization of

the synthetic pathway for biodegradation

of toxic recalcitrant pollutant

1,2,3-trichloropropane

Chapter 1

- 50 -

INTRODUCTION

1,2,3-trichloropropane (TCP) is an anthropogenic compound recently recognized as

an emerging contaminant of groundwater [186]. TCP is produced worldwide in quantities

reaching 50,000 metric tons annually and used by chemical companies as a solvent,

precursor of soil fumigants, and building block for synthesis of other chemicals, e.g.,

dichloropropene or polysulfone liquid polymers [183]. TCP is also formed as a by-product

during the synthesis of epichlorohydrin. Due to its massive production, TCP can often be

found at industrial and hazardous waste sites [229]. Recent incidents with drinking water

sources in California polluted by TCP emphasized the need to develop efficient

technologies for removing this toxic and carcinogenic compound from the environment.

Conventional remediation techniques are relatively inefficient, due to the physical

and chemical properties of the compound [230]. The exception is promissing reductive

conversion by zero-valent zinc [194]. In addition to abiotic transformations,

biotransformations are extensively studied for their ability to decontaminate sites

polluted with chemical contaminants [231]. Recently, isolated bacterial strains were found

to transform TCP under anaerobic conditions [197,198]. However, these

biotransformations often result in products toxic for surrounding environments and due

to the low efficiency can be applied only at limited (<1 mg.L-1) TCP concentrations [195].

No organism capable of aerobic TCP biodegradation has yet been isolated, probably due to

the anthropogenic nature of TCP and its recent introduction into the environment.

The absence of natural catabolic pathways for aerobic TCP utilization was addressed

by Bosma and co-workers [211,213]. The authors assembled a synthetic metabolic

pathway with the heterologous expression of haloalkane dehalogenase DhaA from

Rhodococcus sp. in the natural host Agrobacterium radiobacter AD1, capable of utilization

of haloalcohols. Engineered AD1 strain, expressing 5 times more active variant of

haloalkane dehalogenase, showed a slow growth on 1 mM TCP in minimal medium.

However, practical utility of this construct is limited by the toxicity of TCP for bacterial

cells above 1 mM concentration, isufficient activity of engineered haloalkane

dehalogenase, accumulation of toxic intermediates and legislative barriers on the use of

genetically modified microorganisms [213]. We recently addressed the problems of the

low catalytic efficiency of the first dehalogenation step and accumulation of intermediates

limiting the productivity of the pathway in vivo by combination of protein and metabolic

engineering [232].

In this study, we report the application of an immobilized enzymatic pathway for a

complete five-step degradation of TCP to the harmless product glycerol (GLY; Figure 1).

We immobilized engineered 32-times more active haloalkane dehalogenase DhaA31 from

Rhodococcus rhodochrous NCIMB 13064 [225], the wild-type haloalcohol dehalogenase

HheC and the wild-type epoxide hydrolase EchA from A. radiobacter AD1 into cross-linked

enzyme aggregates (CLEAs) and polyvinylalcohol (PVA) LentiKats lenses [233,234]. The

immobilized biocatalysts were used to convert TCP into the desirable commodity

chemical GLY in both batch and continuous systems. A comparison of the dynamic

Chapter 1

- 51 -

behavior of the complex multi-enzyme system in both soluble and immobilized forms is

provided. To the best of our knowledge, this is the first report on the use of an

immobilized synthetic metabolic pathway employing engineered enzyme for the

biotransformation of an environmental pollutant. The established immobilization strategy

is robust and suitable for scale-up. The developed biocatalyst is recyclable, resistant to

biodegradation, compatible with high input loads of TCP, and can operate under mild non-

sterile conditions. Moreover, the possibility to operate the process without the use of

genetically modified microorganisms makes this biotechnology suitable for environmental

applications.

Cl

Cl Cl

Cl

Cl OHO

Cl

OH

OH

Cl

OH

O

OH

OH OH

1,2,3-trichloropropane (R,S)-2,3-dichloropropane-1-ol DCP

(R,S)-epichlorohydrin ECH

(R,S)-glycidol GDL

(R,S)-3-chloropropane-1,2-diol CPD

glycerol GLY

(glycolysis)

DhaA HheC

HheCEchA

EchA

TCP

Figure 1. Scheme of the five-step synthetic metabolic pathway for biotransformation of 1,2,3-trichloropropane to glycerol. Used abbreviations of individual metabolites are shown. DhaA, haloalkane dehalogenase from Rhodococcus rhodochrous NCIMB 13064; HheC, haloalcohol dehalogenase; and EchA, epoxide hydrolase, both from Agrobacterium radiobacter AD1. The final product glycerol can be utilized in glycolysis when the pathway is assembled in vivo.

Chapter 1

- 52 -

MATERIAL AND METHODS

Reagents. TCP, 2,3-dichloropropane-1-ol (DCP), epichlorohydrin (ECH), 3-chloropropane-

1,2-diol (CPD), glycidol (GDL), and GLY standards were purchased from Sigma-Aldrich

(USA). All of the chemicals used in this study were of analytical grade. Bovine serum

albumin (BSA) for preparation of CLEAs was purchased from Sigma-Aldrich (USA). PVA

and polyethylene glycol of MW 1,000 were provided by LentiKat’s a.s. (Czech Republic).

The solution of the cross-linker dextran polyaldehyde (DPA) was prepared according to a

procedure described elsewhere [235]. A Free Glycerol Assay Kit was purchased from

BioVision (USA). The work with toxic compounds was conducted in a fume hood and with

protective equipment to minimize safety risks.

Gene synthesis and cloning. The nucleotide sequences of genes encoding wild-type

haloalcohol dehalogenase HheC and epoxide hydrolase EchA from Agrobacterium

radiobacter AD1 were downloaded from GenBank database (accession numbers

AF397296.1, Y12804.1, respectively). The tag sequence of six histidine codons was

attached downstream of the echA gene. Restriction sites for cloning into pET21b (NdeI,

BamHI) (Merck, Germany) and pET28b (NcoI, HindIII) were attached to the sequences of

the hheC and echA genes, respectively. Due to the creation of a NcoI restriction site, the

second codon of the echA gene (ACT) encoding threonine was substituted for a GCA codon

encoding alanine. Genes of HheC and EchA together with genes of the wild-type and

mutant haloalkane dehalogenase DhaA and DhaA31 were commercially synthesized

(Geneart, Germany). Sequences of all genes were optimized for expression in Escherichia

coli during synthesis. Synthetic genes were subcloned into the NdeI and BamHI restriction

sites of pET21b (dhaA, dhaA31 and hheC), and NcoI and HindIII restriction sites of pET28b

(echA). The resulting constructs pET21b-dhaA, pET21b-dhaA31, pET21b-hheC and

pET28b::echA were transformed into E. coli DH5α using the heat-shock method for

plasmid propagation.

Cultivation conditions and purification of enzymes. Competent cells of the E. coli

BL21(DE3) strain were transformed with pET21b-dhaA, pET21b-dhaA31, pET21b-hheC or

pET28b-echA using the heat shock method, and plated on LB agar plates with ampicillin

(100 µg.ml-1) or kanamycin (50 µg.ml-1). Plates were incubated overnight at 37°C. Single

colonies were used to inoculate 10 ml of LB medium with the respective antibiotic, and

cells were grown overnight at 37°C. Overnight cultures were used to inoculate 1 l of LB

medium with the respective antibiotic. Cells were cultivated at 37°C with shaking until an

OD600 of 0.4–0.6, after which expression was induced with 0.5 mM IPTG. Cells were then

cultivated overnight at 20°C. Biomass was harvested by centrifugation. Cells were washed

and resuspended in purification buffer A (20 mM K2HPO4 and KH2PO4, 0.5 M NaCl, 10 mM

imidazole, pH 7.5). 1 U of DNaseI (New England Biolabs, USA) per 1 ml of cell suspension

was added. Cells were disrupted by sonication using the ultrasonic processor Hielscher

UP200S (Teltow, Germany) with 0.3 s pulses and 85% amplitude. Cell lysate was

Chapter 1

- 53 -

centrifuged for 1 h at 21,000 g at 4°C, and the resulting cell-free extract was decanted.

DhaA, DhaA31 and EchA were purified using single-step nickel affinity chromatography.

Crude extract was applied to a 5 ml Ni-NTA Superflow column (Qiagen, Germany). The

column was attached to a BioLogic Duo Flow (Bio-Rad, USA). The buffer system consisted

of buffer A and buffer B (20 mM K2HPO4 and KH2PO4, 0.5 M NaCl, 500 mM imidazole, pH

7.5). Recombinant enzymes were eluted at 60% of buffer B during a two-step gradient

method: 0–10% in 5 column volumes and 10–60% of buffer B in 10 column volumes.

Fractions containing DhaA, DhaA31 or EchA were pooled and proteins were concentrated

using a stirred ultrafiltration cell (Millipore, USA). Enzymes were dialyzed against 50 mM

phosphate buffer (pH 7.5). HheC was purified using anion-exchange chromatography.

Crude extract was applied to a 35 ml glass Econo-Column (Bio-Rad, USA) packed with 25

ml of Q Sepharose Fast Flow (GE Healthcare, USA). The column was attached to a BioLogic

Duo Flow (Bio-Rad, USA). The buffer system consisted of buffer A (20 mM Tris-SO4, 1 mM

EDTA, 1 mM β-mercaptoethanol, pH 7.5) and buffer B (20 mM Tris-SO4, 1 mM EDTA, 1 mM

β-mercaptoethanol, 0.45 M (NH4)2SO4, pH 7.5 ). HheC was eluted at 20–25% of buffer B

during a two-step increasing linear gradient: 0–45% buffer B in 20 column volumes and

45– 100% buffer B in 5 column volumes. Fractions containing HheC were pooled and the

protein was concentrated. Enzyme was dialyzed against PD buffer (50 mM phosphate

buffer of pH 7.5 with 2 mM dithiothreitol). Concentrations of DhaA31, EchA and HheC

were determined by Bradford Reagent (Sigma-Aldrich, USA). Enzyme purity was judged

by SDS-PAGE analysis. Purified enzymes were stored for further use at 4°C.

Preparation of CLEAs. CLEAs of DhaA31 and EchA were prepared by dissolving 25 mg of

enzyme and 25 mg of BSA in 1 ml of 10 mM phosphate buffer (pH 7.5). The solution was

added to 9 ml of saturated ammonium sulfate (pH 8.0). After 1 h of incubation in an ice

bath with stirring, 1.3 ml of DPA was added and cross-linking occurred for another 1 h.

The resulting suspension was centrifuged at 4,000 g for 15 min at 4°C. The supernatant

was stored for the determination of residual enzymatic activity; CLEAs were re-suspended

in 20 ml of saturated sodium hydrogencarbonate and incubated with stirring in an ice

bath for 30 min. The suspension was centrifuged at 4,000 g for 15 min at 4°C and the

supernatant was removed. CLEAs were washed with a 50 mM phosphate buffer (pH 7.5)

and stored in 1 ml of this buffer at 4°C before further use.

Encapsulation of enzymes into PVA particles LentiKats. PVA (0.56 g) and polyethylene

glycol (0.34 g) were mixed with 3.7 ml of distilled water and heated at 98°C until PVA

dissolved completely. The liquid was cooled to 35°C, and 10 mg of free HheC in a PD buffer

(50 mM phosphate buffer of pH 7.5 with 2 mM dithiothreitol) or 1 g of CLEAs of EchA or

DhaA31 was added and mixed thoroughly. Small droplets (3-4 mm) of mixture were

dripped on plastic plates and incubated at 37°C until LentiKats lost 70% of their initial wet

weight. Dried LentiKats were soaked in 50 ml of 0.1 M sodium sulfate for 2 hrs to reswell.

HheC LentiKats were washed with a PD buffer and stored in the same buffer at 4°C.

Chapter 1

- 54 -

LentiKats of EchA and DhaA31 were washed with a 50 mM phosphate buffer (pH 7.5) and

stored at 4°C.

Preparation of combi-LentiKats from cell-free extracts. Cell-free extracts with DhaA31,

HheC or EchA were prepared from 200 ml of cell cultures using the same cultivation

conditions as described previously. Biomass was harvested by centrifugation. Cells were

washed and resuspended in 3 ml of 50 mM sodium phosphate buffer of pH 7.0. 1 U of

DNaseI (New England Biolabs, USA) per 1 ml of cell suspension was added. Cells were

disrupted by sonication using the ultrasonic processor Hielscher UP200S (Teltow,

Germany) with 0.3 s pulses and 85% amplitude. Cell lysates were centrifuged for 1 h at

21,000 g at 4°C, and the resulting cell-free extracts were decanted. Dithiothreitol was

added to the cell-free extract containing HheC to a final concentration of 2 mM. The total

protein concentration in the cell-free extracts was determined by Bradford Reagent.

Samples of cell-free extracts (5 µg of total protein) containing DhaA31, HheC or EchA were

loaded onto SDS-polyacrylamide gel. Gels were analyzed using a GS-800 Calibrated

Densitometer (Bio-Rad, USA). The amount of enzyme in the cell-free extract was

estimated according to a comparison of the trace density of its respective band with the

trace density of the standard sample band. CLEAs were prepared from cell-free extract

containing approximately 25 mg of DhaA31 or EchA. Combi-LentiKats were prepared by

mixing cell-free extract containing approximately 5 mg of HheC, CLEAs containing

approximately 5 mg of DhaA31 and aproximately 5 mg of EchA together with solubilized

PVA hydrogel with a mass ratio 1:4 (mixture of CLEAs and cell-free extract : PVA

hydrogel). Small droplets (3–4 mm) of mixture were dripped on plastic plates and

incubated at 37°C until combi-LentiKats lost 70% of their initial wet weight. Dried

combi-LentiKats were soaked in 50 ml of 0.1 M sodium sulfate for 2 hrs to re-swell.

Combi-LentiKats were washed with PD buffer and stored in the same buffer at 4°C.

Enzyme assays. Specific activities of soluble and immobilized DhaA31 with TCP and HheC

with DCP and CPD were assayed in 10 ml of a 50 mM Tris-SO4 buffer (pH 8.5) at 37°C with

a 10 mM substrate. The concentration of the reaction product (chloride ions) was

measured by the method of Iwasaki [236]. The specific activity of EchA was assayed by

following the substrate depletion in 10 μ of a 50 mM Tris-SO4 buffer (pH 8.5) at 37°C with

5 mM ECH or 10 mM GDL. Samples of the reaction mixture were mixed with acetone (1:1)

with internal standard hexanol and analyzed by gas chromatography (GC). The same GC

method was used for the quantitative analyses of all metabolites of the TCP pathway

except for GLY. The rates of abiotic conversions of all metabolites at selected time

intervals were negligible.

Storage stability of free and immobilized enzymes. Storage stability of free and

immobilized DhaA31, HheC and EchA was investigated at 4°C and in the case of free

enzymes, also at room temperature (22±2°C). Free and immobilized DhaA31 and EchA

were stored in 50 mM phosphate buffer (pH 7.5) without additives. Free and immobilized

Chapter 1

- 55 -

HheC was stored in PD buffer (pH 7.5; measurements at 4°C) or in 50 mM phosphate

buffer (pH 7.5) without additives (measurements at room temperature). The residual

activities of DhaA31, HheC and EchA stored at 4°C were measured with 10 mM TCP, DCP

and ECH, respectively, at certain time intervals using conditions described in the Enzyme

assays. The residual activities of DhaA31, HheC and EchA stored at 22±2°C were measured

with 10 mM TCP, DCP and ECH, respectively, in 10 ml of Tris-SO4 buffer of pH 8.2 at

22±2°C.

Confocal microscopy of immobilized enzymes. CLEAs of DhaA31 (200 µl) were

resuspended in 1 ml of carbonate/bicarbonate buffer (pH 9.5). The fluorescent dye

fluorescein 5(6)-isothiocyanate (FITC, Invitrogen, USA) was dissolved in dimethyl

sulfoxide (final concentration of 10 mg.ml-1). A solution of FITC (1 µl) was added to the

CLEAs, which were stained for 60 min at room temperature with shaking. After separation

by centrifugation at 4,024 g for 15 min at 4 °C, CLEAs were washed three times with

50 mM phosphate buffer (pH 7.5). EchA CLEAs (200 µl) were resuspended in 1 ml of

carbonate/bicarbonate buffer buffer (pH 8.3). The fluorescent dye pacific blue

succinimidyl ester (PBSE, Invitrogen, USA) was dissolved in dimethyl sulfoxide (final

concentration of 10 mg.ml-1). A solution of PBSE (5 µl) was added to the CLEAs, which

were stained for 60 min at room temperature with shaking. After separation by

centrifugation at 4,024 g for 15 min at 4 °C, CLEAs were washed three times with 50 mM

phosphate buffer (pH 7.5). The fluorescent dye X-rhodamine-5(6)-isothiocyanate

(5(6)-XRITC, Invitrogen, USA) was dissolved in dimethyl sulfoxide (final concentration of

1 mg.ml-1). A solution of 5(6)-XRITC (20 µl) was added to 200 µL of soluble HheC

(10 mg.ml-1) in carbonate/bicarbonate buffer (15 mM sodium carbonate, 30 mM sodium

bicarbonate, pH 9.5). HheC was stained for 60 min at room temperature with shaking.

Labelled enzyme was separated from unreacted dye using a disposable PD MiniTrap

column containing 2.1 ml of Sephadex G-25 medium (GE Healthcare, UK). All three

labelled enzymes (0.33 ml of solution of free HheC, 0.33 g of DhaA31 CLEAs and 0.33 g of

EchA CLEAs) were encapsulated in LentiKats by mixing together with 4 ml of PVA and PEG

solubilized in water. The resulting combi-LentiKats were washed with 200 ml of 50 mM

phosphate buffer (pH 7.5) and frozen in tissue freezing medium Jung at -26°C using a

Leica Cryocut 1800 Cryostat (Leica Microsystems, Germany). Samples were then

sectioned vertically and horizontally with a thickness of 5 μm and mounted on glass slides

with Mowiol 40-88 (Sigma-Aldrich, USA). Microscopy observations of labelled free HheC,

CLEAs of DhaA31 and EchA, and combi-LentiKats were performed with the Olympus IX81

imaging station FluoView500 (Olympus C&S Ltd., Czech Republic) using 10x, 20x and 40x

objectives. Images were analyzed using Olympus Fluoview (Olympus, Japan) and Imaris

(Bitplane, Switzerland).

Multi-enzyme conversion of TCP in batch system. The multi-enzyme conversion of TCP

to GLY was assayed in 15 ml of a 50 mM Tris-SO4 buffer (pH 8.5) in gas-tight glass vials

(Sigma-Aldrich, USA) incubated at 37°C. The reaction was initiated by adding 1 mg of

Chapter 1

- 56 -

DhaA31, HheC, and EchA in a soluble or immobilized form, into the reaction mixture with

5 mM TCP. The samples withdrawn from the reaction mixture (0.5 ml for soluble enzymes

or CLEAs and 0.1 ml for LentiKats) were mixed with acetone (1:1) containing hexanol as

an internal standard and analyzed by GC. Selected samples were analyzed by gas

chromatograph and mass spectrometer (GC-MS) (Agilent, USA) for identification of

metabolites otherwise routinely detected by GC. The concentration of GLY was

determined spectrophotometrically by the Free Glycerol Assay Kit. Samples of the

reaction mixture (0.1 ml) were heated at 95°C for 5 min, centrifuged at 18,000 g for 1 min,

diluted in an assay buffer and analyzed according to the manufacturer’s protocol.

Concentrations of GLY were calculated from absorbance at 570 nm. For evaluation of the

effects of pH and lower temperature, conversion of TCP was assayed in 25 ml glass vials

with a screw cap mininert valve (Sigma-Aldrich, USA) containing 15 ml of 50 mM sodium

phosphate buffer of pH 7.0 or 50 mM NaOH/glycine buffer of pH 10. Vials were incubated

in the water bath GLS Aqua Plus (Grant Instruments, UK) with shaking (200 rpm) at 20°C.

Recycling of immobilized enzymes in batch system. The reusability of DhaA31, EchA and

HheC immobilized in LentiKats was assayed in 25 ml glass vials with a screw cap mininert

valve (Sigma-Aldrich, USA) with 10 ml of 50 mM Tris-SO4 buffer (pH 8.5) and 1 mM TCP.

Each of ten cycles was started by the addition of LentiKats containing 1 mg of DhaA31,

2 mg of HheC, and 2 mg of EchA to the reaction mixture. After 100 min of incubation at

30°C, LentiKats were separated from the reaction mixture by decantation, washed in

25 ml of the reaction buffer and used for a new cycle. Samples of the reaction mixture

(0.5 ml) were collected at the beginning and end of each cycle, mixed with acetone (1:1)

containing an internal standard and analyzed by GC to determine the concentration of

TCP. In parallel, samples (0.1 ml) were taken for quantification of GLY under standard

conditions. The relative efficiency of the biocatalyst in each cycle was calculated from the

conversion of TCP to GLY, utilizing the conversion in the first cycle as 100%.

Multi-enzyme conversion of TCP in continuous system. Glass column 1 (28 cm in height,

1.5 cm internal diameter, 50 ml working volume) with 100 mg of DhaA31 immobilized in

47 g of wet LentiKats and glass column 2 (25 cm in height, 2.5 cm internal diameter,

100 ml working volume) with 100 mg of both HheC and EchA immobilized in 45 and 43 g

of LentiKats, respectively, were used for the removal of TCP in a ten-week continuous

operation of a packed bed reactor placed in fume hood at 22±2°C. A feed vessel contained

TCP dissolved under stirring in 1 l of distilled water buffered with 0.1 M Tris-SO4 (pH 8.2)

to a final theoretical concentration of 5 mM (week 1) or 10 mM (weeks 2-10). The

experimental concentrations of TCP, ranging from 2.25 to 7.97 mM, for evaluation of

system efficiency were determined in the input of column 1 (Figure 5 and Table S7 of

Supplementary tables and figures). The operation conditions for the packed bed reactor

are summarized in Table S2. Influent and effluent lines were constructed from

polytetrafluoroethylene tubing. Samples from the feed vessel, inlet of column 1 and

effluent vessels 1 and 2 were withdrawn periodically and analyzed by GC and the Free

Chapter 1

- 57 -

Glycerol Assay Kit. A new cycle of operation was started whenever the content of the feed

vessel was completely transferred to the effluent vessel 2. To evaluate possible leaching of

the immobilized enzymes from the reactor, samples from effluent vessel 2 were

withdrawn, concentrated 25 times using stirred ultrafiltration cells (Millipore, USA) and

analyzed by SDS-polyacrylamide gel electrophoresis (SDS-PAGE).

Operational conditions for GC and GC-MS analyses. A Gas Chromatograph 6890N with

flame ionization detector (Agilent Technologies, USA) and a capillary column ZB-FFAP

30 m x 0.25 mm x 0.25 µm (Phenomenex, USA) was used for routine analysis of samples

from multi-enzyme reactions. A Gas Chromatograph 7890A and Mass Spectrometer 5975C

MSD (Agilent Technologies, USA) equipped with a capillary column ZB-FFAP 30 m x

0.25 mm x 0.25 µm (Phenomenex, USA) was used for verification of the presence of

individual metabolites from the TCP pathway in selected samples from the multi-enzyme

reactions. Samples (2 μl) were injected into the GC with an inlet temperature of 250°C and

split ratio 20:1. The operational conditions for the column were: helium carrier gas with

an initial flow of 0.6 ml.min-1 for 1 min, followed by a flow gradient from 0.6 to

1.8 ml.min-1 (ramp 0.2 ml.min-1), temperature program set to give an initial column

temperature of 50°C for 1 min, followed by a temperature gradient from 50 to 220°C hold

for 2 min (ramp 25°C.min-1). The temperature of the detector was 250°C. MS scan speed

was 6.9 s-1. This method was used for all GC analyses. For that purpose, calibration curve

of 0 – 5 mM of TCP, DCP, ECH, CPD and GDL with internal standard hexan-1-ol was

prepared. Detection limits calculated using the software OriginPro v8 (OriginLab

Corporation, USA) were 3 μM, 5 μM, 6 μM, 186 μM and 22 μM for TCP, DCP, ECH, CPD and

GDL, respectively.

Chapter 1

- 58 -

RESULTS AND DISCUSSION

Multi-enzyme conversion of TCP using free enzymes.

We initially tested the ability of an in vitro assembled pathway to fully convert 5 mM

TCP to GLY in one-pot reaction at a time interval of 30 h. Purified DhaA, HheC, and EchA

were mixed in the mass ratio of 1:1:1 mg and incubated with TCP. Because the molecular

weights of DhaA, HheC, and EchA are similar (34, 29, and 35 kDa, respectively), the

proposed 1:1:1 ratio corresponded closely with the same molar ratio of enzymes. The pH

8.5 and temperature 37°C were selected to approach the reaction optima of all three

enzymes [209,216,237]. Using wild-type enzymes, TCP and the intermediates DCP and

GDL were degraded from 73% within 30 h of the reaction (Figure 2A and Table S3 in

Supplementary tables and figures). The percentage of degradation correlated well with

amount of GLY produced from TCP during the same time interval.

The time course of the reaction showed the major bottleneck of the pathway – slow

consumption of TCP by DhaA. Additionally, significant accumulations of two

intermediates, DCP and GDL, were observed. The accumulation of DCP was caused by the

high enantioselectivity (E≥100) of HheC [238]. Non-selective DhaA converts the prochiral

TCP into both enantiomers of DCP in an almost equimolar ratio. Since HheC prefers

(R)-DCP, (S)-DCP tends to accumulate. The specific activities of EchA with ECH and GDL

(29.5 µmol.min-1.mg-1 and 6.5 µmol.min-1.mg-1, respectively) in combination with the

previously reported kinetic parameters indicate that ECH is a much better substrate for

EchA than GDL [214]. Therefore, the substrate specificity of EchA is probably the main

cause of GDL accumulation during the multi-enzyme conversion of TCP.

To overcome the limitation of the first reaction step, the wild-type DhaA

(kcat = 0.04 s-1, kcat/Km = 40 s-1.M-1) was substitued with the mutant DhaA31 (kcat = 1.26 s-1,

kcat/Km = 1050 s-1.M-1), constructed in our laboratory using a computer-assisted design

[225]. The selectivity of DhaA31 with TCP remained unchanged. The benefit of engineered

DhaA31 in the multi-enzyme conversion of TCP was verified during the second

experiment with free enzymes. The time course of the reaction shows that TCP was

completely converted into its metabolites within the first 200 min of the reaction, and the

degradation of TCP, DCP, and GDL reached 99% of the theoretical maximum within 30 h of

the reaction (Figure 2B and Table S4). The conversion of TCP to GLY reached 95% of the

theoretical maximum. We conclude that complete in vitro biodegradation of TCP and its

biotransformation to the final product GLY is possible despite the suboptimal

stereochemistry and specificity in the pathway resulting in the accumulation of DCP and

GDL in the initial phase of reaction.

Chapter 1

- 59 -

Figure 2. Time courses of multi-enzyme conversions of 1,2,3-trichloropropane with free and immobilized enzymes. A) free purified enzymes DhaAwt, HheC, and EchA; B) free purified enzymes DhaA31, HheC, and EchA; C) immobilized purified enzymes DhaA31, HheC, and EchA; D) immobilized cell-free extracts containing DhaA31, HheC, and EchA. All reactions were performed using enzymes of mass ratio of 1:1:1 mg in 15 ml of reaction mixture. TCP, 1,2,3-trichloropropane; DCP, 2,3-dichloropropane-1-ol; ECH, epichlorohydrin; CPD, 3-chloropropane-1,2-diol; GDL, glycidol; GLY, glycerol. Each data point represents the mean value ± standard deviation from three independent experiments.

Immobilization of DhaA31, HheC and EchA.

Application of free enzymes in biotransformation processes is not practical due to

their complicated recycling, limited use in bioreactors, and low stability in harsh process

conditions such as elevated temperatures or the presence of organic co-solvents [162]. We therefore immobilized DhaA31, HheC, and EchA to avoid such limitations. Various

strategies for immobilization of individual biocatalysts are available [234]. Multi-enzyme

systems also benefit from immobilization, but development of joint immobilization

protocols for all employed catalysts is challenging. There is currently no protocol available

for immobilization of DhaA, HheC and EchA. Therefore, we tested enacapsulation of three

enzymes into PVA, which has proven its utility in immobilization of haloalkane

dehalogenase LinB, close relative of DhaA [240].

Chapter 1

- 60 -

The encapsulation of biocatalysts into PVA hydrogel is a widely used immobilization

technique [241,242]. Lens-shaped PVA hydrogel particles, known as LentiKats, are

promising matrices for biocatalysis due to the low cost, resistance to biodegradation and

good mechanical properties [234]. Their favorable geometry (thickness 200-400 μm)

allowes better mass transfer than spherical microbeads [243,244]. The biotechnology

based on LentiKats has already been used in large scale applications, e.g., synthesis of

beta-lactam antibiotics or denitrification in wastewater treatment. The size of LentiKats

allows easy separation from the reaction mixture and is suitable for application in packed

bed reactors, which can suffer from large pressure drops over the column when packed

with particles of inadequate size [245]. LentiKats provide a hydrophilic environment with

the pores sufficiently small to entrap whole cells, cross-linked enzymes, cross-linked

enzyme aggregates (CLEAs), or free enzymes of molecular weights higher than 80 kDa

[240,243,246].

In contrast to the HheC tetramer with a molecular weight of 116 kDa, smaller

monomeric molecules of DhaA31 and EchA are not suitable for direct encapsulation in

LentiKats. Therefore, aggregation and cross-linking of DhaA31 and EchA was carried out

to produce CLEA particles. Immobilization in CLEAs is straightforward technique which

has been used for many enzymes including hydrolases [99,233,247]. The protocols for

preparation of CLEAs from LinB and for miscroscopic monitoring of immobilized

biocatalyst were previsouly established in our laboratory [240,248]. Here we expand

these protocols also for immobilization of DhaA31 and EchA. To enable detailed

characterization of assembled pathway, immobilization was initially performed separately

for each of the three purified enzymes. CLEAs of DhaA31 and EchA were prepared by

mixing an enzyme solution with lyophilized bovine serum albumin, serving as a proteic

feeder [249]. The protein mixture was precipitated using ammonium sulfate, which is

widely used for preparing CLEAs due to its low cost and easy treatment [250]. The

suspension of aggregated enzymes was cross-linked using DPA, which had been shown to

have a less detrimental effect on enzymatic activity than the widely used glutaraldehyde

[251]. CLEAs and free HheC were labeled by fluorescent dyes to allow microscopic

monitoring of biocatalysts during the immobilization procedure (Figure 3). Free HheC and

wet CLEAs of DhaA31 and EchA were separately encapsulated in LentiKats in mass ratio

of 1:4 (enzyme or CLEAs: PVA gel).

Chapter 1

- 61 -

Figure 3. Confocal microscopy of immobilized enzymes from the synthetic pathway for biodegradation of 1,2,3-trichloropropane. A) Cross-linked enzyme aggregates (CLEAs) of haloalkane dehalogenase DhaA31 labeled with fluorescein 5(6)-isothiocyanate; B) free haloalcohol dehalogenase HheC labeled with X-rhodamine-5(6)-isothiocyanate; C) CLEAs of epoxide hydrolase EchA labeled with Pacific Blue succinimidyl ester; D) section through combi-LentiKat containing CLEAs of DhaA31, EchA, and free HheC. Visible fissures are caused by the fragility of PVA after freezing and appear after treatment with Leica Cryocut 1800 Cryostat (Leica Microsystems, Germany). Characterization of immobilized enzymes.

Immobilized enzymes were characterized in terms of their activity, storage and

operational stability, and distribution in PVA matrix. Immobilization resulted in the

decrease of specific activities of all three enzymes with their five corresponding substrates

(Figure S4). The catalytic performance of CLEAs of DhaA31 was 80% of the initial activity

of the soluble enzyme. CLEAs of EchA showed reduced activity with both GDL (50%) and

ECH (37%). The encapsulation of free HheC and CLEAs of DhaA31 and EchA into PVA

hydrogel led to activity retention of approximately 73/69% with DCP/CPD, 54% with TCP,

and 17/36% with ECH/GDL, respectively. Observed activitiy retention for immobilized

DhaA and HheC are comparable with studies describing formation of CLEAs or LentiKats

using other hydrolases, like penicillin acylase, acetyl xylan esterase or naringinase

[243,246,252]. Some epoxide hydrolases have been immobilized with high activity

retention using alternative methods, e.g., interaction with carrier through His-tag or

covalent binding [253,254]. We verified these methods with EchA using Ni-NTA Agarose

and activated CH Sepharose 4B as carriers, achieving 93% and 68% activity retention with

ECH. Nevertheless, high activity retention in these matrices is compromised by higher

Chapter 1

- 62 -

price and lower mechanical stability, which makes them less suitable for full-scale

processes. The specific activity of soluble EchA with ECH (29.5 µmol.min-1.mg-1) is one

order of magnitude higher than the specific activity of DhaA31 with TCP (1.1 µmol.min-

1.mg-1) and HheC with DCP (1.6 µmol.min-1.mg-1) and CPD (3.0 µmol.min-1.mg-1). Despite

the reduction of the catalytic performance of EchA after immobilization, the pattern of

specific activities of three enzymes remained unchanged (Figure S1 in Supplementary

tables and figures). We therefore concluded that even 80% loss of the activity of EchA

towards ECH after immobilization in CLEAs and LentiKats would not result in the

accumulation of intermediates in the TCP pathway and the analogous immobilization

strategy was maintained also for this enzyme.

No enzymatic activities were detected in the supernatants obtained during the

preparation of CLEAs and LentiKats, besides low activity of HheC corresponding to

approximately 10% loss of enzyme during immobilization. Therefore, we expect that the

decreased activities of DhaA31, HheC and EchA are predominantly due to the partial

deactivation caused by aggregation, cross-linking, or encapsulation in PVA hydrogel,

rather than due to the leaching of enzymes during immobilization process.

