Two Lactococcus lactis thioredoxin paralogues play different roles in responses to arsenate and...

Downloaded from www.microbiologyresearch.org by

IP: 54.224.121.223

On: Mon, 06 Jun 2016 05:50:37

Two Lactococcus lactis thioredoxin paralogues playdifferent roles in responses to arsenate andoxidative stress

Petr Efler,1 Mogens Kilstrup,2 Stig Johnsen,1 Birte Svensson1

and Per Hagglund1

Correspondence

Mogens Kilstrup

[email protected]

Per Hagglund

[email protected]

Received 19 December 2014

Accepted 30 December 2014

1Enzyme and Protein Chemistry, Søltofts Plads Building 224, Department of Systems Biology,Technical University of Denmark, DK-2800 Kgs. Lyngby, Denmark

2Center for Systems Microbiology, Matematiktorvet Building 301, Department of Systems Biology,Technical University of Denmark, DK-2800 Kgs. Lyngby, Denmark

Thioredoxin (Trx) maintains intracellular thiol groups in a reduced state and is involved in a wide

range of cellular processes, including ribonucleotide reduction, sulphur assimilation, oxidative

stress responses and arsenate detoxification. The industrially important lactic acid bacterium

Lactococcus lactis contains two Trxs. TrxA is similar to the well-characterized Trx homologue from

Escherichia coli and contains the common WCGPC active site motif, while TrxD is atypical and

contains an aspartate residue in the active site (WCGDC). To elucidate the physiological roles of

the two Trx paralogues, deletion mutants DtrxA, DtrxD and DtrxADtrxD were constructed. In

general, the DtrxADtrxD strain was significantly more sensitive than either of the DtrxA and DtrxD

mutants. Upon exposure to oxidative stress, growth of the DtrxA strain was diminished while that

of the DtrxD mutant was similar to the wild-type. The lack of TrxA also appears to impair

methionine sulphoxide reduction. Both DtrxA and DtrxD strains displayed growth inhibition after

treatment with sodium arsenate and tellurite as compared with the wild-type, suggesting partially

overlapping functions of TrxA and TrxD. Overall the phenotype of the DtrxA mutant matches

established functions of WCGPC-type Trx while TrxD appears to play a more restricted role in

stress resistance of Lac. lactis.

INTRODUCTION

The ubiquitous thioredoxin (Trx) reduces target proteindisulphides and is recycled by NADPH-dependent thior-edoxin reductase (TrxR) through thiol–disulphide exchangereactions (Lu & Holmgren, 2014). The ‘classical’ Trx with aredox active CGPC motif was discovered 50 years ago asan electron donor to ribonucleotide reductase (RNR) inEscherichia coli (Laurent et al., 1964). Subsequently, Trx hasbeen demonstrated to provide reducing equivalents to awide range of substrates, including peroxiredoxin, methio-nine sulphoxide reductase and 39-phosphoadenylsulphatereductase (Arner & Holmgren, 2000). In Gram-positivebacteria such as Bacillus subtilis and Staphylococcus aureus,Trx-dependent arsenate reductase is part of the arsenic

resistance operon (Ji & Silver, 1992). In addition to thesespecific target proteins, Trx is also proposed to act as ageneral protein disulphide reductase maintaining intracel-lular thiol groups in a reduced state (Arner & Holmgren,2000).

Microbial genomes typically encode two or more Trxgenes. The physiological importance of Trx paralogueshave been investigated for example in Helicobacter pylori,Bacteroides fragilis, Mycobacterium tuberculosis and Lacto-bacillus casei, which contain two, six, three and four Trxgenes, respectively (Akif et al., 2008; Comtois et al., 2003;Reott et al., 2009; Serata et al., 2012). It is, however, notevident to what extent these Trx isoforms are redundant orhave unique activity profiles. In addition to the ‘classical’Trx1, E. coli contains Trx2 that provides electrons to RNRand is implicated in oxidative stress resistance (Miranda-Vizuete et al., 1997; Ritz et al., 2000). Bcl. subtilis lacks aTrx2 homologue but contains ORFs for at least eight non-essential Trx-like proteins besides the predominant ‘classical’TrxA (Kunst et al., 1997). Among these, the extracellularResA and StoA are involved in cytochrome c maturationand endospore cortex synthesis, respectively, and receive

Abbreviations: DIGE, difference gel electrophoresis; EP, exponentialphase; INT, iodonitrotetrazolium chloride; MetSO, methionine sulphoxide;Msr, MetSO reductase; RNR, ribonucleotide reductase; SP, stationaryphase; Trx, thioredoxin; TrxR, NADPH-dependent thioredoxin reductase;TV, tetrazolium violet.

Two supplementary figures and three supplementary tables are availablewith the online Supplementary Material.

Microbiology (2015), 161, 528–538 DOI 10.1099/mic.0.000029

528 000029 G 2015 The Authors Printed in Great Britain


IP: 54.224.121.223

On: Mon, 06 Jun 2016 05:50:37

electrons from TrxA via the membrane protein CcdA(Moller & Hederstedt, 2008). In addition, YtpP is pro-posed to be involved in amino acid metabolism (Berka et al.,2003).

Escherichia coli and several other Gram-negative bacteriautilize the disulphide reductase glutaredoxin coupled to thetripeptide glutathione (GSH) and glutathione reductaseas an alternative pathway to provide electrons for RNR andother target proteins (Lillig et al., 2008). In contrast, mostGram-positive bacteria lack GSH and deletion of genesencoding Trx and/or TrxR results in severe growth defectsor lethal phenotypes in Bcl. subtilis and S. aureus(Kobayashi et al., 2003; Moller & Hederstedt, 2008;Scharf et al., 1998; Uziel et al., 2004). Some species lackingGSH produce alternative low-molecular-mass thiols suchas mycothiol or bacillithiol but the biological significanceof these compounds has not yet been firmly established(Fahey et al., 1978; Newton et al., 1996, 2009).

Similar to most Gram-positive bacteria, the industriallyimportant lactic acid bacterium Lactococcus lactis lacks thebiosynthetic pathway for GSH but some strains can utilizeexogenously supplied GSH (Fernandes & Steele, 1993; Liet al., 2003; Newton et al., 1996). Furthermore, Lac. lactislacks catalase and is thus likely to rely on thiol-dependentperoxiredoxins and NADH peroxidase for reduction ofhydrogen peroxide (H2O2). Lac. lactis contains two Trxparalogues (TrxA, TrxD) and a glutaredoxin-like protein(NrdH), which acts as electron donor for a particular typeof RNR (class Ib) that is mainly found among prokaryotesand appears to be functionally important under micro-aerophilic conditions (Jordan et al., 1996). TrxA containsthe WCGPC active site motif and other conserved residuesimportant for Trx function. In contrast, TrxD displayslow similarity to TrxA and contains an unconventionalWCGDC active site motif (Bjornberg et al., 2014). Moreover,TrxD has a higher redox potential than TrxA and lacksactivity towards the model substrate insulin. TrxA, TrxDand NrdH are all recycled by the endogenous Lac. lactisTrxR (named TrxB) in vitro (Bjornberg et al., 2014).In contrast to TrxA and NrdH, TrxD is not efficientlyreduced by E. coli TrxR, suggesting significant differencesin substrate recognition between EcTrxR and LlTrxR.TrxR is important for oxidative stress resistance in Lac.lactis but not essential for viability under mild oxidativeconditions (Vido et al., 2005). In contrast, TrxR from therelated Lactobacillus casei is essential for aerobic growth(Serata et al., 2012).

Here the physiological roles of TrxA and TrxD in Lac. lactiswere investigated using strains lacking either one or bothTrx (DtrxA, DtrxD and DtrxADtrxD). Comparison ofgrowth rates of these mutant strains and wild-type afterexposure to various stress conditions suggests a partialoverlap in function between TrxA and TrxD. TrxA, however,appears to be of major importance for oxidative stressresistance whereas TrxD seems to play a role in arsenatedetoxification.

