The kdgRKAT operon of Bacillus subtilis:detection of the transcript and regulation by the kdgR and...

Downloaded from www.microbiologyresearch.org by

IP: 54.90.167.105

On: Sat, 28 May 2016 22:28:23

Mkrobiology (1 998), 144,3 1 1 1-3 1 1 8 Printed in Great Britain

The kdgRKAT operon of Bacillus subtilis: detection of the transcript and regulation by the k d R and ccpA genes

Petar Pujic, Rozenn Dervyn, Alexei Sorokin and S. Dusko Ehrlich

Author for correspondence: Alexei Sorokin. Tel: +33 1 34 65 25 19. Fax: +33 1 34 65 25 21. e-mail : sorokinea biotec.jouy.inra.fr

Laboratoire de GenCtique Microbienne, lnstitut National de la Recherche Agronomique, Domaine de Vilvert, 78352 Jouy en Josas cedex, France

Transcription of a new catabolic operon in Bacillus subtilis, involved in the late stages of galacturonic acid utilization, has been studied. The operon consists of four genes: kdgR, encoding the putative regulator protein; kdgK, encoding 2- keto-3-deoxyg I uconate kinase ; kdgA, encoding 2-keto-3-deoxyg luconate-6- phosphate aldolase; and kdgr, encoding a transporter. These four genes are organized in one transcriptional unit and map at 198" of the B. subtik chromosome. Primer extension experiments and Northern blot analysis show that an active &-dependent promoter precedes kdgR and transcription is terminated at the putative pindependent terminator downstream of kdgr. The operon is negatively regulated by the kdgR and ccpA gene products, which belong to the Lac1 family of transcription regulators. The expression of the genes in this operon can be induced by galacturonate and strongly repressed when glucose is present in the growth medium. Knockout mutations in genes kdgR and ccpA remove, respectively, the effects of galacturonate and glucose on the transcription of this operon.

Keywords : kdg operon, galacturonate metabolism, Bacillus subtilis, KdgR regulator, CcpA regulator

INTRODUCTION

Bacillus subtilis is a Gram-positive sporulating bacterium that inhabits soil and rotting plant materials, from where it can use different organic compounds as carbon and energy source. The plant cell wall is constituted from different polymers, such as cellulose, consisting of glucose units, o r pectin, composed of galacturonic acid residues, partially as methyl ester. The uronic acids are also an important part of several other polymers, for example, alginic acid in bacteria (Govan & Deretic, 1996), and chondroitin sulphate and hyaluronic acid in animals (Lehninger et al., 1993). Bacteria have evolved in different ways to use these polymers. One of the genetically well-studied pathways is pectin degradation in Erwinia chrysanthemi (Hugou- vieux-Cotte-Pattat et al., 1996). This bacterium produces several extracellular pectinases that degrade pectin into unsaturated and saturated oligogalacturonides and

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Abbreviations: KDG, 2-keto-3-deoxygluconate; KDGP, 2-keto-3-deoxygluconate 6-phosphate; DKI, 5-keto-4-deoxyuronate; DKII, 2,5-diketo-3- deoxygluconate; MU, Miller units.

galacturonate. These compounds are transported into the cell and metabolized by two known pathways (Fig. 1). Galacturonate is isomerized into tagaturonate by the product of the uxaC gene, and transformed into altronate by a specific oxidoreductase, the product of uxaB. Altronate dehydratase, the product of uxaA, produces 2-keto-3-deoxygluconate (KDG) . Unsaturated digalacturonate is metabolized into 5-keto-4-deoxyuronate (DKI) and galacturonate by a specific oligogalacturonate lyase, the product of the ogl gene. DKI is then isomerized into 2,5-diketo-3-deoxygluconate (DKII) by an isomerase encoded by kduZ, and reduced into KDG by the reductase encoded by kduD. KDG is therefore a common intermediate of both pathways. KDG is a key intermediate in the degradation of gluconate in the modified Entner-Doudoroff pathway (Szymona & Doudoroff, 1958), and also in the degradation of galacturonate and glucuronate in Escherichia coli (Ashwell et al., 1960; Lin, 1996). KDG is phosphorylated by the specific kinase KdgK into 2-keto-3- deoxygluconate 6-phosphate (KDGP) , followed by aldolase cleavage (Entner-Doudoroff aldolase), which produces pyruvate and ~-glyceraldehyde-3-phosphate, thus entering the central glycolytic pathway.