Figure 4. Relative activities of free enzymes and enzymes immobilized in cross-linked enzyme aggregates (CLEAs) and LentiKats (LKs) with their corresponding substrates from the 1,2,3-trichloropropane pathway: TCP, 1,2,3-trichloropropane; DCP, 2,3-dichloropropane-1-ol; CPD, 3-chloropropane-1,2-diol; ECH, epichlorohydrin; GDL, glycidol. Colours indicate different enzymes: haloalkane dehalogenase DhaA31 (in green), haloalcohol dehalogenase HheC (red) and epoxide hydrolase EchA (blue). Error bars represent the standard deviation from three independent measurements. Specific activities of free enzyme with corresponding substrate used as the reference for 100% relative activity were: 1.09±0.02 μmol.min

-1.mg

-1 (DhaA31 with

TCP), 1.61±0.10 μmol.min-1

.mg-1

(HheC with DCP), 3.04±0.09 μmol.min-1

.mg-1

(HheC with CPD), 29.45±1.73 μmol.min

-1.mg

-1 (EchA with ECH), 6.46±0.64 μmol.min

-1.mg

-1 (EchA with GDL).

Chapter 1

- 63 -

The storage stability of LentiKats of DhaA31 and EchA in a phosphate buffer without

additives at 4°C was assayed during a three-month period and compared with the storage

stability of free enzymes (Figure S2). After this period, changes in the activity of

immobilized and free enzymes correlated for both DhaA31 and EchA. Free and

immobilized DhaA31 retained 93% and 86% of its initial activity with TCP, respectively.

EchA showed 121% and 122% of its initial activity in free and immobilized form,

respectively. The increase was statistically significant for free EchA (t-test, p<0.05).

Similar moderate increases in specific activity during storage were reported also for some

other enzymes [240,243].

HheC showed significantly lower storage stability than DhaA31 and EchA. Free HheC

stored in a phosphate buffer at 4°C lost all its activity with DCP within two months (data

not shown). Tang and co-workers previously suggested that inactivation of HheC is caused

by the monomerization of the enzyme and intramolecular disulfide bond formation under

oxidizing conditions [219]. Confirming this suggestion, the stability of HheC was

significantly improved by its storage in a buffer containing a reducing agent (Figure S2A).

Free and immobilized HheC retained 63% and 49% of its initial activity, respectively, after

two months of storage in the presence of 2 mM dithiothreitol. The lower activity of HheC

immobilized in LentiKats is due to partial leaching (Figure S2B). The storage stability of

free enzymes was tested also at room temperature (22±2°C), which was later applied

during the continuous removal of TCP in a packed bed reactor. In contrast to DhaA31 and

EchA, HheC showed better storage stability at room temperature than at 4°C (Figures S3).

The enzyme retained 70% of its initial activity after 10 weeks of storage at room

temperature in phosphate buffer without any additive.

Multi-enzyme conversion of TCP using immobilized enzymes.

Numerous studies describe the immobilization of individual enzymes or whole cells,

but reports on the effects of immobilization on the performance of complete biochemical

pathways are scarce [100]. For synthetic biodegradation pathways employing engineered

enzymes such reports are to the best of our knowledge missing. We applied TCP pathway

immobilized in LentiKats in a one-pot multi-enzyme reaction to study how the modified

activities of individual immobilized enzymes affected the time course of TCP

biodegradation. Immobilized enzymes were mixed in a ratio 1:1:1 mg and incubated with

5 mM TCP under the same conditions as had been used for soluble enzymes (Figure 2C

and Table S5). Despite the fact that activity of individual enzymes after immobilization

was reduced by 27–83%, the efficiency of the whole pathway was lower only about 11%

when compared to the conversion with free enzymes. The conversion of TCP to GLY after

30 h reached 84% and conversion of all intermediates 89% of the theoretical maximum.

We also studied the reusability of the immobilized TCP pathway in a batch system (Figure

S4). The immobilized pathway retained 77% of its initial efficiency of TCP conversion to

GLY after 10 successive cycles of batch operation.

Chapter 1

- 64 -

The application of purified enzymes immobilized individually provides better control

over the reaction by enabling the tuning of enzyme stoichiometry and by compensating

for the activity loss of individual enzymes. However, the purification and separate

treatment of all three enzymes increases the cost of the biocatalyst. We therefore

demonstrated that cell-free extracts obtained from E. coli BL21(DE3) cells expressing

DhaA31, HheC, or EchA can be utilized. For the preparation of the LentiKats, cell-free

extracts were mixed to provide a ratio of enzymes corresponding to 1:1:1. The

combi-LentiKats containing approximately 1 mg of each of the three enzymes were used

for the biodegradation of 5 mM TCP. The degradation of TCP and the intermediates DCP

and GDL reached 95% of the theoretical maximum within 30 h of reaction (Figure 2D,

Table S6). The overall reaction profile was very similar to the profile obtained with

purified immobilized enzymes (Figure 2C). The conversion of TCP to GLY reached 89% of

the theoretical maximum. Thus, the co-immobilization of enzymes and their improved

proximity resulted in conversion comparable by efficiency with free enzymes. The

proximity and homogeneous distribution of DhaA31, HheC, and EchA in combi-LentiKats

was verified by confocal microscopy (Figure 3D).

Multi-enzyme conversion of TCP using packed bed reactor.

The performance of immobilized pathway was also tested under continuous

operation using a packed bed reactor composed of two columns (Figure 5). Column 1 was

packed with LentiKats of DhaA31, column 2 with a mixture of LentiKats of HheC and EchA.

The total mass ratio of enzymes 1:1:1 corresponded to the previous batch experiments.

DhaA31 was separated from HheC and EchA in order to prevent the inhibition of the last

step in the pathway, the conversion of GLD to GLY by TCP (Ki = 0.21 mM). EchA was

combined with HheC in column 2 in order to prevent a reverse reaction and push the

reaction equilibrium during the conversion of DCP and CPD by HheC towards the product

[238]. The separation of enzymes into two columns also enabled better control over the

individual reaction steps and evaluation of their efficiency.

The packed bed reactor was operated at room temperature 22±2°C and pH 8.2 for 10

weeks under the conditions described in Table S1 in Supplementary tables and figures.

During that period, the reactor with an effective volume of 0.15 L processed 11 L of

contaminated water. Experimental concentrations of TCP for evaluating column 1

efficiency were determined in the input of the column 1 due to the leakage of TCP from the

pumping system before reaching the column (Table S7). The average leakage caused by

hydrophobic nature of TCP (log P = 2.24) and its tendency to penetrate through rubber

tubing of peristaltic pump was approximately 29%. The experimental concentrations of

TCP in the column 1 input for each of ten weeks of operation were 2.25, 3.03, 5.91, 6.54,

7.97, 6.31, 7.94, 6.65, 6.80, and 7.02 mM. The levels of residual TCP and produced DCP

were determined in the effluent of column 1. The levels of residual DCP and produced GDL

and GLY were determined in the effluent of column 2. These values were used to evaluate

the efficiency of both columns and the efficiency of GLY production from TCP (Figure 6).

Chapter 1

- 65 -

No leakage of intermediates was observed. Neither ECH nor CPD was detected in the

system.

The efficiencies of column 1 and 2 were higher than 95% during the first eight and

four weeks of operation, respectively (Figure 6 and Table S7). The efficiency of the overall

conversion of TCP to GLY decreased from values above 90% observed during the first four

weeks to values lower than 60% at the end of two months of operation. The decreased

reactor efficiency accompanied by an accumulation of unreacted TCP, DCP and GDL in

effluent vessels could be attributed to: (i) slow enzyme inactivation caused by oxidation of

HheC and thermal unfolding of DhaA31 and EchA and (ii) slow leaching of enzymes from

the reactor, which resulted in about 10% loss of HheC activity and about 5% loss of

DhaA31 and EchA activities during the operation (Figure 6B). In total, 65.5 mmol (cca.

9.7 g) of the 67.7 mmol of TCP that entered the packed bed reactor during 10 weeks of

operation were converted to intermediates (efficiency 97%), and 52.6 mmol of GLY were

produced (efficiency 78%).

Figure 5. Schematic overview and photograph of the packed bed reactor used for removing 1,2,3-trichloropropane from the water under continuous operation. The system consisted of two glass columns (C1 and C2) packed with immobilized biocatalysts, two peristaltic pumps (P1 and P2), and three glass vessels connected with polytetrafluoroethylene tubing. A feed vessel (FV) contained 1 L of 0.1 M Tris-SO4 with 5 mM (week 1) or 10 mM (week 2-10) of TCP. Effluent vessel 1 (EV1) was used to collect samples downstream of column 1 and as a feeding bottle for column 2; effluent vessel 2 (EV2) was used to collect final samples downstream of column 2. The site for determining the TCP concentration in the input of column 1 is indicated with a red arrow. The part of tubing with connector was unscrewed from the column and sample of water with TCP was withdrawn from the tubing downstream the pump 1.

Chapter 1

- 66 -

Figure 6. A) Efficiency of column 1 (filled) and column 2 (cross-hatched) of the packed bed reactor during the continuous biodegradation of 1,2,3-trichloropropane (TCP). Black diamonds with a solid line show the overall efficiency of TCP conversion to glycerol by immobilized biocatalysts in the two-step packed bed reactor. Efficiency was calculated from the concentration of TCP measured in the inlet of column 1 and the concentration of glycerol measured in effluent vessel 2. B) Residual concentrations of the metabolites from the synthetic pathway for biodegradation of TCP detected in effluent vessel 1 (TCP) and effluent vessel 2 (2,3-dichloropropane-1-ol, DCP; glycidol, GDL; glycerol, GLY) during 10 weeks of operating the packed bed reactor.

Chapter 1

- 67 -

CONCLUSIONS

Recent studies on bacterial utilization of 1,3-dichloroprop-1-ene and chlorinated

ethenes suggest that the biodegradation of chlorinated aliphatic pollutants is associated

with the accumulation of reactive intermediates and strong oxidative stress, representing

a significant barrier for the evolution of corresponding aerobic metabolic pathways in vivo

[170,202]. The absence of natural microorganisms carrying aerobic pathways for the

biodegradation of TCP and experiences gained during attempts to engineer

microorganisms growing on this toxic compound seem to support this view

[213,228,232]. Engineering synthetic pathway for converting toxic TCP to the harmless

GLY in vitro is an alternative approach, which does not suffer from the limitations of

engineered bacterium. It is now widely accepted that in vitro multi-enzyme systems

represent an emerging field of biocatalysis due to their simplicity, predictability, and

controllability [67,93]. This study shows that an in vitro assembly of natural or synthetic

enzymatic pathways can be a promising concept for the biodegradation of environmental

pollutants and can provide promising results especially when combined with protein

engineering techniques.

Developed biotechnology requires further validation before it can be scaled-up and

used in real conditions. The performance of the immobilized biocatalyst should be tested

in real water samples contaminated with TCP. Pretreatment of the contaminated water by

adjusting high salinity or eliminating possible enzyme-inhibiting constituents might be

necessary. Conversion of TCP to GLY was demonstrated to be efficient at 22±2°C and is

also possible in broad pH range 7-10 (Figure S5). Yet, buffering of the contaminated water

to pH close to the reaction optima of enzymes can be beneficial to achieve maximal

efficiency of the process. The favorable features of the presented biotechnology are: (i)

degradation of TCP using immobilized cell-free extracts instead of purified enzymes is

possible, (ii) material used for immobilization of enzymes is affordable, safe and

non-biodegradable, (iii) protocol for disposal of used LentiKats by burning is well

established, and (iv) the amount of enzymes necessary to degrade almost 10 g of TCP

during the operation of reactor can be obtained from less than 2 l of cell culture yet

without optimized cultivation conditions. An immobilized synthetic pathway works at TCP

concentrations that are close to the water solubility limit of TCP (10 mM), which is one

order of magnitude higher than concentrations tolerated by engineered microorganisms.

At the same time, the system has the potential to cope with significantly lower

concentrations of TCP. Kinetic parameters of DhaA31 for TCP (kcat = 1.26 s-1,

kcat/Km = 1050 s-1.M-1) are of the same order of magnitude as those of the haloalkane

dehalogenase DhlA for 1,2-dichloroethane (kcat = 3.3 s-1, kcat/Km = 6200 s-1.M-1), which has

already proven its utility in a groundwater purification plant treating micro- to nanomolar

concentrations of the pollutant [125].

The remaining bottlenecks of the pathway are: (i) lower activity of HheC with

non-preferred (S)-DCP, (ii) accumulation of GDL due to the substrate preference of EchA,

and (iii) gradual inactivation of HheC. These important, but not critical, limitations can be

Chapter 1

- 68 -

overcome by another round of protein engineering or modification of the immobilization

protocol. The flux of intermediates through the immobilized pathway can be further tuned

using kinetic modeling and optimization of enzyme stoichimetry, which is an objective of

the follow-up research described in Chapter 2 of this Thesis.

Chapter 1

- 69 -

SUPPLEMENTARY TABLES AND FIGURES Table S1. Nucleotide sequences of genes used in this study.[a] Synthetic gene dhaA with 6xHis tag, codon optimized for expression in E. coli

CATATGAGCGAAATTGGCACCGGTTTTCCGTTTGATCCGCATTATGTTGAAGTTCTGGGTGAACGTATGCATTATGTGGATGTTGGTCCGCGTGATGGTACACCGGTTCTGTTTCTGCATGGTAATCCGACCAGCAGCTATCTGTGGCGTAACATTATTCCGCATGTTGCACCGAGCCATCGTTGTATTGCACCGGATCTGATTGGTATGGGTAAAAGCGATAAACCTGATCTGGATTATTTCTTCGATGATCATGTGCGTTATCTGGATGCATTTATTGAAGCACTGGGTCTGGAAGAAGTTGTGCTGGTTATTCATGATTGGGGTAGCGCACTGGGTTTTCATTGGGCAAAACGTAATCCGGAACGTGTTAAAGGTATTGCCTGCATGGAATTTATTCGTCCGATTCCGACCTGGGATGAATGGCCTGAATTTGCACGTGAAACCTTTCAGGCATTTCGTACCGCAGATGTGGGTCGTGAACTGATTATTGATCAGAACGCATTTATCGAAGGTGCACTGCCGAAATGTGTTGTTCGTCCGCTGACCGAAGTTGAAATGGATCATTATCGTGAACCGTTTCTGAAACCGGTTGATCGCGAACCGCTGTGGCGTTTTCCGAATGAACTGCCGATTGCCGGTGAACCTGCAAATATTGTTGCACTGGTTGAAGCCTATATGAATTGGCTGCATCAGAGTCCGGTTCCGAAACTGCTGTTTTGGGGCACACCGGGTGTTCTGATTCCGCCTGCAGAAGCAGCACGTCTGGCAGAAAGCCTGCCGAATTGTAAAACCGTTGATATTGGTCCGGGTCTGCATTATCTGCAAGAAGATAATCCGGACCTGATCGGTAGTGAAATTGCACGTTGGCTGCCTGCACTGCATCATCATCACCATCATTAAGGATCC

Synthetic gene dhaA31 with 6xHis tag, codon optimized for expression in E. coli

CATATGTCCGAAATTGGCACCGGCTTCCCGTTTGATCCGCACTATGTTGAAGTTCTGGGCGAACGTATGCACTATGTTGATGTTGGTCCGCGTGATGGCACCCCGGTTCTGTTTCTGCACGGTAACCCGACGAGCTCTTATCTGTGGCGTAATATTATCCCGCATGTCGCCCCGAGTCACCGCTGCATTGCACCGGATCTGATCGGCATGGGTAAATCCGACAAACCGGATCTGGACTATTTCTTTGATGACCATGTCCGCTACCTGGATGCATTTATTGAAGCTCTGGGCCTGGAAGAAGTGGTTCTGGTGATCCATGACTGGGGCTCTGCGCTGGGTTTCCACTGGGCCAAACGTAATCCGGAACGCGTGAAAGGTATTGCGTGTATGGAATTTATCCGTCCGTTCCCGACCTGGGATGAATGGCCGGAATTTGCCCGCGAAACCTTTCAGGCGTTCCGTACGGCCGATGTTGGCCGCGAACTGATTATCGACCAAAACGCATTCATTGAAGGTGCTCTGCCGAAATATGTCGTGCGTCCGCTGACGGAAGTGGAAATGGATCATTACCGCGAACCGTTTCTGAAACCGGTTGACCGTGAACCGCTGTGGCGCTTCCCGAACGAACTGCCGATTGCAGGCGAACCGGCTAATATCGTTGCGCTGGTCGAAGCCTACATGAACTGGCTGCACCAGTCACCGGTGCCGAAACTGCTGTTTTGGGGCACCCCGGGTTTCATTATCCCGCCGGCAGAAGCAGCACGTCTGGCTGAATCGCTGCCGAATTGCAAAACGGTTGATATCGGCCCGGGTCTGCATTTTCTGCAAGAAGATAACCCGGACCTGATTGGCTCTGAAATTGCCCGCTGGCTGCCGGCTCTGCACCACCACCACCACCACTAAGGATCC

Synthetic gene dhaA90R with 6xHis tag, codon optimized for expression in E. coli

CATATGAGCGAAATTGGCACCGGTTTTCCGTTTGATCCGCATTATGTTGAAGTTCTGGGTGAACGTATGCATTATGTGGATGTTGGTCCGCGTGATGGTACACCGGTTCTGTTTCTGCATGGTAATCCGACCAGCAGCTATCTGTGGCGTAACATTATTCCGCATGTTGCACCGAGCCATCGTTGTATTGCACCGGATCTGATTGGTATGGGTAAAAGCGATAAACCTGATCTGGATTATTTCTTCGATGATCATGTGCGTTATCTGGATGCATTTATTGAAGCACTGGGTCTGGAAGAAGTTGTGCTGGTTATTCATGATTGGGGTAGCGCACTGGGTTTTCATTGGGCAAAACGTAATCCGGAACGTGTTAAAGGTATTGCCTGCATGGAATTTATTCGTCCGCTGACCACCTGGGATGAATGGCCTGAATTTGCACGTGAAACCTTTCAGGCATTTCGTACCGCAGATGTGGGTCGTGAACTGATTATTGATCAGAACATGTGGATTGAAGGTCTGATTCCGGCAGGCGTGATTCGCCCTCTGACCGAAGTTGAAATGGATCATTATCGTGAACCGTTTCTGAAACCGGTTGATCGCGAACCGCTGTGGCGTTTTCCGAATGAACTGCCGATTGCCGGTGAACCTGCAAATATTGTTGCACTGGTTGAAGCCTATATGAATTGGCTGCATCAGAGTCCGGTTCCGAAACTGCTGTTTTGGGGTAATCCGGGTTATCTGATTACACCGGCAGAAGCAGCACGTCTGGCAGAAAGCCTGCCGAATTGTAAAACCGTTGATATTGGTCCGGGTCTGCATTTTCTGCAAGAAGATAATCCGGACCTGATCGGTAGTGAAATTGCACGTTGGCTGCCTGCACTGCATCATCATCACCATCATTAAGGATCC

Wild-type gene hheC[b]

CATATGTCAACCGCAATTGTAACAAACGTTAAGCATTTTGGGGGAATGGGGTCTGCACTTCGTCTCTCGGAAGCAGGACATACAGTGGCTTGCCACGATGAAAGCTTCAAACAAAAGGACGAACTTGAAGCCTTTGCCGAAACCTATCCACAACTCAAACCAATGTCGGAACAAGAACCAGCGGAACTCATCGAGGCAGTTACCTCCGCTTATGGTCAAGTTGATGTACTTGTGAGCAACGACATATTCGCACCAGAGTTCCAACCCATAGATAAATACGCTGTAGAGGACTATCGCGGTGCGGTCGAGGCGCTACAAATTAGACCATTTGCACTGGTCAACGCCGTTGCAAGTCAAATGAAGAAGCGCAAAAGCGGACATATTATCTTTATTACCTCTGCAACGCCCTTCGGGCCTTGGAAGGAACTTTCTACCTACACGTCAGCCCGAGCAGGTGCATGCACCTTGGCAAATGCCCTTTCGAAGGAACTCGGTGAATACAACATTCCGGTGTTCGCAATAGGACCCAATTATCTTCACAGTGAAGATAGTCCCTACTTCTACCCCACAGAACCGTGGAAAACGAATCCAGAACACGTTGCCCATGTCAAAAAAGTCACT

Chapter 1

- 70 -

GCGCTCCAGCGGTTAGGTACACAGAAAGAATTGGGAGAACTCGTCGCGTTTCTCGCGTCTGGTAGTTGTGACTATCTGACCGGCCAGGTGTTCTGGTTGGCCGGCGGATTCCCAATGATCGAGCGTTGGCCTGGTATGCCCGAGTAGGGATCC

Synthetic gene echA with 6xHis tag, codon optimized for expression in E. coli

CCATGGCAATCCGTCGTCCTGAAGATTTCAAACACTATGAGGTCCAGCTGCCTGATGTTAAAATTCATTATGTCCGTGAGGGAGCCGGTCCGACACTGCTGCTGCTGCATGGTTGGCCTGGTTTTTGGTGGGAATGGTCGAAAGTCATCGGTCCACTGGCCGAGCACTATGATGTTATTGTGCCTGATCTGCGCGGTTTTGGTGATAGCGAGAAACCTGACCTGAACGATCTGAGCAAATATAGCCTGGATAAAGCCGCTGATGATCAAGCTGCCCTGCTGGATGCTCTGGGTATCGAAAAAGCCTATGTCGTGGGCCATGATTTTGCCGCTATTGTGCTGCACAAATTCATCCGTAAATATTCCGACCGTGTCATCAAAGCAGCCATCTTTGACCCGATTCAACCAGACTTTGGGCCGGTGTATTTTGGACTGGGCCATGTTCATGAAAGCTGGTATAGCCAGTTTCACCAACTGGACATGGCTGTTGAAGTGGTAGGCTCTTCACGTGAAGTGTGTAAAAAATATTTCAAACATTTCTTCGATCACTGGTCCTATCGTGACGAACTGCTGACAGAGGAGGAACTGGAAGTCCACGTGGACAATTGTATGAAACCGGATAATATCCACGGCGGGTTCAACTATTATCGTGCCAACATTCGTCCTGATGCTGCTCTGTGGACAGACCTGGATCATACCATGAGTGACCTGCCGGTTACTATGATTTGGGGTCTGGGCGACACATGTGTTCCTTATGCCCCACTGATTGAGTTTGTTCCGAAATATTATAGCAACTATACGATGGAAACCATCGAGGATTGTGGCCATTTTCTGATGGTGGAGAAACCGGAAATCGCCATCGACCGTATTAAAACGGCCTTCCGTCACCACCACCACCACCATTAAAAGCTT

[a] Restriction sites used for cloning are underlined and the 6xHis tag is shown in bold. [b] GenBank accession number AF397296.1.

Table S2. Operational conditions of the packed bed reactor. week of operation 1 1

a 2 3 4 5 6 7 8 9 10

operational period (h) 33 66 166 166 166 166 166 166 166 166 166

volumetric flow column 1 (ml.min

-1)

0.50 0.50 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25

volumetric flow column 2 (ml.min

-1)

0.50 0.25 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10

[a] Two cycles were conducted during the first week of operation to optimize running conditions.

Table S3. Concentrations of metabolites (mM) from the time course of the multi-enzyme conversion of 1,2,3-trichloropropane catalyzed by free purified DhaAwt, HheC and EchA. time (min)

TCP std. DCP std. ECH CPD GDL std. GLY std. sum

0 4.68 0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 4.68

50 3.94 0.03 0.18 0.05 0.00 0.00 0.07 0.01 0.03 0.00 4.22

100 3.51 0.05 0.56 0.11 0.00 0.00 0.21 0.02 0.10 0.02 4.38

200 2.86 0.04 0.91 0.16 0.00 0.00 0.48 0.05 0.17 0.02 4.42

400 1.94 0.07 1.17 0.11 0.00 0.00 0.72 0.03 0.56 0.09 4.39

750 0.99 0.11 1.29 0.13 0.00 0.00 0.62 0.02 1.49 0.20 4.39

1260 0.34 0.10 1.12 0.16 0.00 0.00 0.43 0.01 2.60 0.21 4.49

1800 0.11 0.06 0.89 0.20 0.00 0.00 0.26 0.03 3.36 0.13 4.62

Abbreviations common for Tables S3-7: TCP, 1,2,3-trichloropropane; DCP, 2,3-dichloropropane-1-ol; ECH, epichlorohydrin; CPD, 3-chloropropane-1,2-diol; GDL, glycidol; GLY, glycerol; std., standard deviation. Each value represents the mean value ± standard deviation from three independent experiments.

Chapter 1

- 71 -

Table S4. Concentrations of metabolites (mM) from the time course of the multi-enzyme conversion of 1,2,3-trichloropropane catalyzed by free purified DhaA31, HheC and EchA. time (min)


0 4.57 0.07 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 4.57

50 1.11 0.00 1.48 0.21 0.00 0.00 1.13 0.08 0.33 0.01 4.05

100 0.24 0.00 1.88 0.04 0.00 0.00 1.57 0.08 0.90 0.03 4.59

200 0.01 0.00 1.63 0.08 0.00 0.00 1.02 0.08 1.94 0.03 4.60

400 0.00 0.00 0.86 0.16 0.00 0.00 0.37 0.08 2.97 0.05 4.20

750 0.00 0.00 0.37 0.06 0.00 0.00 0.16 0.08 3.86 0.13 4.39

1260 0.00 0.00 0.11 0.01 0.00 0.00 0.03 0.01 4.20 0.02 4.34

1800 0.00 0.00 0.04 0.01 0.00 0.00 0.00 0.00 4.32 0.06 4.36

Table S5. Concentrations of metabolites (mM) from the time course of the multi-enzyme conversion of 1,2,3-trichloropropane catalyzed by immobilized purified DhaA31, HheC and EchA. time (min)


0 4.71 0.27 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 4.71

50 1.86 0.19 1.14 0.03 0.00 0.00 0.33 0.05 0.11 0.04 3.44

100 0.89 0.06 1.54 0.18 0.00 0.00 0.88 0.05 0.28 0.02 3.59

200 0.13 0.01 1.54 0.16 0.00 0.00 1.46 0.15 1.06 0.14 4.19

400 0.00 0.00 1.30 0.12 0.00 0.00 1.04 0.15 2.05 0.06 4.39

700 0.00 0.00 1.02 0.16 0.00 0.00 0.53 0.04 2.77 0.16 4.32

1320 0.00 0.00 0.52 0.07 0.00 0.00 0.27 0.04 3.48 0.19 4.27

1800 0.00 0.00 0.34 0.07 0.00 0.00 0.16 0.03 3.94 0.07 4.44

Table S6. Concentrations of metabolites (mM) from the time course of the multi-enzyme conversion of 1,2,3-trichloropropane catalyzed by immobilized cell-free extracts with DhaA31, HheC and EchA. time (min)


0 4.86 0.10 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 4.86

50 2.19 0.06 0.92 0.04 0.00 0.00 0.32 0.32 0.15 0.02 3.58

100 1.04 0.14 1.28 0.06 0.00 0.00 0.90 0.83 0.59 0.13 3.81

200 0.25 0.08 1.34 0.08 0.00 0.00 0.83 0.96 1.49 0.37 3.91

400 0.00 0.00 1.21 0.06 0.00 0.00 0.37 0.42 2.48 0.35 4.06

700 0.00 0.00 0.88 0.06 0.00 0.00 0.12 0.12 3.06 0.27 4.06

1320 0.00 0.00 0.47 0.02 0.00 0.00 0.00 0.00 4.14 0.24 4.61

1800 0.00 0.00 0.25 0.02 0.00 0.00 0.00 0.00 4.32 0.14 4.57

Chapter 1

- 72 -

Table S7. Concentrations of metabolites in a packed bed reactor and the calculated efficiencies of columns 1 and 2. weeks of operation

1 1 2 3 4 5 6 7 8 9 10

TCP in IC1 (mM)

2.25 3.03 5.91 6.54 7.33 7.97 6.31 7.94 6.65 6.80 7.02

TCP in EV1 (mM)

0.00 0.00 0.00 0.00 0.12 0.15 0.11 0.12 0.36 0.61 0.74

DCP in EV1 (mM)

2.12 2.89 5.55 6.67 6.64 7.45 5.78 8.03 6.52 6.14 7.52

DCP in EV2 (mM)

0.15 0.11 0.34 0.35 0.34 0.64 0.74 1.33 1.37 1.51 2.50

GDL in EV2 (mM)

0.00 0.00 0.00 0.00 0.00 0.17 0.26 0.47 0.52 0.64 1.09

GLY in EV2 (mM)

1.82 2.81 5.69 6.87 6.97 6.73 5.60 5.19 4.45 3.93 2.56

efficiency of C1 (%)

a

100 100 100 100 98 98 98 98 95 91 89

efficiency of C2 (%)

b

93 96 94 95 95 89 83 78 71 65 52

efficiency of GLY production (%)

c

81 93 96 100 95 84 89 65 67 58 36

Each number in mM represents a mean value from at least two technical replicates. Abbreviations: IC1, input of column 1; EV1, effluent vessel 1; EV2, effluent vessel 2; C1, column 1; C2, column 2.

[a] Calculated as:

[b] Calculated as:

[c] Calculated as:

Chapter 1

- 73 -

Figure S1. Specific activities of free enzymes and enzymes immobilized in LentiKats (LKs) with their corresponding substrates from the 1,2,3-trichloropropane pathway: TCP, 1,2,3-trichloropropane; DCP, 2,3-dichloropropane-1-ol; CPD, 3-chloropropane-1,2-diol; ECH, epichlorohydrin; GDL, glycidol. Error bars represent the standard deviation calculated from three independent measurements.

Chapter 1

- 74 -

Figure S2. A) Storage stability of free and immobilized haloalkane dehalogenase DhaA31, haloalcohol dehalogenase HheC, and epoxide hydrolase EchA at 4°C measured during three-month interval. Activities measured at the beginning of storage, after 2 and 3 months of storage are diversified by fading colors. Free enzymes and LentiKats (LKs) were stored in 50 mM phosphate buffer of pH 7.5 without additives (DhaA31 and EchA), or with 2 mM dithiothreitol (HheC). Residual activities of DhaA31, HheC and EchA were measured with their corresponding substrates from the TCP pathway (1,2,3-trichloropropane, 2,3-dichloropropane-1-ol, and epichlorohydrin, respectively) at 37°C. A relative activity of 100% corresponds to the specific activity of: 1.13±0.03 µmol.min

-1.mg

-1 (free DhaA31); 0.58±0.03 µmol.min

-1.mg

-1 (LKs of

DhaA31); 1.61±0.20 µmol.min-1

.mg-1

(free HheC); 1.19±0.13 µmol.min-1

.mg-1

(LKs of HheC); 23.37±0.39 µmol.min

-1.mg

-1 (free EchA); 4.96±0.87 µmol.min

-1.mg

-1 (LKs of EchA). Error bars

represent the standard deviation calculated from three independent measurements. B) SDS-PAGE analysis of the buffer from storage of LKs of HheC, lane 1, standard sample of 1 µg of purified HheC; lane 2, buffer with fresh LKs; lane 3, 1 week of storage; lane 4, 2 weeks of storage; lane 5, 1 month of storage; lane 6, 2 months of storage; lane 7, 3 months of storage. The amount of leached enzyme after 3 months of storage was approximately 1.1 mg, which corresponds to 22% of the original amount of HheC immobilized in the LentiKats.

Chapter 1

- 75 -

Figure S3. Storage stability of free haloalkane dehalogenase DhaA31 (green columns), haloalcohol dehalogenase HheC (red columns), and epoxide hydrolase EchA (blue columns) at room temperature (22±2°C). Residual activities of DhaA31, HheC and EchA were measured with their corresponding substrates from the TCP pathway (1,2,3-trichloropropane, 2,3-dichloropropane-1-ol, and epichlorohydrin, respectively) at room temperature (22±2°C). Error bars represent the standard deviation from three independent measurements. Specific activities of fresh free enzyme with corresponding substrate used as the reference for 100% relative activity were: 0.44±0.04 μmol.min

-1.mg

-1 (DhaA31 with 1,2,3-trichloropropane),

0.29±0.02 μmol.min-1

.mg-1

(HheC with 2,3-dichloropropane-1-ol), and 14.11±0.07 μmol.min

-1.mg

-1 (EchA with epichlorohydrin).

Chapter 1

- 76 -

Figure S4. Recycling of the 1,2,3-trichloropropane pathway in LentiKats. The relative activity of the immobilized biocatalyst during the first cycle is 100% and corresponds to the production of 0.39 mM of glycerol from 1 mM of 1,2,3-trichloropropane during 100 minutes of reaction time in 50 mM Tris-SO4 buffer of pH 8.2 at 30°C. Error bars represent the standard deviation from three independent measurements.

Figure S5. Time courses of multi-enzyme conversions of 1,2,3-trichloropropane (TCP) with the pathway immobilized in LentiKats. A) reaction in 50 mM sodium phosphate buffer of pH 7.0; and B) 50 mM NaOH/glycine buffer of pH 10. All reactions were performed using enzymes with a mass ratio of 1:1:1 mg in 15 ml of reaction mixture at 20°C. DCP, 2,3-dichloropropane-1-ol; ECH, epichlorohydrin; CPD, 3-chloropropane-1,2-diol; GDL, glycidol; GLY, glycerol. Each data point represents the mean value ± standard deviation from three independent experiments.

- 77 -

- 78 -

- 79 -

CHAPTER 2

Maximizing the efficiency of in vitro

multi-enzyme process by stoichiometry

optimization

Chapter 2

- 80 -

INTRODUCTION

Multi-enzyme processes have great potential for the biosynthesis of fine and bulk

chemicals, bioremediation and biosensing [94,255,256]. Studies on two-enzyme systems

dominate the literature, but systems of three [95], four [257] and even twelve [258] or

thirteen [255] enzymes are known. Multi-enzyme systems are superior to single-enzyme

biocatalysis in that they can catalyze more complex synthetic schemes. In vitro

multi-enzyme networks enable simpler process control than analogous in vivo systems

and suffer less from reactant toxicity [29,259,260]. However, both system types often

exhibit reaction bottlenecks due to imbalanced enzyme properties.

Protein engineering is often used to improve enzymes’ catalytic properties and

stability [102,261]. Many engineered enzymes can be used in multi-enzyme processes but

methods for predicting their impact on productivity are lacking. Kinetic modelling is

essential for analyzing enzymatic reactions and can enable their rational optimization

[94,96,262,263]. However, there are only few accurate kinetic models of in vitro multi-

enzyme systems [95,96,255,257]. Available models rarely have experimental support and

their use in optimizing processes with engineered enzymes has not been adequately

explored.

