Oxidative Stress Response in Lactobacillus Plantarum WCFS1: a Functional Genomics Approach

Oxidative Stress Response in Lactobacillus plantarum WCFS1: A Functional Genomics Approach

L. Mariela Serrano

Promotor: Prof. Dr. Willem M. de Vos Hoogleraar Microbiologie Wageningen Universiteit

Co-Promotor: Dr. Eddy J. Smid Projectleider NIZO Food Research

Leden van de promotiecomissie: Prof. Dr. Tjakko Abee Food Microbiology Laboratory Wageningen University, The Netherlands

Dr. Philippe Gaudu INRA, Unité Bactéries Lactiques et pathogènes Jouy en Josas, France

Prof. Dr. Jeroen Hugenholtz NIZO Food Research Ede, The Netherlands

Prof. Dr. R. Paul Ross Moorepark Food Research Centre Cork, Ireland

Dit onderzoek is utigevoerd binnen de onderzoekschool VLAG. Oxidative Stress Response in Lactobacillus plantarum WCFS1: A Functional Genomics Approach

L. Mariela Serrano

Proefschrift Ter verkrijging van de graad van doctor op gezag van de rector magnificus van Wageningen Universiteit, Prof. Dr. M. J. Kropff in het openbaar te verdedigen op vrijdag 18 april 2008 des ochtends om elf uur in de Aula Oxidative Stress Response in Lactobacillus plantarum WCFS1: A Functional Genomics Approach L. Mariela Serrano Ph.D. thesis Wageningen University and Research Centre, The Netherlands, 2008. With summary in dutch ISBN 978-90-8504-920-3 The road not taken

Two roads diverged in a yellow wood And sorry I could not travel both And be one traveller, long I stood And looked down one as far as I could To where it bent in the undergrowth;

Then took the other, as just as fair, And having perhaps the better claim, Because it was grassy and wanted wear; Though as for that the passing there Had worn them really about the same,

And both that morning equally lay In leaves no step had trodden black. Oh, I kept the first for another day! Yet knowing how way leads on to way, I doubted if I should ever come back.

I shall be telling this with a sigh Somewhere ages and ages hence: Two roads diverged in a wood, and I- I took the one less travelled by, And that has made all the difference.

Robert Frost.

Para mis papis Julio y María Sin ellos no habría encontrado mi camino

Table of Contents

Abstract 9

Chapter 1 11 Introduction and Outline of this Thesis

Chapter 2 31 Thioredoxin Reductase is a Key Factor in the Oxidative Stress Response of Lactobacillus plantarum WCFS1

Chapter 3 59 Global Transcriptional Analysis Reveals the Specific Role of Thioredoxin Reductase in Oxidative Stress Response in Lactobacillus plantarum WCFS1

Chapter 4 85 The Thioredoxin System Plays an Important Role in Adaptation of Lactobacillus plantarum WCFS1 to Aerobic and Respiratory Growth.

Chapter 5 113 Glutathione Protects Lactobacillus plantarum WCFS1 Against Hydrogen Peroxide Stress

Chapter 6 137 Summary, Concluding Remarks, and Future perspectives

Appendix 149 Color figures and Supplementary materials

Samenvatting 219 Resúmen, comentarios finales, y perspectivas futuras 223 Training and Supervision Plan (VLAG) 233 List of publications 234 About the author 235 Acknowledgements 237

Abstract

Control of activity and functionality of microbial starter and probiotic cultures under industrial fermentation conditions is essential in order to provide a tasty, attractive, healthy, and safe product. Oxidative stress is one of the harsh conditions that fermentative microbes have managed to endure during their use in industrial fermentation processes. A widely- used lactic acid bacterium in food fermentations is Lactobacillus plantarum. Hence, understanding oxidative stress response in this micro-organism can be used for engineering robustness towards oxidative stress. There are two systems known to be involved in oxidative stress response and redox homeostasis in bacteria: the thioredoxin and glutaredoxin systems. In this study, we constructed a set of L. plantarum WCFS1 strains with alterations in these systems. Using the constructed strains under different oxidative stress conditions (hydrogen peroxide, diamide (thiol-stress), aerobic growth, and respiratory growth), global transcriptome analysis was performed. Subsequently, the functional role of the thioredoxin and glutaredoxin system was analyzed and validated using a number of different techniques including comparative genomics, enzyme assays, and q-PCR. The main results obtained in this study include the role of the thioredoxin system in oxidative stress response in L. plantarum as well as in adaptation to aerobic cultivation; characterization and overlap of the thioredoxin and glutaredoxin system in this bacterium; insight into transcriptome regulation of oxidative stress response, and unravelling of stress-response networks present in L. plantarum WCFS1. The comparative genomics and global transcriptome analysis of the oxidative stress response of L. plantarum WCFS1 presented in this thesis can be used for optimization of the performance of lactic acid bacteria in industrial fermentations.