METHODS

Strains and growth conditions. The strains used in this study arelisted in Table 1. Unless stated otherwise, Lac. lactis strains weremaintained under aerobic conditions on agar plates containing M17medium (Difco) with 1 % (w/v) glucose (GM17), and grown in

chemically defined SA medium (Jensen & Hammer, 1993) containing1 % (w/v) glucose and 4 mg lipoic acid ml21 (GSAL medium). Toobtain synchronized balanced cultures, colonies from fresh GM17plates were inoculated into liquid GSAL medium, serially diluted(102, 103, 104, 105, 106) and grown under static conditions at 30 uCovernight. The dilution with exponentially growing cells (OD450 of 0.3–0.6) was used for further experiments. When performing phenotypescreening on solid GSAL media, synchronized exponentially growingovernight cultures were used for making serial dilutions (102, 103, 104,105) in pre-warmed GSAL medium in a 96-well plate. From each well,

10 ml was spotted on pre-warmed GSAL agar plates containing theparticular stress compound (0.5 mM Na2HAsO4 or 0.3 mM K2TeO4)and incubated at 30 uC for 24 h. Alternatively, single colonies freshlygrown on GM17 plates were streaked directly on GSAL agar platesfollowed by incubation at 30 uC for 72 h. E. coli MC1061 was grown in

Luria-Broth medium (LB) at 28, 30 or 37 uC. When relevant, LB wassupplemented with erythromycin (150 mg ml21) and GM17 byerythromycin (5 mg ml21) plus 1 % NaCl.

Bioscreen assays. A Bioscreen C instrument (Oy Growth Curves)was used to monitor growth of Lac. lactis wt and trx mutants exposedto a range of different stress conditions. Synchronized exponentiallygrowing cultures were diluted in preheated GSAL medium to an

OD450 of 0.01, and then 360 ml was mixed with 40 ml of a stresscompound solution (listed in Table S1, available in the online Sup-plementary Material) or H2O in a well of a pre-warmed honeycombplate. To monitor methionine sulphoxide (MetSO) assimilation, afreshly grown single colony from a GM17 plate was resuspended in

5 ml GSAL medium without methionine and diluted ten times in thesame medium. From this culture 360 ml aliquots were pipetted intowells of a pre-warmed honeycomb plate containing 40 ml of eithermethionine or MetSO at 1 mg ml21. The plates were incubated at30 uC without shaking. OD450 was monitored at 40 min intervals

with 10 s of intense shaking prior to measurements.

Construction of Lac. lactis DtrxA, DtrxD, DtrxADtrxD mutants

and complemented strains. DNA isolation, amplification and

cloning were performed according to standard procedures (Sambrook& Russell, 2001) or the manufacturers’ instructions. Upstream anddownstream regions flanking the trxA and trxD genes were amplifiedfrom genomic DNA of Lac. lactis subsp. cremoris MG1363 by PCR(deletion by overlap extension) using the primers listed in Table S2

and a HotStar HiFidelity PCR kit (Qiagen). The PCR products of theupstream and downstream regions were fused and used as templatesfor PCR using the forward primers for the upstream regions togetherwith the reverse primers for the downstream regions (Table S2). ThePCR products were digested with BamHI and XhoI and ligated into

pGHost4 (Appligene). The resulting plasmids were used to transformE. coli MC1061, and the correct sequences were confirmed by DNAsequencing (Eurofins). Plasmids were electroporated into Lac. lactisand the transformants were selected on GM17 plates containingerythromycin at 28 uC. After homologous recombination into the

chromosome, and clearing of the plasmid as described by Biswas et al.(1993), the deletions were confirmed by colony PCR amplificationusing the flanking primers binding to the chromosome outside thetargeted region (Table S2). The DtrxADtrxD double mutant wasprepared using the DtrxA strain as the template for homologous

recombination of DtrxD as described above. Complemented strainswere constructed by site-specific integration of trxA or trxDcontaining derivatives of plasmid pLB86 (Breuner et al., 2001) intothe phage attachment site (attB) of the DtrxD mutant MK556 or the

Different roles of two Lactococcus lactis thioredoxins

http://mic.sgmjournals.org 529


IP: 54.224.121.223

On: Mon, 06 Jun 2016 05:50:37

DtrxADtrxD double mutant MK562. The mutants were transformed

with plasmid pLB95 containing the gene encoding bacteriophage

TP901-1 integrase, as described by Breuner et al. (2001). The com-

plementing plasmids were constructed by inserting XhoI- and PstI-

digested PCR fragments into pLB86 digested with the same enzymes.

The genotypes of the complemented strains are shown in Table 1, and

the primers used for amplification of the trxA and trxD genes with

their own promoters, as well as the trxD gene lacking a functional

promoter, are shown in Table S2.

Preparation of polyclonal antibodies against TrxA and TrxD.Purified recombinant Lac. lactis TrxA or TrxD produced in E. coli

(Bjornberg et al., 2014) were used for raising anti-TrxA or anti-TrxD

antibodies. Prior to immunization the N-terminal His6 tags of

recombinant TrxA and TrxD were removed by proteolytic digestion

overnight with immobilized thrombin (Calbiochem). Cleaved His6-tags

and uncleaved His6-Trx were subsequently removed on a HisTrap

column (GE Healthcare) and 1.5 ml of non-His-tagged TrxA (120 mM)

or TrxD (65 mM) equilibrated in PBS (137 mM NaCl, 2.7 mM KCl,

10 mM Na2HPO4, 2 mM KH2PO4, pH 7.4) was used for immunization

of New Zealand white rabbits (3–3.5 kg). In the first immunization 500 ml

of the antigen was mixed with 500 ml of complete Freund’s adjuvant and

the solution was injected subcutaneously on five different spots on the

back (0.2 ml per spot). The second and third boosters (given with 2 week

intervals) were performed similarly but using incomplete Freund’s

adjuvant instead. Blood sera containing anti-TrxA or anti-TrxD antibodies

were collected 1 week after the third booster and stored at 280 uC.

Western blot analysis. Synchronized cultures of Lac. lactis wt,

DtrxA and DtrxD strains were grown under static conditions in liquid

GSAL medium at 30 uC. From a total culture volume of 100 ml,

40 ml was harvested in the mid-exponential phase (EP; OD450 of 0.4)

and in the stationary phase (SP; OD450 of #2), respectively. Cell

metabolism was quenched by pouring culture samples into pre-chilled

flasks on ice and incubating for 15 min. Cultures were then centrifuged

10 min at 5000 g at 4 uC, and supernatants were removed. Pellets were

washed in 1 ml of an ice-cold sterile 0.9 % (w/v) NaCl solution,

transferred into Eppendorf tubes and centrifuged again. Supernatants

were discarded and pellets were stored at 220 uC until extraction.

Frozen pellets were dried in a SpeedVac SPD1010 concentrator

(Thermo Scientific) for 1–2 h. Then, 100 and 300 ml of glass beads

(diameter ¡106 mm; Sigma) was added to the dry pellets from EP and

SP cultures, respectively, followed by homogenization by aid of a

micropestle (Eppendorf). Extraction buffer (0.2 M Tris/HCl, 0.2 M

NaCl, 5 % glycerol, 1 mM EDTA, pH 7.6) was added to obtain a final

volume of 200 and 900 ml for the EP and SP samples, respectively.

Following centrifugation (15 min at 14 000 r.p.m., 4 uC) supernatants

were collected and protein concentration was determined (Coomassie

plus protein assay reagent kit; Pierce Biotechnology) with BSA as

standard. SDS-PAGE was performed with 25 mg of total protein from

each cell extract and positive controls with 200 and 100 ng of His6-

tagged TrxA and TrxD, respectively. Western blotting was performed

using a X-Cell II Blot Module (Invitrogen) and Amersham Hybond

ECL nitrocellulose membrane (GE Healthcare). Membranes were

incubated with non-purified rabbit sera containing anti-TrxA and anti-

TrxD antibodies (see above) diluted 1 : 2000 in TBS buffer (100 mM

Tris/HCl, pH 7.5, 150 mM NaCl) containing 0.1 % (v/v) Tween-20

(TBST) for 1 h at room temperature. After several washes in TBST,

alkaline phosphatase-conjugated polyclonal goat anti-rabbit IgG

(0.64 mg ml21; Dako) diluted 1 : 2000 in TBST was added and

incubated for 30 min. The membrane was again washed in the same

buffer followed by incubation for 10 min in 0.015 % (w/v) 5-bromo-4-

chloro-3-indolyl phosphate and 0.030 % (w/v) nitro blue tetrazolium

chloride in 100 mM NaCl, 5 mM MgCl2 and 100 mM Tris/HCl,

pH 9.5, at room temperature. Reactions were stopped by transferring

the membrane into 20 mM EDTA.