The aim of this study was to develop a workflow for optimizing in vitro multi-enzyme

processes by using kinetic modelling to predict the effects of varying enzyme

stoichiometry and introducing available engineered enzymes. Our model system (Figure

1) was a synthetic metabolic pathway for the five-step biotransformation of toxic

industrial waste product 1,2,3-trichloropropane (TCP) into the commodity chemical

glycerol (GLY). This pathway has been previously assembled inside living cells

[211,213,232] and is based on haloalkane dehalogenase DhaA from Rhodococcus

rhodochroust NCIMB 13064 [209], haloalcohol dehalogenase HheC [216], and epoxide

hydrolase EchA from Agrobacterium radiobacter AD1 [237]. Three different haloalkane

dehalogenase variants were evaluated: (i) wild-type DhaA; previously constructed

mutants (ii) DhaA31with improved activity [225]; and (iii) DhaA90R with increased

enantioselectivity [226]. Kinetic models were built for pathways using each DhaA variant

and their enzyme stoichiometry was optimized under defined constraints.

Chapter 2

- 81 -


Reagents. TCP, (R,S)-2,3-dichloropropane-1-ol (DCP), (R,S)-epichlorohydrin (ECH), (R,S)-

3-chloropropane-1,2-diol (CPD), (R,S)-glycidol (GDL), glycerol (GLY), acetone, ethyl

acetate, and hexanol were purchased from Sigma-Aldrich (USA). All chemicals used in this

study were of analytical grade. The Free Glycerol Assay Kit was purchased from BioVision

(USA).

Gene synthesis, cloning, cultivations and purification of enzymes. These procedures

were performed as described in Material and Methods section of Chapter 1.

Analytical techniques. Analytical techniques used in this study were described in

Material and Methods section of Chapter 1 of this Thesis.

Production of (S)-DCP by preparative kinetic resolution of (R,S)-DCP. Kinetics of HheC

with (S)-DCP was determined with the substrate prepared by kinetic resolution of

commercially available racemate using (R)-selective HheC. Kinetic resolution of (R,S)-DCP

was performed in 4 l of 50 mM Tris-SO4 buffer (pH 8.5) at 30°C using a 10 l screw cap

bottle. Substrate of purity ≥ 97% (Sigma Aldrich, USA) was added to a final concentration

of 22 mM. The enzymatic reaction was initiated by adding approximately 115 mg of HheC

as a crude extract and 125 mg of purified EchA. The excess of EchA prevented the reverse

reaction during the conversion of DCP and CPD by pushing the equilibrium toward the

product. The reaction was periodically monitored for 8 h and samples were analysed using

a Network GC System 6890N (Agilent Technologies, USA) equipped with a flame

ionisation detector and an Astec CHIRALDEX B-TA 30 m x 0.25 mm x 0.12 µm capillary

column (Sigma-Aldrich, USA). Samples (0.5 mL) were collected from the reaction mixture,

extracted with 1 ml of ethyl acetate, and dried with anhydrous sodium sulphate. Aliquots

(1 µl) were injected into the GC at an injector temperature of 200°C and a split ratio of

83:1. The operating column temperature was held at 100°C for 15 min. Helium was used

as the carrier gas at a continuous flow rate of 0.8 ml.min-1. The flame ionisation detector

was operated at 250°C. Reaction was terminated and unconverted (S) enantiomer was

extracted twice with 0.5 and 1 l of ethyl acetate. The organic phase was dried with sodium

sulphate. Ethyl acetate was removed at 40°C (water bath temperature) and 240 mbar and

then at 60°C (bath temperature) and 60 mbar using a Vacuum Rotavapor R-215 fitted

with a vacuum controller, vacuum pump and heating bath (Büchi, Switzerland). After the

evaporation of most of the ethyl acetate, the chemical composition of the residue was

determined by NMR after dissolution in CDCl3. NMR spectra were acquired using a 600

MHz AVANCE spectrometer (Bruker, Germany) with a double resonance TCI probe (1H, 13C). The resulting NMR spectra were compared to predicted NMR data calculated using

Advanced Chemistry Development Software v11.01 (ACD/Labs, Canada). The purity

(85%) of the synthesized (S)-DCP with residual ethyl acetate was determined by GC using

(R,S)-DCP as a reference material. The (S)-DCP was obtained obtained with e.e. ≥ 99%.

Chapter 2

- 82 -

Enzyme kinetics. Steady-state kinetic parameters for DhaA, DhaA31 and DhaA90R with

TCP and for HheC with (R,S)-DCP, (S)-DCP and (R,S)-CPD were determined in 25 ml micro-

flasks sealed with Mininert valves (Alltech, USA) in 10 ml of 50 mM Tris-SO4 buffer (pH

8.5) at 37°C. Because HheC is strongly (R)-selective, the kinetic constants for (R)-DCP,

which was commercially unavailable, were determined by measuring the initial velocity of

racemic DCP conversion. The initial velocity measurements were carried out at various

substrate concentrations that were verified by GC. The reaction was initiated by adding

the purified enzyme and terminated by mixing samples of the reaction mixture with 35%

(v/v) nitric acid. The concentration of the reaction product (chloride ions) was measured

using a Sunrise spectrophotometer (Tecan, Switzerland) at 460 nm after the addition of

mercuric thiocyanate and ferric ammonium sulfate. Specific activities were quantified by

measuring the slope of the initially-linear region in a plot of halide concentration against

time. The kinetic constants were calculated by non-linear fitting using Origin 7.5

(OriginLab Corporation, USA) or DynaFit 4 (BioKin, USA) based on the Michaelis-Menten

equation (1) where v is the reaction velocity, E0 is the enzyme concentration, kcat is the

turnover number, S is substrate concentration and Km is the Michaelis constant. Steady-

state kinetic parameters for EchA with (R,S)-ECH and (R,S)-GDL were determined by

monitoring the depletion of the substrate (initial concentration: 5 mM) in 10 ml of 50 mM

Tris-SO4 buffer (pH 8.5) at 37°C. The substrate concentration was verified by GC. The

reaction was initiated by adding the purified enzyme. Samples of the reaction mixture

(0.5 ml) were mixed with 0.5 ml of acetone containing the internal standard hexanol and

analyzed by GC. The numerical integration program Scientist 1.0 (MicroMath, USA) was

used to fit the ECH concentration progress curve to equation (1). The GDL concentration

progress curve was fitted to equation (2), where P is the concentration of the reaction

product and Ki is a product inhibition constant. The inhibition constant Kc was determined

by measuring the inhibitory effects of TCP at various concentrations (0, 0.5, 1.0 and

2.5 mM) on the conversion of 5 mM GDL by EchA in 15 ml of 50 mM Tris-SO4 buffer (pH

8.5) at 37°C (Figure S2 in Supplementary tables and figures). Experimental concentrations

of GDL and TCP were verified by GC. The numerical integration software Scientist 1.0

(MicroMath, USA) was used to fit the GDL concentration progress curves for each separate

inhibitor concentration to equation (3), where I is the inhibitor concentration. The

resulting expressions were compared to obtain a consensus value for the equilibrium

inhibition constant Kc.

Chapter 2

- 83 -

Enzyme enantioselectivity. The formation of the (R) and (S) enantiomers of DCP during

the conversion of TCP (2 mM) by DhaA, DhaA31 or DhaA90R was measured in 15 ml of

Tris-SO4 buffer (pH 8.5) at 37°C. Samples (0.5 ml) were taken periodically, mixed with

1 ml of ethyl acetate and vortexed for 30 s. The organic phase was then dried over sodium

sulphate and analysed by chiral GC as described previously.

Kinetic model development. The kinetic model of the TCP pathway was assembled based

on the reaction scheme shown in Figure 1 using the kinetic parameters listed in Table 1.

All five reaction steps in the multi-enzyme conversion of TCP exhibited Michaelis-Menten

kinetics. The conversion of the prochiral TCP into either (R)-DCP or (S)-DCP by DhaA,

DhaA31 or DhaA90R was described using equation (4). The rate constants kcat,TCP,(R)-DCP and

kcat,TCP,(S)-DCP were determined from the overall kcat for the appropriate DhaA variant with

TCP and the ratio of (R)- and (S)-DCP produced from the prochiral TCP by DhaA, DhaA31

or Dha90R. The formation of (R)-DCP and (S)-DCP and the subsequent conversion of both

enantiomers into (R,S)-ECH was described using equations (5) and (6). The remaining

reaction steps were assumed to be non-selective. The consecutive conversions of

(R,S)-ECH, (R,S)-CPD and (R,S)-GDL by HheC and EchA were described using equations

(7), (8) and (9), respectively. The final reaction step, i.e. the conversion of (R,S)-GDL into

the final product GLY by EchA, was described using Michaelis-Menten equations (9) and

(10). These expression feature the equilibrium constants Ki and Kc, which describe the

competitive inhibition caused by GLY and TCP, respectively.

Chapter 2

- 84 -

Optimizing the multi-enzyme conversion of TCP. The initial constraints used when

modelling the multi-enzyme conversion of TCP in order to verify the newly developed

kinetic model were as follows: the starting concentration of TCP was 2 mM and the initial

experimental concentration was as close to 2 mM as possible (Tables S1-S3); the

multi-enzyme reaction was allowed to proceed for 300 min; the concentrations of each of

the three enzymes (i.e. the DhaA variant, HheC and EchA) were 0.1 mg.ml-1; the total

enzyme loading in 10 ml of reaction mixture was 3 mg. Dynamic simulations of the multi-

enzyme system based on a series of Michaelis-Menten equations were performed using

code written in the Python 2.7 programming language. The equations were expressed in

differential form and integrated using Euler's method with a fixed step size of 0.18 s. The

effects of using different DhaA variants on the performance of the multi-enzyme process

were evaluated under the following constraints: at least 95% conversion of TCP into GLY

was to be achieved within 300 min while using the DhaA variant, HheC, and EchA in an

unoptimized mass ratio of 1:1:1 at a total enzyme loading of 3.0 mg. The optimization

algorithm increased the total enzyme loading in a stepwise fashion, using increments of

around 0.1 mg (i.e. increasing the loading of each individual enzyme by 0.066 mg in each

step) until the threshold of 95% conversion was surpassed. The enzyme stoichiometry in

each of three pathway variants was then optimized in order to identify the ratio of DhaA

variant, HheC and EchA that would yield 95% conversion of 2 mM TCP into GLY within

300 min while minimizing the combined loading of the three enzymes. The algorithm

searched for the most efficient ratio of three enzymes in the system starting with a total

enzyme loading of 3 mg. The total enzyme loading within the system was increased in

0.1 mg increments until the most efficient ratio surpassed 95% conversion. For each total

enzyme loading, 496 enzyme ratios were evaluated with step of 3% of a given total

amount.

Multi-enzyme conversion of TCP in batch experiments. The multi-enzyme conversion of

TCP into the final product GLY was assayed in 10 ml of 50 mM Tris-SO4 buffer (pH 8.5) in

25 ml Micro-flasks sealed with Mininert valves (Alltech, USA) and incubated in water bath

with shaking at 37°C. The reaction was initiated by adding a specific amount of purified

DhaA variant, HheC and EchA into the reaction mixture along with TCP at an initial

concentration of 2 mM. Samples were periodically taken from the reaction mixture, mixed

with acetone (1:1) containing hexanol as an internal standard, and analysed by GC to

determine the concentrations of TCP, DCP, ECH, CPD and GDL. Calibration curves for TCP,

Chapter 2

- 85 -

DCP, ECH, CPD and GDL concentrations of 0 – 5 mM were constructed to facilitate data

analysis. Selected samples were analysed by GC-MS to verify the identities of the

metabolites; in other cases, metabolites were identified based on flame ionisation

detection. The concentration of GLY in the reaction mixture was determined

spectrophotometrically using the Free Glycerol Assay Kit. Samples of the reaction mixture

(0.1 ml) were heated at 95°C for 5 min to terminate the enzymatic reaction and then

centrifuged at 18,000 g for 1 min, after which they were diluted in an assay buffer and

analysed according to the manufacturer's protocol. The GLY concentration was calculated

from the sample’s absorbance at 570 nm. A calibration curve for GLY concentrations of

10 – 80 µM was constructed by analyzing samples prepared from a 1 mM GLY standard

solution.

Figure 1. Synthetic pathway for the three-enzyme biotransformation of 1,2,3-trichloropropane. Five consecutive steps are catalyzed by the haloalkane dehalogenase DhaA, from Rhodococcus rhodochrous NCIMB 13064; the haloalcohol dehalogenase HheC and the epoxide hydrolase EchA from Agrobacterium radiobacter AD1. 1,2,3-trichloropropane (TCP) is converted via (R)-2,3-dichloropropane-1-ol ((R)-DCP) and (S)-2,3-dichloropropane-1-ol ((S)-DCP), epichlorohydrin (ECH), 3-chloropropane-1,2-diol (CPD), glycidol (GDL) to glycerol (GLY). Arrows in bold indicate key bottlenecks.

Chapter 2

- 86 -


We initially prepared soluble enzymes with purities of ≥95% for DhaA and EchA and

≥85% for HheC. To validate the in vitro biotransformation of TCP into GLY, 1 mg each of

purified DhaA, HheC and EchA were mixed in 10 ml of Tris-SO4 buffer (pH 8.5) and

incubated with 2 mM TCP at 37°C for 300 min. The enzymes’ molecular weights are

similar (34.1, 29.3 and 36.5 kDa, respectively), so their mass ratio roughly equals their

molar ratio. A gas chromatography (GC) method for detecting and quantifying TCP and all

pathway intermediates in a single analysis was developed and used to monitor the

five-step process (Figure S1 in Supplementary tables and figures). The time course for the

conversion confirmed the pathway’s viability and revealed two major bottlenecks: (i) the

slow initial conversion of TCP to 2,3-dichloropropan-1-ol (DCP); and (ii) the mismatched

selectivity of DhaA and HheC, which caused (S)-DCP to accumulate (Figures 1 and 2).

Similar bottlenecks were identified in vivo [213,232].

We then investigated two DhaA mutants with properties tuned to address these

bottlenecks. The mutant DhaA31 was previously constructed in our laboratory using

computer-aided focused directed evolution of protein tunnels [225]. Its catalytic rate

towards TCP is 32 times that of wild-type DhaA. The mutant DhaAr5-90R (henceforth

“DhaA90R”) was obtained by van Leeuwen and co-workers via the focused directed

evolution of DhaA31. It has the same activity as wild-type DhaA but 7 times higher

selectivity for (R)-DCP formation [226]. Time courses for the three-enzyme conversion of

TCP using purified DhaA31 or DhaA90R were recorded under the conditions described for

the wild-type pathway. The differences between the resulting conversion profiles were

consistent with the kinetic properties of DhaA, DhaA31 and DhaA90R (Figure 2 and Tables

S1-S3 in Supplementary tables and figures).

Sixteen steady-state kinetic parameters were determined for each purified enzyme to

establish a kinetic model of the pathway (Table 1). All of the studied reactions exhibit

Michaelis-Menten (MM) kinetics. The conversion of glycidol (GDL) into GLY, was

described by a MM equation with two inhibition constants defining the inhibitory effects

of GLY (Ki = 1.00 mM) and TCP (Kc = 0.21 mM). These effects together with the substrate

preference of EchA for epichlorohydrin (ECH) caused GDL accumulation in vitro (Figure

2).

Chapter 2

- 87 -

Figure 2. Three-enzyme conversion of 1,2,3-trichloropropane (TCP) to glycerol (GLY) catalyzed by the pathway employing (a) wild-type DhaA, (b) engineered DhaA31 and (c) engineered DhaA90R, respectively. Experimental and simulated metabolite concentrations are indicated by symbols and solid lines, respectively. The reaction’s intermediates are 2,3-dichloropropane-1-ol (DCP), epichlorohydrin (ECH), 3-chloropropane-1,2-diol (CPD), and glycidol (GDL). The following parameters were constrained during the in vitro and in silico experiments: reaction volume (10 ml), initial TCP concentration (~2 mM; experimental concentrations are listed in Tables S1-S3), reaction time interval (300 min), the concentrations of each enzyme (DhaA variant, HheC and EchA; 0.1 mg.ml

-1), and the total loading of all three enzymes in the reaction

mixture (3 mg). Data points represent mean values from three independent measurements. Error bars are omitted for clarity; standard deviations are provided in Tables S1-S3 of Supplementary tables and figures.

Chapter 2

- 88 -

Table 1. Experimental steady-state kinetic parameters used in the kinetic model.

DhaA HheC

Km,TCP (mM) 1.01±0.08 Km,(R)-DCP (mM) 2.49±0.16

kcat,TCP,(R)-DCP (s-1

) 0.04a Km,(S)-DCP (mM) 3.33±0.51

kcat,TCP,(S)-DCP (s-1

) 0.03a Km,CPD (mM) 0.86±0.07

DhaA31 kcat,(R)-DCP (s-1

) 1.81±0.05

Km,TCP (mM) 1.79±0.09 kcat,(S)-DCP (s-1

) 0.08±0.00


) 0.58b kcat,CPD (s

-1) 2.38±0.06


) 0.47b EchA

DhaA90R Km,ECH (mM) 0.09±0.08

Km,TCP (mM) 12.56±2.99 Km,GDL (mM) 3.54±0.09


) 0.19c kcat,ECH (s

-1) 14.37±0.52


) 0.02c kcat,GDL (s

-1) 3.96±0.08

[a] Determined from the ratio of (R)- and (S)-DCP production as 56 % and 44 % of 0.07±0.00 s-1 for kcat,TCP,(R)-DCP and kcat,TCP,(S)-DCP, respectively (Material and Methods). [b] Determined as 55 % and 45 % of 1.05±0.02 s-1 for kcat,TCP,(R)-DCP and kcat,TCP,(S)-DCP, respectively (Material and Methods). [c] Determined as 90 % and 10 % of 0.21±0.03 s-1 for kcat,TCP,(R)-DCP and kcat,TCP,(S)-DCP, respectively (Material and Methods).

The kinetic model was validated against experimental data for the three-enzyme

conversion of 2 mM TCP (Figure 2). The two datasets were in good agreement – using

1:1:1 mixtures of the three enzymes with total enzyme loads of 3 mg, the predicted (and

measured) productivities of the DhaA, DhaA31 and DhaA90R pathways were 72% (62%),

85% (85%) and 45% (42%), respectively. No enzyme inactivation occurred, but the model

could be extended to include inactivation constants if necessary.

The model was then used to predict the DhaA, HheC and EchA loadings needed to

achieve 95% conversion of TCP to GLY under the chosen conditions by simulating

stepwise increases in enzyme loading within the reaction system until the productivity

goal was reached. At a fixed DhaA:HheC:EchA mass ratio of 1:1:1, the wild-type haloalkane

dehalogenase pathway reached the productivity goal using 2.4 mg of each enzyme, i.e. a

total enzyme load of 7.2 mg (Figure 3). The DhaA31 and DhaA90R pathways required

1.8 mg and 4 mg of each enzyme, respectively, (i.e. total enzyme loads of 5.4 or 12 mg) to

reach the goal. The differences in the modelled time courses and quantities of enzyme

required to achieve 95% conversion for the three pathway variants demonstrate the

profound effects of introducing engineered DhaA variants (Figure 3). Despite a

Chapter 2

- 89 -

pronounced accumulation of DCP and GDL during the initial 25 min, the DhaA31 pathway

was around 25% more efficient than the wild-type version. DhaA31 significantly

accelerated the conversion of TCP and thus accelerated the consumption of accumulated

GDL by suppressing TCP’s inhibitory effect. Conversely, DhaA90R reduced the system’s

efficiency despite effectively minimizing DCP accumulation. This selective but catalytically

inefficient mutant was thus not beneficial to the process.

We then evaluated the effects of enzyme stoichiometry on efficiency. An algorithm

was used to minimize the total enzyme load without sacrificing productivity. Optimal

enzyme mass ratios were calculated for each pathway. The modelled time courses for

individual reactions were similar in each case, but the optimized ratios and total enzyme

loadings differed significantly between pathways (Figure 3). Enzyme stoichiometry

optimization increased the efficiencies of the DhaA, DhaA31 and DhaA90R pathways and

reduced their total enzyme loadings required for >95% conversion by 21%, 41%, and

38%, respectively.

All of the optimization simulations were validated by testing the enzyme mass ratios

in vitro. The resulting data agreed very closely with the predictions (Figure 3). The

optimized DhaA90R pathway was around 10% less productive than predicted, possibly

because the Km of DhaA90R for TCP was underestimated due to the limited water

solubility of TCP (cca. 10 mM). In all other cases, the optimized systems achieved

productivities of 94% - 98% (Table S1-S3). Because the experimental time courses only

reflect optimal cases based on pre-defined constraints, we used the simulated data to

create 3D isoproductivity charts showing the effects of varying the loadings of the DhaA

variants, HheC and EchA (Figure 4). These charts show the limiting components for each

pathway and can be used to identify solutions with similar productivities.

Chapter 2

- 90 -

Figure 3. Optimization of three-enzyme conversion of 1,2,3-trichloropropane by kinetic modelling and employment of engineered enzyme variants. Calculated and measured results are indicated by solid lines and symbols, respectively. The following parameters were constrained: reaction volume (10 ml), initial TCP concentration (2 mM), and reaction duration (300 min). The initial optimization goal was 95% conversion of TCP into GLY within 300 min using a 1:1:1 ratio of wild-type enzymes. The calculated and experimentally verified total enzyme loading required to achieve this is 7.2 mg. The wild-type DhaA enzyme was then replaced with the mutants DhaA31 or DhaA90R to study the effect of their kinetics on the pathway (white arrows). Further optimization was achieved by tuning the enzyme ratios (grey arrows). Reductions in total enzyme loading achieved by employment of mutants or stoichiometry optimization alone and stoichiometry optimization together with mutants's effect are shown in circles and squares, respectively. Experimental concentrations of

Chapter 2

- 91 -

1,2,3-trichloropropane (TCP), 2,3-dichloropropane-1-ol (DCP), epichlorohydrin (ECH), 3-chloropropane-1,2-diol (CPD), glycidol (GDL) and glycerol (GLY) were determined by GC and using a glycerol assay kit. Data points represent mean values from three independent experiments. Error bars are omitted for clarity; standard deviations are provided in the Supporting Information (Tables S1-S3).

Figure 4. Isoproductivity color charts showing how biocatalyst concentration affects productivity in the three-enzyme conversion of 1,2,3-trichloropropane into glycerol. Applied constraints are described in the caption of Figure 3; the optimal solutions shown in that figure are indicated by white dots. Values on axes represent the loading of the corresponding enzyme in the reaction mixture. The black line indicates the threshold productivity of 95%.

Chapter 2

- 92 -

CONCLUSIONS

Our results demonstrate that modifying enzyme kinetic parameters and optimizing

enzyme stoichiometry both improved the efficiency of the studied multi-enzyme system.

However, the far simpler process of stoichiometry optimization had a greater impact than

introducing engineered enzymes. The optimized pathway using wild-type DhaA required

a similar total enzyme load to the non-optimized pathway using the engineered DhaA31

(5.7 mg versus 5.4 mg), showing that kinetic modelling alone can provide excellent

solutions in certain cases. Naturally, the best result was achieved with an optimized

pathway using engineered DhaA31: additive effects in this case reduced the catalyst load

required for 95% productivity by 56% relative to the unoptimized wild-type pathway

(Figure 3). This would be very economically beneficial in a large-scale industrial process.

In summary, we present an experimentally validated in silico optimization of a multi-

enzyme process by employing engineered enzyme variant and biocatalyst stoichiometry

tuning. Our workflow entails: (i) experimental verification of process viability, (ii)

determination of enzyme kinetics, (iii) identification of pathway bottlenecks and selection

of suitable engineered enzymes, (iv) development of a robust and accurate kinetic model,

(iv) experimental model validation, and (v) process optimization by in silico enzyme

stoichiometry modelling. Recent progress in efficient enzyme stabilization [264,265] and

ex-vivo cofactor regeneration [100,266] together with the development of integrated

analytical techniques [267], increases in computational power, and the growing

availability of kinetic data will enable the refinement of many useful biotransformations in

vitro and in silico. We believe that the workflow described herein and implemented in the

provided computer code represents a widely applicable strategy for rapid optimization of

multi-enzyme processes.

Chapter 2

- 93 -

SUPPLEMENTARY TABLES AND FIGURES Table S1. Concentrations of metabolites from time courses of 1,2,3-trichloropropane conversion catalyzed by wild-type DhaA, HheC and EchA.a

1 : 1 : 1 (SUM 3 mg)b; productivity 72 %

c

time (min)

TCP DCP ECH CPD GDL GLY SUM

0 2.03±0.06 0.00±0.00 0.00±0.00 0.00±0.00 0.00±0.00 0.00±0.00 2.03 10 1.80±0.01 0.08±0.02 0.00±0.00 0.00±0.00 0.00±0.00 n.d. n.d. 25 1.62±0.02 0.15±0.01 0.00±0.00 0.00±0.00 0.08±0.01 0.13±0.01 1.98 50 1.36±0.02 0.23±0.01 0.00±0.00 0.00±0.00 0.13±0.01 0.24±0.01 1.96 75 1.14±0.02 0.28±0.01 0.00±0.00 0.00±0.00 0.12±0.02 0.41±0.01 1.95

100 0.97±0.02 0.32±0.02 0.00±0.00 0.00±0.00 0.12±0.02 0.56±0.03 1.97 150 0.71±0.02 0.35±0.03 0.00±0.00 0.00±0.00 0.09±0.01 0.88±0.00 2.03 200 0.51±0.03 0.36±0.03 0.00±0.00 0.00±0.00 0.06±0.02 1.01±0.08 1.94 300 0.25±0.02 0.30±0.03 0.00±0.00 0.00±0.00 0.00±0.00 1.47±0.06 2.02

2.4 : 2.4 : 2.4 (SUM 7.2 mg)b; productivity 94 %

c

time (min)


0 1.95±0.06 0.00±0.00 0.00±0.00 0.00±0.00 0.00±0.00 0.00±0.00 1.95 10 1.54±0.06 0.09±0.01 0.00±0.00 0.00±0.00 0.06±0.01 n.d. n.d. 25 1.29±0.05 0.19±0.01 0.00±0.00 0.00±0.00 0.13±0.00 0.14±0.01 1.75 50 0.96±0.03 0.29±0.02 0.00±0.00 0.00±0.00 0.14±0.01 0.40±0.01 1.79 75 0.72±0.02 0.33±0.03 0.00±0.00 0.00±0.00 0.10±0.02 0.70±0.06 1.85

100 0.53±0.01 0.35±0.03 0.00±0.00 0.00±0.00 0.07±0.01 1.07±0.03 2.02 150 0.26±0.00 0.32±0.02 0.00±0.00 0.00±0.00 0.04±0.01 1.29±0.02 1.91 200 0.11±0.01 0.25±0.03 0.00±0.00 0.00±0.00 0.00±0.00 1.54±0.01 1.90 300 0.01±0.01 0.12±0.02 0.00±0.00 0.00±0.00 0.00±0.00 1.83±0.03 1.96


c

time (min)


0 1.95±0.11 0.00±0.00 0.00±0.00 0.00±0.00 0.00±0.00 0.00±0.00 1.95 10 1.62±0.10 0.08±0.02 0.00±0.00 0.00±0.00 0.06±0.00 n.d. n.d. 25 1.33±0.09 0.17±0.01 0.00±0.00 0.00±0.00 0.21±0.01 0.07±0.01 1.78 50 1.00±0.06 0.27±0.02 0.00±0.00 0.00±0.00 0.33±0.02 0.26±0.02 1.86 75 0.75±0.04 0.31±0.03 0.00±0.00 0.00±0.00 0.36±0.04 0.52±0.02 1.94

100 0.54±0.05 0.31±0.02 0.00±0.00 0.00±0.00 0.30±0.01 0.86±0.10 2.01 150 0.27±0.03 0.28±0.02 0.00±0.00 0.00±0.00 0.18±0.01 1.18±0.10 1.91 200 0.11±0.01 0.20±0.03 0.00±0.00 0.00±0.00 0.08±0.01 1.56±0.08 1.95 300 0.00±0.00 0.09±0.01 0.00±0.00 0.00±0.00 0.00±0.00 1.92±0.12 2.01

[a] Data represent mean values ± standard deviation calculated from three independent experiments. [b] Mass ratio of DhaA : HheC : EchA in mg of enzyme dissolved in 10 ml of reaction mixture. n.d., not determined

[c] Productivity was calculated using the formula:

Chapter 2

- 94 -

Table S2. Concentrations of metabolites from time courses of 1,2,3-trichloropropane conversion catalyzed by DhaA31, HheC and EchA.a


c

time (min)


0 2.09±0.10 0.00±0.00 0.00±0.00 0.00±0.00 0.00±0.00 0.00±0.00 2.09 5 1.18±0.03 0.46±0.02 0.00±0.00 0.00±0.00 0.00±0.00 n.d. n.d. 10 0.75±0.03 0.65±0.01 0.00±0.00 0.00±0.00 0.23±0.05 n.d. n.d. 15 0.48±0.03 0.75±0.01 0.00±0.00 0.00±0.00 0.43±0.04 n.d. n.d. 25 0.21±0.00 0.80±0.03 0.00±0.00 0.00±0.00 0.58±0.10 0.43±0.04 2.02 35 0.08±0.01 0.82±0.01 0.00±0.00 0.00±0.00 0.63±0.06 n.d. n.d. 50 0.02±0.02 0.78±0.02 0.00±0.00 0.00±0.00 0.37±0.05 0.87±0.05 2.04 75 0.00±0.00 0.69±0.01 0.00±0.00 0.00±0.00 0.00±0.00 1.19±0.05 1.88

100 0.00±0.00 0.63±0.02 0.00±0.00 0.00±0.00 0.00±0.00 1.37±0.18 2.00 150 0.00±0.00 0.49±0.01 0.00±0.00 0.00±0.00 0.00±0.00 1.44±0.00 1.93 200 0.00±0.00 0.38±0.01 0.00±0.00 0.00±0.00 0.00±0.00 1.53±0.09 1.91 300 0.00±0.00 0.24±0.00 0.00±0.00 0.00±0.00 0.00±0.00 1.78±0.09 2.02


c

time (min)


0 2.06±0.22 0.00±0.00 0.00±0.00 0.00±0.00 0.00±0.00 0.00±0.00 2.06 5 0.96±0.12 0.64±0.06 0.00±0.00 0.00±0.00 0.15±0.03 n.d. n.d. 10 0.45±0.06 0.79±0.06 0.00±0.00 0.00±0.00 0.45±0.09 n.d. n.d. 15 0.21±0.03 0.84±0.08 0.00±0.00 0.00±0.00 0.56±0.02 n.d. n.d. 25 0.05±0.02 0.79±0.08 0.00±0.00 0.00±0.00 0.53±0.02 0.67±0.02 2.04 35 0.01±0.01 0.77±0.06 0.00±0.00 0.00±0.00 0.35±0.01 n.d. n.d. 50 0.00±0.00 0.69±0.07 0.00±0.00 0.00±0.00 0.14±0.02 1.27±0.05 2.10 75 0.00±0.00 0.59±0.03 0.00±0.00 0.00±0.00 0.00±0.00 1.60±0.08 2.19

100 0.00±0.00 0.50±0.03 0.00±0.00 0.00±0.00 0.00±0.00 1.68±0.03 2.18 150 0.00±0.00 0.37±0.02 0.00±0.00 0.00±0.00 0.00±0.00 1.79±0.04 2.16 200 0.00±0.00 0.26±0.02 0.00±0.00 0.00±0.00 0.00±0.00 1.88±0.02 2.14 300 0.00±0.00 0.13±0.02 0.00±0.00 0.00±0.00 0.00±0.00 1.98±0.08 2.11


c

time (min)


0 2.02±0.12 0.00±0.00 0.00±0.00 0.00±0.00 0.00±0.00 0.00±0.00 2.02 10 1.39±0.11 0.27±0.02 0.00±0.00 0.00±0.00 0.16±0.01 n.d. n.d. 25 0.85±0.07 0.45±0.03 0.00±0.00 0.00±0.00 0.48±0.01 0.21±0.01 1.99 50 0.37±0.03 0.56±0.04 0.00±0.00 0.00±0.00 0.65±0.02 0.47±0.01 2.05 75 0.14±0.03 0.54±0.03 0.00±0.00 0.00±0.00 0.57±0.01 0.85±0.10 2.10

100 0.05±0.01 0.47±0.02 0.00±0.00 0.00±0.00 0.45±0.05 1.08±0.03 2.05 150 0.00±0.00 0.33±0.02 0.00±0.00 0.00±0.00 0.19±0.02 1.53±0.02 2.05 200 0.00±0.00 0.22±0.01 0.00±0.00 0.00±0.00 0.08±0.02 1.82±0.04 2.12 300 0.00±0.00 0.09±0.02 0.00±0.00 0.00±0.00 0.00±0.00 1.98±0.05 2.08

[a] Data represent mean values ± standard deviation calculated from three independent experiments. [b] Mass ratio of DhaA31 : HheC : EchA in mg of enzyme dissolved in 10 ml of reaction mixture. n.d., not determined


Chapter 2

- 95 -

Table S3. Concentrations of metabolites from time courses of 1,2,3-trichloropropane conversion catalyzed by DhaA90R, HheC and EchA.[a]


c

time (min)


0 1.80±0.07 0.00±0.00 0.00±0.00 0.00±0.00 0.00±0.00 0.00±0.00 1.80 10 1.68±0.04 0.05±0.01 0.00±0.00 0.00±0.00 0.00±0.00 n.d. n.d. 25 1.60±0.04 0.05±0.00 0.00±0.00 0.00±0.00 0.00±0.00 0.10±0.01 1.75 50 1.51±0.01 0.05±0.01 0.00±0.00 0.00±0.00 0.00±0.00 0.16±0.01 1.82 75 1.45±0.01 0.05±0.01 0.00±0.00 0.00±0.00 0.00±0.00 0.25±0.00 1.75

100 1.35±0.01 0.06±0.01 0.00±0.00 0.00±0.00 0.00±0.00 0.31±0.02 1.72 150 1.22±0.02 0.06±0.01 0.00±0.00 0.00±0.00 0.00±0.00 0.45±0.00 1.73 200 1.13±0.03 0.06±0.01 0.00±0.00 0.00±0.00 0.00±0.00 0.53±0.06 1.72 300 0.91±0.02 0.06±0.00 0.00±0.00 0.00±0.00 0.00±0.00 0.75±0.03 1.72


c

time (min)


0 2.06±0.04 0.00±0.00 0.00±0.00 0.00±0.00 0.00±0.00 0.00±0.00 2.06 10 1.69±0.07 0.03±0.00 0.00±0.00 0.00±0.00 0.10±0.01 n.d. n.d. 25 1.40±0.02 0.05±0.02 0.00±0.00 0.00±0.00 0.13±0.05 0.21±0.01 1.79 50 1.27±0.09 0.04±0.01 0.00±0.00 0.00±0.00 0.13±0.02 0.51±0.01 1.95 75 1.06±0.08 0.06±0.01 0.00±0.00 0.00±0.00 0.09±0.04 0.82±0.04 2.03

100 0.88±0.04 0.04±0.01 0.00±0.00 0.00±0.00 0.05±0.05 0.97±0.04 1.94 150 0.67±0.03 0.04±0.01 0.00±0.00 0.00±0.00 0.00±0.00 1.39±0.10 2.10 200 0.44±0.04 0.04±0.02 0.00±0.00 0.00±0.00 0.00±0.00 1.56±0.03 2.04 300 0.28±0.02 0.01±0.01 0.00±0.00 0.00±0.00 0.00±0.00 1.75±0.06 2.04


c

time (min)


0 1.96±0.01 0.00±0.00 0.00±0.00 0.00±0.00 0.00±0.00 0.00±0.00 1.96 10 1.56±0.07 0.10±0.01 0.00±0.00 0.00±0.00 0.00±0.00 n.d. n.d. 25 1.30±0.05 0.10±0.01 0.00±0.00 0.00±0.00 0.24±0.05 0.09±0.02 1.73 50 1.00±0.04 0.10±0.00 0.00±0.00 0.00±0.00 0.47±0.04 0.32±0.02 1.89 75 0.83±0.02 0.10±0.01 0.00±0.00 0.00±0.00 0.50±0.04 0.62±0.02 2.05

100 0.66±0.03 0.09±0.01 0.00±0.00 0.00±0.00 0.34±0.05 0.84±0.03 1.93 150 0.46±0.03 0.09±0.01 0.00±0.00 0.00±0.00 0.17±0.02 1.31±0.09 2.03 200 0.32±0.01 0.08±0.03 0.00±0.00 0.00±0.00 0.09±0.02 1.56±0.07 2.05 300 0.19±0.01 0.06±0.01 0.00±0.00 0.00±0.00 0.00±0.00 1.72±0.10 1.97

[a] Data represent mean values ± standard deviation calculated from three independent experiments. [b] Mass ratio of DhaA31 : HheC : EchA in mg of enzyme dissolved in 10 ml of reaction mixture. n.d., not determined


Chapter 2

- 96 -

Figure S1. Chromatogram from a gas chromatography analysis of 5 mM standards of 1,2,3-trichloropropane (TCP), 2,3-dichloropropane-1-ol (DCP), epichlorohydrin (ECH), 3-chloropropane-1,2-diol (CPD), and glycidol (GDL). Hexan-1-ol of 4 mM concentration was used as an internal standard (ISTD).