0002-2209 0 1998 SGM 3111


IP: 54.90.167.105

On: Sat, 28 May 2016 22:28:23

P. P U J l C a n d OTHERS

(a)

palacturonatc

glucurc

pectin

JJ a? (pem)

'? (peh) G-l poIyga[tu;;;;;

out saturated digalacturonatc unsaturated digalacturonate

saturated digalacturonate unsaturated digalacturonate I

5-kcto-deoxyuronate (DKI)

galacturonatc tagaturonate %yjmI(uuB) 4 kdul

? yjmG (exnT)

mate glucurc

altr'.onate 2,5-diketo-3-deoxygluconate (DKII)

yjmJ (uxaA) 4 kduD

2-keto-3-deoxygluconate (KDG) yjmA (im C)

yjmE (uxirA)

2-kcto-3-deoxygluconate 6-phosphate (KDGP) .mate fructuronate 'nann"nate

? (uxuB)

kdgT

@

I I pyruvate + glyceraldehyde 3-phosphate I

uxaC uxuA exuT uxaR uxaB uxa.4 I I

1300 kb I

KDG

1312 kb

C D E F J

aldolase

kduD kdul kdgR kdgK kdgA kdgT

2327 kbT @@ 1)1)1)1) 2321 kb

Fig. 1. Possible degradative pathways of pectin and galacturonate in Erwinia chrysanthemi, Escherichia coli and Bacillus subtilis. (a) Different steps are catalysed by the products of the genes specified near the corresponding arrow. Biochemical pathways are based on genetic and biochemical studies in Gram-negative bacteria (Lin, 1996; Hugouvieux- Cotte-Pattat et a/., 1996). Corresponding genes found by homology in B. subtilis are in bold print. The genes whose symbols in Gram-negative bacteria do not coincide with the B. subtilis homologues are indicated in parentheses. Question marks indicate functions whose attribution to a B. subtilis gene is unclear. pem A, 6, genes encoding pectin methylesterase; pel, pectate lyase; ogl, oligogalacturonate lyase; kdul, 5-keto-4-deoxyuronate isomerase; kduD, 2-keto-3- deoxygluconate oxidoreductase; exuT, exuronate permease; uxaC, uronate isomerase; uxaB, altronate oxidoreductase; uxaA, altronate dehydratase; UXUB, mannonate oxidoreductase; UXUA, mannonate dehydratase; kdgT, 2-keto-3- deoxygluconate permease; kdgK, 2-keto-3-deoxygluconate kinase; kdgA, 2-keto-3-deoxygluconate-6-phosphate aldolase. (b, c) Organization of the yjm, kdu and kdg operons of B. subtilis. Highest homology hits with known proteins are indicated within ovals. Putative gene functions corresponding to proteins characterized in Gram-negative bacteria are shown above or below arrows depicting ORFs found by DNA sequencing (Rivolta eta/ . , 1998; Kunst eta/ . , 1997; Sorokin et a/., 1996). 'T' indicates positions of putative p-independent transcription terminators. Chromosomal positions of operons are indicated in kb. Systematically updated homology data for B. subtilis proteins can be consulted on specialized WWW sites (http://pc13mi.biologie.uni-greifswald.de/sub2d/bacillus.htm contains links to many of them).


IP: 54.90.167.105

On: Sat, 28 May 2016 22:28:23

kdgRKAT operon of Bacillus subtilis

The genetics of pectin utilization in B. subtilis is not well known. This bacterium can synthesize pectate lyase, encoded by the pel gene, which has been cloned and sequenced (Nasser et al., 1993). The corresponding protein was purified (Nasser et al., 1990) and its X-ray structure established (Pickersgill et al., 1994). Another potential pectin-depolymerizing-enzyme gene, pel& was found during the B. subtilis sequencing project (Kunst et a[., 1997).

A putative operon designated yjmABCDEFGHZJ was discovered in the course of systematic sequencing of the entire B. subtilis genome (Rivolta et al., 1998; Kunst et al., 1997). This operon contains four open reading frames: yjmG, A, Z and J , which encode proteins homologous, respectively, to ExuT, UxaC, B and A from E. coli (Fig. 1) . Thus the complete biochemical pathway from galacturonate to KDG is present in B. subtilis. YjmE has high homology to E. coli mannonate hydrolase UxuA. yjmF was assumed to encode mannonate oxidoreductase UxuB, although in this case homology to the enzyme from Gram-negative bacteria is not very convincing. yjmB was postulated to encode b- glucuronide transporter UidB (Rivolta et al., 1998). It is not clear from sequence analysis if putative dehydro- genases YjmC and YjmD are also involved in KDG metabolism.

Another two divergent operons, designated kduZD and kdgRKAT, were also identified by homology in the course of systematic sequencing of the lysA-iluA region of the B. subtilis genome (Sorokin et al., 1996; Kunst et al., 1997). These operons encode genes presumably involved in conversion of DKI into KDG and KDG into pyruvate and glyceraldehyde 3-phosphate (Fig. 1).

Since putative counterparts of almost all the components known for pectin utilization in Gram-negative bacteria exist in B. subtilis, the latter can serve as a model to study the regulation of these metabolic pathways in Gram-positive bacteria. Three genes encoding DNA- binding regulators of transcription can be postulated to be involved in this regulation. The first is yjmH, encoding the LacI-family transcription regulator. The corresponding protein can negatively regulate expression of the yjm operon, whose genes are involved in galacturonate or glucuronate degradation, thus pro- viding an inducer (which is KDG in Gram-negative bacteria : Nasser et al., 1991) for the kdg operon. KdgR is probably a repressor of transcription of the kdu and kdg operons. The expression of pectin-degradation genes should be a target for the general catabolite- repression system which, in B. subtilis, usually transfers a metabolic signal to genes by modulating the DNA- binding activity of the CcpA protein, which is also a LacI-family regulator.