Tetrazolium salt reduction assay. Samples of synchronized cultures

(0.9 ml) of Lac. lactis wt, DtrxA and DtrxD grown under static

conditions at 30 uC were collected in the mid-EP (OD450 of 0.4) and in

SP (OD450 of # 2), mixed with 100 ml of 5 mM tetrazolium violet (TV)

or iodonitrotetrazolium chloride (INT) and incubated for 15 min at

room temperature in the dark. Samples were centrifuged (20 000 g,

15 min, room temperature) and supernatants discarded. Pellets were

resuspended in 1 ml DMSO and centrifuged again (20 000 g, 15 min,

room temperature). The absorbances at 510 nm (reduced TV) and

468 nm (reduced INT) in the supernatants were determined and

divided by cell density (OD450 values of the cultures at harvest).

Difference gel electrophoresis (DIGE). Synchronized cultures of

Lac. lactis wt, DtrxA and DtrxD strains were grown in GSAL medium

under static conditions at 30 uC and samples were harvested in the

mid-EP (OD450 of 0.4). Cell pellets from 80 ml cultures washed in

0.9 % NaCl were freeze-dried (Scanvac CoolSafe; LaboGene) for 2 h.

Thereafter, 500 ml extraction buffer (0.2 M Tris/HCl, 0.2 M NaCl,

5 % glycerol, 1 mM EDTA, pH 7.6) and 500 ml of glass beads

(diameter ¡106 mm) were added and cells were disrupted by three

cycles in a FastPrep FP120 homogenizer (Qbiogene) set up at speed 4

and time 45 s (samples were kept on ice for 2 min between the cycles).

The extracts were centrifuged (15 min at 14 000 r.p.m. at 4 uC),

supernatants were collected, treated using Benzonase (0.25 U extract

ml21) and proteins concentrations were determined (Coomassie plus

protein assay reagent kit; Pierce Biotechnology) with BSA as standard.

Four biological replicates each of Lac. lactis wt, DtrxA and DtrxD were

compared. For each replicate, 30 mg protein was precipitated by

chloroform/methanol extraction (Wessel & Flugge, 1984). Pellets were

dissolved in 105 ml rehydration buffer (7 M urea, 2 M thiourea,

10 mM Tris/HCl, pH 8.5, 4 % CHAPS) and 70 ml of each sample was

labelled with 100 pmol (1 ml of 100 mM) of either of the the fluorescent

dyes Cy3 or Cy5 (CyDye DIGE Fluor minimal; GE Healthcare) in

Table 1. Details of the strains used

Strain Relevant genotype Reference

MG1363 Plasmid and prophage cured wt strain of Lactococcus lactis subsp. cremoris Gasson (1983)

LB504 MG1363/pLB95 Breuner et al. (2001)

MK555 MG1363 DtrxA This work

MK556 MG1363 DtrxD This work

MK562 MG1363 DtrxA DtrxD This work

MK740 MG1363 DtrxA DtrxD attB : : pMK1222 (trxA+) This work

MK741 MG1363 DtrxA DtrxD attB : : pMK1221(trxD with no promoter) This work

MK742 MG1363 DtrxA DtrxD attB : : pMK1220 (trxD+) This work

MK745 MG1363 DtrxD attB : : pMK1220 (trxD+) This work

P. Efler and others

530 Microbiology 161


IP: 54.224.121.223

On: Mon, 06 Jun 2016 05:50:37

anhydrous N,N-dimethylformamide. In addition, an internal standardcontaining 35 ml from each sample was labelled with 600 pmol (6 ml of100 mM) Cy2 dye (CyDye DIGE Fluor minimal) in anhydrous N,N-dimethylformamide. Fluorophore labelling was carried out on ice inthe dark for 30 min followed by addition of 2 ml lysine (100 mg ml21)and incubation for 10 min on ice in the dark. Samples were mixedaccording to the set-up detailed in Table S3, 6 ml of 100 mg DTT ml21

and 1 ml IPG buffer, pH 4–7 (GE Healthcare), was added andisoelectric focusing with Immobiline DryStrip, pH 4–7, 11 cm strips(GE Healthcare) was performed according to the following pro-gramme: 6 h at 30 V, 6 h at 60 V, 1 h at 200 V, 1 h at 500 V, 1 h at1000 V, 1 h gradient from 1000 to 8000 V followed by a constant8000 V until reaching 20 000 Vh. Prior to the second dimension, stripswere incubated for 15 min in equilibration buffer [6 M urea, 30 % (v/v)glycerol, 0.01 % bromophenol blue, 2 % (w/v) SDS, 100 mg DTT ml21,50 mM Tris/HCl, pH 8.8] followed by 15 min in the same buffercontaining iodoacetamide (250 mg ml21) instead of DTT. The seconddimension was performed using Criterion Precast 12.5 % polyacryla-mide gels (Bio-Rad) with MES running buffer (50 mM MES, 50 mMTris base, 0.1 % SDS, 1 mM EDTA, pH 7.3). Gels were fixed for 30 min in30 % (v/v) ethanol containing 2 % (v/v) phosphoric acid, scanned using aTyphoon Trio (GE Healthcare) at 100 mm resolution at excitation/emission wavelengths of 488/520 nm (Cy2), 532/580 nm (Cy3) and633/670 nm (Cy5), and subsequently stained by Coomassie brilliantblue G-250 (Merck) as described previously (Candiano et al., 2004).Fluorescence images were analysed by Progenesis SameSpots software(Non-linear Dynamics). Only spots displaying volume fold change .1.5and an ANOVA P value ,0.05 were selected for identification by MS.

In-gel trypsin digestion and MS. Spot gel-plugs were manuallypicked from Coomassie-stained gels and subjected to in-gel trypsindigestion as described previously (Majumder et al., 2011). Briefly, thegel-plugs were washed using 40 % ethanol, dried with 100 % aceto-nitrile, added to 250 ng porcine trypsin (Promega) in 10 ml 10 mMNH4HCO3 and incubated overnight at 37 uC. Samples of 1 or 2 ml wereloaded on an AnchorChip target plate (Bruker Daltonics) followed by1 ml of 0.5 mg matrix solution ml21 (a-cyano-4-hydroxycinnamic acidin 70 % acetonitrile, 0.1 % trifluoroacetic acid). In some cases, sampleswere desalted and concentrated by using a POROS R2 (Applied Biosystems)microcolumn prior to analysis (Gobom et al., 1999). Samples wereanalysed using an Ultraflex II MALDI-TOF/TOF mass spectrometer(Bruker Daltonics), and spectra were processed by FlexAnalysis (v3.3)and BioTools (v3.2) software provided by the instrument manufac-turer. Combinations of MS and MS/MS data were used as input for theMascot search engine (www.matrixscience.com) with the followingparameters: NCBInr database, trypsin digestion (one partial cleavage),carbamidomethylation of Cys (global modification), oxidation ofMet (variable modification), and MS and MS/MS mass tolerance of80 p.p.m. and 0.6 Da, respectively. Alternatively, the trypsin digestswere analysed on an LC-MS system composed of an EASY nLC 1000chromatograph coupled online to a Q-Exactive spectrometer (ThermoScientific) and spectra were processed using Proteome Discoverer(Thermo Scientific). The set-up of the Mascot database searching forLC/MS data was as follows: Swiss-Prot database, trypsin digestion (onepartial cleavage), carbamidomethylation of Cys (global modification),oxidation of Met (variable modification), and peptide and fragmentmass tolerance of 10 p.p.m. and 20 mDa, respectively. The significancethreshold for protein identifications was P,0.05.

RESULTS AND DISCUSSION

TrxA can compensate for the loss of TrxD undernon-stressed growth conditions

Lac. lactis MG1363 contains two thioredoxins encoded bytrxA (llmg_0779) and trxD (llmg_0406; annotated as trxH),

organized in two separate operons (Fig. S1). Expression ofthe genes was confirmed by Western blot analysis, whichfurther demonstrated that TrxA and TrxD were present inthe mid-EP and in the SP (Fig. 1). Deletions of trxA andtrxD were constructed by overlap extension PCR, followedby homologous recombination into the chromosome, asverified by colony PCR (data not shown). The success ofthe deletions and the correct identification of TrxA andTrxD in the Western blots were confirmed by the absenceof signal in protein extracts from DtrxD and DtrxA mutants,respectively using the appropriate antibodies (Fig. 1).