Figure S2. Determination of inhibition constant Kc by measuring the inhibitory effect of TCP on the conversion of 5 mM GDL. Symbols represent different TCP concentrations: Diamonds, 0 mM TCP; squares, 0.5 mM TCP; triangles, 1 mM TCP; spheres, 2.5 mM TCP.

- 97 -

- 98 -

- 99 -

CHAPTER 3

Computer-assisted engineering of the

synthetic pathway for biodegradation of

1,2,3-trichloropropane in heterologous

host Escherichia coli

Chapter 3

- 100 -

INTRODUCTION

Halogenated hydrocarbons, which are often anthropogenic compounds, are widely

used for agricultural, industrial and military purposes [268]. Once introduced into the

environment, these halogenated hydrocarbons often persist and cause a serious threat to

natural ecosystems and human health. Natural catabolic pathways for their mineralization

are inefficient or lacking, primarily due to the fact that most of the discussed chemicals

were present in the environment until last century. Engineering bacteria towards

enhanced biodegradation capacities has been intensively studied, because the

self-replicating whole-cell catalysts represent the cheapest option for bioremediation or

biotransformation processes. However only limited success has been achieved so far due

to the reasons described in previous two chapters of this Thesis [126,128]. One of the

most common failures is because of an imbalance in the expression or catalytic properties

of enzymes employed in synthetic routes. This factor often leads to an accumulation of

toxic intermediates, insufficient flux through the pathway, and limited fitness of the host

organism [145,146,269–271].

A number of approaches dealing with the unbalanced properties of enzymes in

engineered pathways have been reported in recent years. Improvement of pathway

performance can be achieved through the introduction of engineered enzyme variants

[116,117], kinetic modeling of the system [267,272] or modular optimization of the

pathway [69,70]. Although these examples come predominantly from the field of

biosynthesis of valuable chemicals, adoption of such synthetic biology tools holds

considerable promise for rational tuning of pathways for biodegradation of industrial

waste [231].

Example of halogenated hydrocarbon discussed here is a man-made

1,2,3-trichloropropane (TCP). TCP is an emerging toxic groundwater pollutant and

suspected carcinogen, which spreads to the environment mainly due to improper waste

management [186,195]. No naturally-occurring bacterial strain capable of aerobic

utilization of TCP has been isolated from nature thus far. However, TCP can be converted

to harmless glycerol (GLY) via previously described five-step catabolic pathway

assembled with haloalkane dehalogenase (DhaA) from Rhodococcus rhodochrous NCIMB

13064, haloalcohol dehalogenase (HheC) and epoxide hydrolase (EchA) from

Agrobacterium radiobacter AD1 [209,216,237]. The first recombinant strain partially

degrading TCP was constructed using heterologous expression of DhaA or an 8-times

more active mutant form of DhaAM2 in its natural host A. radiobacter AD1, possessing the

remainder of the pathway [211,213]. Although some increase in optical density of cultures

were observed in the latter case, the efficiency of TCP mineralization was insufficient for

supporting significant growth. It was proposed that the poor activity of haloalkane

dehalogenase towards TCP, and formation of toxic intermediates represent primary

bottlenecks of the pathway. Furthermore, equimolar production of the (S) and (R)

enantiomers of 2,3-dichloropropane-1-ol (DCP) from prochiral TCP by non-selective DhaA

and the resulting accumulation of (S)-DCP caused by high enantioselectivity of HheC

Chapter 3

- 101 -

towards (R)-DCP was observed and discussed as the factor limiting the flux of carbon

through the pathway. The effects of the described bottlenecks in the context of the whole

synthetic pathway have not been studied in detail.

Significant efforts have been invested in the past few years in engineering the first

enzyme of the TCP pathway; haloalkane dehalogenase (DhaA) [225,226,265,273].

Constructed mutants can possibly help overcome the previously mentioned limitations of

the TCP pathway. Mutant DhaA31, constructed in our laboratory using computer-assisted

directed evolution, showed 26-fold higher catalytic efficiency towards TCP than the wild

type enzyme (DhaAwt: kcat = 0.04 s-1, kcat/Km = 40 s-1 M-1; DhaA31: kcat = 1.26 s-1,

kcat/Km = 1050 s-1 M-1), while its enatioselectivity with the same substrate remained

unchanged [225]. Mutant r5-90R (referred here as DhaA90R), obtained after five rounds

of directed evolution of DhaA31 by van Leeuwen et al., converted prochiral TCP

predominantly into (R)-DCP (ee 90%), which is the preferred substrate for the selective

HheC [226]. DhaA90R possessed similar catalytic efficiency towards TCP as the wild-type

enzyme (kcat = 0.16 s-1; kcat/Km = 25 M-1 s-1). Thus, each of the mutants, DhaA31 and

DhaA90R, show improvement in one of the limiting factors of the TCP pathway.

After gaining a detailed knowledge about the performance and kinetics of the

synthetic TCP pathway in in vitro conditions, in this study, we carried out a systematic

analysis and rational optimization of the pathway also in heterologous host organism

Escherichia coli. Both engineered variants of DhaA were integrated to evaluate the effects

of increased activity and improved selectivity on the performance of the pathway. A

developed and in vitro verified kinetic model described in Chapter 2 of the Thesis was

supplied with new biological parameters and employed for rational selection of suitable

plasmid combinations and balancing the stoichiometry of the enzyme

DhaA/DhaA31/DhaA90R:HheC:EchA. The pathway was optimized towards faster removal

of toxic metabolites and higher production of GLY. Several E. coli constructs were

prepared based on the predictions and characterized for their viability and ability to

degrade TCP. Advantages and limitations of the engineered DhaA variants in the context of

the pathway assembled in vivo were analyzed and obtained information was used to

propose further steps for optimizing the pathway in heterologous host.

Chapter 3

- 102 -


Chemicals, media, strains, and plasmids. 1,2,3-Trichloropropane (TCP),

2,3-dichloropropane-1-ol (DCP), epichlorohydrin (ECH), 3-chloropropane-1,2-diol (CPD),

glycidol (GDL) and glycerol (GLY) standards were purchased from Sigma-Aldrich (USA).

All chemicals used in this study were of analytical grade. All restriction enzymes and DNA

ligase were purchased from New England Biolabs (USA). A free Glycerol Assay Kit was

acquired from BioVision (USA). Luria Broth (LB) (Sigma Aldrich, USA) was used for

routine cultures. A synthetic mineral medium (SMM) containing 5.4 g of Na2HPO4.12H2O,

1.4 g of KH2PO4, 0.5 g of (NH4)2SO4, 0.2 g of MgSO4.7H2O, 2 ml of trace elements solution

and 1 ml of vitamin B1 (10 g.ml-1) per 1 l was used for selection experiments [274]. M9

minimal medium (Sigma Aldrich, USA) containing 0.2 g of MgSO4.7H2O and 2 ml of trace

elements per 1 l and 10 mM glucose were used for toxicity tests. Escherichia coli DH5α

(Life Technologies, USA) was used in cloning and plasmid propagation. E. coli BL21 (DE3)

(Life Technologies, USA) was used as a heterologous host for expression of the synthetic

pathway for the biodegradation of TCP. Plasmids pET21b, pACYCDuet-1, pETDuet-1,

pCDFDuet-1 (Merck Millipore, Germany) were used for subcloning and modular

construction of pathway variants.

Molecular techniques and culture conditions.The genes of the haloalcohol dehalogenase

encoding HheC and the epoxide hydrolase encoding EchA from Agrobacterium radiobacter

AD1, together with genes of the wild-type haloalkane dehalogenase from Rhodococcus

rhodochrous NCIMB 13064 encoding DhaAwt and the mutants DhaA90R and DhaA31 were

commercially synthesized (GeneArt, Germany). A tag sequence of six histidine codons was

attached downstream from the gene in all cases except for hheC. Sequences of all genes

except for hheC were codon-optimized for expression in E. coli during gene synthesis.

Synthetic genes were subcloned into the NdeI and BamHI restriction sites of pET21b. An

alternative NcoI restriction site was introduced at the beginning of the echA gene to enable

cloning into the first multiple cloning sites of Duet vectors. Upon introduction of the NcoI

site, the second codon of the echA gene, ACT encoding threonine, was substituted for GCA

encoding alanine. The constructs pET21b-dhaA, pET21b-dhaA31, pET21b-dhaA90R,

pET21b-hheC, and pET21b-echA were transformed into competent cells of E. coli DH5α

using the heat-shock method for plasmid propagation. Isolated plasmids were

transformed into E. coli BL21 (DE3) for evaluation of individual gene expression under

similar conditions. Subcloning of the genes coding for the synthetic pathway into Duet

plasmids is summarized in Table 1. E. coli BL21 (DE3) co-transformants, prepared by

modular combination of recombinant Duet vectors are summarized in Table 2. All plasmid

constructs were verified by sequencing (GATC, Germany).

Precultures of E. coli DH5α, BL21 (DE3) host cells, and co-transformants were

prepared by growth on LB medium with or without antibiotics at 37°C overnight. Final

concentrations of respective antibiotics (ampicillin 100 µg.ml-1, chloramphenicol

34 µg.ml-1, streptomycin 50 µg.ml-1) were used in the cultures with cells containing a

Chapter 3

- 103 -

single plasmid. Half concentrations of two relevant antibiotics were used in precultures,

and cultures of co-transformants. A standard protocol was developed for the cultivation of

the degraders. Precultures were used to inoculate fresh LB medium and cultures were

grown at 37°C until the cell density reached 1 at OD600. Expression of recombinant

enzymes was induced by 0.2 mM isopropyl-β-D-thiogalactopyranoside (IPTG) and

cultivation continued at 20°C. Cells were harvested during the late exponential phase by

centrifugation (5000 g, 15 min, 4°C), washed three times with sterile ice-cold SMM or M9

medium 50 mM sodium phosphate buffer pH 7.0, and used in further experiments.

Table 1. Plasmids used and recombinants constructed.

Duet vectorsa Origin Copy no. Recombinant

plasmids Restriction sites

pACYCDuet-1b P15A 10-12 pACYC-echA-hheC echA (NcoI/HindIII)

hheC (NdeI/KpnI) pCDFDuet-1

c CloDF13 20-40 pCDF-dhaAwt NdeI/KpnI

pCDF-dhaA31 NdeI/XhoI

pCDF-dhaA90R NdeI/KpnI pCDF-echA-hheC echA (NcoI/HindIII)

hheC (NdeI/KpnI) pETDuet-1

d ColE1 40 pETDuet-dhaAwt NdeI/KpnI

pETDuet-echA-hheC echA (NcoI/HindIII) hheC (NdeI/KpnI)

pETDuet-dhaA90R NdeI/KpnI

[a] Source – Merck Millipore [b] CmR, chloramphenicol resistance [c] SmR, streptomycin/spectinomycin resistance [d] ApR, ampicillin resistance.

Analytical techniques. A gas chromatograph 6890N with a flame ionisation detector

(GC-FID) and mass spectrometer (GC-MS) 5975C MSD (Agilent Technologies, USA), with

the capillary column ZB-FFAP 30 m x 0.25 mm x 0.25 µm (Phenomenex, USA) were used

for routine analysis and quantification of TCP and its metabolites. The separation method

used for both GC-FID and GC-MS used an inlet temperature of 250°C, split ratio 20:1,

helium carrier gas with an initial flow of 0.6 ml.min-1 for 1 min, followed by a flow

gradient of 0.2 ml.min-1 from 0.6 to 1.8 ml.min-1, and an oven temperature program set

initially to 50°C for 1 min, followed by a temperature gradient of 25°C.min-1 from 50 to

220°C with holding for 2 min. Different concentrations of these compounds were used to

prepare calibration curves for quantification.

Determination of toxicity of the substrate and intermediate metabolites. A growth test

in M9 medium containing 10 mM glucose was used to determine the toxicity of TCP and its

intermediates. The growth of E. coli cultures was evaluated at 37°C for 6 hours using 0.5,

1, 2, 4, 5, 10, 15 and 20 mM of TCP, DCP, ECH, CPD and GLD. To avoid the evaporation of

volatile compounds, cultivation was done in 25 ml glass vials with a screw cap mininert

valve (Sigma-Aldrich, USA). The concentration of TCP and intermediate compounds was

Chapter 3

- 104 -

monitored by GC. Samples (0.5 ml) were withdrawn and extracted with acetone (1:1),

containing hexan-1-ol as an internal standard and centrifuged for 2 min at 18,000 g. The

acetone extracts (2 μl) were injected directly into GC. E. coli cultures without the addition

of the tested compounds was used as a negative control. For growth monitoring, 1 ml

samples were withdrawn in 1 h intervals and measured at OD600. Toxicity data were fit

into polynomial equations from which the inhibitory concentration of individual

compounds (IC20 value) was determined.

Determination of minimal GLY concentration required for host cells growth. Single

colony of E. coli BL21 (DE3) was cultivated overnight in 10 ml of M9 medium containing

20 mM of GLY at 37°C. The preculture was washed two times with ice-cold sterile M9

medium to remove the residual GLY. Resuspension was concentrated to OD600 = 10. The

cultivation was started by addition of 100 l of washed preculture achieving the starting

OD600 0.1 in 5 ml of M9 medium with GLY concentrations varying from 0.01 to 2 mM. The

growth was monitored by measuring OD600 at 0, 3, 24 and 28 h time intervals. After 24 h of

cultivation, GLY was added to the cultures at the concentrations corresponding to the

starting ones. The growth curve with 1 and 2 mM GLY in M9 minimal medium was

determined once again to collect sufficient data between 3 h to 24 h. The seed culture of E.

coli BL21 (DE3) was prepared in M9 medium containing 20 mM of GLY at 37°C. Cell pellets

were collected, washed and concentrated to OD600 = 10. The cultivation started by

addition of 100 l of washed cells achieving the starting OD600 0.1 in 10 ml of M9

medium containing: (i) no carbon source as a negative control, (ii) 1 and 2 mM of GLY, and

(iii) 10 mM of glucose as a positive control. The growth was monitored by measuring

OD600 at 0, 3, 6, 9, 12 and 24 h after cultivation.

Determination of expression levels of the enzymes of the TCP pathway. The expression

levels of DhaAwt, DhaA31, DhaA90R, HheC and EchA were determined in E. coli BL21

(DE3) cells transformed with a pET21b vector and subcloned with corresponding genes,

as well as in the degraders prepared by standardized cultivation procedures. Washed cells

were resuspended in 10 ml of 50 mM sodium phosphate buffer and cell density adjusted

to 3.5 at OD600. 1 U of DNaseI per 1 ml of cell suspension was added. Cells were disrupted

with 5 cycles of sonication using a Hielscher UP200S (Teltow, Germany) ultrasonic

processor with 0.3 s pulses and an amplitude 85%. Each cycle consisted of 5 min

sonication followed by 5 min cooling at 4°C. The cell lysate was centrifuged for 1 h at

18,000 g at 4°C and the resulting cell-free extract (CFE) was decanted. The concentration

of total protein in CFEs was determined using Bradford reagent (Sigma Aldrich, USA).

Samples of CFE containing 5 μg of total protein were separated by sodium dodecyl sulfate

polyacrylamide gel electrophoresis (SDS-PAGE). CFE prepared from E. coli BL21 (DE3)

cells without plasmids were used as controls. For determination of the mass ratio, the

method was calibrated with standard samples containing 0.25, 0.5, 1, 1.5 and 2 μg of each

enzyme DhaA, EchA and HheC in purified form. Purification of DhaA, DhaA31, DhaA90R,

HheC and EchA was described in previous chapters of the Thesis. Gels were stained with

Chapter 3

- 105 -

Coomassie Brilliant Blue R-250 (Fluka, Switzerland) and analyzed using a GS-800

Calibrated Imaging Densitometer (Bio-Rad, USA). The amounts of DhaAwt, DhaA90R,

DhaA31, HheC and EchA in samples of CFE applied to the gel, and the corresponding

relative and mass ratios of the enzymes in the individual degraders were estimated from

trace densities of selected bands.

Verification of expression levels by activity measurements. The expression levels of

DhaAwt, DhaA31, HheC and EchA estimated from SDS-PAGE analysis of CFE samples using

deg31 and deg31-opt was verified by the measurement of individual enzyme activity in

CFE. The activity of DhaA, HheC and EchA was estimated using 10 mM TCP, 20 mM DCP

and 10 mM ECH, respectively. Substrates were dissolved in 10 ml of 50 mM Tris-SO4

buffer (pH 8.5) at 37°C. The reaction was initiated with the addition of CFE to a volume

corresponding to 1 mg of DhaA, 1 mg of HheC or 1 mg of EchA, as was estimated from the

expression levels analyzed by SDS-PAGE. CFE from E. coli BL21 (DE3) without plasmids

was used as a negative control. Samples (0.5 ml) of the reaction mixtures were withdrawn

at certain time intervals and mixed with a 0.5 ml solution of acetone in hexan-1-ol and

analyzed by GC. Specific activity was calculated as a decrease in substrate concentration in

μmol per minute per mg of the enzyme.

Mathematical modeling and optimization of the TCP pathway. The mathematical model

describing the TCP pathway was constructed using kinetic equations and parameters

obtained from in vitro enzyme kinetics with the purified enzymes and corresponding

substrates (TCP, DCP, ECH, CPD and GDL). All kinetic experiments were performed in

50 mM Tris-SO4 buffer at pH 8.5 and 37°C. Determination of kinetic parameters, and

construction of the model and its experimental verification for in vitro conditions is

described in Chapter 2 of this Thesis. Initial parameters for pathway optimization were:

starting concentration of TCP, 2 mM; time interval of multi-enzyme reaction, 300 min;

total mass of enzymes in the pathway, 3 mg. Allowed combinations of Duet vectors were

prepared using the following criteria: only two out of three employed Duet vector

derivatives (pACYC, pCDF and pETDuet) can be combined; either one or two genes coding

for enzymes from the TCP pathway can be expressed from one vector; only the echA gene

possesses the NcoI restriction site for subcloning into the first multiple cloning site of Duet

vectors, thus it can be expressed only from a vector with another subcloned gene and not

separately. The relative ratio of enzymes in a simulation was then determined by a

particular combination of plasmids carrying individual enzymes and a copy number of

individual plasmids (10 for pACYC, 20 for pCDF, and 40 for pETDuet). For all allowed

combinations of plasmids, dynamic simulations of the multi-enzyme system based on a

series of Michaelis-Menten equations were performed using the Python 2.7 programming

language. The differential form of equations was integrated using Euler's method with a

fixed step size of 0.05 min. The toxicity effect at individual times along degradation

profiles was calculated using previously obtained polynomial equations fitted to

experimental toxicity data and the actual concentration of each compound at a given time.

Chapter 3

- 106 -

The overall toxicity effect was calculated by integration of individual effects along the

profile.

Multi-enzyme reactions including variants of the DhaA enzyme with modified

catalytic properties were modeled using the following adjustments of the previously

described process. The total mass of enzymes was set to 0.1 mg, the time-interval of the

multi-enzyme reaction was set to 24 hours. A modification of the catalytic performance of

DhaA was as follows: KM was set to 1 mM, while the overall kcat was varied across the

range of 1 to 1000 min-1. The enantioselectivity of DhaA expressed as a ratio of catalytic

rate constants leading towards production of individual enantiomers of DCP was assigned

for values ranging from 1 to 200. To avoid the limitation of the relative ratio of enzymes in

the simulation by the employed plasmids, the mass of individual enzymes were assigned a

value from 0.0 to 0.1 mg with intervals of 0.0125 mg. The ratio resulting in the highest

end-point production of GLY was considered as representative of the optimized pathway.

Characterization of the degraders in minimal medium with TCP. Pre-induced cells of

the six constructed degraders (Table 2) were washed with sterile ice cold SMM medium

and resuspended in the same medium. A final concentration of 2 mM TCP was added to

15 ml of sterile SMM in 25 ml glass vials with a screw cap mininert valve (Sigma-Aldrich,

USA) and incubated at 30°C with shaking for 2 h to allow for complete dissolution.

Washed cells were adjusted to a final cell density 0.1 at OD600. Vials containing BL21 (DE3)

cells with and without TCP were used as controls. Flasks were continuously incubated for

5 days in a shaking incubator NB-205 (N-Biotek, South Korea) at 200 rpm and 30°C.

Samples for analysis of TCP and metabolites were withdrawn at 0, 6, 24, 48, 72, 96 and

120 h. Viability of cells at the beginning and end of their incubation with TCP was tested

by plating a sample of cell suspension diluted 2x104-times with ice-cold sterile SMM onto

LB agar plates. Plates were incubated overnight at 37°C and grown colonies were counted

using the Colony Picker CP7200 (Norgren Systems, USA).

Recovery of degraders in LB medium. Pre-induced cells of the six constructed degraders

were incubated at 37°C with shaking in 25 ml glass vials with a screw cap mininert valve

containing 10 ml of 50 mM sodium phosphate buffer (pH 7.0) with 3.5 mM TCP. The OD600

of cell suspensions was 3.5. BL21 (DE3) cells without plasmids were used as controls.

After 5 h of incubation, cell suspension samples (10 μl) were transferred into wells of a

96-well microtiter plate containing 100 μl of LB medium. The plate was incubated at 37°C

with shaking (900 rpm) in a shaking incubator (PST-100 HL Thermo Shaker, BIOSAN).

Cell growth was monitored by measuring OD600 at 30 min intervals using a microtiter-

plate reader (Tecan, Switzerland).

Degradation of TCP in buffer by pre-induced resting cells. Cell suspensions of pre-

induced deg31, deg31-opt and E. coli BL21 (DE3) without plasmids (negative control)

were diluted with sterile 50 mM sodium phosphate buffer to a final OD600 7. Glass vials

(25 ml) with a screw cap mininert valve containing 7.5 ml of the sterile 50 mM sodium

Chapter 3

- 107 -

phosphate buffer of pH 7.0 and 4 mM TCP were prepared separately and incubated for 1 h

at 37°C with shaking. The reaction was initiated by mixing 7.5 ml of the cell suspension

with 7.5 ml of buffer and dissolved TCP. The final concentration of TCP in 15 ml of the cell

suspension was 2 mM and the final theoretical OD600 was 3.5. The vials were incubated for

300 min at 30°C with shaking. Cell suspension samples (0.5 ml) were quenched in 0.5 ml

acetone with hexan-1-ol, vortexed and centrifuged at 18,000 g for 2 min. The

concentration of metabolites in the supernatant was analyzed using GC.

Detection of GLY produced from TCP degradation pathway. GLY concentration was

analysed by Free Glycerol Assay Kit (BioVision, USA) according to the manufacturer's

protocol. The pre-induced cells of deg31 and deg31-opt were prepared in LB medium. The

cells were washed two times with ice-cold buffer to remove residual LB. The cells with

final OD600 of 3.5 were incubated with 2 mM theoretical concentration of TCP for

300 minutes in 10 ml phosphate buffer of pH 7.0 supplied with 5 mM of glucose. The

concentration of TCP and DCP were monitored throughout the reaction course by GC.

Production of GLY at certain time points (0, 100, 200 and 300 min) was tested by Free

Glycerol Assay Kit. Samples of cell suspension (50 μl) were heated for 10 min at 95°C and

centrifuged at 18,000 g for 2 min. The supernatants were diluted and analysed using the

kit.

Determination of accumulation of TCP and metabolites in host cells. Accumulation of

TCP and intermediates in the cell biomass was evaluated after 3 h of incubation of cell

suspension of E. coli BL21 (DE3) with 2 mM of TCP, DCP or GDL. All experiments were

performed in 10 ml of 50 mM sodium phosphate buffer at 37°C using the cell density of

3.5 at OD600. After incubation, the cells were harvested by centrifugation at 4000 g for

15 min. Supernatant containing buffer was removed; cell pellet was resuspended in 0.5 ml

of acetone with hexan-1-ol, vortexed for 1 min and centrifuged. All the samples, i.e., initial

cell suspension, buffer and acetone supernatant, were analysed by GC. Proportional

concentrations of TCP, DCP and GDL in cell suspension, cell biomass and buffer were

calculated and compared.

Chapter 3

- 108 -


The haloalkane dehalogenase DhaA, the haloalcohol dehalogenase HheC, and the

epoxide hydrolase EchA were transferred into a heterologous host, Escherichia coli BL21

(DE3), for systematic analysis and rational optimization of a synthetic metabolic pathway

for the biodegradation of TCP. The three enzymes and five metabolites of the pathway do

not naturally occur in E. coli. The possibility of interactions between original metabolism

of E. coli and the synthetic route (so called "metabolic cross-talk") is thus minimized,

which makes the TCP pathway in E. coli an optimal orthogonal system suitable for the

rational analysis of pathway bottlenecks [91].

Toxicity of TCP and intermediate metabolites for E. coli.

Both toxicity of metabolites and the flux of carbon through the pathway represent

the factors dictating the viability of the host organism. Toxicity of TCP and its metabolites

towards E. coli has not been reported. Thus, we tested the effects of various

concentrations of TCP, DCP, ECH, CPD and GDL on growth of the E. coli BL21 (DE3) culture

in mineral medium supplied with glucose. The concentration at which 20 percent of the

total cell population was inhibited (IC20) was calculated for each compound

(Supplementary Figure S1). Results revealed that TCP and epoxide ECH are the most toxic

among all the tested compounds with IC20 values of 1.35 and 1.41 mM, respectively

(Figure 1). The toxicity of both halogenated alcohols, out of which DCP tends to

accumulate in the pathway, was one order of magnitude lower. The second epoxide, GDL,

had an intermediate toxic effect with an IC20 of 6.25 mM. The analysis of TCP, DCP and GDL

accumulation in E. coli showed that TCP is the only compound which has the potential for

cellular accumulation due to its higher hydrophobicity (Supplementary Table S1). The

hydrophobicity and accumulation of the other metabolites was significantly lower due to

the introduced polar hydroxyl groups. These results suggest that concentrations of

metabolites from the TCP pathway detected in the supernatant of the buffer or medium

are a good representation of their intracellular concentrations.

Figure 1. Toxicity of TCP and intermediate metabolites for E. coli BL21 (DE3). Toxicity is expressed as inhibition concentration IC20.

Chapter 3

- 109 -

Minimal GLY requirement for the growth of E. coli.

Aerobic mineralization of TCP should provide sufficient energy for sustainable

growth based on thermodynamic calculations [200]. however, TCP cannot be used at high

concentration during the cultivation due to its toxicity towards the host cells. Thus, only a

limited amount of GLY can be achieved in a culture medium from the degradation of TCP.

GLY is one of the suitable carbon sources that can be utilized by E. coli for energy

generation and cellular biosynthesis. Paliy et al reported the effect of different minimal

media and gluconeogenic carbon sources on the growth of E. coli BL21 and showed that

the strain grew reasonably well in mineral medium supplied with 2 g.l-1 GLY (~21.7 mM)

[275]. Paalme et al. also observed the growth rate of E coli K-12 on various carbon sources

at a concentration of 2 g.l-1 and reported the growth rate varying from 0.22 h-1 on acetate

to 0.77 h-1 on glucose with casamino acids [276]. The growth rate on GLY was 0.32 h-1. In

this study, we determined the minimal concentration of GLY necessary for observable

growth of E. coli BL21 (DE3) cell culture of OD600 0.1 in the mineral medium spiked with

GLY ranging from 0.01 mM to 2 mM (0.922 mg.l-1 to 184.3 mg.l-1). 1mM and 2 mM

concentration of GLY provided carbon and energy for significant increase in OD600 during

the first 3 h of cultivation, but were not sufficient to support the growth continuously for

24 h (Supplementary Figure S2). Repeated feeding of GLY at similar concentrations to the

cultures after 24 h resulted significant increase in OD600 at the concentration ≥ 1 mM.

Construction of TCP degraders by the assembly of an engineered pathway in E. coli.

The genes coding for enzyme of TCP pathway including two mutant variants of dhaA

- dhaA90R and dhaA31 - were commercially synthesized and codon-optimized for

expression in E. coli except for hheC which showed good expression in E. coli even without

optimization. Initially, all genes were subcloned separately into NdeI and BamHI

restriction sites of pET21b vector to compare the expression of individual enzymes under

the T7 promoter. SDS-PAGE analysis of cell-free extracts (CFE) obtained from pre-induced

cells prepared with standardized cultivation protocol showed comparable expression of

all enzymes in E. coli BL21 (DE3) (Supplementary Figure S3). Expression of enzymes

represented by the densities of corresponding bands was related to the expression of

DhaAwt which was defined as 1.0. The expression of the other enzymes was 1.3, 0.8, 1.3,

and 1.3 for DhaA90R, DhaA31, HheC, and EchA, respectively. To allow subcloning of echA

into the first multiple cloning site of Duet vectors, NcoI restriction site was introduced at

the start of the gene altering its second codon ACT (threonine) for GCA (alanine). This

single amino acid substitution had no effect either on enzyme expression or its activity.

The TCP pathway was assembled in two modules, separating expression of the first

enzyme (DhaA) from expression of the second and third enzymes (HheC and EchA). The

two major bottlenecks of the pathway: (i) poor activity of the first enzyme with TCP and

(ii) formation of toxic intermediates, were thus dissected. Modular assembly of the

pathway was carried out using combinations of two out of three Duet vectors: pACYC,

Chapter 3

- 110 -

pCDF and pETDuet. The Duet vector system has already proven its utility in modular

engineering of biosynthetic multi-enzyme pathways [50–53]. Individual Duet vectors can

be combined in a single host cell due to the different origins of the replication and

antibiotic resistance marker (Table 1) [49]. The modular expression strength can be

calculated using the promoter strength and plasmid copy number [69,70]. The expression

levels of subcloned genes can be estimated from the copy numbers of combined plasmids,

since all Duet vectors contain the same T7 promoter and the differences in expression of

individual genes from the TCP pathway under this promoter were found to be negligible

(Supplementary Figure S3).

The starting variants of the TCP pathway were constructed by subcloning dhaAwt,

dhaA90R or dhaA31 into the second multiple cloning site of pCDF, and echA and hheC into

the first and second multiple cloning sites of pETDuet, respectively. This combination

should provide an almost equal ratio of subcloned enzymes based on their reported copy

numbers. The combination of plasmids, consisting of the first enzyme (DhaA) in pCDF, and

the second and third (HheC and EchA) enzymes in pETDuet, was the same in all three

variants. Plasmid pairs were co-transformed into E. coli, resulting in three constructs

denoted degWT, deg90R, and deg31 (Table 2). Degraders were cultivated in LB medium

using a standardized cultivation protocol. An approximate estimation of expression

profiles obtained from SDS-PAGE analysis of CFE showed that the relative ratio of

enzymes in the three degraders was close to 0.2:0.4:0.4 (Supplementary Figure S4).

Table 2. Theoretical and experimental (in bold) ratios of enzymes in the constructed degraders. Degrader Plasmid combinations Theoretical ratio

Experimental ratio DhaA:HheC:EchA

Sum of enzymes

a

degWT pCDF-dhaAwt + pETDuet-echA-hheC 0.20 : 0.40 : 0.40 0.28 : 0.38 : 0.34 1.00

degWT-opt pETDuet-dhaAwt + pCDF-echA-hheC 0.50 : 0.25 : 0.25 0.56 : 0.25 : 0.19 0.94

deg90R pCDF-dhaA90R + pETDuet-echA-hheC 0.20 : 0.40 : 0.40 0.16 : 0.40 : 0.44 1.03

deg90R-opt pETDuet-dhaA90R + pACYC-echA-hheC 0.67 : 0.16 : 0.17 0.70 : 0.12 : 0.18 0.95

deg31 pCDF-dhaA31 + pETDuet-echA-hheC 0.20 : 0.40 : 0.40 0.12 : 0.42 : 0.46 1.06

deg31-opt pCDF-dhaA31 + pACYC-echA-hheC 0:50 : 0.25 : 0.25 0.60 : 0.16 : 0.24 0.74

[a] The sum of the three enzymes in each degrader was calculated from the density of corresponding bands on SDS-PAGE gels and compared to the degWT value (set as 1.00).