In this paper we present data on the regulation of kdg- operon transcription by the KdgR and CcpA regulatory proteins. We show that kdg-operon transcription can be induced by galacturonate and repressed by glucose. KdgR and CcpA are, respectively, the terminal molecular transmitters of these two regulatory signals.

METHODS

Bacterial strains and growth conditions. Escherichia coli cells harbouring plasmids were grown on Luria broth (LB) (Sam- brook et al., 1989) supplemented with ampicillin (50-100 pg ml-’). B. subtilis strain 168 trpC was grown on LB or o n minimal medium containing: (NH,),SO, (2 g 1-’), K,HPO, (14 g 1-’), KH,PO, (6 g l-’), sodium citrate (1 g 1-’) and 20 pM FeCl,. Carbon sources used were added at concentrations of 10-30mM. Polygalacturonate was added to a final concentration of 4 g 1-’. Plasmid pDT1, which enables transcriptional fusions to be made with the B-galactosidase gene as a reporter, has been described elsewhere (Lapidus et al., 1997). Plasmid pMUTIN- 2mcs, which contains the P-galactosidase gene with the ribosome-binding site of spoVG of B. subtilis and the IPTG- inducible P,,,, promoter, was kindly provided by Dr V. Vagner (Gknktique Microbienne, INRA, Jouy-en- Josas) . This plasmid enables the creation of chromosomal integrations in B. subtilis, thereby placing the B-galactosidase gene under the control of the studied B. subtilis promoter, separating it from the downstream genes, which are under P,,,, regulation in the same strain. The list of constructed strains is presented in Table 1.

To construct pDTKDGRI and pDTKDGKI the fragments starting from sequences 5’ dTGAATGTGCAGG . . . 3 ’ and 5’ GGAGCAGAAAGC . . .3’, respectively, and ending in sequences 5’ ... ATGCTCTTAAAG 3’ and 5’ ... TAGAAGG- CTGCC 3’, respectively, were cut from M13mp19 phages carrying these fragments. These phages were constructed in the course of the sequencing of this region (Sorokin et af., 1996). Primers KRHl (5’ dGCCAAGCTTAGCTGAATGT- GCAGGAG 3’) and KRB2 (5’ dCGGGATCCCGGAGTA- CATCCTTGTT 3’) were used to amplify the corresponding fragment from the B. subtilis chromosome. The PCR fragment was then digested by BamHI and HindIII and cloned in pMUTIN2mcs to construct pM2KDGRI. In a similar way, the primers KRH2 (5’ dGCCAAGCTTAACAAGGAGCAGCT- TAAG 3’) and KRBl (5’ dCGGGATCCAACCTCCTCTT- AGTTGA 3’) were used to construct pM2KDGRF, and the primers KTHl (5’ dGCCAAGCTTGAGACTTATTCAGC- AAG 3’) and KTBl (5’ dCGGATCCTCAGTTGATCGTC- ATTT 3’) to construct pM2KDGTF. The primers KDGAEl (5’ dGGAATTCGTGCAGTTGAAGT 3’) and KDGAHl(5’ dAAAAGCTTTAATGTCGTAAAAC 3’) were used for PCR amplification to construct pDTKDGAI. The PCR fragment was then digested by EcoRI and HindIII and inserted into the correspondingly cut pDTl.

Plasmids pDTKDGRI (pDT1 as vector) and pM2KDGRI (pMUTIN2mcs as vector) contain internal fragments of the kdgR gene. Their integration into the B. subtilis chromosome destroys this gene. The expression of the kdgKAT genes can presumably be regulated by IPTG in the BSPP2 strain, but we did not use this property of the pMUTIN2mcs-based con- structs during this work. pM2KDGRF (pMUTIN2mcs as vector) contains the 3’ part of kdgR and the 5’ part of kdgK. The strain BSPP3 therefore carries active kdgR and the expression of kdgKAT depends on IPTG. As mentioned above, all experiments described below were performed without IPTG. pDTKDGKI and pDTKDGAI (pDT1 as vector) contain internal parts of kdgK and kdgA, respectively, and therefore their integration into the chromosome in- activates these genes. pM2KDGTF contains the 3’ terminal fragment of the kdg operon. Integration of this plasmid into the chromosome does not modify any gene of the operon and measuring of B-galactosidase activity in BSPP7 allows moni-

3113


IP: 54.90.167.105

On: Sat, 28 May 2016 22:28:23

P. P U J I C a n d OTHERS

Table 1. Bacterial strains used in this study

Strain Genotype Source or reference

E. coli TG1

B. subtilis

168 BSPPl BSPP2 BSPP3 BSPP4 BSPPS BSPP7 GM1096

BSPPS BSPPlO

supE hsdDS thi A(lac-proAB) F"traD36 proAB+ ladq lacZAMlS]

trpC2 168 kdgR : : pDTKDGRI [Cm"] 168 kdgR : : pM2KDGRI [Em"] 168 kdgR : : pM2KDGRF [Em"] 168 kdgK : : pDTKDGKI [Cm"] 168 kdgA : : pDTKDGAI [Cm"] 168 kdgT: : pM2KDGTF [Em"] trpC2 leuB6 sacB'-'lacZ ccpA : : Tn91 7A(lacZ-erm)SpR