Growth of the wt strain and the trx mutants in chemicallydefined GSAL medium under microaerophilic conditionswas compared in Bioscreen assays (Fig. 2a; Table 2). TheDtrxA mutant showed clear growth defects while the DtrxDmutant was unaffected, indicating that TrxA can compens-ate for the loss of TrxD under normal growth conditions,but not vice versa. The DtrxADtrxD mutant, however, grewmuch more slowly than the DtrxA mutant. Thus, a pro-nounced effect of the TrxD deletion is observed only inthe DtrxA background. The observed poor growth ofthe DtrxADtrxD mutant is in agreement with previouslypublished results demonstrating that a TrxR-deficient Lac.lactis strain is strongly affected but viable under micro-aerophilic conditions (Vido et al., 2005). A major difference,however, is that in the absence of TrxA and TrxD, Lac. lactiscan still utilize TrxR to recycle NrdH as an electron donorfor RNR (Jordan et al., 1996, 1997).

Changes in proteome profiles of DtrxA and DtrxDmutants

Protein profiles of the wt strain and DtrxA and DtrxDmutants in mid-EP under standard (non-stressed) condi-tions were compared using DIGE. In general most differenceswere observed between the DtrxA mutant and wt (Fig. 3).Several proteins involved in the oxidative stress responsewere upregulated in the DtrxA proteome compared with thewt, including TrxR (TrxB) and glutathione peroxidase (Table3). Despite its name, glutathione peroxidase is likely to be

1

anti-

TrxD

anti-

TrxA

2 3 4 5 6 7 8

Fig. 1. Western blot of protein extracts from wt, DtrxA and DtrxD

strains harvested in mid-EP or SP. Recombinant TrxA and TrxDwere used as positive controls. 1, wt EP; 2, wt SP; 3, DtrxA EP; 4,DtrxA SP; 5, DtrxD EP; 6, DtrxD SP; 7, 200 ng TrxA; 8, 100 ngTrxD. No apparent cross-reactivity between anti-TrxA and anti-TrxD was observed. The observed mass shift is attributed to the N-terminal His-tags on the recombinant proteins.




IP: 54.224.121.223

On: Mon, 06 Jun 2016 05:50:37

thioredoxin-dependent, as has been demonstrated for thehomologous proteins from plants, fungi and bacteria (Leeet al., 2008). The protein encoded by llmg_1475 was the mosthighly upregulated (3.7-fold) in DtrxA versus the wt (Table3). YnzC, a homologous protein from Bcl. subtilis (41 %identity, 58 % similarity), was previously suggested to beinvolved in the SOS DNA damage response and wasupregulated upon H2O2 treatment in Bacillus licheniformis(Kawai et al., 2003; Schroeter et al., 2011). The upregulatedprotein encoded by llmg_2273 contains a histidine triadactive site motif and is tentatively annotated as a diadenosinetetraphosphate hydrolase. No bacterial homologue of Llmg_

2273 has been characterized, but eukaryotic proteinscontaining histidine triad motifs influence the cell cyclethrough interactions with regulatory proteins (Huebner et al.,2011; Nishizaki et al., 2004), and are associated withoxidative stress defence and DNA repair.

Downregulated proteins in DtrxA versus the wt includepyvuvate kinase (Pyk), formate-tetrahydrofolate ligase (Fhs),tyrosyl-tRNA synthetase (TyrS), as well as a putative

tellurium resistance protein (TelB) and Llmg_0304, anno-tated as a potential RNA-binding protein. Both Pyk and Fhsappear in two spots with different pI, suggesting that theseproteins are subjected to some type of post-translationalmodification (Fig. 3). It is noteworthy that Fhs is over-expressed in Lac. lactis under respiratory conditions, andwas highly overexpressed in Porphyromonas gingivalis sub-jected to oxidative stress (Lewis et al., 2009; Vido et al., 2004).The most downregulated protein in the trxA mutant (twofold)was TelB. Lac. lactis TelB is a homologue of TerD, a metal-binding protein proposed to be involved in calcium signalling(Pan et al., 2011). Overall, DIGE analysis demonstrated thatdeletion of TrxA influences the expression profiles of proteinsimplicated in oxidative stress resistance.

Reduction of TV and INT is dependent on TrxA

Tetrazolium salts are weakly coloured compounds thatturn into strongly coloured formazans upon reduction andare widely used for spectrophotometric determinations ofcellular redox activities (Berridge et al., 2005). Here, Lac.

(a)

10

1 1

0.10 0.0 10.0 20.0 30.0 40.0 0 5 10 15 205 10

Time (h)

Wild type ΔtrxA ΔtrxD

Time (h) Time (h)15 20

OD

45

010

0.1

OD

45

0

10

1

0.1

OD

45

0

Non-stressed 0.3 mM hydrogen peroxide 1.25 mM sodium arsenate(b) (c)

(d)

Control

10–2 10–3 10–4 10–5 10–2 10–3 10–4 10–5 10–2 10–3 10–4 10–5

0.5 mM

Na2HAsO4

0.3 mM

K2TeO4

Fig. 2. Trx mutant phenotypes. Growth curves display wt (X), DtrxA (m), DtrxD (#) and DtrxADtrxD (�) strains in an unstressedcontrol (a), or exposed to 0.3 mM H2O2 (b) or 1.25 mM sodium arsenate (c). (d) Serial dilutions of three biological replicates ofexponentially growing wt, DtrxA and DtrxD strains spotted on regular GSAL agar plates (control) or corresponding platessupplemented with 0.5 mM sodium arsenate and 0.3 mM potassium tellurite (see Methods for details). Dilution factors areindicated at the bottom of the figures. In the presence of tellurite, black colonies appear due to formation of metallic Te0.

P. Efler and others



IP: 54.224.121.223

On: Mon, 06 Jun 2016 05:50:37

lactis DtrxA and DtrxD mutants were exposed to TV andINT, which are correlated with dehydrogenase activity inbacterial cells (Smith & McFeters, 1996). Overall, the extentof INT and TV reduction was similar in the wt and theDtrxD mutant (Fig. 4). By contrast, the DtrxA mutantexhibited significantly increased reduction of INT in mid-EP compared with the wt. Both TV and INT were reducedsignificantly more efficiently by the DtrxA mutant in SP,but TV reduction during exponential growth was similarfor the wt and the DtrxA mutant (Fig. 4). It is also note-worthy that the extent of tetrazolium salt reduction wasfive- to 10-fold higher in mid-EP than in SP for the threestrains, probably reflecting an overall higher rate of meta-bolic turnover in the EP. Reduction of TV in Lac. lactis hasbeen linked to menaquinones and the membrane-boundNADH dehydrogenases NoxA and NoxB in an NADH-dependent manner (Tachon et al., 2009). It may thusbe suggested that depletion of TrxA causes an increasedavailability of NADH, menaquinones or NoxA/NoxBthrough: (i) shifts in metabolic fluxes, (ii) transcriptionalregulation and/or (iii) thiol/disulphide exchange controlmechanisms.T

ab

le2.

Gro

wth

rate

san

dla

gp

hase

so

fw

t,D

trxA

,D

trxD

andD

trxA

Dtr

xDst

rain

sex

po

sed

tose

lect

edco

mp

oun

ds

inth

eB

iosc

reen

assa

yfo

rmat

Sta

nd

ard

dev

iati

on

sar

eb

ased

on

thre

eb

iolo

gica

lre

pli

cate

s.T

able

S1

dis

pla

ysco

rres

po

nd

ing

dat

afo

ral

lst

ress

com

po

un

ds

test

ed.

NA

,N

ot

app

lica

ble

.

Co

mp

ou

nd

(mM

)W

ild

-typ

eD

trxA

Dtr

xDD

trxA

Dtr

xD

Rel

ativ

em

*

(%ct

rl)

Lag

ph

ase

(h)

Rel

ativ

em

*L

ag

ph

ase

(h)

Rel

ativ

em

*L

ag

ph

ase

(h)

Rel

ativ

em

*L

ag

ph

ase

(h)

%ct

rl%

wt

%ct

rl%

wt

%ct

rl%

wt

Co

ntr

ol

( NA

)1

00±

3.9

3±

11

00±

2.8

72

4±

11

00±

2.9

95

3±

11

00±

30

.53

38±

1

Met

SO

(NA

)D9

6±

3.3

4±

13

0±

5.6

22

12±

39

5±

8.5

97

4±

15

7±

12

.01

82

4±

5

H2O

2(0

.31

3)

95±

9.8

8±

29

4±

10

.47

1.