Optimization of the ratio of the three enzymes in the TCP pathway was conducted

using an extended version of the previously developed kinetic model. The model is based

on the Michaelis-Menten steady-state kinetic parameters of DhaA variants, HheC, and

EchA, determined from their corresponding substrates in vitro. The concentration of each

enzyme is kept constant at 0.1 mg.ml-1. New constraints defining the toxicity of individual

Chapter 3

- 111 -

metabolites and copy numbers of the used plasmids determining the expression level of

enzymes were introduced for optimization of the pathway in vivo. The model was used to

calculate all possible combinations of the two plasmids within the defined constraints

(Supplementary Tables S2-S4). Calculated time courses of the modeled multi-enzyme

conversion of 2 mM TCP at a time interval of 300 min were visualized using Gnuplot.

Twelve resulting combinations for each of the pathway variants and the corresponding

twelve relative expression ratios were ranked according to: (i) the efficiency of GLY

production and (ii) the level of the overall toxicity in the system, assuming additivity of

the metabolites’ toxic effects. Plasmid combinations used in the non-optimized degraders

degWT, deg90R and deg31 (pCDF-dhaA variant+pETDuet-echA-hheC) were ranked lower

compared to highly ranked degraders with optimized expression levels.

A single combination of plasmids (pETDuet-dhaA90R+pACYC-echA-hheC) showed

the best rank, leading to the highest GLY production and the lowest toxicity for the

degrader employing the DhaA90R variant (Supplementary Table S3). Variants with

DhaAwt and DhaA31 showed three and four equally good combinations, respectively,

from which one combination was selected for experimental construction (Supplementary

Tables S2 and S4). The plasmid combination pETDuet-dhaAwt+pCDF-echA-hheC was

selected for degWT-opt and pCDF-dhaA31+pACYC-echA-hheC for deg31-opt (Table 2).

Comparison of the calculated values representing GLY production and overall toxicity

showed that both degWT-opt and deg90R-opt produced 2-fold more GLY and less toxic

metabolites, in comparison to degWT and deg90R (Supplementary Tables S2 and S3).

Calculated values for plasmid combinations bearing catalytically efficient DhaA31 plus

HheC and EchA, showed that this biochemical pathway cannot be effectively optimized for

GLY production within the defined constraints (Supplementary Table S4), and deg31-opt

was expected to provide improvement only in terms of a lower level of toxic metabolites.

Experimental characterization of the TCP degraders.

The resting cells of six degraders were experimentally characterized in terms of

their: (i) expression profile, (ii) viability, and (iii) degradation capacity. The degraders

were cultivated in LB medium using a standardized cultivation protocol with heterologous

protein expression induced by IPTG. The degraders with over-expressed enzymes of the

TCP pathway were harvested and disintegrated using sonication. Equal amounts of CFEs

were loaded on SDS-PAGE and the expression profiles of six degraders were analyzed by

densitometry (Figure 2). Relative ratios of DhaA variants, HheC and EchA were calculated

from band densities and compared with the theoretical relative ratios predicted by

modeling (Table 2). An excellent agreement between theoretical and experimental values

was observed. The relative sum of the three enzymes from the TCP pathway was

calculated as a sum of the band densities, and revealed that degraders with an optimized

ratio contained equal or lower total amounts of the enzymes than non-optimized ones.

Chapter 3

- 112 -

Figure 2. SDS-PAGE analysis of the expression profiles of the six constructed degraders. M: protein marker (14.4, 18.4, 25, 35, 45, 66.2, and 116 kDa), 1: degWT, 2: degWT-opt, 3: deg90R, 4: deg90R-opt, 5: deg31, 6: deg31-opt and Std: standard sample with purified DhaAwt, HheC and EchA containing 0.25 μg of each enzyme. The theoretical molecular weights of the DhaA variants, HheC and Echa are 34, 29 and 35 kDa, respectively.

The viability and ability to degrade TCP was tested with resting cells of the degraders

resuspended in SMM medium with a final OD600 of 0.1. Cells were incubated at 30°C with

2 mM TCP and no additional carbon source. Time course of TCP and DCP concentrations

was monitored by GC analysis over 120 h (Figures 3A-C). No significant accumulation of

other metabolites of the TCP pathway was observed. Viabilities of the six degraders and

host with and without TCP as controls are presented in Figure 3D. No GLY was detected in

cell cultures suggesting its utilization by the cells. The production of GLY via the TCP

pathway was confirmed in parallel by incubation of induced cells of deg31 and deg31-opt

in the presence of 2 mM TCP and 5 mM glucose (Supplementary Figure S5). Theoretical

concentrations of produced GLY in cultures without glucose and relative toxicity for each

degrader were calculated from recorded time course concentrations of TCP and DCP

(Figures 3E and F).

The averaged data from the three independent experiments revealed that all three

degraders with rationally optimized ratios of pathway enzymes degraded TCP faster than

non-optimized constructs. Experimental time courses of TCP and DCP concentrations

correlated with the relative ratios of enzymes in the degraders. Differences in profiles

corresponded to the catalytic parameters of the DhaA variant and the relative ratio of

haloalkane dehalogenase with respect to the other two associated enzymes (Figure 3A-C

and Table 2). Deg90R and deg90R-opt with selective DhaA90R showed minimal

accumulation of DCP, but their degradation potential was limited by a low catalytic

efficiency with TCP. Deg31 and deg31-opt, both containing DhaA31, the non-selective

variant with improved catalytic efficiency towards TCP, showed the fastest conversion of

Chapter 3

- 113 -

TCP, but also the most significant accumulation of DCP. These two effects were even more

pronounced in the deg31-opt with higher expression of DhaA31.

Viability reflects the physiological state of the cells heterologously expressing the

synthetic pathway after their exposure to toxic TCP (Figure 3D). DegWT-opt showed a

small, but statistically significant difference in cell viability compared to degWT,

suggesting that optimization of the expression levels provided benefits to the cells. The

difference in viability of deg90R and deg90R-opt was statistically insignificant – the low

production of GLY was not balanced by a higher selectivity of DhaA90R and reduced

accumulation of DCP. The most striking result was the superior viability of both degraders

containing the DhaA31 enzyme, connected to a higher production of GLY and/or reduced

toxicity compared to the degraders expressing DhaAwt and DhaA90R. Deg31-opt removes

the toxic metabolites faster than deg31, but the effect of the reduced toxicity on the

viability of this degrader is not visible, suggesting that both degraders surpassed some

threshold level beyond which toxicity does not play a role and only the amount of

produced GLY is the key factor determining cell viability. Further improvement in viability

should be achieved by a more efficient production of GLY. Approximately 1 mM

concentration of GLY produced within 5 days by the best variants, deg31 and deg31-opt, is

probably not sufficient to provide energy for growth and compensation for the oxidative

stress induced during the aerobic mineralization of toxic chlorinated substrates [202].

The ability of the degraders to recover from short-term (300 min) exposure to a

higher concentration (3.5 mM) of TCP was tested (Supplementary Figure S6). Resulting

growth curves corresponded well with the viabilities of degraders calculated from plating

after 120 h incubation with a lower concentration of TCP. Only deg31 and deg31-opt

recovered faster than the host BL21 (DE3) without the synthetic pathway (control) and

with comparable growth. Degraders like degWT and deg90R, with a non-optimized ratio

of pathway enzymes, and deg90R-opt, with an optimized ratio but poor conversion of TCP,

recovered slower than the control. Only degWT-opt showed comparable growth with host

cells without the introduced pathway, again confirming the benefits of pathway

optimization. The data confirm that the constructs containing DhaA31 do not have an

increased metabolic burden [277].

Identification of pathway bottlenecks.

Two constructs carrying DhaA31 and showing the highest viability were selected for

further characterization. Pre-induced cells of elevated cell densities (OD600 3.5) were used

to secure complete degradation of 2 mM TCP during a shorter incubation time (300 min).

The mass ratio of enzymes in both degraders was estimated by SDS-PAGE and verified

with measurements of catalytic activities in CFE (Supplementary Table S5). Obtained data

were used for determination of the enzyme concentrations and applied to calculations of

the degradation profiles of deg31 and deg31-opt. In both cases the experimental profiles

showed a good agreement with predictions (Figure 4). Accumulation of DCP and GDL was

observed for both constructs. A more pronounced accumulation of DCP was observed for

Chapter 3

- 114 -

deg31-opt, with faster removal of TCP and a corresponding lower toxicity of the system.

The accumulated DCP was analyzed by chiral GC and found to be mainly composed of the

(S)-enantiomer. Theoretical conversions of TCP to GLY at the end of incubation calculated

from experimental concentrations of TCP at time 0 min and concentrations of the

accumulated metabolites at 300 min was similar for both deg31 and deg31-opt (75% and

69%). This is in agreement with the data previously collected over 120 h for conversion of

2 mM TCP, and suggests that the degraders containing DhaA31 cannot be further

optimized for better production of GLY without further engineering of DhaA properties

(Supplementary Table S4).

Chapter 3

- 115 -

Figure 3. Degradation profiles and viability of the six degraders. (A-C) Degradation profiles of the degraders containing DhaAwt, DhaA90R and DhaA31, respectively. The diamonds and squares refer to concentrations of TCP and DCP of non-optimized (red) and optimized (blue) variants. D) Viability of the host and degraders after 120 h incubation in SMM medium with TCP, relative to the viability at the beginning of the experiment. E) GLY production calculated from concentrations of TCP and DCP detected at the end of incubation. F) Overall toxicity of metabolites to degraders presented in arbitrary units (AU). Standard deviations were obtained from three independent experiments.

Chapter 3

- 116 -

Figure 4. Predicted and experimentally determined degradation profiles of constructs containing the engineered haloalkane dehalogenase DhaA31. Theoretical (A) and experimental (B) profiles of deg31; theoretical (C) and experimental (D) profiles of deg31-opt. Toxicity is denoted as the grey area under the graph and given in arbitrary units (AU). Concentrations of GLY were calculated from concentrations of other detected metabolites. Experimental data points are the mean values of two replicates. Currently available DhaA variants with increased activity and improved

enantioselectivity provided only partial improvement of the overall efficiency of TCP

degradation, even after optimization in the ratio of enzymes of the synthetic pathway. The

level of GLY production required for cellular maintenance and growth of the degrader

continued to be limited by insufficient selectivity of the DhaA variant with enhanced

activity, and vice versa by insufficient activity of the variant with increased

enantioselectivity. Following these findings, we applied our mathematical model to

estimate the effect of catalytic activity and enantioselectivity on the production of GLY

(Figure 5). We measured the minimum concentration of GLY required for cellular

maintenance and observable growth of the host cells for OD600 0.1, and determined that at

least 1 mM of GLY should be supplied to the culture within the time interval of 24 h

(Supplementary Figure S7). The amount of GLY required for maintenance and growth

could be higher in a culture containing toxic substances like TCP and intermediate

metabolites due to the extra burden caused by their toxicity. The theoretical production of

GLY by the culture of the best degrader deg31-opt with OD600 of 0.1 was calculated to be

Chapter 3

- 117 -

0.6 mM per 24 h, suggesting that DhaA31 does not possess the required catalytic

properties. To increase GLY production to 1 mM per 24 h, the model suggested the need

for further improvement of the catalytic efficiency of DhaA31 by 4-fold, or improvement of

the enantioselectivity by 20-fold. Alternatively, the employment of a DhaA variant with

combined improvement of catalytic efficiency by 1.2-fold and enantioselectivity by 10-fold

in favor of production of (R)-DCP would produce 1 mM GLY within 24 h. A DhaA variant

possessing the properties enabling the highest possible pathway efficiency (1.6 mM GLY

per 24 h) should have 4-fold improvement in catalytic efficiency and 20-fold higher

enantioselectivity compared to DhaA31.

Figure 5. Hypersurface plot describing the effect of catalytic efficiency (kcat/Km) and enantioselectivity (E-value) of DhaA on the production of GLY. The positions of the three variants of DhaA are indicated by red dots. The substrate TCP is supplied at 2 mM. The kinetic constants of DhaAwt, DhaAR90 and DhaA31 were experimentally determined and differ slightly from data reported in the literature.

Chapter 3

- 118 -

CONCLUSIONS

The toxicity of the substrate/metabolites and insufficient carbon/energy flow can

limit viability of the host organism and represent possible bottlenecks in the TCP

degradation pathway. The most toxic substances within the pathway are TCP

(IC20 = 1.35 mM) and ECH (IC20 = 1.41 mM). Other intermediates of the pathway are

significantly less toxic. Since accumulation of ECH inside the cells does not take place, the

hydrophobic substrate TCP represents the single most important toxic substance of the

pathway. A concentration of GLY ≥ 1 mM supports cellular maintenance and growth of E.

coli BL21 (DE3) with an initial OD600 0.1.

A kinetic model was employed to improve the TCP degradation pathway using

engineered enzyme variants and balanced enzyme ratios. Predictions were evaluated in

terms of minimized toxicity of intermediates and maximized production of GLY, and

showed excellent correspondence with experimental data. Expression levels of the

enzymes can also be reliably predicted from a combination of plasmids with different

copy numbers. A statistically significant increase in the viability of degWT-opt compared

to degWT demonstrates that optimization of the enzyme ratio is a useful approach for

improvement of overall pathway performance, and that the kinetic model employing in

vitro measured kinetic parameters of individual enzymes is applicable for rational

pathway engineering also in suitable heterologous host organism. The model can be used

for the design of constructs with optimized expression levels and prediction of minimum

requirements for catalyst properties.

Toxicity of the substrate does not limit viability of the cells employing the highly

active variant of the first enzyme, haloalkane dehalogenase DhaA31, in the pathway. This

is due to rapid conversion of the highly toxic TCP to the less toxic DCP. However, the

growth and viability of the cells are limited by the insufficient production of GLY.

Production of GLY can be improved by further engineering the first enzyme of the

pathway towards a higher catalytic efficiency (kcat/Km >2300 s-1M-1) or higher

enantioselectivy (E-value >20), or by a combination of both properties (kcat/Km >700 s-1M-1

and E-value >10). Removal of the second enzyme selectivity of the pathway, haloalcohol

dehalogenase HheC, represents another option for improvement of the carbon/energy

flow. Construction of a new generation of enzyme catalysts for degradation of TCP is

currently on-going in our laboratory. In parallel, further options for adaptive laboratory

evolution of the whole synthetic pathway in alternative heterologous host are studied in

order to obtain efficient, self-reproducing, and stress-resistant biocatalyst suitable for

biotechnological applications. These options are outlined in the last chapter of the Thesis.

Chapter 3

- 119 -

SUPPLEMENTARY TABLES AND FIGURES

Table S1. Accumulation of selected metabolites of TCP pathway inside the E. coli cells. concentration (mM) in factor

logPa cell suspension buffer cell biomass biomass/buffer

TCP 2.29 2.18 ± 0.15 2.17 ± 0.15 10.30 ± 1.39 4.7

DCP 0.78 2.27 ± 0.09 2.25 ± 0.10 2.23 ± 0.19 1.0

GDL -0.93 2.24 ± 0.09 2.20 ± 0.10 1.53 ± 0.51 0.7

[a] Calculated using ALOGPS 2.1 program [196].

Ch

apte

r 3

- 1

20

-

Ta

ble

S2

. Pre

dic

tio

n o

f G

LY

pro

du

ctio

n a

nd

to

xici

ty e

xpo

sure

fo

r tw

elv

e p

oss

ible

co

nst

ruct

s u

sin

g se

lect

ed c

om

bin

atio

ns

of

pla

smid

s an

d

Dh

aAw

t en

zym

e.

Exp

erim

enta

lly

p

rep

ared

in

itia

l (r

ed)

and

o

pti

miz

ed

(gre

en)

con

stru

cts

are

hig

hli

ghte

d.

pla

sm

id 1

pla

sm

id 2

norm

aliz

ed r

atio

DhaA

: H

heC

: E

ch

A

convers

ion

to G

LY

rank

GL

Y

toxic

ity

rank

toxic

ity

avera

ge

rank

pD

UE

T-d

ha

A

pC

DF

-hhe

C-e

chA

0.5

0

0.2

5

0.2

5

70.9

%

1

14.0

0

2

1

pD

UE

T-d

ha

A

pA

CY

C-h

heC

-ech

A

0.6

7

0.1

7

0.1

7

69.0

%

2

11.1

2

1

1

pC

DF

-dha

A

pA

CY

C-h

heC

-ech

A

0.5

0

0.2

5

0.2

5

70.9

%

1

14.0

0

2

1

pA

CY

C-h

heC

pD

UE

T-d

ha

A-e

chA

0.4

4

0.1

1

0.4

4

59.9

%

4

15.3

2

3

2

pC

DF

-hhe

C

pD

UE

T-d

ha

A-e

chA

0.4

0

0.2

0

0.4

0

63.4

%

3

16.6

2

4

2

pA

CY

C-h

heC

pC

DF

-dha

A-e

ch

A

0.4

0

0.2

0

0.4

0

63.4

%

3

16.6

2

4

2

pD

UE

T-h

heC

pC

DF

-dha

A-e

ch

A

0.2

5

0.5

0

0.2

5

52.9

%

5

23.1

9

5

3

pC

DF

-hhe

C

pA

CY

C-d

haA

-ech

A

0.2

5

0.5

0

0.2

5

52.9

%

5

23.1

9

5

3

pC

DF

-dha

A

pD

UE

T-h

heC

-ech

A

0.2

0

0.4

0

0.4

0

44.4

%

6

25.7

1

6

4

pA

CY

C-d

haA

pC

DF

-hhe

C-e

chA

0.2

0

0.4

0

0.4

0

44.4

%

6

25.7

1

6

4

pD

UE

T-h

heC

pA

CY

C-d

haA

-ech

A

0.1

7

0.6

7

0.1

7

34.3

%

7

28.2

1

7

5

pA

CY

C-d

haA

pD

UE

T-h

heC

-ech

A

0.1

1

0.4

4

0.4

4

26.5

%

8

31.5

4

8

6

Ch

apte

r 3

- 1

21

-

Ta

ble

S3

. Pre

dic

tio

n o

f G

LY

pro

du

ctio

n a

nd

to

xici

ty e

xpo

sure

fo

r tw

elv

e p

oss

ible

co

nst

ruct

s u

sin

g se

lect

ed c

om

bin

atio

ns

of

pla

smid

s an

d

Dh

aA9

0R

en

zym

e.

Exp

erim

enta

lly

p

rep

ared

in

itia

l (r

ed)

and

o

pti

miz

ed

(gre

en)

con

stru

cts

are

hig

hli

ghte

d.

pla

sm

id 1

pla

sm

id 2

norm

aliz

ed r

atio

DhaA

: H

heC

: E

ch

A

convers

ion

to G

LY

rank

GL

Y

toxic

ity

rank

toxic

ity

avera

ge

rank

pD

UE

T-d

ha

A

pA

CY

C-h

heC

-ech

A

0.6

7

0.1

6

0.1

7

62.9

%

1

20.4

2

1

1

pD

UE

T-d

ha

A

pC

DF

-hhe

C-e

chA

0.5

0

0.2

5

0.2

5

56.8

%

2

23.3

1

2

2

pC

DF

-dha

A

pA

CY

C-h

heC

-ech

A

0.5

0

0.2

5

0.2

5

56.8

%

2

23.3

1

2

2

pA

CY

C-h

heC

pD

UE

T-d

ha

A-e

chA

0.4

4

0.1

1

0.4

4

52.2

%

3

24.4

5

3

3

pC

DF

-hhe

C

pD

UE

T-d

ha

A-e

chA

0.4

0

0.2

0

0.4

0

50.3

%

4

25.5

0

4

4

pA

CY

C-h

heC

pC

DF

-dha

A-e

ch

A

0.4

0

0.2

0

0.4

0

50.3

%

4

25.5

0

4

4

pD

UE

T-h

heC

pC

DF

-dha

A-e

ch

A

0.2

5

0.5

0

0.2

5

34.1

%

5

30.1

5

5

5

pC

DF

-hhe

C

pA

CY

C-d

haA

-ech

A

0.2

5

0.5

0

0.2

5

34.1

%

5

30.1

5

5

5

pC

DF

-dha

A

pD

UE

T-h

heC

-ech

A

0.2

0

0.4

0

0.4

0

30.0

%

6

31.7

3

6

6

pA

CY

C-d

haA

pC

DF

-hhe

C-e

chA

0.2

0

0.4

0

0.4

0

30.0

%

6

31.7

3

6

6

pD

UE

T-h

heC

pA

CY

C-d

haA

-ech

A

0.1

7

0.6

7

0.1

7

21.2

%

7

33.2

9

7

7

pA

CY

C-d

haA

pD

UE

T-h

heC

-ech

A

0.1

1

0.4

4

0.4

4

18.0

%

8

35.2

6

8

8

Ch

apte

r 3

- 1

22

-

Ta

ble

S4

. Pre

dic

tio

n o

f G

LY

pro

du

ctio

n a

nd

to

xici

ty e

xpo

sure

fo

r tw

elv

e p

oss

ible

co

nst

ruct

s u

sin

g se

lect

ed c

om

bin

atio

ns

of

pla

smid

s an

d

Dh

aA3

1

enzy

me.

E

xper

imen

tall

y

pre

par

ed

init

ial

(red

) an

d

op

tim

ized

(g

reen

) co

nst

ruct

s ar

e h

igh

ligh

ted

.

pla

sm

id 1

pla

sm

id 2

norm

aliz

ed r

atio

DhaA

: H

heC

: E

ch

A

convers

ion

to G

LY

rank

GL

Y

toxic

ity

rank

toxic

ity

avera

ge

rank

pC

DF

-dha

A

pA

CY

C-h

heC

-ech

A

0.5

0

0.2

5

0.2

5

80.4

%

5

2.0

1

2

1

pD

UE

T-h

heC

pC

DF

-dha

A-e

ch

A

0.2

5

0.5

0

0.2

5

91.5

%

2

3.3

2

5

1

pC

DF

-hhe

C

pA

CY

C-d

haA

-ech

A

0.2

5

0.5

0

0.2

5

91.5

%

2

3.3

2

5

1

pD

UE

T-d

ha

A

pC

DF

-hhe

C-e

chA

0.5

0

0.2

5

0.2

5

80.4

%

5

2.0

1

2

1

pD

UE

T-h

heC

pA

CY

C-d

haA

-ech

A

0.1

7

0.6

7

0.1

7

94.1

%

1

4.9

2

7

2

pD

UE

T-d

ha

A

pA

CY

C-h

heC

-ech

A

0.6

7

0.1

7

0.1

7

73.4

%

7

1.8

9

1

2

pC

DF

-dha

A

pD

UE

T-h

heC

-ech

A

0.2

0

0.4

0

0.4

0

88.1

%

4

3.7

9

6

3

pA

CY

C-d

haA

pC

DF

-hhe

C-e

chA

0.2

0

0.4

0

0.4

0

88.1

%

4

3.7

9

6

3

pC

DF

-hhe

C

pD

UE

T-d

ha

A-e

chA

0.4

0

0.2

0

0.4

0

76.9

%

6

2.2

4

4

3

pA

CY

C-h

heC

pC

DF

-dha

A-e

ch

A

0.4

0

0.2

0

0.4

0

76.9

%

6

2.2

4

4

3

pA

CY

C-d

haA

pD

UE

T-h

heC

-ech

A

0.1

1

0.4

4

0.4

4

88.5

%

3

6.3

2

8

4

pA

CY

C-h

heC

pD

UE

T-d

ha

A-e

chA

0.4

4

0.1

1

0.4

4

68.6

%

8

2.2

2

3

4

Chapter 3

- 123 -

Table S5. Determination of relative and mass ratio of enzymes of TCP pathway in two selected degraders deg31 and deg31-opt. deg31 DhaA31 HheC EchA sum

relative ratioa 0.15 0.38 0.47 1

mass ratio (mg)a 0.62 ± 0.09 1.50 ±0.22 1.90 ± 0.23 4.02 ± 0.53

activity in CFE (μmol.min-1

.mg-1

) 1.27 ± 0.03 1.45 ± 0.02 31.17 ± 2.01

activity of ctrl. (μmol.min-1

.mg-1

)b 1.47 ± 0.14 1.22 ± 0.08 33.35 ± 2.76

activity in CFE/control (%) 86 119 93

corrected mass ratio (mg) 0.53 1.79 1.77 4.09

deg31-opt DhaA31 HheC EchA sum

relative ratioa 0.64 0.17 0.19 1

mass ratio (mg)a 2.20 ± 0.10 0.56 ± 0.07 0.66 ± 0.08 3.43 ± 0.16

activity in CFE (μmol.min-1

.mg-1

) 1.12 ± 0.06 1.04 ± 0.07 32.55 ± 0.92

activity of ctrl. (μmol.min-1

.mg-1

)b 1.47 ± 0.14 1.22 ± 0.08 33.35 ± 2.76

activity in CFE/control (%) 76 85 98

corrected mass ratio (mg) 1.67 0.48 0.65 2.8

[a] Mass ratio and sum of enzymes in mg corresponds to the amount of enzymes in CFE obtained from 10 ml of cell suspension of OD600 of 3.5 by sonication. Ratios were calculated from band densities on SDS-polyacrylamide gel. [b] Control activities were determined with purified enzymes.

Chapter 3

- 124 -

Figure S1. Polynomial equations describing the dependence of E. coli growth on concentration of individual metabolites: A) TCP, B) DCP, C) ECH, D) CPD, and E) GLD.

Chapter 3

- 125 -

Figure S2. The minimal concentration of GLY required for growth and maintenance of E. coli BL21 (DE3). A) The optical density of E. coli culture grown on minimal medium with 0-2 mM of GLY determined after 0, 3 and 24h of cultivation. The same as initial concentration of GLY was spiked to the minimal medium after 24h. B) The growth curves of E. coli cultures grown on minimal medium with no carbon source, 1 mM GLY, 2 mM GLY and 10 mM glucose. All data points represent the mean value from two independent experiments.

Chapter 3

- 126 -

Figure S3. Comparison of the expression of DhaA variants, HheC and EchA from plasmid pET21b. 5 μg of protein were loaded on the gel. Expression of DhaAwt represented by the band density was set as 1.0, expression of the other enzymes related to this value was 1.3 for DhaA90R, 0.8 for DhaA31, 1.3 for HheC and 1.3 for EchA. M, protein marker (14.4, 18.4, 25, 35, 45, 66.2, 116 kDa).

Figure S4. Comparison of the expression profiles of unoptimized degraders. M: protein marker (units in kDa), 1: detWT, 2: deg90R, 3: deg31 and Std: standard sample with 0.25 μg of purified DhaAwt, HheC and EchA.

Chapter 3

- 127 -

Figure S5. Verification of GLY production during whole-cell conversion of TCP with deg31-opt (A) and deg31 (B) of cell density corresponding to the OD600 of 3.5. Glucose (5 mM) was added to the minimal medium as a preferred carbon source. Decrease in concentration of GLY in later phase of reaction was most likely caused by early depletion of glucose. Chiral analysis of 2,3-dichloropropane-1-ol accumulated during conversion of TCP by deg31-opt (C).

Figure S6. Recovery of degraders in LB medium after exposure to 3.5 mM TCP for 300 min. The data represent the mean value from two independent experiments.

Chapter 3

- 128 -

Figure S7. Prediction of DhaA variants optimized for the production of GLY using the engineered TCP pathway in the host cells of OD600 0.1 incubated with 2 mM TCP as a sole carbon and energy source. The existing variants are in red, variants with improved production of GLY (1 mM within 24 h period) are in yellow and optimal variant is in green.

- 129 -

- 130 -

- 131 -

CHAPTER 4

Assembly of the synthetic pathway for

biodegradation of 1,2,3-trichloropropane

in Pseudomonas putida KT2440 CF1

Chapter 4

- 132 -

INTRODUCTION

The performance of the synthetic pathway for biodegradation of

1,2,3-trichloropropane (TCP) was previously tested and optimized in laboratory strain

Escherichia coli BL21 (DE3). This strain is routinely used for overexpression of

recombinant proteins and often serves as a model organism also for metabolic

engineering or synthetic biology studies [15]. However, applicability of E. coli degraders in

adaptive laboratory evolution or continuous biodegradation process is limited due to: (i)

instability of plasmid-based expression system in prolonged cultivations, (ii) sensitivity of

E. coli to the toxic effect of TCP, and (iii) strong overexpression of heterologous genes from

T7 promoter, which may result in extensive metabolic burden for the host cell

[58,60,277,278]. Therefore, we decided to transfer TCP pathway into the host organism

and expression system that could be better suited for further evolution of the synthetic

route and potential biotechnological applications.

Pseudomonas putida KT2440 is a saprophytic non-pathogenic soil bacterium with

ability to naturally mineralize aromatic hydrocarbons (e.g., toluene or xylene), chloro- and

nitro- organic compounds including some pesticides, herbicides and even explosive

chemicals [134]. Recent studies revealed that its alternative variant of Embden-Meyerhof-

Parnas pathway (archetypal glycolysis) known as Entner-Doudoroff pathway empowers

P. putida KT2440 with higher tolerance to the oxidative stress which often accompanies

aerobic bacterial utilization of chlorinated pollutants [202,279]. The metabolic versatility

and robustness made P. putida a host of choice for several projects focused on engineering

of biodegradation pathways [141,145,146]. Increasing popularity of P. putida as a host

organism is accompanied by rising number of genetic tools developed for the purpose of

strain engineering. SEVA plasmids with standardized architecture and nomenclature

represent state-of-the-art vector system with regulatory elements tailored to allow

cloning or expression of heterologous genes in P. putida and other Gram-negative bacteria

[54]. They provide unique variability in terms of antibiotic markers, origins of

replications, and cargos that can be combined in a single plasmid. The pSEVA vectors were

recently modified to allow also scarless deletions of genes or implementation of

regulatory and metabolic modules in a chromosome of Gram-negative bacteria through

specialized mini-transposon-based delivery systems [138,42]. Some of these tools were

used to engineer strain KT2440 Cell Factory 1 (CF1) in a way that a prophages and genes

encoding proteins of flagellum were deleted [280,281]. These modifications provided P.

putida KT2440 with increased energy charge and genomic stability, making it a suitable

chassis for heterologous expression of recombinant enzymes and synthetic metabolic

pathways.

The aim of this study was to transplant the most efficient variant of synthetic TCP

pathway from E. coli into P. putida KT2440 CF1 and express genes for haloalkane

dehalogenase DhaA31, haloalcohol dehalogenase HheC, and epoxide hydrolase EchA from

suitable SEVA vector and from P. putida chromosome. The degrading capacity of obtained

Chapter 4

- 133 -

P. putida constructs was compared with one of two most promissing E. coli degraders,

deg31 described in the Chapter 3. The conducted experiments represent a good starting

point for a broader study focused on comparison of E. coli and P. putida degraders and

development of a robust microbial cell factory for biodegradation of TCP.

Chapter 4

- 134 -


Chemicals, media, strains, and plasmids. TCP, DCP, ECH, CPD, GDL, and GLY standards

were purchased from Sigma-Aldrich (USA). All chemicals used in this study were of

analytical grade. All restriction enzymes were purchased from New England Biolabs

(USA). Phusion High-Fidelity DNA Polymerase was from New England Biolabs (USA) and

GoTag DNA Polymerase was from Promega (USA). T4 DNA ligase and dNTPs were from

Roche Applied Science (USA). Oligonucleotides were synthesized by Sigma Aldrich (USA).

NucleoSpin Gel and PCR Clean-up kit was purchased from Macherey-Nagel (Germany). A

free Glycerol Assay Kit was acquired from BioVision (USA). Luria Broth (LB) (Sigma

Aldrich, USA) was used for routine cultures. M9 minimal medium (Sigma Aldrich, USA)

containing 2 mM MgSO4.7H2O and 2 ml of trace elements per 1 l and 10 mM glucose were

used for toxicity tests [274]. E. coli DH5α (Life Technologies, USA) or CC118λpir was used

in cloning and plasmid propagation. E. coli BL21 (DE3) (Life Technologies, USA) or P.

putida KT2440 CF1 [281] were used as a heterologous hosts and plasmids pCDF and

pETDuet (Merck Millipore, Germany) or pSEVA238 and pBEXK [54,138] were used for

expression of the synthetic pathway for the biodegradation of TCP. Properties of

employed strains and plasmids are listed in Table 1.

Determination of toxicity of TCP for P. putida. A growth test in M9 medium containing

10 mM glucose was used to determine the toxicity of TCP for P. putida KT2440 CF1. The

experiment was carried out as described for E. coli in Material and Methods section of

Chapter 3 of this Thesis with minor modifications. The growth of P. putida cultures was

evaluated at 30°C for 6 hours using 1, 2, 4, and 6 mM of TCP. The concentration of TCP was

monitored by GC. To minimize the evaporation of TCP, cultivation was done in 100 ml

Erlenmeyer flasks with penetrable Suba Seal rubber septa (Sigma-Aldrich, USA).

PCR amplification of dhaA31, hheC and echA from pET21b and pET28b. The genes

dhaA31, hheC, and echA were amplified from the template plasmid constructs

pET21b-dhaA31, pET21b-hheC and pET28b-echA together with the ribosome binding site

(RBS) and C-terminal 6xHis tag (DhaA31 and EchA) by PCR using oligonucleotide primers

dhaA31-F/R, hheC-F/R, and echA-F/R listed in Table 2. To amplify genes, approximately

2 ng of template DNA was mixed in 50 μl reaction mixture containing Phusion HF Buffer

with MgCl2, 3% DMSO, dNTPs of final concentration 0.2 mM each, 0.5 μM of each primer,

and 1 U of Phusion High-Fidelity DNA Polymerase. PCR cycling conditions were as follows:

after an initial DNA denaturation step of 3 min at 94°C, the reaction mixtures were

subjected to 30 cycles consisting of 10 s at 98°C, 30 s at 57°C and 30 s at 72°C. Additional

5 min annealing step at 72°C followed. PCR products were verified by agarose

electrophoresis and purified from gel using NucleoSpin Gel and PCR Clean-up.