BSPP2 ccpA : : SpR BSPP7 ccpA : : Sp"

Sambrook et al. (1989)

Laboratory stock This work This work This work This work This work This work Deutscher et al. (1994)

This work This work

toring of native regulation. Strains BSPP9 and BSPPlO were constructed by transforming, respectively, BSPP2 and BSPP7 by DNA of strain GM1096 (Deutscher et al., 1994), containing the SpR marker inactivating the ccpA gene, and selecting for spectinomycin resistance. All plasmid constructions were tested by sequencing and all B. subtilis chromosomal integrations were verified by Southern hybridization.

Chloramphenicol was used at 5 pg ml-l for growth of strains BSPP1,4 and 5, erythromycin at 1 pg ml-' for strains BSPP2,3 and 7, and spectinomycin at 100 pg ml-l for BSPP9 and 10.

DNA manipulations. Chromosomal and plasmid DNA preparations, standard cloning procedures and DNA sequencing were applied as previously described (Sorokin et al., 1995).

RNA extraction. Cells from 50 ml culture were centrifuged and resuspended in 0-2 ml TEI buffer (10 mM Tris/HCl, pH 7.5; 1 mM EDTA; 10 mM sodium iodoacetate). The cells were then transferred to a 2 ml tube containing: 0.5 ml glass beads (diameter 100 pm) ; 0.4 ml phenol saturated with 0-1 M sodium citrate, pH 4.3; 0.8 ml 15 mM hexadecyltrimethylammonium bromide solution in 50 mM sodium acetate, pH 4 5 ; 1 mM DTT. The tube was shaken on a Biospec Homogenizer (Biospec Products) at the maximum speed setting for 1-2 min. After centrifugation, the supernatant was treated once with 200 pl of chloroform and subsequently with an equal volume of phenol/chloroform/isoamyl alcohol (ratio 125 :24: 1, by vol.; pH 4.7). RNA was precipitated by adding 0.1 vol. 3 M sodium acetate, pH 5, and an equal volume of 2-propanol. After 15 min on ice, precipitated RNA was collected by centrifugation in a microcentrifuge, washed with 95 '/o etha- nol, air-dried and stored in TE (10 mM Tris/HCl, pH 7; 1 mM EDTA) or water at -80 OC.

Northern analysis. A 30 pg (3-6 pl) sample of the total RNA preparation was mixed with 100 mM sodium phosphate buffer (3 pl) pH 7.0, then 5-5 p140% glyoxal was added, followed by DMSO to adjust the volume to 30 pl. The samples were denatured at 55 "C for 1 h and mixed with 3 p1 gel loading buffer containing: 80% formamide, 2 mM EDTA pH 7.0, 0.1 O h xylene cyanol, and 0.1 O/O bromophenol blue. Electro- phoresis was performed in 1 % Sea Kem GTG agarose gel (Gibco) using 10 mM sodium phosphate buffer pH 7.0, with 10 mM sodium iodoacetate. Electrophoresis conditions were :

50-60 V, 50 mA, for 6 h. After electrophoresis, RNA was transferred to a NYTRAN 0.2 pm membrane (Schleicher & Schuell) by capillary transfer using 20 x SSC to create a salt concentration gradient. RNA was then fixed by UV cross- linking (Stratagene) for 10 s and baked for 20 min at 80 "C in an oven. Hybridization was performed in 50 '/o formamide, 2 x Denhardt's reagent, 0.1 '/o SDS, 100-200 pg denatured salmon sperm DNA ml-' and 5xSSC. Hybridization was performed overnight a t 42 "C with slow shaking. The membrane was then washed twice for 15 min each time in 10 x SSC, 1 % SDS at room temperature, and then twice for at least 15 min in 1 x SSC, 1 '/o SDS at 42 "C, and then again for 15 min in 0.1 x SSC, 1 '/o SDS a t 42 "C with shaking. The membrane was then exposed to X-ray film or scanned with a Phos- phoImager (Molecular Dynamics). Primer extension experiments were done as described by Soldo et al. (1993).

PGalactosidase activity. B-Galactosidase activity was determined according to the method of Miller (1972), and expressed in Miller units (MU). The cells were permeabilized for ONPG by adding a drop of toluene to a culture sample.

RESULTS

Detection of the transcript corresponding to the kdg operon

Northern analysis of total RNA with probes specific to the kdgRKAT region was performed to detect the corresponding transcript. RNA was prepared from the cells growing exponentially in minimal media containing polygalacturonate, galacturonate, glucuronate, ribose, glucose, gluconate, or succinate as a carbon source, at the middle of exponential growth. Since it was shown in Gram-negative bacteria that the direct intracellular effector of the regulator of the pectin- utilization-system genes is KDG (Nasser et al., 1991), we expected that the first three carbon sources would provide a metabolic inducer for expression of the kdg operon in B . subtilis. A transcript of about 3.7 kb was detected in cells grown on galacturonate, glucuronate and gluconate, with probe specific to gene kdgA (Fig.