24

94±

7.9

94

9±

1,

10

NA

.2

4

So

diu

mar

sen

ate

(1.2

5)

92±

1.6

6±

19

4±

3.7

73

9±

29

0±

6.4

93

12±

1,

10

NA

.2

4

Po

tass

ium

tell

uri

te(1

.25

)4

9±

2.5

4±

25

3±

4.2

78

7±

23

3±

1.4

64

4±

11

4±

0.9

10

.2

4

Dia

mid

e(1

.25

)8

1±

11

.08±

46

1±

55

.15

41

8±

27

8±

11

.99

19±

4,

10

NA

.2

4

Par

aqu

at(2

0)

61±

10

.18±

32

9±

30

.93

51

1±

16

2±

6.1

96

10±

2,

10

NA

.2

4

*Rel

ativ

egr

ow

thra

tes

(m)

wer

eca

lcu

late

db

yd

ivid

ingm

of

ast

rain

ata

stre

ssco

nd

itio

nb

ym

of

no

n-s

tres

sed

con

tro

l(c

trl)

orm

of

wt

atth

esa

me

stre

ssco

nd

itio

n.

At

no

n-s

tres

sed

con

dit

ion

s,1

00

%

(wt)

50

.50

6h

21,

10

0%

(Dtr

xA)5

0.3

65

h2

1,

10

0%

(Dtr

xD)5

0.4

79

h2

1an

d1

00

%(D

trxA

Dtr

xD)5

0.1

67

h2

1.

DM

eth

ion

ine

rep

lace

db

yM

etS

Oin

gro

wth

med

ium

.

200 kDapl 4 7

5 kDa

8 10 12 7

200 kDa

5 4

9

13

32

11 1

6

5 kDa

Fig. 3. Proteome profile of DtrxA versus wt. Representative DIGEgel of soluble cytosolic proteins in the acidic range (pI 4–7)displaying an overlay of Cy5 (red, wt) and Cy3 (green, DtrxA)fluorescence signals (top), and a greyscale image of the Cy2signal (bottom). Proteins up- and downregulated in the DtrxA

mutant versus wt are marked in red and purple, respectively.




IP: 54.224.121.223

On: Mon, 06 Jun 2016 05:50:37

Table 3. Up- and downregulated proteins identified by DIGE of non-stressed DtrxA mutant versus wt in mid-EP

Spot

no.

Protein name Fold

change*

pI/Mr

[kDa]DGene Accession

no.

Pd Score§ SC

(%)||

M Peptide sequences identified

by MS/MS

G#

1 Hypothetical protein +3.7 5.6/9.2 llmg_1475 gi|125624282 0.0002 87 21 2 KAEGLSEAELEEQALLRR 4

2 Glutathione peroxidase +2.1 5.2/18.1 gpo gi|125623919 0.0002 84 28 6 KFLIDRDGQVIERF 1

3 30S ribosomal protein S5 +1.7 10.2/17.6 rpsE gi|125625124 0.0030 204 56 11 RFAALVVVGDRN 3

KAQEVPEAIRKA

KSLGSNTPINVVRA

4 Pyruvate dehydrogenase E1 +1.6 4.8/35.1 pdhB gi|125622952 0.0003 292 63 19 KDKDALIFGEDVGQNGGVFRA 2

RVVVVQEAQRT

5 Thioredoxin reductase +1.5 4.8/34 trxB1 gi|125624390 0.0008 156 18 4 RNQEILVIGGGDSAVEEALYLTRF 1

6 Hypothetical protein +1.5 5.4/14.9 llmg_2273 gi|125625038 0.0030 114 49 5 KFTAHDYDLAEIAKQ 4

7 Formate-tetrahydrofolate

ligase

21.6 5.7/59.7 fhs gi|125623054 0.0050 615 53 34 KSTVTVGLADAFARQ ; RIVIAQNYDRK;

KTQYSFSDQANLLAAPEGFEVTVRE

4

8 Pyruvate kinase 21.6 5.2/54.3 pyk gi|125623950 0.0050 465 53 26 KIVSTLGPAVEIRG 2

RTELFTDGADSISVVTGDKFRV

KLIVALTESGNTARL

9 Tyrosyl-tRNA synthetase 21.6 5.38/47.3 tyrS gi|125624396 0.0090 323 37 16 KTSEILFGGGDLRQ 3

RVQELDYVLTDSDKIENRL

10 Pyruvate kinase 21.6 5.2/54.3 pyk gi|125623950 0.0060 335 57 31 KLIVALTESGNTARL 2

KIPFPALAERDDADIRF

11 Hypothetical protein 21.7 4.5/8.8 llmg_0304 gi|125623269 0.0110 236 88 8 RETTDALYLDLDVATKEEGVILARE 4

12 Formate-tetrahydrofolate

ligase

21.7 5.7/59.7 fhs gi|125623950 0.0010 169 28 15 K.STVTVGLADAFARQ 4

RIVIAQNYDRK

13 Putative tellurium resistance

protein

21.9 4.38/21.1 telB gi|125624170 0.0310 241 48 9 KVRNDDDFIFYNHKI 1

RNDDDFIFYNHKI

*Spot volume ratio of DtrxA to wt; plus and minus indicate up- and downregulation, respectively.

DCalculated values.

dANOVA P value from the image analysis; applied threshold was P,0.05.

§Mascot score.

||Sequence coverage in peptide mass fingerprinting (PMF).

Number of matched peptides in PMF.

#Groups: 1, stress; 2, carbon metabolism; 3, translation; 4, other.

P.E

flerand

others

53

4M

icrob

iolo

gy1

61


IP: 54.224.121.223

On: Mon, 06 Jun 2016 05:50:37

MetSO reduction is dependent on TrxA

Methionine residues are highly susceptible to oxidativedamage resulting in a racemic mixture of (S)- and (R)-enantiomers of MetSO. This reaction is reversed by Trx-coupled MetSO reductase (Msr) activity. (S)- and (R)-MetSOare reduced by two distinct types of MetSO reductases, MsrAand MsrB, respectively (Boschi-Muller et al., 2008). In E. coli,MsrA efficiently reduces both bound and free (S)-MetSOwhile MsrB reduces only peptide- or protein-bound (R)-MetSO (Grimaud et al., 2001). In addition, MsrA-independ-ent reduction of free (S)-MetSO is catalysed in a Trx-independent manner by BisC in E. coli (Ezraty et al., 2005).The genome of Lac. lactis contains genes encoding putativeMsrA, MsrB and free methionine-(R)-sulphoxide reductase(llmg_2480), but no BisC homologue. Lac. lactis MG1363 isauxotrophic for methionine (Jensen & Hammer, 1993;Seefeldt & Weimer, 2000) and the capacity of the trx mutantsto generate Msr activity was thus tested in a Bioscreen assayusing (R/S)-MetSO as sole methionine source. No significantdifference was observed between growth of the wt and theDtrxD mutant (Table 2). However, both the DtrxA and theDtrxADtrxD mutants exhibited significantly prolonged lagphases, and growth rates were reduced by 78 and 82 %,respectively, compared with the wt. Thus, it appears likely thatTrxA functions as the major electron donor for Msr in L.lactis. However, because all the trx mutants were viable, Lac.lactis might also utilize a less efficient alternative Trx-independent MetSO reduction pathway.

TrxA is important while TrxD is dispensable forrecovery from oxidative stress.