Construction of synthetic TCP operon. Restriction analysis of dhaA31, hheC and echA was

conducted using the on-line tool NEBcutter V2.0 (http://tools.neb.com/NEBcutter2/, New

Chapter 4

- 135 -

England Biolabs, USA). Purified PCR products were double-digested with KpnI/BamHI

(dhaA31His), BamHI/XbaI (hheC) and XbaI/PstI (echAHis) and subcloned in a step-wise

manner into the polylinker of vector pSEVA238 digested with corresponding restriction

enzymes (Figure 1). Ligation mixtures were transformed into the competent E. coli DH5α

using heat shock method. Clones were checked for plasmids with inserts by colony PCR

using 1 U of GoTag DNA Polymerase or by restriction analysis. Primers p470, PS2, TCPo1

and TCPo2 listed in Table 2 were used for sequencing of the final constructs including

synthetic TCP operon (Secugen, Spain). Verified plasmid construct pSEVA238-dhaA31-

hheC-echA (signed here as pSEVA238-TCPop) was double digested with KpnI and PstI and

TCP operon was subcloned also into the pBEXK vector (Figure 1) cut with the same

restrictases. Gene cloning techniques and other molecular biology procedures followed

standard protocols [56] or manufacturer's instructions.

Transformation of plasmid constructs into electrocompetent P. putida KT2440 CF1.

Constructs pSEVA238-TCPop, pBEXK-TCPop, and empty pSEVA238 were transformed into

competent P. putida KT2440 CF1 by electroporation. Electrocompetent cells were

prepared as described by Choi et al. by washing the cell biomass twice with 0.3 M sucrose

[282]. Plasmid DNA was added to a 100 μl aliquot of the cell suspension. The mixture was

transferred to a 0.2-cm gap width cuvette and electroporated using the program EC2 of a

MicroPulserTM apparatus (Bio-Rad Laboratories, USA), which corresponds to a single

pulse of 2.5 kV (field strength 12.5 kV.cm-1) and a time constant of ca. 5 ms. Following

electropulsing, 1 ml of LB medium was added quickly to the cells and incubated for 90 min

at 30°C. Adequate dilutions of the resulting culture were then plated onto LB agar plates

with Km (50 μg.ml-1) and plates were incubated overnight at 30°C. Transformants were

checked for pSEVA238-TCPop by colony PCR and restriction analysis and verified clones

were used for expression tests. Single colonies of transformants after electroporation of

pBEXK-TCPop were checked for the loss of the plasmid marker ApR and chromosomal

insertions were verified also by colony PCR using the primers cFRT-Ab-F and cKm-R.

Test of expression of enzymes from TCP pathway in P. putida KT2440 CF1. Single

colonies of KT2440 CF1 transformants were used for inoculation of overnight culture in

LB medium containing Km. Cell were grown at 30°C with shaking (180 rpm). 200 μl of

overnight culture was used to inoculate 20 ml of LB medium with Km. Cultures were

grown at 30°C with shaking (180 rpm) until OD600 of about 0.6. Expression of recombinant

enzymes was induced with 5 mM 3-methylbenzoate (0.5 M stock in milliQ water was

prepared by dissolution of 3-methylbenzoate in presence of sodium hydroxide). Induction

continued overnight (13 – 14 h) at 20°C with shaking (180 RPM). Cells were pelleted by

centrifugation at 4,000 g at 4°C. Cell biomass was resuspended in 50 mM sodium

phosphate buffer (pH 7.0) and cells were disrupted by sonication. Cell lysates were

centrifuged at 21,000 g for 45 min at 4°C. Resulting cell-free extracts (CFE) were decanted

and samples of CFE and cell pellets were loaded on 15 % sodium dodecyl sulfate

polyacrylamide gel. After SDS-PAGE, gels were stained with Coomassie Brilliant Blue

Chapter 4

- 136 -

R-250 (Fluka, Switzerland) and analyzed using a GS-800 Calibrated Imaging Densitometer

(Bio-Rad, USA).

Degradation of TCP in buffer by pre-induced resting cells. Cell suspensions of

pre-induced deg31, pre-induced P. putida KT2440 CF1 with pSEVA238-TCPop or empty

pSEVA238 (negative control) and three randomly picked clones of P. putida KT2440 CF1

with TCP operon in chromosome were diluted with sterile 50 mM sodium phosphate

buffer to a final OD600 7.0. Glass vials (25 ml) with a screw cap mininert valve containing

7.5 ml of the sterile 50 mM sodium phosphate buffer of pH 7.0 and 4 mM TCP were

prepared separately and incubated for 1 h at 37°C with shaking. The reaction was initiated

by mixing 7.5 ml of the cell suspension with 7.5 ml of buffer and dissolved TCP. The final

concentration of TCP in 15 ml of the cell suspension was 2 mM and the final theoretical

OD600 was 3.5. The vials were incubated for 300 min at 30°C with shaking. Cell suspension

samples (0.5 ml) were quenched in 0.5 ml acetone with hexan-1-ol, vortexed and

centrifuged at 18,000 g for 2 min. The concentration of metabolites in the supernatant was

analyzed using GC and the production of GLY was monitored by Glycerol Assay Kit as

described in Material and Method sections of previous chapters of this Thesis. The

stability of TCP degradation phenotype of P. putida constructs with synthetic operon

inserted in chromosome was verified by twenty rounds of serial plating of clones on LB

agar plates without antibiotics and repeated determination of TCP degradation in buffer

with pre-induced cells.

Determination of growth curve of E. coli BL21 (DE3) and P. putida KT2440 CF1 on GLY.

The overnight cultures of E. coli BL21 (DE3) and P. putida KT2440 CF1 grown in LB

medium were washed by ice cold M9 minimal medium with 2 mM MgSO4.7H2O and trace

elements and used for inoculation of 20 ml of the same medium but with 0.2% GLY to the

final OD600 of 0.1. The growth curves of cultures grown with shaking (180 rpm) at 30°C

were determined by measuring optical density at 600 nm for 30 h.

Analytical techniques. Analytical techniques used in this study were described in

Material and Methods section of Chapter 3 of this Thesis.

Chapter 4

- 137 -

Table 1. Bacterial strains and plasmids used in this study

Strain or plasmid Relevant characteristicsa Source or reference

Escherichia coli

Dh5α Cloning host; Φ80lacZΔM15 recA1 endA1 gyrA96

thi-1 hsdR17(rK−mK

+) supE44 relA1

deoR Δ(lacZYA-argF)U169

Life Technologies, USA

CC118λpir Cloning host; araD139 Δ(ara-leu)7697 ΔlacX74

galE galK phoA20 thi-1 rpsE

rpoB(RifR) argE(Am) recA1 λpir lysogen

[55]

BL21 (DE3) Expression host; F– ompT gal dcm lon hsdSB(rB

-

mB-) λ(DE3 [lacI lacUV5-T7 gene 1 ind1 sam7

nin5])

Life Technologies, USA

Pseudomonas putida

KT2440 Cell Factory 1

(CF1)

Engineered KT2440 derived from P. putida mt-2

cured of the TOL plasmid pWW0;

Δprophages1,2,3,4 ΔTn7 ΔendA1 ΔendA2

ΔhsdRMS Δflagella ΔTn4652

[281]

Plasmids

pCDF Expression vector; CDF(CloDF13) T7lac SmR Merck Millipore, Germany

pETDuet Expression vector; ColE1(pBR322) T7lac ApR Merck Millipore, Germany

pSEVA238b Expression vector; oriV(pBBR1) xylS Pm aphA

KmR

[54]

pBEXK Mini-Tn5 delivery vector derived from

pBAM1; tnpA oriV(R6Kγ) oriT(RK2) xylS Pm bla

FRT-aphA-FRT ApR Km

R

[138]

[a] Antibiotic markers: Ap, ampicillin; Sm, streptomycin; Km, kanamycin. [b] Plasmids belonging to the Standard European Vector Architecture (SEVA) collection.

Table 2. Nucleotide sequences of primers used in this study. dhaA31-F

1 5' TCGGTACCCTTTAAGaaggagATATACATATGTCCGAAATTG 3'

dhaA31-R2 5' AATGGATCCTTAGTGGTGGTGGTGG 3'

hheC-F2 5' TTCGGATCCCTTTAAGaaggagATATACATATGTCAACCGCA 3'

hheC-R3 5' TCTCTAGATTACTCGGGCATACCAGGCCA 3'

echA-F3 5' TCTCTAGACTTTAAGaaggagATATACCATGGCAATC 3'

echA-R4 5' TCCTGCAGTTAATGGTGGTGGTGGTGGTG 3'

p470 5' GGTTTGATAGGGATAAGTCCAG 3'

PS2 5' GCGGCAACCGAGCGTTC 3'

TCPo1 5' GAACGAACTGCCGATTGCAG 3'

TCPo2 5' CGAATCCAGAACACGTTGC 3'

cFRT-F 5' CTTGCAATGGTACAATCCTCC 3'

cKm-R 5' CGGAATGCTATGCAGACG 3'

Introduced restriction sites are underlined: 1KpnI; 2 BamHI; 3XbaI; 4PstI. Sequence of original ribosome binding site from pET vector is represented by small letters.

Chapter 4

- 138 -

Figure 1. Schematic representation of the Tn5-based insertional vector pBEX for regulated expression of heterologous genes in Gram-negative bacteria. The functional elements of the plasmid backbone include an antibiotic resistance marker (bla encoding ß-lactamase), a hyperactive transposase (tnpA), a conditional vegetative origin of replication (oriR6Kγ), an origin of transfer region (oriT), and two mosaic elements (ME-O and ME-I). The expression modules span edited copy of the P. putida xylS gene together with the inducible Pm promoter. Promoter is in front of a multiple cloning site followed by strong T0 terminator from phage λ. Expression module is transferred into the chromosome of recipient bacterium along with an removable cassette consisting of the antibiotic resistance marker (aphA-encoded aminoglycoside 3'-phosphotransferase conferring kanamycin resistance) bracketed by FRT sequences. Adopted from [138] and modified.

Chapter 4

- 139 -


The response of P. putida KT2440 CF1 to the toxic effect of TCP was initially tested

and compared with E. coli BL21 (DE3). TCP was previously showed to be the most toxic

compound from all the metabolites in TCP pathway. Effect of various concentrations of

TCP on growth of the P. putida KT2440 CF1 culture in mineral medium supplied with

glucose was determined using the protocol described for E. coli BL21 (DE3) in Chapter 3

of the Thesis. The concentration of TCP at which the growth of total cell population was

reduced by 20% (IC20) was calculated (Supplementary Figure S1). The IC20 value for P.

putida KT2440 CF1 (1.32 mM) was almost identical to the value previously published for

E. coli BL21 (DE3) (1.35 mM). The result shows that TCP is equally detrimental for both

bacterial hosts suggesting the same toxicity mode of action.

The TCP pathway genes were then subcloned from pET to SEVA vector system to

allow transfer of the pathway from E. coli to P. putida. Genes dhaA31, hheC and echA with

their RBSs were amplified from plasmid constructs pET21b-dhaA31, pET21b-hheC, and

pET28b-echA by PCR (Figure 2) and new restriction sites for subcloning into the

polylinker of SEVA vectors were introduced upstream and downstream of the genes.

Restriction sites were selected with care to allow serial subcloning of all three genes in the

polylinker and construction of a synthetic operon, which could be easily transferred in

between various SEVA vectors.

Plasmid pSEVA238 was used for the assembly of synthetic operon (Figure 3). The

plasmid carries inducible Pm/xylS promoter from P. putida TOL plasmid pWW0,

kanamycin resistance marker, and the origin of replication from pBBR1 that confers the

vectors with medium-copy number (30 – 40 molecules in a single P. putida cell). The

genes coding for enzymes of TCP pathway were subcloned into pSEVA238 in step-wise

manner resulting in constructs pSEVA238-hheC, pSEVA238-hheC-echA and pSEVA238-

TCPop. New constructs were verified by colony PCR, restriction analysis (Figure 2) and

sequencing. The plasmid constructs were transformed into the electrocompetent P. putida

CF1 cells and the expression of individual genes of TCP pathway after induction with 5

mM 3-methylbenzoate was evaluated by SDS-PAGE analysis of CFEs obtained from

disrupted cell biomass (Figure 4A). Interestingly, analysis revealed that the relative ratio

of DhaA31, HheC and EchA expressed from pSEVA238-TCPop in P. putida under selected

conditions (0.11:0.49:0.40) was very similar to the ratio achieved previously in E. coli

deg31 after induction with 0.2 mM IPTG (0.12:0.42:0.46). The expression of DhaA31 in P.

putida was significantly lower than the expression of HheC and EchA although the front

position of dhaA31 in the synthetic operon and long transcription distance from the

beginning of the gene to the end of the operon predetermined it rather for higher

expression [64,283]. The relative amount of all three enzymes in total soluble protein of P.

putida was about 38% while in E. coli deg31 the content of DhaA31, HheC, and EchA in cell

had reached 50% of the total soluble protein.

Chapter 4

- 140 -

Figure 2. Electrophoretic analysis of products of PCR amplification of dhaA31 (935 bp), hheC (801 bp), and echA (937 bp) from pET plasmids (gel on left) and restriction analysis of pSEVA238-TCPop (7744 bp) cut with KpnI and PstI. Length of synthetic TCP operon (bottom band) is 2685 bp. The 500 bp molecular ruler (Bio-Rad, USA) was used for analyses.

Figure 3. Map of plasmid construct pSEVA238-TCPop with synthetic operon encoding three enzymes from pathway for biodegradation of TCP. Operon is expressed from inducible xylS-Pm promoter and transcription is terminated at single rho-independent transcriptional terminator T0 from phageλ. Each gene in operon possesses its own ribosome binding site (RBS). The primers used for sequencing are shown at their annealing sites (in blue).

Chapter 4

- 141 -

The TCP operon was subsequently subcloned also into the suicide pBEXK plasmid

with minitransposon system for delivery of selected cargo into the chromosome of host

bacterium (Figure 1). The synthetic operon was integrated into the chromosome of P.

putida KT2440 CF1 as described by Nikel and de Lorenzo to ensure genetic stability of

engineered strain even in prolonged cultivations [138]. The integration in six selected

clones was verified by parallel plating on LB agar plates with kanamycin and ampicillin

and by colony PCR with insert- and plasmid backbone-specific primers. Expression of

individual genes of TCP pathway in E. coli deg31, P. putida CF1 pSEVA238-TCPop and one

of the clones of P. putida CF1 with TCP operon was compared after induction with 0.2 mM

IPTG (E. coli) or 5 mM 3-methylbenzoate (P. putida; Figure 4B). SDS-PAGE analysis

showed that the relative ratio of DhaA31, HheC and EchA expressed from operon in

chromosome of P. putida remained unchanged when compared with expression from

pSEVA238. Also the relative amount of all three enzymes in total soluble protein of P.

putida (35%) was similar to the value determined for recombinant strain expressing

enzymes from plasmid.

Figure 4. A) SDS-PAGE analysis of expression of DhaA31 (34 kDa), HheC (29 kDa) and EchA (35 kDa) in Pseudomonas putida CF1. M, protein marker (6.9, 21, 29, 35.8, 56.2, 101, 125, 210 kDa); 1, cell-free extract (CFE) of CF1 with empty plasmid pSEVA238; 2, CFE of CF1 with pSEVA238-hheC; 3, CFE of CF1 with pSEVA238-hheC-echA; 4, CFE of CF1 with pSEVA238-TCPop. B) Comparison of expression of individual enzymes in: 1, Escherichia coli deg31; 2, P. putida CF1 pSEVA238-TCPop; 3, P. putida CF1 with synthetic TCP operon introduced into the chromosome. Preinduced cells of P. putida KT2440 with pSEVA238-TCPop and six selected clones

with TCP operon in the chromosome were incubated with 2 mM TCP in buffer for 300 min

and the degradation profiles of P. putida degraders were determined and compared with

reaction time course collected for E. coli deg31 (Figure 5). The degradation profile of P.

putida carrying pSEVA238-TCPop was very similar to the profile of deg31. This result

Chapter 4

- 142 -

corresponded well with the similarity of relative ratios of DhaA31, HheC, and EchA in E.

coli and P. putida constructs. The theoretical conversion of TCP to GLY reached 68% in

both constructs. Interestingly, the TCP degradation profile of six P. putida clones with

single copy of synthetic operon in chromosome (Figure 5C – average data) was

comparable with the profile of cells expressing TCP pathway from medium copy plasmid.

Nevertheless, conversion of TCP varied in between individual constructs with operon

integrated in chromosome and was in some degree slower, reaching theoretical

conversion of TCP to GLY of 61% on average. Slight variations in profiles of individual

clones can be attributed to the different sites of integration, because the context of

integration position might influence the level of expression [138]. Specification of

integration sites of six selected P. putida constructs by arbitrary primed PCR and

sequencing is in progress. Stability of phenotype of six constructs with chromosomal

insertion was verified by repeated determination of ability to produce GLY from TCP after

12 successive transfers of cells grown for 12 h in liquid LB medium without TCP or

antibiotic (data not shown).

Although there was not much variation in degradation profiles of E. coli and P. putida

recombinants, their potential to utilize produced GLY differed substantially (Figure 5D).

While deg31 utilized all of generated GLY within 300 min interval both P. putida

constructs showed continuous accumulation of GLY. Our observation is in agreement with

recent study of Nikel and co-workers showing that P. putida KT2440 grown on GLY has

considerably long lag phase, which may last up to 18 h [284]. Both E. coli and P. putida are

genetically equipped with the functions required for GLY metabolism. The prolonged lag

phase of P. putida can be attributed to the differences in central carbon metabolism and

rearrangements of the whole metabolic network that is necessary to start growth on GLY

[284].

On the other hand, Nikel and co-workers claim that the growth of P. putida on GLY as

a sole carbon source eventually results in a specific metabolic mode of bacterium that is

characterized by: (i) remarkably low level of physiological stress, when compared with

growth on other carbon sources, and (ii) very efficient conversion of substrate into

biomass [284]. We verified the latter by recording the growth curve of P. putida KT2440

CF1 in minimal medium with 0.2% GLY (Supplementary Figure S2). Compared to E. coli

BL21 (DE3), the P. putida culture showed longer lag phase but achieved higher optical

density at the end of exponential phase of growth. Our future research will focus on

comparison of the levels of physiological stress accompanying the degradation of TCP and

utilization of produced GLY in E. coli deg31 and P. putida constructs by viability testing

and flow cytometry measurements with cells labelled by stress-specific fluorophores.

Chapter 4

- 143 -

Figure 5. TCP degradation profiles of A) E. coli deg31, B) Pseudomonas putida CF1 pSEVA238-TCPop, and C) P. putida CF1 with synthetic TCP operon introduced into the chromosome (average profile of six different constructs is presented). The theoretical concentrations of produced glycerol (GLY calc.) were calculated from the experimentally determined concentrations of accumulated intermediates and starting concentration of TCP. D) Comparison of experimentally determined concentrations of GLY produced by individual constructs during TCP degradation. Each data point represents the mean value ± standard deviation from three or six (in case of CF1 TCPop in chromosome) independent experiments.

Chapter 4

- 144 -

CONCLUSIONS AND OUTLOOK Genes encoding enzymes from syntetic TCP pathway were successfully subcloned

from pET to SEVA vector system with inducible Pm/xylS promoter, which enabled

expression of TCP pathway in P. putida KT2440 CF1. The strain showed similar respons to

the toxicity of TCP as E. coli BL21 (DE3). Synthetic operon consisting of dhaA31, hheC, and

echA genes was assembled in pSEVA238 and subcloned into plasmid vector pBEXK, which

was utilized for implantation of the construct into the chromosome of KT2440 CF1. P.

putida recombinants with ratio of enzymes and TCP degradation profiles resembling

those of E. coli deg31 were obtained. These recombinants will be used for thorough

comparative study of two different microbial strains carrying TCP pathway of same

productivity and for further in vivo evolution of TCP pathway.

We expect that shortening of the lag phase by additional engineering interventions

targeting GLY metabolism will make P. putida attractive host for biodegradation of TCP. As

was demonstrated in several bacterial species, shorter lag phase on GLY can be reached

e.g., by deleting glpR represor that regulates expression of glp gene cluster for GLY

metabolism [285,286]. On the other hand, certain applications can take advantage from

GLY accumulating in KT2440 CF1 strain during growth on preffered co-substrate like

glucose or succinate. In such cases, GLY can be streamed in alternative biosynthetic

pathways for production of some value-added compounds, e.g., 1,3-propanediol, citric

acid, or bioplastics like polyhydroxyalkanoates [287]. Thus, toxic industrial waste TCP

could be converted into useful chemicals using complex biochemical networks of whole-

cell biocatalyst that cannot be, with recent state of knowledge, reconstructed in vitro.

Alternatively, accumulated GLY as a direct product of TCP conversion could be

detected and quantified using available enzymatic kits with specific fluorescence probes.

Such system could be employed in fluorescence activated cell sorting (FACS) of clones

resulting from adaptive laboratory evolution of P. putida recombinants incubated with

TCP or clones containing variants of synthetic TCP operon randomized by error-prone

PCR, oligonucleotide-directed randomization or some other mutant library generating

methods [119,288].

We believe, that newly constructed bacterial recombinants and proposed methods

will be benefitial for evolution of more efficient TCP pathway lacking the bottlenecks

described in previous chapters of the Thesis, namely, unbalanced enantioselectivity of

DhaA31 and HheC resulting in accumulation of (S)-DCP or unfavourable substrate

preference of EchA causing accumulation of GDL. In our future research, we aim to further

study the effects of TCP on the host cell physiology, to remove discussed bottlenecks and

to gain bacterial degrader capable of sustainable growth on TCP as the sole carbon and

energy source.

Chapter 4

- 145 -

SUPPLEMENTARY FIGURES

Figure S1. Linear dependence of inhibition of Pseudomonas putida KT2440 CF1 growth on concentration of TCP. Data points represent a mean value from three independent experiments ± standard deviation.

Figure S2. Growth curves of Escherichia coli BL21 (DE3) and Pseudomonas putida KT2440 CF1 obtained during cultivation in minimal medium with 0.2% glycerol. The data points represent a mean value from duplicate measurement.

- 146 -

Summary

- 147 -

SUMMARY

Recent chemical pollution of soil, ground water and surface water constitutes a

major threat to the ecosystems and the public health. Reports of renowned environment

protection agencies predicate that the situation will become worse and the number of

sites requiring remediation will further increase. Biotransformations using either natural

or recombinant microorganisms are extensively studied for their ability to decontaminate

polluted sites. However, rational engineering of microorganisms towards novel or

improved biodegradative capacities performed in the complex environments remains a

challenge. Broader employment of cutting edge techniques and approaches of emerging

biological engineering disciplines can help with both better understanding of studied

biological systems and exploitation of obtained knowledge in practical applications.

This Thesis summarizes the work in which synthetic biology and metabolic

engineering approaches were applied for study and rational redesign of metabolic

pathway for biodegradation of important environmental pollutant TCP. We first

reconstructed the pathway under in vitro conditions and developed a novel biotechnology

for TCP transformation based on immobilized enzymes. Then, we improved the efficiency

of the three-enzyme system with help of engineered biocatalyst and kinetic model using

the data obtained from detailed characterization of individual enzymes. We have

successfully assembled the pathway in heterologous host E. coli BL21 (DE3) and showed

that the same kinetic model can be applied also for tuning of the in vivo orthogonal system.

Eventually, one of the most promising pathway variants was transferred to the new host,

P. putida KT2440 CF1, whose capacity to become a suitable chassis for further evolution of

TCP route is under investigation.

The Thesis presents an innovative concept for rational engineering of a synthetic

metabolic pathway for degradation of toxic recalcitrant compounds. This concept is based

on detailed understanding of studied system and exploitation of in vitro, in silico and in

vivo tools of synthetic biology and metabolic engineering. Conducted research generated a

deeper insight into a studied model system, allowed rational dissection of pathway

bottlenecks, resulted in development of new in vitro biocatalysts applicable for TCP

biodegradation and will hopefully lead to the design of better microbial degrader capable

of sustainable growth on TCP. The developed approaches and molecular tools should be

generally applicable also to other environmental pollutants.

References

- 148 -

REFERENCES

(1) De Lorenzo, V. From the selfish gene to selfish metabolism: revisiting the central dogma. BioEssays

News Rev. Mol. Cell. Dev. Biol. 2014, 36, 226–235. (2) Watson, J. D.; Crick, F. H. The structure of DNA. Cold Spring Harb. Symp. Quant. Biol. 1953, 18, 123–

131. (3) Rich, A. Recent studies on the structure of ribonucleic acid. Prog. Neurobiol. 1956, 1, 114–121. (4) Pauling, L.; Corey, R. B.; Branson, H. R. The structure of proteins; two hydrogen-bonded helical

configurations of the polypeptide chain. Proc. Natl. Acad. Sci. U. S. A. 1951, 37, 205–211. (5) Arber, W.; Linn, S. DNA modification and restriction. Annu. Rev. Biochem. 1969, 38, 467–500. (6) Meselson, M.; Yuan, R. DNA restriction enzyme from E. coli. Nature 1968, 217, 1110–1114. (7) Nielsen, J.; Fussenegger, M.; Keasling, J.; Lee, S. Y.; Liao, J. C.; Prather, K.; Palsson, B. Engineering

synergy in biotechnology. Nat. Chem. Biol. 2014, 10, 319–322. (8) Pirie, C. M.; De Mey, M.; Jones Prather, K. L.; Ajikumar, P. K. Integrating the protein and metabolic

engineering toolkits for next-generation chemical biosynthesis. ACS Chem. Biol. 2013, 8, 662–672. (9) Gibson, D. G.; Glass, J. I.; Lartigue, C.; Noskov, V. N.; Chuang, R.-Y.; Algire, M. A.; Benders, G. A.;

Montague, M. G.; Ma, L.; Moodie, M. M.; et al. Creation of a bacterial cell controlled by a chemically synthesized genome. Science 2010, 329, 52–56.

(10) Jiang, L.; Althoff, E. A.; Clemente, F. R.; Doyle, L.; Röthlisberger, D.; Zanghellini, A.; Gallaher, J. L.; Betker, J. L.; Tanaka, F.; Barbas, C. F.; et al. De novo computational design of retro-aldol enzymes. Science 2008, 319, 1387–1391.

(11) Bailey, J. E. Toward a science of metabolic engineering. Science 1991, 252, 1668–1675. (12) Ghisalba, O.; Meyer, H.-P.; Wohlgemuth, R.; Flickinger, M. C. Industrial Biotransformation. In

Encyclopedia of Industrial Biotechnology; John Wiley & Sons, Inc., 2009. (13) Yang, Y.-T.; Bennett, G. N.; San, K.-Y. Genetic and metabolic engineering. Electron. J. Biotechnol. 1998, 1,

134–141. (14) Stephanopoulos, G. Synthetic biology and metabolic engineering. ACS Synth. Biol. 2012, 1, 514–525. (15) Smolke, C. The Metabolic Pathway Engineering Handbook: Fundamentals; 1 edition.; CRC Press: Boca

Raton, 2009. (16) Alexeyev, M. F.; Shokolenko, I. N.; Croughan, T. P. New mini-Tn5 derivatives for insertion mutagenesis

and genetic engineering in gram-negative bacteria. Can. J. Microbiol. 1995, 41, 1053–1055. (17) Zhang, Y.-X.; Perry, K.; Vinci, V. A.; Powell, K.; Stemmer, W. P. C.; del Cardayré, S. B. Genome shuffling

leads to rapid phenotypic improvement in bacteria. Nature 2002, 415, 644–646. (18) Thykaer, J.; Nielsen, J. Metabolic engineering of beta-lactam production. Metab. Eng. 2003, 5, 56–69. (19) De Graaf, A. A.; Eggeling, L.; Sahm, H. Metabolic engineering for L-lysine production by

Corynebacterium glutamicum. Adv. Biochem. Eng. Biotechnol. 2001, 73, 9–29. (20) Dhamankar, H.; Prather, K. L. J. Microbial chemical factories: recent advances in pathway engineering

for synthesis of value added chemicals. Curr. Opin. Struct. Biol. 2011, 21, 488–494. (21) Copeland, W. B.; Bartley, B. A.; Chandran, D.; Galdzicki, M.; Kim, K. H.; Sleight, S. C.; Maranas, C. D.;

Sauro, H. M. Computational tools for metabolic engineering. Metab. Eng. 2012, 14, 270–280. (22) Medema, M. H.; van Raaphorst, R.; Takano, E.; Breitling, R. Computational tools for the synthetic

design of biochemical pathways. Nat. Rev. Microbiol. 2012, 10, 191–202. (23) Santos, C. N. S.; Stephanopoulos, G. Combinatorial engineering of microbes for optimizing cellular

phenotype. Curr. Opin. Chem. Biol. 2008, 12, 168–176. (24) Hatzimanikatis, V.; Li, C.; Ionita, J. A.; Henry, C. S.; Jankowski, M. D.; Broadbelt, L. J. Exploring the

diversity of complex metabolic networks. Bioinforma. Oxf. Engl. 2005, 21, 1603–1609. (25) Funahashi, A.; Matsuoka, Y.; Jouraku, A.; Morohashi, M.; Kikuchi, N.; Kitano, H. CellDesigner 3.5: A

Versatile Modeling Tool for Biochemical Networks. Proc. IEEE 2008, 96, 1254–1265. (26) Shannon, P.; Markiel, A.; Ozier, O.; Baliga, N. S.; Wang, J. T.; Ramage, D.; Amin, N.; Schwikowski, B.;

Ideker, T. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003, 13, 2498–2504.

(27) Tang, W. L.; Zhao, H. Industrial biotechnology: tools and applications. Biotechnol. J. 2009, 4, 1725–1739.

(28) Stephanopoulos, G. Metabolic fluxes and metabolic engineering. Metab. Eng. 1999, 1, 1–11. (29) Meyer, A.; Pellaux, R.; Panke, S. Bioengineering novel in vitro metabolic pathways using synthetic

biology. Curr. Opin. Microbiol. 2007, 10, 246–253.

References

- 149 -

(30) Becker, J.; Klopprogge, C.; Herold, A.; Zelder, O.; Bolten, C. J.; Wittmann, C. Metabolic flux engineering of L-lysine production in Corynebacterium glutamicum--over expression and modification of G6P dehydrogenase. J. Biotechnol. 2007, 132, 99–109.

(31) Schellenberger, J.; Que, R.; Fleming, R. M. T.; Thiele, I.; Orth, J. D.; Feist, A. M.; Zielinski, D. C.; Bordbar, A.; Lewis, N. E.; Rahmanian, S.; et al. Quantitative prediction of cellular metabolism with constraint-based models: the COBRA Toolbox v2.0. Nat. Protoc. 2011, 6, 1290–1307.

(32) Rocha, I.; Maia, P.; Evangelista, P.; Vilaça, P.; Soares, S.; Pinto, J. P.; Nielsen, J.; Patil, K. R.; Ferreira, E. C.; Rocha, M. OptFlux: an open-source software platform for in silico metabolic engineering. BMC Syst. Biol. 2010, 4, 45.

(33) Burgard, A. P.; Pharkya, P.; Maranas, C. D. Optknock: a bilevel programming framework for identifying gene knockout strategies for microbial strain optimization. Biotechnol. Bioeng. 2003, 84, 647–657.

(34) Parachin, N. S.; Bergdahl, B.; van Niel, E. W. J.; Gorwa-Grauslund, M. F. Kinetic modelling reveals current limitations in the production of ethanol from xylose by recombinant Saccharomyces cerevisiae. Metab. Eng. 2011, 13, 508–517.

(35) Costa, R. S.; Machado, D.; Rocha, I.; Ferreira, E. C. Critical perspective on the consequences of the limited availability of kinetic data in metabolic dynamic modelling. IET Syst. Biol. 2011, 5, 157–163.

(36) Van Eunen, K.; Kiewiet, J. A. L.; Westerhoff, H. V.; Bakker, B. M. Testing biochemistry revisited: how in vivo metabolism can be understood from in vitro enzyme kinetics. PLoS Comput. Biol. 2012, 8, e1002483.

(37) Mendes, P.; Hoops, S.; Sahle, S.; Gauges, R.; Dada, J.; Kummer, U. Computational modeling of biochemical networks using COPASI. Methods Mol. Biol. Clifton NJ 2009, 500, 17–59.

(38) Tomita, null; Hashimoto, null; Takahashi, null; Shimizu, null; Matsuzaki, null; Miyoshi, null; Saito, null; Tanida, null; Yugi, null; Venter, null; et al. E-CELL: Software Environment for Whole Cell Simulation. Genome Inform. Workshop Genome Inform. 1997, 8, 147–155.

(39) Ma, S. M.; Garcia, D. E.; Redding-Johanson, A. M.; Friedland, G. D.; Chan, R.; Batth, T. S.; Haliburton, J. R.; Chivian, D.; Keasling, J. D.; Petzold, C. J.; et al. Optimization of a heterologous mevalonate pathway through the use of variant HMG-CoA reductases. Metab. Eng. 2011, 13, 588–597.

(40) Koffas, M. A. G.; Jung, G. Y.; Stephanopoulos, G. Engineering metabolism and product formation in Corynebacterium glutamicum by coordinated gene overexpression. Metab. Eng. 2003, 5, 32–41.

(41) Datsenko, K. A.; Wanner, B. L. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci. U. S. A. 2000, 97, 6640–6645.

(42) Martínez-García, E.; de Lorenzo, V. Transposon-based and plasmid-based genetic tools for editing genomes of gram-negative bacteria. Methods Mol. Biol. Clifton NJ 2012, 813, 267–283.

(43) Blazeck, J.; Alper, H. S. Promoter engineering: recent advances in controlling transcription at the most fundamental level. Biotechnol. J. 2013, 8, 46–58.

(44) Pfleger, B. F.; Pitera, D. J.; Smolke, C. D.; Keasling, J. D. Combinatorial engineering of intergenic regions in operons tunes expression of multiple genes. Nat. Biotechnol. 2006, 24, 1027–1032.

(45) Hebert, C. G.; Valdes, J. J.; Bentley, W. E. Beyond silencing--engineering applications of RNA interference and antisense technology for altering cellular phenotype. Curr. Opin. Biotechnol. 2008, 19, 500–505.

(46) Salis, H. M.; Mirsky, E. A.; Voigt, C. A. Automated design of synthetic ribosome binding sites to control protein expression. Nat. Biotechnol. 2009, 27, 946–950.

(47) Dueber, J. E.; Wu, G. C.; Malmirchegini, G. R.; Moon, T. S.; Petzold, C. J.; Ullal, A. V.; Prather, K. L. J.; Keasling, J. D. Synthetic protein scaffolds provide modular control over metabolic flux. Nat. Biotechnol. 2009, 27, 753–759.

(48) You, C.; Myung, S.; Zhang, Y.-H. P. Facilitated substrate channeling in a self-assembled trifunctional enzyme complex. Angew. Chem. Int. Ed Engl. 2012, 51, 8787–8790.