31 14


IP: 54.90.167.105

On: Sat, 28 May 2016 22:28:23


(c) TATTAAAAAAACTCTTCCCAAATTTTAAAATTATGGTATTCTTTTGTTGTMCGGTTTTGMATCGG~TCWTA - - b e -

-lo mRNA -35

TATGAGATACAGGAGCACT - RBS

T G x G y j M C A G G C C ATACAACCATCWGACGTAGCTGMTGT 3 '-TQTTGGTAGTTTCTGCATC-5 '

KdgR start primer R1

GCAGGAGllTCGMATCGACAGTTTCCCGCTATAT 3 '-TCCTCWGCTTTAGCTGTC-5 '

primer R2

..................................................................................................... Fig. 2. Structural characterization of the kdg-operon transcript. (a) The results of Northern hybridization with kdgA gene fragment as probe. RNA was extracted from cells grown on minimal media with the following carbon sources: 1, ribose; 2, galacturonate; 3, glucuronate; 4, polygalacturonate; 5, succinate; 6, glucose; 7, gluconate. (b) Primer extension mapping of the Pkd,. promoter. Sequencing ladders show reactions performed on M13mp18 ssDNA as template with ' -40' standard primer. Ladders R1 and R2 correspond to the reverse-transcriptase reactions performed with primers R1 and R2. (c) Promoter region of gene kdgR. Transcription start indicated by bent arrow. ' -35' and '-10' boxes are marked. An inverted repeat is marked by arrows below the sequence. The repeat may represent a CcpA-binding area since it has high homology to the consensus sequence of catabolite-responsive elements, which is (T/A)G N AA( UG)CG N (T/A) (T/A) N CA (H uec k e t a/., 1994). The ribosome-binding site and the putative start of the kdgR coding sequence are marked. Sequences of the primers R1 and R2 are also shown.

2a). The transcription was abundant in the medium containing galacturonate and was much lower in the presence of glucuronate and gluconate. No signal was detected when glucose, ribose, polygalacturonate or succinate was used. A similar result was obtained (not shown) when kdgR- or kdgK-specific probes were used. These results indicate that the genes kdgRKAT are cotranscribed, as suggested from the nucleotide sequence (Sorokin et al., 1996). These results also indicate that galacturonate is a better source of the kdg-operon inducer in B. subtilis than glucuronate, although the last can also be used as a carbon source. The absence of operon induction by polygalacturonate and a weak induction by gluconate were surprising. At the moment we do not have an explanation for these results.

The 5' end of the transcript was determined by primer extension. Two oligonucleotides were used in these experiments (Fig. 2b). The results show that the kdgRKAT mRNA starts from the sequence 5'- TGTAACG . . .3', 51 bp upstream of the start codon of gene kdgR. Two sequences can be indicated upstream of this start point as potential sites for recognition by oA. These are TATTCT, as the ' - 10' box, and TTCCCA, as the ' -35' box (Fig. 2c). Other elements important for B. subtilis promoters, including the dinucleotide T G (at -15, -14) and an A-rich region near -43 (Helmann, 1995), are present in this region. We des- ignate this promoter Pkd,.

kdgR encodes a negative regulator of transcription of the kdg operon

Reporter-gene expression and Northern-hybridization experiments for the cells with active and inactive kdgR were performed to verify the influence of gene kdgR, encoding a LacI-family regulator, on the transcription

kdgRkdgKkdgA kdgT 1 1 1 1 lac2 lac1 ApREmR T T

- 4 . m pMUTIN2mcs

memce) 'I pMUTIN2mcs

Fig. 3. Induction of the Pkd, promoter by inactivation of the kdgR gene. (a) Chromosome structures of strains BSPP7 and BSPP2 obtained upon integration, respectively, of pM2KDGTF and pM2KDGRl into the 6. subtilis chromosome. Chromosomal sequences are shown in black and pMUTlN2mcs sequences in white. Transcription terminators are marked T. P,,,, is not shown. (b) Northern hybridization of RNA isolated from the strains: 1, BSPP2; 2, BSPP1; 3, 168. RNA was extracted in the middle of exponential growth in LB liquid medium. kdgR internal fragment was used as a hybridization probe.

from the Pkdg promoter. For this purpose B. subtilis strains BSPPl to BSPP7 were constructed (Table 1, Fig. 3a). Measurements of P-galactosidase activity in BSPPl and BSPP2, where kdgR is inactivated by inserted plasmids, on LB showed that it could exceed up to 1000 MU (not shown). By contrast, Pkdg was essentially inactive in these growth conditions in the strain carrying a fusion which leaves kdgR intact (strain, BSPP3, not shown). Contrary to the mutations produced by plasmids inserted in the chromosomes of strains BSPPl

31 15


IP: 54.90.167.105

On: Sat, 28 May 2016 22:28:23

P. P U J I C a n d OTHERS

(a) 1000, 110

250.