It is well established that thioredoxins provide reducingequivalents to thiol-based peroxiredoxins that in turn

catalyse reduction of H2O2 (Hofmann et al., 2002). Thisfunction is likely to be of particular importance for stressresistance in organisms such as Lac. lactis lacking catalase, ahighly efficient haem-based peroxidase. To analyse theinvolvement of TrxA and TrxD in the response towardsoxidative stress, the Lac. lactis wt strain and the trx mutantswere exposed to oxidizing reagents. In the presence of313 mM H2O2, the lag phases before reaching maximalgrowth rate were prolonged by 5 h for the wild-type andDtrxD mutant, and .24 h for the DtrxA and DtrxADtrxDmutants (Fig. 2b). Following the lag phase the DtrxA andDtrxD mutants showed similar growth rates as prior toexposure while the DtrxADtrxD mutant grew very slowly.With 1.25 mM diamide a similar pattern was observedexcept that the DtrxA mutant displayed a decreased growthrate after the lag phase while the DtrxD mutant and wtdid not (Table 2). At high concentrations of paraquat(5220 mM), the DtrxA mutant was impaired to a greaterextent than the wt and DtrxD cultures, and the DtrxADtrxDmutant was the most severely affected (Table 2). Surpris-ingly, concentrations of the superoxide-inducing reagentparaquat ,1 mM had no effect on growth of the wt strainand the DtrxA and DtrxD mutants, but a positive effect ongrowth of the DtrxADtrxD mutant was observed.Formaldehyde is a reactive electrophilic species and hasbeen shown to interact with thiol-based redox sensors andinduces a disulphide stress response, including upregulationof Trx and TrxR (Antelmann & Helmann, 2011; Nguyen etal., 2009). No significant difference between the effects ongrowth of DtrxA, DtrxD and wt strains was, however,observed upon exposure to formaldehyde (8025000 mM),but the DtrxADtrxD mutant was severely impaired (TableS1). In conclusion, TrxA appears to be more important foroxidative stress resistance than TrxD. These results suggest

0.81.01.2 TV

10

15 INT(a)

00.20.40.6

0

5

0.12TV

2.0INT(b)

0.020.040.060.080.10

0.5

1.0

1.5

00.02

0

A51

0/O

D45

0A

510/

OD

450

A46

8/O

D45

0A

468/

OD

450

Wild-type ΔtrxA ΔtrxD Wild-type ΔtrxA ΔtrxD

Wild-type ΔtrxA ΔtrxD Wild-type ΔtrxA ΔtrxD

Fig. 4. Reduction of tetrazolium salts. TV and INT were added to the wt, DtrxA and DtrxD strains harvested in mid-EP (a) and SP(b). Reduction was monitored spectrophotometrically at 510 nm (TV) and 468 nm (INT). Error bars represent SD of at leastthree biological replicates.




IP: 54.224.121.223

On: Mon, 06 Jun 2016 05:50:37

that TrxA rather than TrxD acts as preferential electrondonor to thiol-based enzymes such as peroxiredoxinsinvolved in decomposition of reactive oxygen species.

The Trx system influences resistance to metal iontoxicity

The toxic effects of various divalent metal ions are relatedto their propensities to react with thiol groups in general(Lemire et al., 2013) and Trx in particular (Hansen et al.,2006). In addition, Fe(II) and Cu(II) react with H2O2

through the Fenton reaction and thus contribute to pro-duction of reactive oxygen species (Imlay, 2003). Selectedmetal ions were investigated to determine their influenceon growth of the trx mutants. Cd(II) is highly toxic andwas found to be lethal for Lac. lactis at 20 mM (Table S1),in good agreement with previous observations in E. coli(Ferianc & Farewell, 1998). At 5 mM Cd(II), the DtrxAmutant displayed a pronounced growth defect while noapparent growth of the DtrxADtrxD mutant was detected.These results suggest that TrxA is important for cadmiumresistance, in agreement with previous observations inSaccharomyces cerevisae (Vido et al., 2001). Compared withCd(II), Lac. lactis was more tolerant to Zn(II) and nopronounced effects of the trx genes were observed at 78 mM(Table S1). At 313 mM Zn(II), all strains were inhibited to asimilar extent and no growth was observed at 1.25 mM.Interestingly, lower concentrations (5220 mM) of Zn(II) hada slightly positive effect on the growth rate of the DtrxADtrxDmutant. In this context it is relevant to mention that Zn(II)has been suggested to have a thiol-protective function andLac. lactis strains with impaired Zn uptake are hypersensitiveto oxidative stress (Scott et al., 2000).

Addition of Fe(II) in the range of concentrations tested(521250 mM) had only minor effects on growth, except thatit led to a prolonged lag phase for the DtrxADtrxD mutant(Table S1). The high tolerance may reflect the presence ofproteins such as Dps sequestering cytosolic Fe ions and thuspreventing formation of hydroxyl radicals through theFenton reaction (Imlay, 2003; Stillman et al., 2005). Cu(II)was more toxic than iron, causing complete growth inhibi-tion at 313 mM for the DtrxA, DtrxD strains and wt and at78 mM for the DtrxADtrxD mutant (Table S1). In general acopper-dependent growth defect was observed for DtrxAcompared with the DtrxD mutant and wt.

Resistance towards arsenate and tellurite isdependent on TrxD

Arsenate is a highly toxic analogue of phosphate that reactswith thiol groups and induces oxidative stress (Hughes,2002). Arsenate(V) is reduced to arsenite(III) by arsenatereductase (ArsC) and exported by ArsAB, an ATPase-driventransporter (Turner et al., 1992). In Gram-positive bacteriasuch as Bcl. subtilis, ArsC receives reducing equivalents fromTrx, but a range of alternative electron donors have beendescribed; for example, ArsC from E. coli is reduced by Grx

(Messens & Silver, 2006). Here, we investigated the influenceof the two Lac. lactis Trxs on arsenate stress. The wt, DtrxAand DtrxD cultures subjected to 1.25 mM arsenate hadsimilar growth rates in the Bioscreen assay but the lag phaseswere prolonged by 3, 6 and 9 h, respectively, compared withan unstressed control (Fig. 2a, c). The DtrxADtrxD mutantdid not recover from exposure to 1.25 mM arsenate, suggestingthat at least one Trx is required to counteract arsenate stress.Sensitivity to arsenate stress was also studied in seriallydiluted cultures plated on GSAL agar plates containingarsenate. Growth was detected in a spot of 103-fold dilutedcultures of both the wt and the DtrxA mutant, but no growthwas detected for the trxD mutant in spots of 102-fold dilutedcultures (Fig. 2d). The increased arsenate sensitivity of thetrx deletion mutants suggests that Lac. lactis TrxD and TrxAplay important roles in detoxification, for example aselectron donors to ArsC.

Tellurite causes intracellular production of superoxide(Perez et al., 2007) and is correlated with arsenate detoxifi-cation in E. coli (Turner et al., 1992). Tellurite-resistantLac. lactis strains were found to contain mutations in, forexample, high-affinity phosphate (particularly pstA and pstD)and iron transporters (mntH), and in trmA, a homologue ofthe disulphide stress sensor spx (Turner et al., 2007). Here,tests with agar plates containing tellurite indicated similarsensitivity of the DtrxD and DtrxA mutants (Fig. 2d). In theBioscreen assay the DtrxA mutant displayed a prolonged lagphase following addition of 1.25 mM potassium telluritecompared with the wt. The DtrxD mutant, by contrast,showed a similar lag phase as the wt but the growth rate wasreduced to 64 % that of the wt following tellurite stress(Table 2). These observations indicate that TrxA is impor-tant in the phase immediately after tellurite exposure,suggesting involvement in adaptive responses such as tran-scriptional regulation. The more long-term effects of theDtrxD mutant strain, by contrast, indicate direct involve-ment in detoxification reactions.

To confirm that the phenotypes were caused by the loss ofthe trxA or trxD genes, the DtrxADtrxD mutant was com-plemented with either trxA or trxD. Either of the two genesrestored growth of single colonies on GSAL plates con-taining 1.5 mM potassium tellurite (Fig. S2). In contrast, aDtrxADtrxD strain complemented with trxD lacking afunctional promoter was not able to form single colonieson the same type of plates.

CONCLUSIONS

The observed phenotypes of the trx mutants suggest that thetwo thioredoxins have different functions in stress resistancein Lac. lactis. TrxA seems to be involved in responses tooxidative stress while TrxD appears to be important forresistance towards arsenate and tellurite. The role of TrxD inthese processes is unknown but it is speculated that TrxDmay act as an alternative electron donor for arsenate reduc-tase, an established Trx target in Bcl. subtilis.