(49) Tolia, N. H.; Joshua-Tor, L. Strategies for protein coexpression in Escherichia coli. Nat. Methods 2006, 3, 55–64.

(50) Nawabi, P.; Bauer, S.; Kyrpides, N.; Lykidis, A. Engineering Escherichia coli for biodiesel production utilizing a bacterial fatty acid methyltransferase. Appl. Environ. Microbiol. 2011, 77, 8052–8061.

(51) Zhang, C.; Liu, L.; Teng, L.; Chen, J.; Liu, J.; Li, J.; Du, G.; Chen, J. Metabolic engineering of Escherichia coli BL21 for biosynthesis of heparosan, a bioengineered heparin precursor. Metab. Eng. 2012, 14, 521–527.

(52) Wu, J.; Du, G.; Zhou, J.; Chen, J. Metabolic engineering of Escherichia coli for (2S)-pinocembrin production from glucose by a modular metabolic strategy. Metab. Eng. 2013, 16, 48–55.

(53) Xu, P.; Vansiri, A.; Bhan, N.; Koffas, M. A. G. ePathBrick: a synthetic biology platform for engineering metabolic pathways in E. coli. ACS Synth. Biol. 2012, 1, 256–266.

(54) Silva-Rocha, R.; Martínez-García, E.; Calles, B.; Chavarría, M.; Arce-Rodríguez, A.; de Las Heras, A.; Páez-Espino, A. D.; Durante-Rodríguez, G.; Kim, J.; Nikel, P. I.; et al. The Standard European Vector

References

- 150 -

Architecture (SEVA): a coherent platform for the analysis and deployment of complex prokaryotic phenotypes. Nucleic Acids Res. 2013, 41, D666–D675.

(55) Herrero, M.; de Lorenzo, V.; Timmis, K. N. Transposon vectors containing non-antibiotic resistance selection markers for cloning and stable chromosomal insertion of foreign genes in gram-negative bacteria. J. Bacteriol. 1990, 172, 6557–6567.

(56) Green, M. R. Molecular Cloning: A Laboratory Manual (Fourth Edition): Three-volume set; 4th edition.; Cold Spring Harbor Laboratory Press: Cold Spring Harbor, N.Y, 2012.

(57) Boyle, P. M.; Silver, P. A. Parts plus pipes: synthetic biology approaches to metabolic engineering. Metab. Eng. 2012, 14, 223–232.

(58) Glick, B. R. Metabolic load and heterologous gene expression. Biotechnol. Adv. 1995, 13, 247–261. (59) San, K.-Y.; Bennett, G. N.; Berríos-Rivera, S. J.; Vadali, R. V.; Yang, Y.-T.; Horton, E.; Rudolph, F. B.;

Sariyar, B.; Blackwood, K. Metabolic engineering through cofactor manipulation and its effects on metabolic flux redistribution in Escherichia coli. Metab. Eng. 2002, 4, 182–192.

(60) Santos, C. N. S.; Yoshikuni, Y. Engineering complex biological systems in bacteria through recombinase-assisted genome engineering. Nat. Protoc. 2014, 9, 1320–1336.

(61) St-Pierre, F.; Cui, L.; Priest, D. G.; Endy, D.; Dodd, I. B.; Shearwin, K. E. One-step cloning and chromosomal integration of DNA. ACS Synth. Biol. 2013, 2, 537–541.

(62) Choi, K.-H.; Gaynor, J. B.; White, K. G.; Lopez, C.; Bosio, C. M.; Karkhoff-Schweizer, R. R.; Schweizer, H. P. A Tn7-based broad-range bacterial cloning and expression system. Nat. Methods 2005, 2, 443–448.

(63) Loeschcke, A.; Markert, A.; Wilhelm, S.; Wirtz, A.; Rosenau, F.; Jaeger, K.-E.; Drepper, T. TREX: a universal tool for the transfer and expression of biosynthetic pathways in bacteria. ACS Synth. Biol. 2013, 2, 22–33.

(64) Lim, H. N.; Lee, Y.; Hussein, R. Fundamental relationship between operon organization and gene expression. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 10626–10631.

(65) Tyo, K. E. J.; Ajikumar, P. K.; Stephanopoulos, G. Stabilized gene duplication enables long-term selection-free heterologous pathway expression. Nat. Biotechnol. 2009, 27, 760–765.

(66) Nakamura, C. E.; Whited, G. M. Metabolic engineering for the microbial production of 1,3-propanediol. Curr. Opin. Biotechnol. 2003, 14, 454–459.

(67) Hodgman, C. E.; Jewett, M. C. Cell-free synthetic biology: thinking outside the cell. Metab. Eng. 2012, 14, 261–269.

(68) Biggs, B. W.; De Paepe, B.; Santos, C. N. S.; De Mey, M.; Kumaran Ajikumar, P. Multivariate modular metabolic engineering for pathway and strain optimization. Curr. Opin. Biotechnol. 2014, 29C, 156–162.

(69) Ajikumar, P. K.; Xiao, W.-H.; Tyo, K. E. J.; Wang, Y.; Simeon, F.; Leonard, E.; Mucha, O.; Phon, T. H.; Pfeifer, B.; Stephanopoulos, G. Isoprenoid pathway optimization for Taxol precursor overproduction in Escherichia coli. Science 2010, 330, 70–74.

(70) Xu, P.; Gu, Q.; Wang, W.; Wong, L.; Bower, A. G. W.; Collins, C. H.; Koffas, M. A. G. Modular optimization of multi-gene pathways for fatty acids production in E. coli. Nat. Commun. 2013, 4, 1409.

(71) Elowitz, M. B.; Leibler, S. A synthetic oscillatory network of transcriptional regulators. Nature 2000, 403, 335–338.

(72) Gardner, T. S.; Cantor, C. R.; Collins, J. J. Construction of a genetic toggle switch in Escherichia coli. Nature 2000, 403, 339–342.

(73) Vilanova, C.; Porcar, M. iGEM 2.0--refoundations for engineering biology. Nat. Biotechnol. 2014, 32, 420–424.

(74) De Lorenzo, V.; Danchin, A. Synthetic biology: discovering new worlds and new words. EMBO Rep. 2008, 9, 822–827.

(75) Paddon, C. J.; Keasling, J. D. Semi-synthetic artemisinin: a model for the use of synthetic biology in pharmaceutical development. Nat. Rev. Microbiol. 2014, 12, 355–367.

(76) Kosuri, S.; Church, G. M. Large-scale de novo DNA synthesis: technologies and applications. Nat. Methods 2014, 11, 499–507.

(77) Røkke, G.; Korvald, E.; Pahr, J.; Oyås, O.; Lale, R. BioBrick assembly standards and techniques and associated software tools. Methods Mol. Biol. Clifton NJ 2014, 1116, 1–24.

(78) Daniel, R.; Rubens, J. R.; Sarpeshkar, R.; Lu, T. K. Synthetic analog computation in living cells. Nature 2013, 497, 619–623.

(79) An, W.; Chin, J. W. Synthesis of orthogonal transcription-translation networks. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 8477–8482.

(80) Glass, J. I.; Assad-Garcia, N.; Alperovich, N.; Yooseph, S.; Lewis, M. R.; Maruf, M.; Hutchison, C. A.; Smith, H. O.; Venter, J. C. Essential genes of a minimal bacterium. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 425–430.

References

- 151 -

(81) Dymond, J. S.; Richardson, S. M.; Coombes, C. E.; Babatz, T.; Muller, H.; Annaluru, N.; Blake, W. J.; Schwerzmann, J. W.; Dai, J.; Lindstrom, D. L.; et al. Synthetic chromosome arms function in yeast and generate phenotypic diversity by design. Nature 2011, 477, 471–476.

(82) Cheng, J. K.; Alper, H. S. The genome editing toolbox: a spectrum of approaches for targeted modification. Curr. Opin. Biotechnol. 2014, 30C, 87–94.

(83) Nakamura, H.; Shimomura-Shimizu, M.; Karube, I. Development of microbial sensors and their application. Adv. Biochem. Eng. Biotechnol. 2008, 109, 351–394.

(84) Silva-Rocha, R.; de Lorenzo, V. Engineering multicellular logic in bacteria with metabolic wires. ACS Synth. Biol. 2014, 3, 204–209.

(85) Siegel, J. B.; Zanghellini, A.; Lovick, H. M.; Kiss, G.; Lambert, A. R.; St Clair, J. L.; Gallaher, J. L.; Hilvert, D.; Gelb, M. H.; Stoddard, B. L.; et al. Computational design of an enzyme catalyst for a stereoselective bimolecular Diels-Alder reaction. Science 2010, 329, 309–313.

(86) Gibson, D. G.; Young, L.; Chuang, R.-Y.; Venter, J. C.; Hutchison, C. A.; Smith, H. O. Enzymatic assembly of DNA molecules up to several hundred kilobases. Nat. Methods 2009, 6, 343–345.

(87) Kok, S. de; Stanton, L. H.; Slaby, T.; Durot, M.; Holmes, V. F.; Patel, K. G.; Platt, D.; Shapland, E. B.; Serber, Z.; Dean, J.; et al. Rapid and Reliable DNA Assembly via Ligase Cycling Reaction. ACS Synth. Biol. 2014, 3, 97–106.

(88) Gustafsson, C.; Minshull, J.; Govindarajan, S.; Ness, J.; Villalobos, A.; Welch, M. Engineering genes for predictable protein expression. Protein Expr. Purif. 2012, 83, 37–46.

(89) Freemont, P. S.; Kitney, R. I. Synthetic Biology - A Primer; 1 edition.; World Scientific Publishing: Singapore ; London, 2012.

(90) Wang, K.; Schmied, W. H.; Chin, J. W. Reprogramming the genetic code: from triplet to quadruplet codes. Angew. Chem. Int. Ed Engl. 2012, 51, 2288–2297.

(91) Kim, J.; Copley, S. D. Inhibitory cross-talk upon introduction of a new metabolic pathway into an existing metabolic network. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, E2856–E2864.

(92) Long, C. P.; Antoniewicz, M. R. Metabolic flux analysis of Escherichia coli knockouts: lessons from the Keio collection and future outlook. Curr. Opin. Biotechnol. 2014, 28C, 127–133.

(93) Hold, C.; Panke, S. Towards the engineering of in vitro systems. J. R. Soc. Interface R. Soc. 2009, 6 Suppl 4, S507–S521.

(94) Santacoloma, P. A.; Sin, G.; Gernaey, K. V.; Woodley, J. M. Multienzyme-Catalyzed Processes: Next-Generation Biocatalysis. Org. Process Res. Dev. 2011, 15, 203–212.

(95) Ishii, N.; Suga, Y.; Hagiya, A.; Watanabe, H.; Mori, H.; Yoshino, M.; Tomita, M. Dynamic simulation of an in vitro multi-enzyme system. FEBS Lett. 2007, 581, 413–420.

(96) Van Hecke, W.; Bhagwat, A.; Ludwig, R.; Dewulf, J.; Haltrich, D.; Van Langenhove, H. Kinetic modeling of a bi-enzymatic system for efficient conversion of lactose to lactobionic acid. Biotechnol. Bioeng. 2009, 102, 1475–1482.

(97) Bujara, M.; Schümperli, M.; Billerbeck, S.; Heinemann, M.; Panke, S. Exploiting cell-free systems: Implementation and debugging of a system of biotransformations. Biotechnol. Bioeng. 2010, 106, 376–389.

(98) Martín del Campo, J. S.; Rollin, J.; Myung, S.; Chun, Y.; Chandrayan, S.; Patiño, R.; Adams, M. W. W.; Zhang, Y.-H. P. High-yield production of dihydrogen from xylose by using a synthetic enzyme cascade in a cell-free system. Angew. Chem. Int. Ed Engl. 2013, 52, 4587–4590.

(99) Sheldon, R. A. Cross-linked enzyme aggregates (CLEAs): stable and recyclable biocatalysts. Biochem. Soc. Trans. 2007, 35, 1583–1587.

(100) Scism, R. A.; Bachmann, B. O. Five-component cascade synthesis of nucleotide analogues in an engineered self-immobilized enzyme aggregate. Chembiochem Eur. J. Chem. Biol. 2010, 11, 67–70.

(101) Paddon, C. J.; Westfall, P. J.; Pitera, D. J.; Benjamin, K.; Fisher, K.; McPhee, D.; Leavell, M. D.; Tai, A.; Main, A.; Eng, D.; et al. High-level semi-synthetic production of the potent antimalarial artemisinin. Nature 2013, 496, 528–532.

(102) Bornscheuer, U. T.; Huisman, G. W.; Kazlauskas, R. J.; Lutz, S.; Moore, J. C.; Robins, K. Engineering the third wave of biocatalysis. Nature 2012, 485, 185–194.

(103) Davids, T.; Schmidt, M.; Böttcher, D.; Bornscheuer, U. T. Strategies for the discovery and engineering of enzymes for biocatalysis. Curr. Opin. Chem. Biol. 2013, 17, 215–220.

(104) Damborsky, J.; Brezovsky, J. Computational tools for designing and engineering enzymes. Curr. Opin. Chem. Biol. 2014, 19, 8–16.

(105) Röthlisberger, D.; Khersonsky, O.; Wollacott, A. M.; Jiang, L.; DeChancie, J.; Betker, J.; Gallaher, J. L.; Althoff, E. A.; Zanghellini, A.; Dym, O.; et al. Kemp elimination catalysts by computational enzyme design. Nature 2008, 453, 190–195.

(106) Brakmann, S. Discovery of superior enzymes by directed molecular evolution. Chembiochem Eur. J. Chem. Biol. 2001, 2, 865–871.

References

- 152 -

(107) Neylon, C. Chemical and biochemical strategies for the randomization of protein encoding DNA sequences: library construction methods for directed evolution. Nucleic Acids Res. 2004, 32, 1448–1459.

(108) Cadwell, R. C.; Joyce, G. F. Randomization of genes by PCR mutagenesis. PCR Methods Appl. 1992, 2, 28–33.

(109) Stemmer, W. P. Rapid evolution of a protein in vitro by DNA shuffling. Nature 1994, 370, 389–391. (110) Eriksen, D. T.; Lian, J.; Zhao, H. Protein design for pathway engineering. J. Struct. Biol. 2014, 185, 234–

242. (111) Hidalgo, A.; Schliessmann, A.; Molina, R.; Hermoso, J.; Bornscheuer, U. T. A one-pot, simple

methodology for cassette randomisation and recombination for focused directed evolution. Protein Eng. Des. Sel. PEDS 2008, 21, 567–576.

(112) Chaparro-Riggers, J. F.; Polizzi, K. M.; Bommarius, A. S. Better library design: data-driven protein engineering. Biotechnol. J. 2007, 2, 180–191.

(113) Pavelka, A.; Chovancova, E.; Damborsky, J. HotSpot Wizard: a web server for identification of hot spots in protein engineering. Nucleic Acids Res. 2009, 37, W376–W383.

(114) Jochens, H.; Bornscheuer, U. T. Natural diversity to guide focused directed evolution. Chembiochem Eur. J. Chem. Biol. 2010, 11, 1861–1866.

(115) Reetz, M. T.; Carballeira, J. D. Iterative saturation mutagenesis (ISM) for rapid directed evolution of functional enzymes. Nat. Protoc. 2007, 2, 891–903.

(116) Zhang, K.; Li, H.; Cho, K. M.; Liao, J. C. Expanding metabolism for total biosynthesis of the nonnatural amino acid L-homoalanine. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 6234–6239.

(117) Leonard, E.; Ajikumar, P. K.; Thayer, K.; Xiao, W.-H.; Mo, J. D.; Tidor, B.; Stephanopoulos, G.; Prather, K. L. J. Combining metabolic and protein engineering of a terpenoid biosynthetic pathway for overproduction and selectivity control. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 13654–13659.

(118) Hwang, I. Y.; Tan, M. H.; Koh, E.; Ho, C. L.; Poh, C. L.; Chang, M. W. Reprogramming microbes to be pathogen-seeking killers. ACS Synth. Biol. 2014, 3, 228–237.

(119) Wang, B. L.; Ghaderi, A.; Zhou, H.; Agresti, J.; Weitz, D. A.; Fink, G. R.; Stephanopoulos, G. Microfluidic high-throughput culturing of single cells for selection based on extracellular metabolite production or consumption. Nat. Biotechnol. 2014, 32, 473–478.

(120) Church, G. M.; Elowitz, M. B.; Smolke, C. D.; Voigt, C. A.; Weiss, R. Realizing the potential of synthetic biology. Nat. Rev. Mol. Cell Biol. 2014, 15, 289–294.

(121) Alexander, M. Biodegradation and Bioremediation, Second Edition; 2 edition.; Academic Press: San Diego, 1999.

(122) El Fantroussi, S.; Agathos, S. N. Is bioaugmentation a feasible strategy for pollutant removal and site remediation? Curr. Opin. Microbiol. 2005, 8, 268–275.

(123) Francis, C. W.; Mankin, J. B. High nitrate denitrification in continuous flow-stirred reactors. Water Res. 1977, 11, 289–294.

(124) Lovley, D. R.; Phillips, E. J. P.; Gorby, Y. A.; Landa, E. R. Microbial reduction of uranium. Nature 1991, 350, 413–416.

(125) Stucki, G.; Thueer, M. Experiences of a Large-Scale Application of 1,2-Dichloroethane Degrading Microorganisms for Groundwater Treatment. Environ. Sci. Technol. 1995, 29, 2339–2345.

(126) De Lorenzo, V. Recombinant bacteria for environmental release: what went wrong and what we have learnt from it. Clin. Microbiol. Infect. Off. Publ. Eur. Soc. Clin. Microbiol. Infect. Dis. 2009, 15 Suppl 1, 63–65.

(127) Ramos, J.-L.; Marqués, S.; van Dillewijn, P.; Espinosa-Urgel, M.; Segura, A.; Duque, E.; Krell, T.; Ramos-González, M.-I.; Bursakov, S.; Roca, A.; et al. Laboratory research aimed at closing the gaps in microbial bioremediation. Trends Biotechnol. 2011, 29, 641–647.

(128) Wittich, R.-M.; Dillewijn, P. van; Ramos*, J.-L. Rational Construction of Bacterial Strains with New/Improved Catabolic Capabilities for the Efficient Breakdown of Environmental Pollutants. In Handbook of Hydrocarbon and Lipid Microbiology; Timmis, K. N., Ed.; Springer Berlin Heidelberg, 2010; pp. 1247–1254.

(129) Kellogg, S. T.; Chatterjee, D. K.; Chakrabarty, A. M. Plasmid-assisted molecular breeding: new technique for enhanced biodegradation of persistent toxic chemicals. Science 1981, 214, 1133–1135.

(130) Devers, M.; Rouard, N.; Martin-Laurent, F. Fitness drift of an atrazine-degrading population under atrazine selection pressure. Environ. Microbiol. 2008, 10, 676–684.

(131) Nakatsu, C. H.; Korona, R.; Lenski, R. E.; de Bruijn, F. J.; Marsh, T. L.; Forney, L. J. Parallel and divergent genotypic evolution in experimental populations of Ralstonia sp. J. Bacteriol. 1998, 180, 4325–4331.

(132) Dai, M.; Copley, S. D. Genome shuffling improves degradation of the anthropogenic pesticide pentachlorophenol by Sphingobium chlorophenolicum ATCC 39723. Appl. Environ. Microbiol. 2004, 70, 2391–2397.

References

- 153 -

(133) De Lorenzo, V.; Timmis, K. N. Analysis and construction of stable phenotypes in gram-negative bacteria with Tn5- and Tn10-derived minitransposons. Methods Enzymol. 1994, 235, 386–405.

(134) Nikel, P. I.; Martínez-García, E.; de Lorenzo, V. Biotechnological domestication of pseudomonads using synthetic biology. Nat. Rev. Microbiol. 2014, 12, 368–379.

(135) Shrivastava, R.; Basu, B.; Godbole, A.; Mathew, M. K.; Apte, S. K.; Phale, P. S. Repression of the glucose-inducible outer-membrane protein OprB during utilization of aromatic compounds and organic acids in Pseudomonas putida CSV86. Microbiol. Read. Engl. 2011, 157, 1531–1540.

(136) Reineke, W.; Knackmuss, H. J. Microbial metabolism of haloaromatics: isolation and properties of a chlorobenzene-degrading bacterium. Appl. Environ. Microbiol. 1984, 47, 395–402.

(137) Brenner, V.; Arensdorf, J. J.; Focht, D. D. Genetic construction of PCB degraders. Biodegradation 1994, 5, 359–377.

(138) Nikel, P. I.; de Lorenzo, V. Implantation of unmarked regulatory and metabolic modules in Gram-negative bacteria with specialised mini-transposon delivery vectors. J. Biotechnol. 2013, 163, 143–154.

(139) Lehrbach, P. R.; Zeyer, J.; Reineke, W.; Knackmuss, H. J.; Timmis, K. N. Enzyme recruitment in vitro: use of cloned genes to extend the range of haloaromatics degraded by Pseudomonas sp. strain B13. J. Bacteriol. 1984, 158, 1025–1032.

(140) Rojo, F.; Pieper, D. H.; Engesser, K. H.; Knackmuss, H. J.; Timmis, K. N. Assemblage of ortho cleavage route for simultaneous degradation of chloro- and methylaromatics. Science 1987, 238, 1395–1398.

(141) Ramos, J. L.; Wasserfallen, A.; Rose, K.; Timmis, K. N. Redesigning metabolic routes: manipulation of TOL plasmid pathway for catabolism of alkylbenzoates. Science 1987, 235, 593–596.

(142) Jiang, J.; Zhang, R.; Li, R.; Gu, J.-D.; Li, S. Simultaneous biodegradation of methyl parathion and carbofuran by a genetically engineered microorganism constructed by mini-Tn5 transposon. Biodegradation 2007, 18, 403–412.

(143) Wittich, R.-M.; Wolff, P. Growth of the genetically engineered strain Cupriavidus necator RW112 with chlorobenzoates and technical chlorobiphenyls. Microbiol. Read. Engl. 2007, 153, 186–195.

(144) Walker, A. W.; Keasling, J. D. Metabolic engineering of Pseudomonas putida for the utilization of parathion as a carbon and energy source. Biotechnol. Bioeng. 2002, 78, 715–721.

(145) Haro, M. A.; de Lorenzo, V. Metabolic engineering of bacteria for environmental applications: construction of Pseudomonas strains for biodegradation of 2-chlorotoluene. J. Biotechnol. 2001, 85, 103–113.

(146) De la Peña Mattozzi, M.; Tehara, S. K.; Hong, T.; Keasling, J. D. Mineralization of paraoxon and its use as a sole C and P source by a rationally designed catabolic pathway in Pseudomonas putida. Appl. Environ. Microbiol. 2006, 72, 6699–6706.

(147) Reichmuth, D. S.; Blanch, H. W.; Keasling, J. D. Dibenzothiophene biodesulfurization pathway improvement using diagnostic GFP fusions. Biotechnol. Bioeng. 2004, 88, 94–99.

(148) Li, G.-Q.; Ma, T.; Li, S.-S.; Li, H.; Liang, F.-L.; Liu, R.-L. Improvement of dibenzothiophene desulfurization activity by removing the gene overlap in the dsz operon. Biosci. Biotechnol. Biochem. 2007, 71, 849–854.

(149) Sohn, S. B.; Kim, T. Y.; Park, J. M.; Lee, S. Y. In silico genome-scale metabolic analysis of Pseudomonas putida KT2440 for polyhydroxyalkanoate synthesis, degradation of aromatics and anaerobic survival. Biotechnol. J. 2010, 5, 739–750.

(150) Kim, J.; Oliveros, J. C.; Nikel, P. I.; de Lorenzo, V.; Silva-Rocha, R. Transcriptomic fingerprinting of Pseudomonas putida under alternative physiological regimes. Environ. Microbiol. Rep. 2013, 5, 883–891.

(151) Shapir, N.; Mongodin, E. F.; Sadowsky, M. J.; Daugherty, S. C.; Nelson, K. E.; Wackett, L. P. Evolution of catabolic pathways: Genomic insights into microbial s-triazine metabolism. J. Bacteriol. 2007, 189, 674–682.

(152) Poblete-Castro, I.; Binger, D.; Rodrigues, A.; Becker, J.; Martins Dos Santos, V. A. P.; Wittmann, C. In-silico-driven metabolic engineering of Pseudomonas putida for enhanced production of poly-hydroxyalkanoates. Metab. Eng. 2013, 15, 113–123.

(153) Izallalen, M.; Mahadevan, R.; Burgard, A.; Postier, B.; Didonato, R.; Sun, J.; Schilling, C. H.; Lovley, D. R. Geobacter sulfurreducens strain engineered for increased rates of respiration. Metab. Eng. 2008, 10, 267–275.

(154) Planson, A.-G.; Carbonell, P.; Paillard, E.; Pollet, N.; Faulon, J.-L. Compound toxicity screening and structure-activity relationship modeling in Escherichia coli. Biotechnol. Bioeng. 2012, 109, 846–850.

(155) Finley, S. D.; Broadbelt, L. J.; Hatzimanikatis, V. In silico feasibility of novel biodegradation pathways for 1,2,4-trichlorobenzene. BMC Syst. Biol. 2010, 4, 7.

(156) Brenner, K.; You, L.; Arnold, F. H. Engineering microbial consortia: a new frontier in synthetic biology. Trends Biotechnol. 2008, 26, 483–489.

References

- 154 -

(157) Gilbert, E. S.; Walker, A. W.; Keasling, J. D. A constructed microbial consortium for biodegradation of the organophosphorus insecticide parathion. Appl. Microbiol. Biotechnol. 2003, 61, 77–81.

(158) Phelan, V. V.; Liu, W.-T.; Pogliano, K.; Dorrestein, P. C. Microbial metabolic exchange--the chemotype-to-phenotype link. Nat. Chem. Biol. 2012, 8, 26–35.

(159) Sinha, J.; Reyes, S. J.; Gallivan, J. P. Reprogramming bacteria to seek and destroy an herbicide. Nat. Chem. Biol. 2010, 6, 464–470.

(160) Aso, Y.; Miyamoto, Y.; Harada, K. M.; Momma, K.; Kawai, S.; Hashimoto, W.; Mikami, B.; Murata, K. Engineered membrane superchannel improves bioremediation potential of dioxin-degrading bacteria. Nat. Biotechnol. 2006, 24, 188–189.

(161) Zhang, H.; Li, Q.; Ye, T.; Zhang, Z.; Li, L. Optimization of the whole-cell catalytic activity of recombinant Escherichia coli cells with surface-immobilized organophosphorus hydrolase. J. Environ. Biol. Acad. Environ. Biol. India 2013, 34, 315–319.

(162) Ba, S.; Arsenault, A.; Hassani, T.; Jones, J. P.; Cabana, H. Laccase immobilization and insolubilization: from fundamentals to applications for the elimination of emerging contaminants in wastewater treatment. Crit. Rev. Biotechnol. 2013, 33, 404–418.

(163) Zhao, H.-P.; Ontiveros-Valencia, A.; Tang, Y.; Kim, B. O.; Ilhan, Z. E.; Krajmalnik-Brown, R.; Rittmann, B. Using a two-stage hydrogen-based membrane biofilm reactor (MBfR) to achieve complete perchlorate reduction in the presence of nitrate and sulfate. Environ. Sci. Technol. 2013, 47, 1565–1572.

(164) Tong, H.-W.; Mutlu, B. R.; Wackett, L. P.; Aksan, A. Manufacturing of bioreactive nanofibers for bioremediation. Biotechnol. Bioeng. 2014, 111, 1483–1493.

(165) Geueke, B.; Garg, N.; Ghosh, S.; Fleischmann, T.; Holliger, C.; Lal, R.; Kohler, H.-P. E. Metabolomics of hexachlorocyclohexane (HCH) transformation: ratio of LinA to LinB determines metabolic fate of HCH isomers. Environ. Microbiol. 2013, 15, 1040–1049.

(166) Khersonsky, O.; Tawfik, D. S. Enzyme promiscuity: a mechanistic and evolutionary perspective. Annu. Rev. Biochem. 2010, 79, 471–505.

(167) Afriat, L.; Roodveldt, C.; Manco, G.; Tawfik, D. S. The latent promiscuity of newly identified microbial lactonases is linked to a recently diverged phosphotriesterase. Biochemistry (Mosc.) 2006, 45, 13677–13686.

(168) Parales, R. E.; Ditty, J. L. Laboratory evolution of catabolic enzymes and pathways. Curr. Opin. Biotechnol. 2005, 16, 315–325.

(169) Iwakiri, R.; Yoshihira, K.; Ngadiman, null; Futagami, T.; Goto, M.; Furukawa, K. Total degradation of pentachloroethane by an engineered Alcaligenes strain expressing a modified camphor monooxygenase and a hybrid dioxygenase. Biosci. Biotechnol. Biochem. 2004, 68, 1353–1356.

(170) Rui, L.; Kwon, Y. M.; Reardon, K. F.; Wood, T. K. Metabolic pathway engineering to enhance aerobic degradation of chlorinated ethenes and to reduce their toxicity by cloning a novel glutathione S-transferase, an evolved toluene o-monooxygenase, and gamma-glutamylcysteine synthetase. Environ. Microbiol. 2004, 6, 491–500.

(171) Lee, J.; Cao, L.; Ow, S. Y.; Barrios-Llerena, M. E.; Chen, W.; Wood, T. K.; Wright, P. C. Proteome changes after metabolic engineering to enhance aerobic mineralization of cis-1,2-dichloroethylene. J. Proteome Res. 2006, 5, 1388–1397.

(172) Strong, L. C.; McTavish, H.; Sadowsky, M. J.; Wackett, L. P. Field-scale remediation of atrazine-contaminated soil using recombinant Escherichia coli expressing atrazine chlorohydrolase. Environ. Microbiol. 2000, 2, 91–98.

(173) De Lorenzo, V. Systems biology approaches to bioremediation. Curr. Opin. Biotechnol. 2008, 19, 579–589.

(174) Liu, X.; Germaine, K. J.; Ryan, D.; Dowling, D. N. Genetically modified Pseudomonas biosensing biodegraders to detect PCB and chlorobenzoate bioavailability and biodegradation in contaminated soils. Bioeng. Bugs 2010, 1, 198–206.

(175) Liss, M.; Daubert, D.; Brunner, K.; Kliche, K.; Hammes, U.; Leiherer, A.; Wagner, R. Embedding permanent watermarks in synthetic genes. PloS One 2012, 7, e42465.

(176) Paul, D.; Pandey, G.; Jain, R. K. Suicidal genetically engineered microorganisms for bioremediation: need and perspectives. BioEssays News Rev. Mol. Cell. Dev. Biol. 2005, 27, 563–573.

(177) Aken, B. V.; Correa, P. A.; Schnoor, J. L. Phytoremediation of polychlorinated biphenyls: new trends and promises. Environ. Sci. Technol. 2010, 44, 2767–2776.

(178) Chakrabarti, A.; Miskovic, L.; Soh, K. C.; Hatzimanikatis, V. Towards kinetic modeling of genome-scale metabolic networks without sacrificing stoichiometric, thermodynamic and physiological constraints. Biotechnol. J. 2013, 8, 1043–1057.

(179) Nikel, P. I.; Silva-Rocha, R.; Benedetti, I.; de Lorenzo, V. The private life of environmental bacteria: pollutant biodegradation at the single cell level. Environ. Microbiol. 2014, 16, 628–642.

References

- 155 -

(180) Carson, R.; Lear, L.; Wilson, E. O. Silent Spring; Anniversary edition.; Houghton Mifflin Company: Boston, 2002.

(181) Oberg, G. The natural chlorine cycle--fitting the scattered pieces. Appl. Microbiol. Biotechnol. 2002, 58, 565–581.

(182) Janssen, D. B.; Dinkla, I. J. T.; Poelarends, G. J.; Terpstra, P. Bacterial degradation of xenobiotic compounds: evolution and distribution of novel enzyme activities. Environ. Microbiol. 2005, 7, 1868–1882.

(183) The Dow Chemical Company. Product Safety Assessment: 1,2,3-Trichloropropane; The Dow Chemical Company, 2008.

(184) Miermans, C. J.; van der Velde, L. E.; Frintrop, P. C. Analysis of volatile organic compounds, using the purge and trap injector coupled to a gas chromatograph/ion-trap mass spectrometer: review of the results in Dutch surface water of the Rhine, Meuse, Northern Delta Area and Westerscheldt, over the period 1992-1997. Chemosphere 2000, 40, 39–48.

(185) Liska, I.; Barceló, D.; Grasserbauer, M. Strategy for the screening of organic pollutants in a river basin — an overview of the Nitra river monitoring programme. TrAC Trends Anal. Chem. 1996, 15, 326–334.

(186) U.S. Environmental Protection Agency. Emerging Contaminant – 1,2,3-Trichloropropane (TCP); U.S. Environmental Protection Agency, 2009.

(187) Yamamoto, K.; Fukushima, M.; Kakutani, N.; Kuroda, K. Volatile organic compounds in urban rivers and their estuaries in Osaka, Japan. Environ. Pollut. Barking Essex 1987 1997, 95, 135–143.

(188) U.S. Environmental Protection Agency. Toxicological Review of 1,2,3-Trichloropropane; CAS No. 96-18-4; U.S. Environmental Protection Agency: Washington, DC, 2009.

(189) La, D. K.; Schoonhoven, R.; Ito, N.; Swenberg, J. A. The effects of exposure route on DNA adduct formation and cellular proliferation by 1,2,3-trichloropropane. Toxicol. Appl. Pharmacol. 1996, 140, 108–114.

(190) Tafazoli, M.; Kirsch-Volders, M. In vitro mutagenicity and genotoxicity study of 1,2-dichloroethylene, 1,1,2-trichloroethane, 1,3-dichloropropane, 1,2,3-trichloropropane and 1,1,3-trichloropropene, using the micronucleus test and the alkaline single cell gel electrophoresis technique (comet assay) in human lymphocytes. Mutat. Res. 1996, 371, 185–202.

(191) Liu, P.; Liang, Y.-G.; Meng, Q.-Y.; Zhang, C.-G.; Wang, H.-C.; Zhang, X.-G.; Li, G.; Liu, Z.-Y.; He, Y.-Z. Successful therapy with hemoperfusion and plasma exchange in acute 1,2,3-trichloropropane poisoning. Hum. Exp. Toxicol. 2012, 31, 523–527.

(192) Tardiff, R. G.; Carson, M. L. Derivation of a reference dose and drinking water equivalent level for 1,2,3-trichloropropane. Food Chem. Toxicol. Int. J. Publ. Br. Ind. Biol. Res. Assoc. 2010, 48, 1488–1510.