200

150

100 n

5 50

.- E > .-

'

I I

250-

200 -

150-

100 -

10

1

0.1

I o . 0 1 2 3 4 5 6 7 8

Time (h)

800

1 600

400 0.1

200 n

8, 0.01 8

1200 10 *z 2 3 4 5 6 7 8 -

E Q, 0

Q, U

- - 1000

800 1

600

400 0.1

2oo* 2 3 4 5 6 7 8 0.01

Time (h)

Fig. 4. Regulation of the P,,, promoter. Strains: (a) BSPP7 (kdgR+ ccpA+), (b) BSPP2 (kdgR- ccpA+), (c) BSPP10 (kdgR' ccpA-) and (d) BSPP9 (kdgR- ccpA-) were grown on minimal media containing 0.5% galacturonate (m, P-galactosidase activity; 0, cell density), 0.5% galacturonate and 2 % glucose (0, P-galactosidase activity; 0, cell density), and 2% glucose (+, P-galactosidase activity; 0, cell density) as carbon sources. The experiments were repeated three times for each strain and medium; representative results are shown.

and BSPP2, fusions which knock out kdgK (strain BSPP4) or kdgA (strain BSPP5) but not kdgR, did not activate P,,,. Addition of galacturonate to a concentration 0.5 Yo was needed to see /I-galactosidase activity in BSPP4 and BSPPS grown on LB medium (not shown). This result was confirmed by Northern hybridization of RNA isolated from strains BSPPl and BSPP2 grown on LB medium (Fig. 3b). An RNA band corresponding to the kdgR-lac2 fusion (3.7 kb) together with a presumed degradation product (0.7 kb) was detected, whereas no kdgR-specific transcription can be seen for the parent strain B. subtilis 168. We conclude from these experiments that inactivation of kdgR, but not of kdgK (strain BSPP4) or kdgA (strain BSPPS) causes high transcription of the kdg operon.

Expression of the kdg operon is induced during growth on galacturonate and repressed by glucose

We did not detect transcription of the kdg operon by Northern analysis when B. subtilis 168 was grown on LB (Fig. 3b, slot 3). Presumably, some amount of the KdgR

repressor is synthesized and this protein reduces the transcription of the operon to an undetectable level. We found also that glucose at concentrations higher than 1 % reduces the expression of P-galactosidase in strains BSPPl and BSPP2, containing knockout mutations in the kdgR repressor gene (not shown). To understand how the operon may be induced we used strain BSPP7, which carries the lac2 reporter inserted between the last gene of the operon kdgT and the p-independent terminator (Fig. 3a). Strains BSPPS and BSPP10, carrying an inactivated ccpA gene, were constructed by transforming BSPP2 and BSPP7 with DNA of GM1096 to see if glucose repression is due to the catabolic CcpA repressor (Table 1). The kinetics of P-galactosidase activity in these strains grown on minimal media, containing galacturonate (0.5%), galacturonate with glucose (0.5 and 2%) and glucose ( 2 % ) as carbon sources, are shown in Fig. 4. It can be seen from Fig. 4(a) that galacturonate induced transcription from P,,, in BSPP7 at the end of exponential growth. This transcription was repressed by glucose to an undetectable level (less than 1 MU). kdgR

31 16


IP: 54.90.167.105

On: Sat, 28 May 2016 22:28:23


knockout mutation increased the highest expression level about 5-fold, but did not prevent glucose repression and did not change the late character of transcription induction (Fig. 4b). It is interesting that in the case of BSPP2, glucose repressed transcription 10-fold (from 100 to 1000 MU), although, in the case of an active kdgR gene, repression of the operon in the absence of galacturonate was higher than 200-fold (from < 1 to 200 MU), which can be explained by higher affinity of the KdgR to the kdg-operon promoter region, compared to that of CcpA. In the case of BSPP7 grown on the medium containing galacturonate and glucose (Fig. 4a), the low level of /I-galactosidase activity ( < 3 MU) can be explained by the absence of intracellular effector of KdgR. This effector must appear as a result of degradation of galacturonate by the enzymes encoded by the yjrn operon, which is probably strongly repressed by CcpA in these conditions (see Discussion).

The ccpA knockout mutation renders the expression of the kdg operon independent of the presence of glucose in the growth medium. It is also remarkable that in cells with inactivated ccpA, transcription of the kdg operon starts early in the exponential phase of growth (Fig. 4c). Finally, to make transcription independent of the presence of galacturonate in the growth medium, it is necessary to inactivate kdgR by knockout mutation (Fig. 4c, d).

DISCUSSION

The experiments described above provide the first information on the transcriptional regulation in B. subtilis of kdg genes, encoding proteins involved in KDG metabolism. We found that transcription of this operon can be induced by galacturonate and strongly repressed by glucose. Knockout mutations in two genes, kdgR and ccpA, encoding LacI-family regulatory proteins, release transcription of the operon from depen- dence on galacturonate and glucose, respectively. CcpA is also responsible for the repression of the kdg-operon transcription in the exponential phase of growth.