P. Efler and others



IP: 54.224.121.223

On: Mon, 06 Jun 2016 05:50:37

ACKNOWLEDGEMENTS

Marzanna Pulka-Amin, Anne Blicher, Aida Curovic and BirgitAndersen are acknowledged for technical assistance and Morten EjbyHansen for assistance with DIGE. This work was supported by theDanish Research Councils for Technology and Production Sciences(FTP, grant no. 274-08-0413) and Natural Sciences (FNU, grant for Q-exactive and Typhoon), the Carlsberg Foundation, and the Centre forAdvanced Food Studies (LMC). The PhD grant to P. E. was in partfinanced by the Technical University of Denmark and the FTP grant.

REFERENCES

Akif, M., Khare, G., Tyagi, A. K., Mande, S. C. & Sardesai, A. A. (2008).Functional studies of multiple thioredoxins from Mycobacteriumtuberculosis. J Bacteriol 190, 7087–7095.

Antelmann, H. & Helmann, J. D. (2011). Thiol-based redox switchesand gene regulation. Antioxid Redox Signal 14, 1049–1063.

Arner, E. S. J. & Holmgren, A. (2000). Physiological functions ofthioredoxin and thioredoxin reductase. Eur J Biochem 267, 6102–6109.

Berka, R. M., Cui, X. & Yanofsky, C. (2003). Genomewidetranscriptional changes associated with genetic alterations andnutritional supplementation affecting tryptophan metabolism inBacillus subtilis. Proc Natl Acad Sci U S A 100, 5682–5687.

Berridge, M. V., Herst, P. M. & Tan, A. S. (2005). Tetrazolium dyes astools in cell biology: new insights into their cellular reduction.Biotechnol Annu Rev 11, 127–152.

Biswas, I., Gruss, A., Ehrlich, S. D. & Maguin, E. (1993). High-efficiency gene inactivation and replacement system for gram-positivebacteria. J Bacteriol 175, 3628–3635.

Bjornberg, O., Efler, P., Ebong, E. D., Svensson, B. & Hagglund, P.(2014). Lactococcus lactis TrxD represents a subgroup of thioredoxinsprevalent in Gram-positive bacteria containing WCXDC active sitemotifs. Arch Biochem Biophys 564, 164–172.

Boschi-Muller, S., Gand, A. & Branlant, G. (2008). The methioninesulfoxide reductases: catalysis and substrate specificities. Arch BiochemBiophys 474, 266–273.

Breuner, A., Brøndsted, L. & Hammer, K. (2001). Resolvase-likerecombination performed by the TP901-1 integrase. Microbiology 147,2051–2063.

Candiano, G., Bruschi, M., Musante, L., Santucci, L., Ghiggeri, G. M.,Carnemolla, B., Orecchia, P., Zardi, L. & Righetti, P. G. (2004). Bluesilver: a very sensitive colloidal Coomassie G-250 staining for proteomeanalysis. Electrophoresis 25, 1327–1333.

Comtois, S. L., Gidley, M. D. & Kelly, D. J. (2003). Role of thethioredoxin system and the thiol-peroxidases Tpx and Bcp in medi-ating resistance to oxidative and nitrosative stress in Helicobacterpylori. Microbiology 149, 121–129.

Ezraty, B., Bos, J., Barras, F. & Aussel, L. (2005). Methioninesulfoxide reduction and assimilation in Escherichia coli: new role forthe biotin sulfoxide reductase BisC. J Bacteriol 187, 231–237.

Fahey, R. C., Brown, W. C., Adams, W. B. & Worsham, M. B. (1978).Occurrence of glutathione in bacteria. J Bacteriol 133, 1126–1129.

Ferianc, P. & Farewell, A. (1998). The cadmium-stress stimulon ofEscherichia coli K-12. Microbiology 144, 1045–1050.

Fernandes, L. & Steele, J. L. (1993). Glutathione content of lactic acidbacteria. J Dairy Sci 76, 1233–1242.

Gasson, M. J. (1983). Plasmid complements of Streptococcus lactisNCDO 712 and other lactic streptococci after protoplast-inducedcuring. J Bacteriol 154, 1–9.

Gobom, J., Nordhoff, E., Mirgorodskaya, E., Ekman, R. & Roepstorff,P. (1999). Sample purification and preparation technique based onnano-scale reversed-phase columns for the sensitive analysis ofcomplex peptide mixtures by matrix-assisted laser desorption/ionization mass spectrometry. J Mass Spectrom 34, 105–116.

Grimaud, R., Ezraty, B., Mitchell, J. K., Lafitte, D., Briand, C., Derrick,P. J. & Barras, F. (2001). Repair of oxidized proteins. Identification ofa new methionine sulfoxide reductase. J Biol Chem 276, 48915–48920.

Hansen, J. M., Zhang, H. & Jones, D. P. (2006). Differential oxidationof thioredoxin-1, thioredoxin-2, and glutathione by metal ions. FreeRadic Biol Med 40, 138–145.

Hofmann, B., Hecht, H. J. & Flohe, L. (2002). Peroxiredoxins. BiolChem 383, 347–364.

Huebner, K., Saldivar, J. C., Sun, J., Shibata, H. & Druck, T. (2011).Hits, Fhits and Nits: beyond enzymatic function. Adv Enzyme Regul51, 208–217.

Hughes, M. F. (2002). Arsenic toxicity and potential mechanisms ofaction. Toxicol Lett 133, 1–16.

Imlay, J. A. (2003). Pathways of oxidative damage. Annu Rev Microbiol57, 395–418.

Jensen, P. R. & Hammer, K. (1993). Minimal requirements forexponential growth of Lactococcus lactis. Appl Environ Microbiol 59,4363–4366.

Ji, G. & Silver, S. (1992). Reduction of arsenate to arsenite by the ArsCprotein of the arsenic resistance operon of Staphylococcus aureusplasmid pI258. Proc Natl Acad Sci U S A 89, 9474–9478.

Jordan, A., Pontis, E., Aslund, F., Hellman, U., Gibert, I. & Reichard, P.(1996). The ribonucleotide reductase system of Lactococcus lactis.Characterization of an NrdEF enzyme and a new electron transportprotein. J Biol Chem 271, 8779–8785.

Jordan, A., Aslund, F., Pontis, E., Reichard, P. & Holmgren, A. (1997).Characterization of Escherichia coli NrdH. A glutaredoxin-like proteinwith a thioredoxin-like activity profile. J Biol Chem 272, 18044–18050.

Kawai, Y., Moriya, S. & Ogasawara, N. (2003). Identification of aprotein, YneA, responsible for cell division suppression during theSOS response in Bacillus subtilis. Mol Microbiol 47, 1113–1122.

Kobayashi, K., Ehrlich, S. D., Albertini, A., Amati, G., Andersen, K. K.,Arnaud, M., Asai, K., Ashikaga, S., Aymerich, S. & other authors(2003). Essential Bacillus subtilis genes. Proc Natl Acad Sci U S A 100,4678–4683.

Kunst, F., Ogasawara, N., Moszer, I., Albertini, A. M., Alloni, G.,Azevedo, V., Bertero, M. G., Bessieres, P., Bolotin, A. & otherauthors (1997). The complete genome sequence of the gram-positivebacterium Bacillus subtilis. Nature 390, 249–256.

Laurent, T. C., Moore, E. C. & Reichard, P. (1964). Enzymaticsynthesis of deoxyribonucleotides. IV. Isolation and characterizationof thioredoxin, the hydrogen donor from Escherichia coli B. J BiolChem 239, 3436–3444.

Lee, S. Y., Song, J. Y., Kwon, E. S. & Roe, J. H. (2008). Gpx1 is astationary phase-specific thioredoxin peroxidase in fission yeast.Biochem Biophys Res Commun 367, 67–71.

Lemire, J. A., Harrison, J. J. & Turner, R. J. (2013). Antimicrobialactivity of metals: mechanisms, molecular targets and applications.Nat Rev Microbiol 11, 371–384.

Lewis, J. P., Iyer, D. & Anaya-Bergman, C. (2009). Adaptation ofPorphyromonas gingivalis to microaerophilic conditions involvesincreased consumption of formate and reduced utilization of lactate.Microbiology 155, 3758–3774.

Li, Y., Hugenholtz, J., Abee, T. & Molenaar, D. (2003). Glutathioneprotects Lactococcus lactis against oxidative stress. Appl EnvironMicrobiol 69, 5739–5745.