(193) Pagan, M.; Cooper, W. J.; Joens, J. A. Kinetic studies of the homogeneous abiotic reactions of several chlorinated aliphatic compounds in aqueous solution. Appl. Geochem. 1998, 13, 779–785.

(194) Salter-Blanc, A. J.; Tratnyek, P. G. Effects of solution chemistry on the dechlorination of 1,2,3-trichloropropane by zero-valent zinc. Environ. Sci. Technol. 2011, 45, 4073–4079.

(195) Samin, G.; Janssen, D. B. Transformation and biodegradation of 1,2,3-trichloropropane (TCP). Environ. Sci. Pollut. Res. Int. 2012, 19, 3067–3078.

(196) Tetko, I. V.; Tanchuk, V. Y. Application of associative neural networks for prediction of lipophilicity in ALOGPS 2.1 program. J. Chem. Inf. Comput. Sci. 2002, 42, 1136–1145.

(197) Yan, J.; Rash, B. A.; Rainey, F. A.; Moe, W. M. Isolation of novel bacteria within the Chloroflexi capable of reductive dechlorination of 1,2,3-trichloropropane. Environ. Microbiol. 2009, 11, 833–843.

(198) Bowman, K. S.; Nobre, M. F.; da Costa, M. S.; Rainey, F. A.; Moe, W. M. Dehalogenimonas alkenigignens sp. nov., a chlorinated-alkane-dehalogenating bacterium isolated from groundwater. Int. J. Syst. Evol. Microbiol. 2013, 63, 1492–1498.

(199) Bosma, T.; Janssen, D. B. Conversion of chlorinated propanes by Methylosinus trichosporium OB3b expressing soluble methane monooxygenase. Appl. Microbiol. Biotechnol. 1998, 50, 105–112.

(200) Dolfing, J.; Janssen, D. B. Estimates of Gibbs free energies of formation of chlorinated aliphatic compounds. Biodegradation 1994, 5, 21–28.

(201) Van Hylckama Vlieg, J. E.; Poelarends, G. J.; Mars, A. E.; Janssen, D. B. Detoxification of reactive intermediates during microbial metabolism of halogenated compounds. Curr. Opin. Microbiol. 2000, 3, 257–262.

(202) Nikel, P. I.; Pérez-Pantoja, D.; de Lorenzo, V. Why are chlorinated pollutants so difficult to degrade aerobically? Redox stress limits 1,3-dichloroprop-1-ene metabolism by Pseudomonas pavonaceae. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 2013, 368, 20120377.

(203) Crebelli, R.; Andreoli, C.; Carere, A.; Conti, L.; Crochi, B.; Cotta-Ramusino, M.; Benigni, R. Toxicology of halogenated aliphatic hydrocarbons: structural and molecular determinants for the disturbance of chromosome segregation and the induction of lipid peroxidation. Chem. Biol. Interact. 1995, 98, 113–129.

References

- 156 -

(204) Wijngaard, A. J. V. D.; Janssen, D. B.; Witholt, B. Degradation of Epichlorohydrin and Halohydrins by Bacterial Cultures Isolated from Freshwater Sediment. J. Gen. Microbiol. 1989, 135, 2199–2208.

(205) Janssen, D. B. Evolving haloalkane dehalogenases. Curr. Opin. Chem. Biol. 2004, 8, 150–159. (206) Koudelakova, T.; Bidmanova, S.; Dvorak, P.; Pavelka, A.; Chaloupkova, R.; Prokop, Z.; Damborsky, J.

Haloalkane dehalogenases: biotechnological applications. Biotechnol. J. 2013, 8, 32–45. (207) Koudelakova, T.; Chovancova, E.; Brezovsky, J.; Monincova, M.; Fortova, A.; Jarkovsky, J.; Damborsky, J.

Substrate specificity of haloalkane dehalogenases. Biochem. J. 2011, 435, 345–354. (208) Yokota, T.; Omori, T.; Kodama, T. Purification and properties of haloalkane dehalogenase from

Corynebacterium sp. strain m15-3. J. Bacteriol. 1987, 169, 4049–4054. (209) Kulakova, A. N.; Larkin, M. J.; Kulakov, L. A. The plasmid-located haloalkane dehalogenase gene from

Rhodococcus rhodochrous NCIMB 13064. Microbiol. Read. Engl. 1997, 143 ( Pt 1), 109–115. (210) Poelarends, G. J.; Zandstra, M.; Bosma, T.; Kulakov, L. A.; Larkin, M. J.; Marchesi, J. R.; Weightman, A. J.;

Janssen, D. B. Haloalkane-utilizing Rhodococcus strains isolated from geographically distinct locations possess a highly conserved gene cluster encoding haloalkane catabolism. J. Bacteriol. 2000, 182, 2725–2731.

(211) Bosma, T.; Kruizinga, E.; de Bruin, E. J.; Poelarends, G. J.; Janssen, D. B. Utilization of trihalogenated propanes by Agrobacterium radiobacter AD1 through heterologous expression of the haloalkane dehalogenase from Rhodococcus sp. strain M15-3. Appl. Environ. Microbiol. 1999, 65, 4575–4581.

(212) Newman, J.; Peat, T. S.; Richard, R.; Kan, L.; Swanson, P. E.; Affholter, J. A.; Holmes, I. H.; Schindler, J. F.; Unkefer, C. J.; Terwilliger, T. C. Haloalkane dehalogenases: structure of a Rhodococcus enzyme. Biochemistry (Mosc.) 1999, 38, 16105–16114.

(213) Bosma, T.; Damborský, J.; Stucki, G.; Janssen, D. B. Biodegradation of 1,2,3-trichloropropane through directed evolution and heterologous expression of a haloalkane dehalogenase gene. Appl. Environ. Microbiol. 2002, 68, 3582–3587.

(214) Rink, R.; Fennema, M.; Smids, M.; Dehmel, U.; Janssen, D. B. Primary structure and catalytic mechanism of the epoxide hydrolase from Agrobacterium radiobacter AD1. J. Biol. Chem. 1997, 272, 14650–14657.

(215) Nardini, M.; Ridder, I. S.; Rozeboom, H. J.; Kalk, K. H.; Rink, R.; Janssen, D. B.; Dijkstra, B. W. The x-ray structure of epoxide hydrolase from Agrobacterium radiobacter AD1. An enzyme to detoxify harmful epoxides. J. Biol. Chem. 1999, 274, 14579–14586.

(216) Van Hylckama Vlieg, J. E.; Tang, L.; Lutje Spelberg, J. H.; Smilda, T.; Poelarends, G. J.; Bosma, T.; van Merode, A. E.; Fraaije, M. W.; Janssen, D. B. Halohydrin dehalogenases are structurally and mechanistically related to short-chain dehydrogenases/reductases. J. Bacteriol. 2001, 183, 5058–5066.

(217) De Jong, R. M.; Tiesinga, J. J. W.; Rozeboom, H. J.; Kalk, K. H.; Tang, L.; Janssen, D. B.; Dijkstra, B. W. Structure and mechanism of a bacterial haloalcohol dehalogenase: a new variation of the short-chain dehydrogenase/reductase fold without an NAD(P)H binding site. EMBO J. 2003, 22, 4933–4944.

(218) Tang, L.; Lutje Spelberg, J. H.; Fraaije, M. W.; Janssen, D. B. Kinetic mechanism and enantioselectivity of halohydrin dehalogenase from Agrobacterium radiobacter. Biochemistry (Mosc.) 2003, 42, 5378–5386.

(219) Tang, L.; van Hylckama Vlieg, J. E. T.; Lutje Spelberg, J. H.; Fraaije, M. W.; Janssen, D. B. Improved stability of halohydrin dehalogenase from Agrobacterium radiobacter AD1 by replacement of cysteine residues. Enzyme Microb. Technol. 2002, 30, 251–258.

(220) Tang, L.; Torres Pazmiño, D. E.; Fraaije, M. W.; de Jong, R. M.; Dijkstra, B. W.; Janssen, D. B. Improved catalytic properties of halohydrin dehalogenase by modification of the halide-binding site. Biochemistry (Mosc.) 2005, 44, 6609–6618.

(221) Rui, L.; Cao, L.; Chen, W.; Reardon, K. F.; Wood, T. K. Active site engineering of the epoxide hydrolase from Agrobacterium radiobacter AD1 to enhance aerobic mineralization of cis-1,2-dichloroethylene in cells expressing an evolved toluene ortho-monooxygenase. J. Biol. Chem. 2004, 279, 46810–46817.

(222) Rui, L.; Cao, L.; Chen, W.; Reardon, K. F.; Wood, T. K. Protein engineering of epoxide hydrolase from Agrobacterium radiobacter AD1 for enhanced activity and enantioselective production of (R)-1-phenylethane-1,2-diol. Appl. Environ. Microbiol. 2005, 71, 3995–4003.

(223) Fox, R. J.; Davis, S. C.; Mundorff, E. C.; Newman, L. M.; Gavrilovic, V.; Ma, S. K.; Chung, L. M.; Ching, C.; Tam, S.; Muley, S.; et al. Improving catalytic function by ProSAR-driven enzyme evolution. Nat. Biotechnol. 2007, 25, 338–344.

(224) Banás, P.; Otyepka, M.; Jerábek, P.; Petrek, M.; Damborský, J. Mechanism of enhanced conversion of 1,2,3-trichloropropane by mutant haloalkane dehalogenase revealed by molecular modeling. J. Comput. Aided Mol. Des. 2006, 20, 375–383.

(225) Pavlova, M.; Klvana, M.; Prokop, Z.; Chaloupkova, R.; Banas, P.; Otyepka, M.; Wade, R. C.; Tsuda, M.; Nagata, Y.; Damborsky, J. Redesigning dehalogenase access tunnels as a strategy for degrading an anthropogenic substrate. Nat. Chem. Biol. 2009, 5, 727–733.

References

- 157 -

(226) Van Leeuwen, J. G. E.; Wijma, H. J.; Floor, R. J.; van der Laan, J.-M.; Janssen, D. B. Directed evolution strategies for enantiocomplementary haloalkane dehalogenases: from chemical waste to enantiopure building blocks. Chembiochem Eur. J. Chem. Biol. 2012, 13, 137–148.

(227) Arif, M. I.; Samin, G.; van Leeuwen, J. G. E.; Oppentocht, J.; Janssen, D. B. Novel dehalogenase mechanism for 2,3-dichloro-1-propanol utilization in Pseudomonas putida strain MC4. Appl. Environ. Microbiol. 2012, 78, 6128–6136.

(228) Samin, G.; Pavlova, M.; Arif, M. I.; Postema, C. P.; Damborsky, J.; Janssen, D. B. A Pseudomonas putida Strain Genetically Engineered for 1,2,3-Trichloropropane Bioremediation. Appl. Environ. Microbiol. 2014, 80, 5467–5476.

(229) (IRIS), U. E. O. N. I. R. I. S. 1,2,3-Trichloropropane (CASRN 96-18-4) | IRIS | US EPA http://www.epa.gov/iris/subst/0200.htm (accessed Aug 15, 2014).

(230) Sarathy, V.; Salter, A. J.; Nurmi, J. T.; O’Brien Johnson, G.; Johnson, R. L.; Tratnyek, P. G. Degradation of 1,2,3-trichloropropane (TCP): hydrolysis, elimination, and reduction by iron and zinc. Environ. Sci. Technol. 2010, 44, 787–793.

(231) Copley, S. D. Evolution of efficient pathways for degradation of anthropogenic chemicals. Nat. Chem. Biol. 2009, 5, 559–566.

(232) Kurumbang, N. P.; Dvorak, P.; Bendl, J.; Brezovsky, J.; Prokop, Z.; Damborsky, J. Computer-assisted engineering of the synthetic pathway for biodegradation of a toxic persistent pollutant. ACS Synth. Biol. 2014, 3, 172–181.

(233) Cao, L.; van Rantwijk, F.; Sheldon, R. A. Cross-linked enzyme aggregates: a simple and effective method for the immobilization of penicillin acylase. Org. Lett. 2000, 2, 1361–1364.

(234) Immobilization of Enzymes and Cells; Guisan, J. M., Ed.; Methods in BiotechnologyTM; Humana Press: Totowa, NJ, 2006; Vol. 22.

(235) Drobchenko, S. N.; Isaeva-Ivanova, L. S.; Kleiner, A. R.; Lomakin, A. V.; Kolker, A. R.; Noskin, V. A. An investigation of the structure of periodate-oxidised dextran. Carbohydr. Res. 1993, 241, 189–199.

(236) Iwaji Iwasaki, S. U. New Colorimetric Determination of Chloride using Mercuric Thiocyanate and Ferric Ion. Bull. Chem. Soc. Jpn. - BULL CHEM SOC JPN 1952, 25, 226–226.

(237) Jacobs, M. H.; Van den Wijngaard, A. J.; Pentenga, M.; Janssen, D. B. Characterization of the epoxide hydrolase from an epichlorohydrin-degrading Pseudomonas sp. Eur. J. Biochem. FEBS 1991, 202, 1217–1222.

(238) Spelberg, L.; Harald, J. Enantioselective biocatalytic conversions of epoxides, University of Groningen, 2003.

(239) Ba, S.; Arsenault, A.; Hassani, T.; Jones, J. P.; Cabana, H. Laccase immobilization and insolubilization: from fundamentals to applications for the elimination of emerging contaminants in wastewater treatment. Crit. Rev. Biotechnol. 2013, 33, 404–418.

(240) Bidmanova, S.; Damborsky, J.; Prokop, Z. Immobilization of Haloalkane dehalogenase LinB from Sphingobium japonicum UT26 for Biotechnological Applications. J. Biocatal. Biotransformation 2013, 02.

(241) Jurchescu, I.-M.; Hamann, J.; Zhou, X.; Ortmann, T.; Kuenz, A.; Prüße, U.; Lang, S. Enhanced 2,3-butanediol production in fed-batch cultures of free and immobilized Bacillus licheniformis DSM 8785. Appl. Microbiol. Biotechnol. 2013, 97, 6715–6723.

(242) Cárdenas-Fernández, M.; Neto, W.; López, C.; Álvaro, G.; Tufvesson, P.; Woodley, J. M. Immobilization of Escherichia coli containing ω-transaminase activity in LentiKats®. Biotechnol. Prog. 2012, 28, 693–698.

(243) Nunes, M. A. P.; Fernandes, P. C. B.; Ribeiro, M. H. L. High-affinity water-soluble system for efficient naringinase immobilization in polyvinyl alcohol-dimethyl sulfoxide lens-shaped particles. J. Mol. Recognit. JMR 2012, 25, 580–594.

(244) Czichocki, G.; Dautzenberg, H.; Capan, E.; Vorlop, K.-D. New and effective entrapment of polyelectrolyte-enzyme-complexes in LentiKats. Biotechnol. Lett. 2001, 23, 1303–1307.

(245) Sheldon, R. A. Characteristic features and biotechnological applications of cross-linked enzyme aggregates (CLEAs). Appl. Microbiol. Biotechnol. 2011, 92, 467–477.

(246) Wilson, L.; Illanes, A.; Pessela, B. C. C.; Abian, O.; Fernández-Lafuente, R.; Guisán, J. M. Encapsulation of crosslinked penicillin G acylase aggregates in lentikats: evaluation of a novel biocatalyst in organic media. Biotechnol. Bioeng. 2004, 86, 558–562.

(247) Tran, D. N.; Balkus, K. J. Perspective of Recent Progress in Immobilization of Enzymes. ACS Catal. 2011, 1, 956–968.

(248) Bidmanova, S.; Hrdlickova, E.; Jaros, J.; Ilkovics, L.; Hampl, A.; Damborsky, J.; Prokop, Z. Microscopic monitoring provides information on structure and properties during biocatalyst immobilization. Biotechnol. J. 2014, 9, 852–860.

References

- 158 -

(249) Shah, S.; Sharma, A.; Gupta, M. N. Preparation of cross-linked enzyme aggregates by using bovine serum albumin as a proteic feeder. Anal. Biochem. 2006, 351, 207–213.

(250) Schoevaart, R.; Wolbers, M. W.; Golubovic, M.; Ottens, M.; Kieboom, A. P. G.; van Rantwijk, F.; van der Wielen, L. a. M.; Sheldon, R. A. Preparation, optimization, and structures of cross-linked enzyme aggregates (CLEAs). Biotechnol. Bioeng. 2004, 87, 754–762.

(251) Mateo, C.; Palomo, J. M.; van Langen, L. M.; van Rantwijk, F.; Sheldon, R. A. A new, mild cross-linking methodology to prepare cross-linked enzyme aggregates. Biotechnol. Bioeng. 2004, 86, 273–276.

(252) Montoro-García, S.; Gil-Ortiz, F.; Navarro-Fernández, J.; Rubio, V.; García-Carmona, F.; Sánchez-Ferrer, A. Improved cross-linked enzyme aggregates for the production of desacetyl beta-lactam antibiotics intermediates. Bioresour. Technol. 2010, 101, 331–336.

(253) Choi, S. H.; Kim, H. S.; Lee, I. S.; Lee, E. Y. Functional expression and magnetic nanoparticle-based Immobilization of a protein-engineered marine fish epoxide hydrolase of Mugil cephalus for enantioselective hydrolysis of racemic styrene oxide. Biotechnol. Lett. 2010, 32, 1685–1691.

(254) Yildirim, D.; Tükel, S. S.; Alagöz, D.; Alptekin, O. Preparative-scale kinetic resolution of racemic styrene oxide by immobilized epoxide hydrolase. Enzyme Microb. Technol. 2011, 49, 555–559.

(255) Ardao, I.; Zeng, A.-P. In silico evaluation of a complex multi-enzymatic system using one-pot and modular approaches: Application to the high-yield production of hydrogen from a synthetic metabolic pathway. Chem. Eng. Sci. 2013, 87, 183–193.

(256) Schoffelen, S.; van Hest, J. C. M. Chemical approaches for the construction of multi-enzyme reaction systems. Curr. Opin. Struct. Biol. 2013, 23, 613–621.

(257) Findrik, Z.; Vasić-Racki, D. Biotransformation of D-methionine into L-methionine in the cascade of four enzymes. Biotechnol. Bioeng. 2007, 98, 956–967.

(258) Itoh, A.; Ohashi, Y.; Soga, T.; Mori, H.; Nishioka, T.; Tomita, M. Application of capillary electrophoresis-mass spectrometry to synthetic in vitro glycolysis studies. Electrophoresis 2004, 25, 1996–2002.

(259) Lopez-Gallego, F.; Schmidt-Dannert, C. Multi-enzymatic synthesis. Curr. Opin. Chem. Biol. 2010, 14, 174–183.

(260) Billerbeck, S.; Härle, J.; Panke, S. The good of two worlds: increasing complexity in cell-free systems. Curr. Opin. Biotechnol. 2013, 24, 1037–1043.

(261) Bommarius, A. S.; Blum, J. K.; Abrahamson, M. J. Status of protein engineering for biocatalysts: how to design an industrially useful biocatalyst. Curr. Opin. Chem. Biol. 2011, 15, 194–200.

(262) Blackmond, D. G. Reaction progress kinetic analysis: a powerful methodology for mechanistic studies of complex catalytic reactions. Angew. Chem. Int. Ed Engl. 2005, 44, 4302–4320.

(263) Xue, R.; Woodley, J. M. Process technology for multi-enzymatic reaction systems. Bioresour. Technol. 2012, 115, 183–195.

(264) Reetz, M. T.; Carballeira, J. D.; Vogel, A. Iterative saturation mutagenesis on the basis of B factors as a strategy for increasing protein thermostability. Angew. Chem. Int. Ed Engl. 2006, 45, 7745–7751.

(265) Koudelakova, T.; Chaloupkova, R.; Brezovsky, J.; Prokop, Z.; Sebestova, E.; Hesseler, M.; Khabiri, M.; Plevaka, M.; Kulik, D.; Kuta Smatanova, I.; et al. Engineering enzyme stability and resistance to an organic cosolvent by modification of residues in the access tunnel. Angew. Chem. Int. Ed Engl. 2013, 52, 1959–1963.

(266) Lauterbach, L.; Lenz, O.; Vincent, K. A. H₂-driven cofactor regeneration with NAD(P)+-reducing hydrogenases. FEBS J. 2013, 280, 3058–3068.

(267) Bujara, M.; Schümperli, M.; Pellaux, R.; Heinemann, M.; Panke, S. Optimization of a blueprint for in vitro glycolysis by metabolic real-time analysis. Nat. Chem. Biol. 2011, 7, 271–277.

(268) Bhatt, P.; Kumar, M. S.; Mudliar, S.; Chakrabarti, T. Biodegradation of Chlorinated Compounds—A Review. Crit. Rev. Environ. Sci. Technol. 2007, 37, 165–198.

(269) Cámara, B.; Herrera, C.; González, M.; Couve, E.; Hofer, B.; Seeger, M. From PCBs to highly toxic metabolites by the biphenyl pathway. Environ. Microbiol. 2004, 6, 842–850.

(270) Pollmann, K.; Wray, V.; Pieper, D. H. Chloromethylmuconolactones as critical metabolites in the degradation of chloromethylcatechols: recalcitrance of 2-chlorotoluene. J. Bacteriol. 2005, 187, 2332–2340.

(271) Martínez, P.; Agulló, L.; Hernández, M.; Seeger, M. Chlorobenzoate inhibits growth and induces stress proteins in the PCB-degrading bacterium Burkholderia xenovorans LB400. Arch. Microbiol. 2007, 188, 289–297.

(272) Li, R.-D.; Li, Y.-Y.; Lu, L.-Y.; Ren, C.; Li, Y.-X.; Liu, L. An improved kinetic model for the acetone-butanol-ethanol pathway of Clostridium acetobutylicum and model-based perturbation analysis. BMC Syst. Biol. 2011, 5 Suppl 1, S12.

(273) Gray, K. A.; Richardson, T. H.; Kretz, K.; Short, J. M.; Bartnek, F.; Knowles, R.; Kan, L.; Swanson, P. E.; Robertson, D. E. Rapid Evolution of Reversible Denaturation and Elevated Melting Temperature in a Microbial Haloalkane Dehalogenase. Adv. Synth. Catal. 2001, 343, 607–617.

References

- 159 -

(274) Assis, H. M. S.; Sallis, P. J.; Bull, A. T.; Hardman, D. J. Biochemical Characterization of a Haloalcohol Dehalogenase from Arthrobacter erithii H10a. Enzyme Microb. Technol. 1998, 22, 568–574.

(275) Paliy, O.; Gunasekera, T. S. Growth of E. coli BL21 in minimal media with different gluconeogenic carbon sources and salt contents. Appl. Microbiol. Biotechnol. 2007, 73, 1169–1172.

(276) Paalme, T.; Elken, R.; Kahru, A.; Vanatalu, K.; Vilu, R. The growth rate control in Escherichia coli at near to maximum growth rates: the A-stat approach. Antonie Van Leeuwenhoek 1997, 71, 217–230.

(277) Diaz Ricci, J. C.; Hernández, M. E. Plasmid effects on Escherichia coli metabolism. Crit. Rev. Biotechnol. 2000, 20, 79–108.

(278) Ramchuran, S. O.; Holst, O.; Karlsson, E. N. Effect of postinduction nutrient feed composition and use of lactose as inducer during production of thermostable xylanase in Escherichia coli glucose-limited fed-batch cultivations. J. Biosci. Bioeng. 2005, 99, 477–484.

(279) Chavarría, M.; Nikel, P. I.; Pérez-Pantoja, D.; de Lorenzo, V. The Entner-Doudoroff pathway empowers Pseudomonas putida KT2440 with a high tolerance to oxidative stress. Environ. Microbiol. 2013, 15, 1772–1785.

(280) Martínez-García, E.; Jatsenko, T.; Kivisaar, M.; de Lorenzo, V. Freeing Pseudomonas putida KT2440 of its proviral load strengthens endurance to environmental stresses. Environ. Microbiol. 2014.

(281) Martínez-García, E.; Nikel, P. I.; Chavarría, M.; de Lorenzo, V. The metabolic cost of flagellar motion in Pseudomonas putida KT2440. Environ. Microbiol. 2014, 16, 291–303.

(282) Choi, K.-H.; Kumar, A.; Schweizer, H. P. A 10-min method for preparation of highly electrocompetent Pseudomonas aeruginosa cells: application for DNA fragment transfer between chromosomes and plasmid transformation. J. Microbiol. Methods 2006, 64, 391–397.

(283) Chizzolini, F.; Forlin, M.; Cecchi, D.; Mansy, S. S. Gene position more strongly influences cell-free protein expression from operons than T7 transcriptional promoter strength. ACS Synth. Biol. 2014, 3, 363–371.

(284) Nikel, P. I.; Kim, J.; de Lorenzo, V. Metabolic and regulatory rearrangements underlying glycerol metabolism in Pseudomonas putida KT2440. Environ. Microbiol. 2014, 16, 239–254.

(285) Schweizer, H. P.; Po, C. Regulation of glycerol metabolism in Pseudomonas aeruginosa: characterization of the glpR repressor gene. J. Bacteriol. 1996, 178, 5215–5221.

(286) Escapa, I. F.; del Cerro, C.; García, J. L.; Prieto, M. A. The role of GlpR repressor in Pseudomonas putida KT2440 growth and PHA production from glycerol. Environ. Microbiol. 2013, 15, 93–110.

(287) Yang, F.; Hanna, M. A.; Sun, R. Value-added uses for crude glycerol--a byproduct of biodiesel production. Biotechnol. Biofuels 2012, 5, 13.

(288) Wang, H. H.; Isaacs, F. J.; Carr, P. A.; Sun, Z. Z.; Xu, G.; Forest, C. R.; Church, G. M. Programming cells by multiplex genome engineering and accelerated evolution. Nature 2009, 460, 894–898.

Curriculum vitae

- 160 -

CURRICULUM VITAE

Name: Pavel Dvořák

Birth date: August 9, 1983

Place of birth: Kroměříž (Czech Republic)

Nationality: Czech

Affiliation:

Loschmidt Laboratories, Department of Experimental Biology, Faculty of Science, Masaryk

University, Kamenice 5/A13, 625 00 Brno, Czech Republic; phone: +420 602 377 908; FAX:

+420-5-49496302; e-mail: [email protected]

Education

2009 – now: Ph.D. student, Molecular and Cell Biology, Faculty of Science,

Masaryk University

2009: MSc. in Genetics and Molecular Biology, Faculty of Science, Masaryk

University (passed with honours)

2007: Bc. in Genetics and Molecular Biology, Faculty of Science, Masaryk

University (passed with honours)

Employment

2013-now: research assistant, Research Centre for Toxic Compounds in the

Environment RECETOX, Faculty of Science, Masaryk University

2011-now: Ph.D. student, International Clinical Research Center (FNUSA-ICRC),

Brno, Czech Republic

2009-2011: research assistant, Mendel's Centre for Education in Biology,

Biomedicine and Bioinformatics, Department of Biology, Medical Faculty,

Masaryk University

Research interests

biocatalysis; rational design and directed evolution of enzymes; mutagenesis,

screening and selection techniques; synthetic biology and metabolic engineering

of bacteria

Teaching activities

Loschmidt Laboratories Summer School of Protein Engineering; supervisor,

mentor and lecturer (lecture Molecular biology in protein engineering)

Molecular Biotechnology, lectures on Metabolic Engineering (in Czech),

Molecular Biotechnology Practice (in Czech), lecturer

Curriculum vitae

- 161 -

Academic stays

9 – 11/2013: Research stay at the Centro Nacional de Biotecnología, Madrid,

Spain, group of prof. Víctor de Lorenzo (contact person), techniques of genetic

engineering applied for metabolic engineering of bacteria.

12/2010: Research stay at the Wissenschafts Zentrum Straubing, Technische

Universität München (Germany), group of dr. Volker Sieber (contact person dr.

Jan K. Gutterl), robotic screening of mutant libraries generated by directed

evolution of industrially relevant enzymes.

International practical courses

6/2014: International Synthetic and Systems Biology Summer School, Taormina,

Sicily, Italy

11/2012: Advanced Course Metabolomics for Microbial Systems Biology, BSDL-

EDU, TU Delft, Delft, Netherlands

9/2010: Protein Engineering - Rational Design & Directed Evolution, Dechema

Summer School, Institute of Biochemistry, Greifswald University, Germany

Awards

2014: Award of the Dean of the Faculty of Science, Masaryk University for

exemplary representation of the faculty in abroad.

2014: The Sigma-Aldrich award of Gerty T. and Carl. F. Cori for the best young

scientist presentation, Interdisciplinary Meeting of Young Biologists, Biochemists,

and Chemists, Milovy, Czech Republic


organisation of The Student Scientific Conference on GMO Research


excellent students of master studies

2008: Award of the Czech Society for Biochemistry and Molecular Biology for the

best presentation at XII. Meeting of Biochemists and Molecular Biologists, Section

of Young Investigators, Brno, Czech Republic

Research projects

2010-2013: Specific Research, Category b) Support of specific research projects

focused on organization of student scientific conferences. Co-founder of student

club Biomania and organizer of The International Student Scientific Conference

on Biotechnology and Biomedicine (www.biomania.cz/conference-2013)

2008-2009: Masaryk University Rector's Program for Support of Creative

Activities of Students. Part A, Support of Excellent Diploma Theses. Semi-rational

design and construction of haloalkane dehalogenases for biotechnological

applications; 20081431A0006; principal investigator.

List of publications

- 162 -

LIST OF PUBLICATIONS

Dvorak, P., Bidmanova, S., Prokop, Z., Damborsky, J., 2014: Immobilized Synthetic

Pathway for Biodegradation of Toxic Recalcitrant Pollutant 1,2,3-Trichloropropane.

Environmental Science and Technology. 48, 6859-6866. (ACS Editors' Choice May 24 2014)

Dvorak, P., Kurumbang, N.P., Bendl, J., Brezovsky, J., Prokop, Z., Damborsky, J., 2014:

Maximizing the Efficiency of Multi-enzyme Processes by Stoichiometry Optimization.

ChemBioChem, DOI: 10.1002/cbic.201402265.

Kurumbang, N.P.*, Dvorak, P.*, Bendl, J., Brezovsky, J., Prokop, Z., Damborsky, J., 2014:

Computer-Assisted Engineering of the Synthetic Pathway for Biodegradation of a Toxic

Persistent Pollutant. ACS Synthetic Biology. 3: 172-181. (*shared first author)

Koudelakova, T., Bidmanova, T., Dvorak, P., Pavelka, A., Chaloupkova, R., Prokop, Z.,

Damborsky, J., 2013: Haloalkane Dehalogenases: Biotechnological Applications.

Biotechnology Journal. 8: 32-45.

Klvana, M., Pavlova, M., Koudelakova, T., Chaloupkova, R., Dvorak, P., Prokop, Z.,

Stsiapanava, A., Kuty, M., Kuta-Smatanova, I., Dohnalek, J., Kulhanek, P., Wade, R.C.,

Damborsky, J., 2009: Pathways and Mechanisms for Product Release in the Engineered

Haloalkane Dehalogenases Explored using Classical and Random Acceleration Molecular

Dynamics Simulations. Journal of Molecular Biology. 392(5): 1339-1356.

http://loschmidt.chemi.muni.cz/peg/abstracts/jmb09b.html



List of contributions at conferences and symposia

- 163 -

LIST OF CONTRIBUTIONS AT CONFERENCES AND SYMPOSIA

Dvorak, P., Kurumbang, N.P., Bendl, J., Brezovsky, J., Prokop, Z., Damborsky, J., Computer-

Assisted Engineering of Synthetic Pathway for Biodegradation of Anthropogenic Pollutant

Under In Vitro and In Vivo Conditions. International Synthetic and Systems Biology Summer

School, 15.-19.6.2014. Taormina, Italy (poster)

Dvorak, P., Kurumbang, N.P., Bendl, J., Brezovsky, J., Prokop, Z., Damborsky, J.,

Engineering of Metabolic Pathway for Biodegradation of Anthropogenic Pollutant Using

Methods of Synthetic Biology (in Czech). Interdisciplinary Meeting of Young Biologists,

Biochemists, and Chemists, 13.-16.5.2014. Milovy, Czech Republic (oral presentation)

Dvorak, P., Kurumbang, N.P., Bendl, J., Brezovsky, J., Prokop, Z., Damborsky, J., In Vitro and

In Silico Engineering of Multi-enzyme Reactions. Biotrans, 21.-25.7.2013. Manchester, UK

(oral presentation)

Dvorak, P., Kurumbang, N.P., Bendl, J., Brezovsky, J., Prokop, Z., Damborsky, J., Rational

Engineering of Synthetic Biodegradation Pathway. Symposium Consistent Bioprocess

Development, 1.3.2013. Technische Universität Berlin, Germany (invited lecture)

Dvorak, P., Prokop, Z., Bednar, D., Brezovsky, J., Bidmanova, S., Damborsky, J., In Vitro

Protein and Metabolic Engineering of Biodegradation Pathway. Biotrans, 2.-6.10.2011,

Giardini Naxos, Italy (oral presentation)

Dvorak, P., Pavlova, M., Klvana, M., Brezovsky, J., Chaloupkova, R., Volfova, I., Damborsky,

J., Increasing Activity of Bacterial Biocatalyst with Toxic Anthropogenic Substrate Using

Methods of Protein Engineering. 25th International Symposium on Microscale

Bioseparations, 21.3.-25.3.2010, Prague, Czech Republic (poster)

Dvorak, P., Pavlova, M., Klvana, M., Brezovsky, J., Chaloupkova, R., Prokop, Z., Damborsky,

J., Increasing the Activity of Haloalkane Dehalogenase DhaA with Anthropogenic Substrate

1,2-Dichloroethane Using Methods of Focused Directed Evolution. XX. Enzyme Egineering,

20.-24.9.2009, Gröningen, Netherlands (poster)

Dvorak, P., Pavlova, M., Klvana, M., Chaloupkova, R., Prokop, Z., Nagata, Y., Uhlirova, R.,

Brezovsky, J., Damborsky, J., Semi-rational Engineering of Haloalkane Dehalogenase DhaA

Towards Improved Activity with 1,2-Dichloroethane. XII. Meeting of Biochemists and

Molecular Biologists, Section of Young Investigators, 6.-7.2.2008, Brno, Czech Republic (oral

presentation)

- 164 -

- 165 -

Engineering of the synthetic metabolic pathway for ... - IS MUNI

Documents

Transcript of Engineering of the synthetic metabolic pathway for ... - IS MUNI