Putative functional counterparts of E. coli galacturonate catabolic enzymes were discovered in the course of the B. subtilis genome-sequencing project (Rivolta et al., 1998; Kunst et al., 1997). Proteins encoded by the B. subtilis genes yjmA, G, I and J can convert galacturonate to KDG as their counterparts do in E. coli. KDG is the starting substrate for the action of enzymes encoded by the kdg operon. It was shown for Gram-negative bacteria that KDG is also one of the effectors for the protein KdgR, which regulates transcription of genes involved in this metabolism. It is reasonable to hypothesize that in B. subtilis KDG can also inhibit KdgR protein for DNA binding.

CcpA can influence kdg-operon transcription by two possible pathways. First, it can bind directly to the kdg- operon promoter region and thus inhibit transcription in the presence of glucose. Second, it can repress transcription of the yjm operon and thus prevent

formation from galacturonate of intracellular inducer, which is probably KDG.

Several promoters for which catabolite repression depends on CcpA have been described for B. subtilis (for a review see Hueck & Hillen, 1995). The efficiency of repression by CcpA varies in these systems from 4- to 100-fold. In the case of the kdg operon the repression is about 10-fold. In all cases where it was studied, the efficiency of repression has been found to depend on correctly phosphorylated Hpr protein, suggesting that the signal for activation of CcpA is transferred by protein-protein interaction between phosphorylated Hpr and CcpA (Deutscher et al., 1994). It would be interesting to test whether the same mechanism is operational in the case of the Pkdg promoter.

The study of the regulation of kdg-operon transcription is simplified by the fact that all potential transcriptional regulators can be easily postulated. We demonstrated in this work that the KdgR protein, encoded by the kdg operon, is indeed the principal regulator of the operon’s transcription. Repression of the transcription by glucose depends exclusively on the activity of another LacI- family regulator, CcpA. YjmH protein, which is also a LacI-family regulator, is probably involved in the regulation of the yjrn operon, which encodes the metabolic pathway from galacturonate to KDG. There- fore, YjmH can also influence the DNA-binding activity of KdgR, by regulating the levels of the relevant enzymes and thus by changing the level of intracellular effector in the cells grown on galacturonate. The yjrn operon was also postulated to be regulated by CcpA, since a perfect catabolite-responsive element consensus was found in the sequence of the promoter region (Rivolta et al., 1998). It would be useful, for detailed understanding of this regulatory network, to separate the regulation of yjm and kdg operons by mutations in different yjm genes. This approach may elucidate the exact nature of the intracellular KdgR effector.

An important question about essentiality of the kdg operon for growth of B. subtilis on galacturonate as the only carbon source remains unanswered. It was surprising to find that strains BSPP2, BSPP3 and BSPP4 were able to grow on galacturonate almost as well as BSPP7 (not shown). Only BSPPS did not grow; apart from an inability to metabolize KDG, this may also be explained by accumulation of KDGP which is toxic to cells. We found that strains BSPP2, BSPP4 and BSPPS, when grown on LB medium containing 15 mM galacturonate, accumulate up to 2 mM of KDG, determined by thiobarbituric acid assay (Weissbach & Hurwitz, 1959), extracellularly (P. Pujic, unpublished). Since this did not happen with wild-type B. subtilis 168 or BSPP7, we can assume that activities of aldolase (KdgA) and kinase (KdgK), which are reduced (BSPP2) or one of which is absent (BSPP4 and BSPP5) in these strains, are needed to metabolize all the KDG produced. However, the ability of BSPP2, BSPP3 and BSPP4 to grow o n galacturonate as the only carbon source remains to be elucidated.

31 17


IP: 54.90.167.105

On: Sat, 28 May 2016 22:28:23

P. P U J I C a n d OTHERS -

We have shown that the level of kdg mRNA production can vary more than 100-fold depending on the activity of the KdgR regulator. This system may therefore be a good candidate for use as a tool to regulate expression of foreign genes in B. subtilis.

ACKNOWLEDGEMENTS

We thank Dr Valerie Vagner for plasmid pMUTIN2mcs, Dr Isabelle Martin-Verstraete for strain GM1096, Professor Jeannine Robert-Baudouy, Drs Nicole Hugouvieux-Cotte- Pattat and Ken-ichi Yoshida for discussions, Florence Haimet and Anne Aucoutourier for help in preparation of the manuscript, and Maelle Dervyn for technical assistance. This work was supported by grant BI04-CT95-0278 from the EU.

REFERENCES

Ashwell, G., Wahba, A. J. & Hickman, J. (1960). Uronic acid metabolism in bacteria. Purification and properties of uronic acid isomerase in Escherichia coli. ] Biol Chem 235, 1559-1565.

Deutscher, J., Reizer, J., Fischer, C., Galinier, A,, Saier, M. H. & Steinmetz, M. (1994). Loss of protein kinase-catalyzed phos- phorylation of HPr, a phosphocarrier protein of the phos- photransferase system, by mutation of the ptsH gene confers catabolite repression resistance to several catabolic genes of Bacillus subtilis. ] Bacteriol 176, 3336-3344.