IP: 54.224.121.223

On: Mon, 06 Jun 2016 05:50:37

Lillig, C. H., Berndt, C. & Holmgren, A. (2008). Glutaredoxin systems.Biochim Biophys Acta 1780, 1304–1317.

Lu, J. & Holmgren, A. (2014). The thioredoxin antioxidant system.Free Radic Biol Med 66, 75–87.

Majumder, A., Sultan, A., Jersie-Christensen, R. R., Ejby, M.,Schmidt, B. G., Lahtinen, S. J., Jacobsen, S. & Svensson, B.(2011). Proteome reference map of Lactobacillus acidophilus NCFMand quantitative proteomics towards understanding the prebioticaction of lactitol. Proteomics 11, 3470–3481.

Messens, J. & Silver, S. (2006). Arsenate reduction: thiol cascadechemistry with convergent evolution. J Mol Biol 362, 1–17.

Miranda-Vizuete, A., Damdimopoulos, A. E., Gustafsson, J. &Spyrou, G. (1997). Cloning, expression, and characterization of anovel Escherichia coli thioredoxin. J Biol Chem 272, 30841–30847.

Moller, M. C. & Hederstedt, L. (2008). Extracytoplasmic processesimpaired by inactivation of trxA (thioredoxin gene) in Bacillussubtilis. J Bacteriol 190, 4660–4665.

Newton, G. L., Arnold, K., Price, M. S., Sherrill, C., Delcardayre, S. B.,Aharonowitz, Y., Cohen, G., Davies, J., Fahey, R. C. & Davis, C.(1996). Distribution of thiols in microorganisms: mycothiol is amajor thiol in most actinomycetes. J Bacteriol 178, 1990–1995.

Newton, G. L., Rawat, M., La Clair, J. J., Jothivasan, V. K., Budiarto, T.,Hamilton, C. J., Claiborne, A., Helmann, J. D. & Fahey, R. C. (2009).Bacillithiol is an antioxidant thiol produced in Bacilli. Nat Chem Biol5, 625–627.

Nguyen, T. T., Eiamphungporn, W., Mader, U., Liebeke, M., Lalk, M.,Hecker, M., Helmann, J. D. & Antelmann, H. (2009). Genome-wideresponses to carbonyl electrophiles in Bacillus subtilis: control of thethiol-dependent formaldehyde dehydrogenase AdhA and cysteineproteinase YraA by the MerR-family regulator YraB (AdhR). MolMicrobiol 71, 876–894.

Nishizaki, M., Sasaki, J., Fang, B., Atkinson, E. N., Minna, J. D., Roth,J. A. & Ji, L. (2004). Synergistic tumor suppression by coexpression ofFHIT and p53 coincides with FHIT-mediated MDM2 inactivationand p53 stabilization in human non-small cell lung cancer cells.Cancer Res 64, 5745–5752.

Pan, Y. R., Lou, Y. C., Seven, A. B., Rizo, J. & Chen, C. (2011). NMRstructure and calcium-binding properties of the tellurite resistanceprotein TerD from Klebsiella pneumoniae. J Mol Biol 405, 1188–1201.

Perez, J. M., Calderon, I. L., Arenas, F. A., Fuentes, D. E., Pradenas,G. A., Fuentes, E. L., Sandoval, J. M., Castro, M. E., Elıas, A. O. &Vasquez, C. C. (2007). Bacterial toxicity of potassium tellurite:unveiling an ancient enigma. PLoS ONE 2, e211.

Reott, M. A., Parker, A. C., Rocha, E. R. & Smith, C. J. (2009).Thioredoxins in redox maintenance and survival during oxidativestress of Bacteroides fragilis. J Bacteriol 191, 3384–3391.

Ritz, D., Patel, H., Doan, B., Zheng, M., Aslund, F., Storz, G. &Beckwith, J. (2000). Thioredoxin 2 is involved in the oxidative stressresponse in Escherichia coli. J Biol Chem 275, 2505–2512.

Sambrook, J. & Russell, D. (2001). Molecular Cloning – A LaboratoryManual, 3rd edn. Cold Spring Harbor, NY: Cold Spring HarborLaboratory Press.

Scharf, C., Riethdorf, S., Ernst, H., Engelmann, S., Volker, U. &Hecker, M. (1998). Thioredoxin is an essential protein induced bymultiple stresses in Bacillus subtilis. J Bacteriol 180, 1869–1877.

Schroeter, R., Voigt, B., Jurgen, B., Methling, K., Pother, D. C.,Schafer, H., Albrecht, D., Mostertz, J., Mader, U. & other authors(2011). The peroxide stress response of Bacillus licheniformis.Proteomics 11, 2851–2866.

Scott, C., Rawsthorne, H., Upadhyay, M., Shearman, C. A., Gasson, M.J., Guest, J. R. & Green, J. (2000). Zinc uptake, oxidative stress and theFNR-like proteins of Lactococcus lactis. FEMS Microbiol Lett 192, 85–89.

Seefeldt, K. E. & Weimer, B. C. (2000). Diversity of sulfur compoundproduction in lactic acid bacteria. J Dairy Sci 83, 2740–2746.

Serata, M., Iino, T., Yasuda, E. & Sako, T. (2012). Roles of thioredoxinand thioredoxin reductase in the resistance to oxidative stress inLactobacillus casei. Microbiology 158, 953–962.

Smith, J. J. & McFeters, G. A. (1996). Effects of substrates andphosphate on INT (2-(4-iodophenyl)-3-(4-nitrophenyl)-5-phenyltetrazolium chloride) and CTC (5-cyano-2,3-ditolyl tetrazoliumchloride) reduction in Escherichia coli. J Appl Bacteriol 80, 209–215.

Stillman, T. J., Upadhyay, M., Norte, V. A., Sedelnikova, S. E.,Carradus, M., Tzokov, S., Bullough, P. A., Shearman, C. A., Gasson,M. J. & other authors (2005). The crystal structures of Lactococcuslactis MG1363 Dps proteins reveal the presence of an N-terminal helixthat is required for DNA binding. Mol Microbiol 57, 1101–1112.

Tachon, S., Michelon, D., Chambellon, E., Cantonnet, M., Mezange,C., Henno, L., Cachon, R. & Yvon, M. (2009). Experimental conditionsaffect the site of tetrazolium violet reduction in the electron transportchain of Lactococcus lactis. Microbiology 155, 2941–2948.

Turner, R. J., Hou, Y., Weiner, J. H. & Taylor, D. E. (1992). Thearsenical ATPase efflux pump mediates tellurite resistance. J Bacteriol174, 3092–3094.

Turner, M. S., Tan, Y. P. & Giffard, P. M. (2007). Inactivation of aniron transporter in Lactococcus lactis results in resistance to telluriteand oxidative stress. Appl Environ Microbiol 73, 6144–6149.

Uziel, O., Borovok, I., Schreiber, R., Cohen, G. & Aharonowitz, Y.(2004). Transcriptional regulation of the Staphylococcus aureusthioredoxin and thioredoxin reductase genes in response to oxygenand disulfide stress. J Bacteriol 186, 326–334.

Vido, K., Spector, D., Lagniel, G., Lopez, S., Toledano, M. B. &Labarre, J. (2001). A proteome analysis of the cadmium response inSaccharomyces cerevisiae. J Biol Chem 276, 8469–8474.

Vido, K., Le Bars, D., Mistou, M. Y., Anglade, P., Gruss, A. & Gaudu, P.(2004). Proteome analyses of heme-dependent respiration inLactococcus lactis: involvement of the proteolytic system. J Bacteriol186, 1648–1657.

Vido, K., Diemer, H., Van Dorsselaer, A., Leize, E., Juillard, V., Gruss,A. & Gaudu, P. (2005). Roles of thioredoxin reductase during theaerobic life of Lactococcus lactis. J Bacteriol 187, 601–610.

Wessel, D. & Flugge, U. I. (1984). A method for the quantitativerecovery of protein in dilute solution in the presence of detergentsand lipids. Anal Biochem 138, 141–143.

Edited by: J. Kok

P. Efler and others


Two Lactococcus lactis thioredoxin paralogues play different roles in responses to arsenate and...

Documents

Transcript of Two Lactococcus lactis thioredoxin paralogues play different roles in responses to arsenate and...