Govan, 1. R. W. 81 Deretic, V. (1996). Microbial pathogenesis in cystic fibrosis : mucoid Pseudomonas aeruginosa and Burk- holderia cepacia. Microbiol Rev 60, 539-574.

Helmann, 1. (1995). Compilation and analysis of Bacillus subtilis 8-dependent promoter sequences : evidence for extended contact between RNA polymerase and upstream promoter DNA. Nucleic Acids Res 23, 2351-2360.

Hueck, C. J. & Hillen, W. (1995). Catabolite repression in Bacillus subtilis : a global regulatory mechanism for the Gram-positive bacteria? Mol Microbiol 15, 395401.

Hueck, C. J., Hillen, W. 81 Saier, M. H., Jr (1994). Analysis of a cis- active sequence mediating catabolite repression in Gram-positive bacteria. Res Microbiol 145, 503-518.

Hugouvieux-Cotte-Pattat, N., Condemine, G., Nasser, W. & Reverchon, 5. (1996). Regulation of pectinolysis in Erwinia chrysanthemi. Annu Rev Microbiol50, 213-257.

Kunst, F., Ogasawara, N., Moszer, 1. & 148 other authors (1997). The complete genome sequence of the Gram-positive bacterium Bacillus subtilis. Nature 390, 249-256.

Lapidus, A., Galleron, N., Sorokin, A. & Ehrlich, S.D. (1997). Sequencing and functional annotation of the Bacillus subtilis genes in the 200 kb rrnB-dnaB region. Microbiology 143, 343 1-3441.

Lehninger, A. L., Nelson, D. L. & Cox, M. M. (1993). Poly- saccharides and proteoglycans. In Principles of Biochemistry, 2nd edn, pp. 308-315. Edited by V. Neal. New York: Worth.

Lin, E. C. C. (1996). Dissimilatory pathways for sugars, polyols, and carboxylates. In Escherichia coli and Salmonella : Cellular and Molecular Biology, 2nd edn, vol. 1, pp. 307-342. Edited by F. C. Neidhardt and others. Washington, DC : American Society for Microbiology.

Miller, J. H. (1972). Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Nasser, W., Chalet, F. & Robert-Baudouy, 1. (1990). Purification and characterization of extracellular pectate lyase from Bacillus subtilis. Biochimie 72, 689-695.

Nasser, W., Condemine, G., Plantier, R., Anker, D. & Robert- Baudouy, J. (1991). Inducing properties of analogs of 2-keto-3- deoxygluconate on the expression of pectinase genes of Erwinia chrysanthemi. FEMS Microbiol Lett 65,73-78.

Nasser, W., Awade, A. C., Reverchon, 5. & Robert-Baudouy, J. (1993). Pectate lyase from Bacillus subtilis : molecular characterization of the gene, and properties of the cloned enzyme. FEBS Lett 335, 319-326.

Pickersgill, R., Jenkins, J., Harris, G., Nasser, W. & Robert- Baudouy, J. (1994). The structure of Bacillus subtilis pectate lyase in complex with calcium. Nut Struct Biol 1, 717-723.

Rivolta, C., Soldo, B., Lazarevic, V., Joris, B., Mauel, C. & Karamata, D. (1998). A 35.7 kb DNA fragment from the Bacillus subtilis chromosome containing a putative 12.3 kb operon involved in hexuronate catabolism and a perfectly symmetrical hypothetical catabolite-responsive element. Microbiology 144, 877-884.

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Soldo, B., Lazarevic, V., Margot, P. & Karamata, D. (1993). Sequencing and analysis of the divergon comprising gtaB, the structural gene of UDP-glucose pyrophosphorylase of Bacillus subtilis 168. ] Gen Microbiol 139, 3185-3195.

Sorokin,A., Serror,P., Pujic, P., Azevedo, V. & Ehrlich, 5. D. (1995). The Bacillus subtilis chromosome region encoding homologues of the Escherichia coli mssA and rpsA gene products. Microbiology

Sorokin, A., Azevedo, v.8 Zumstein, E., Galleron, N., Ehrlich, 5. D. & Serror, P. (1996). Sequence analysis of the Bacillus subtilis chromosome region between the serA and kdg loci cloned in a yeast artificial chromosome. Microbiology 142, 2005-2016.

Szymona, M. & Doudoroff, M. (1958). Carbohydrate metabolism in Rhodopseudomonas spheroides. ] Gen Microbiol22,167-183.

Weissbach, A. & Hurwitz, J. (1959). The formation of 2-keto-3- deoxyheptonic acid in extracts of Escherichia coli B. ] Biol Chem

141,311-319.

234,705-709.

Received 2 June 1998; accepted 13 July 1998.

31 18

The kdgRKAT operon of Bacillus subtilis:detection of the transcript and regulation by the kdgR and...

Documents

Transcript of The kdgRKAT operon of Bacillus subtilis:detection of the transcript and regulation by the kdgR and...