1 Introduction and Outline of this Thesis

11 12 1 Introduction and Outline of this Thesis

Lactobacillus plantarum is a versatile lactic acid bacterium encountered in a variety of niches including industrial food fermentations. The diversity of ecological niches where this bacterium survives and prospers alludes to its flexibility. In an industrial fermentation process the flexibility of L. plantarum is reflected in its capacity to counteract adverse conditions during process including acid, osmotic, temperature, and oxidative stress. Therefore, a major challenge for improving and controlling industrial fermentations is to understand the adaptation mechanisms present in this organism towards stress. A stress may lead to activation of genes required to cope with the altered condition, while repress- ing other genes that do not fulfill an important function in the novel condition. The total set of RNAs, transcriptome, under a stress condition contains information about the biological state of the cell and the genes that play a role under this specific stress condition. Hence, transcriptome analysis is a powerful tool in obtaining new information on gene-regulatory circuits and the effect of stress on the cell.

This introduction will provide an overview of the characteristics and applications of L. plantarum WCFS1, the model organism used in this thesis. In addition, oxidative stress in bacteria will be discussed with special focus on the bacteria’s response upon this stress and the antioxidants glutathione and thioredoxin. Furthermore in this introduction the defi- nition and application of global transcriptional analysis will be presented to the reader.

1313 Lactobacillus plantarum zymes that catalyze the synthesis of important Lactic acid bacteria (LAB) are low C+G Gram- fermentation-end-products from pyruvate in- positive, non-sporulating, fermentative organ- cluding flavor and conservation metabolites isms that grow in anaerobic or micro-aerobic (lactate, acetate, acetoine, ethanol, formate, 2, habitats. The growth of these organisms requires 3-butanodiol). The high abundance of transport- high-nutrient containing niches. Their main func- encoding genes (411 proteins; 13.47% of total tion is to convert sugars -present in the raw ma- proteins) reflects the preference of L. plantarum terial- into lactic acid. Lactobacillus plantarum is WCFS1 to reside in nutrient-rich niches. On the a facultative heterofermentative member of the other hand, the capacity of L. plantarum WCFS1 LAB found in a large number of niches including to grow and persist in a range of habitats in- dairy, meat, vegetable fermentations as well as cluding the mammalian and other GI tracts is in the gastro-intestinal (GI) tract (3). Both with reflected by the high number of putative extra- respect to its distribution and its applications, L. cellular proteins (231; 7.47 % of total proteins) plantarum is one of the predominant species of encoded by the genome. A large proportion of the Lactobacillus genus, comprising more than genes encoding enzymes for substrate utilization 80 different species (67). It has been shown that are clustered in a 600-kb region near the origin L. plantarum survives the conditions encountered of replication. This zone was later denominated in the human GI tract (74). In addition, health- the sugar island because most of these genes improving properties have been attributed to are predicted to be involved in sugar metabo- various L. plantarum strains and some are mar- lism and may represent a life-style adaptation is- keted as probiotic (4, 14, 37, 44). land (43). The latter observation was confirmed in a DNA microarray comparison between 20 L. plantarum strains where it was shown that The annotated genome of Lactobacillus the sugar island significantly varied within the plantarum WCFS1 strains suggesting that several genes had been From a single colony isolate of strain L. plan- acquired through interactions with varying envi- tarum NCIMB 8826, originally isolated from ronments, and supports its persistence and sur- human saliva, the complete genome has been vival in diverse ecological niches (43). Moreover, sequenced and annotated (36). The circular 3.3- the genome sequence of L. plantarum WCFS1 Mb chromosome was predicted to contain 3052 reveals the presence of a high number of genes protein-encoding genes of which 70% have a predicted to encode regulatory functions (261 putative function (Table 1). The annotated ge- proteins; 8.58% of total proteins) including three nome confirmed that L. plantarum WCFS1 has sigma-factor encoding genes (rpoD, rpoN, and the capacity for uptake and utilization of many sigH); over 200 transcription regulators and 21 sugars, uptake of peptides, and biosynthesis genes associated with two-component regula- of amino acids as well as DNA/RNA building tor systems. In total, 7.6% of the genome of L. blocks. The genome is predicted to encode all plantarum is assigned to regulatory functions. enzymes involved in glycolysis as well as en- For comparison in genomes of other LAB such

14 1 Introduction and Outline of this Thesis

as Streptococcus thermophilus and Lactobacillus they contribute to the conservation, flavor, and delbrueckii regulation processes are predicted texture of fermented foods (30). Growth perfor- to be encoded by 2.9 % and 4% of their 2-Mb mance and robustness of this bacterium are es- genome, respectively. The relative high propor- sential factors that determine the characteristics tion of regulatory genes present in L. plantarum of the final products. The use of this bacterium WCFS1 suggests that this bacterium has the ca- has developed contradictory views. For example pacity to adapt and survive to a variety of rapidly in meat products, L plantarum was shown to ei- changing environmental conditions. All together, ther suppress (62) or enhance spoilage (11) of the genome of L. plantarum WCFS1 reflects the the products. The same is the case in wine pro- known ability of this bacterium to grown under duction, where studies have shown that L. plan- rapidly changing environments and the flexibility tarum enhances spoilage (9) while it can also to adapt to different niches. contribute to the quality of the end product due its malolactic properties (56). Consequently, understanding the different adaptation mecha- Lactobacillus plantarum industrial nisms present in this organism is the only way to applications secure the proper use of the bacterium in indus- L. plantarum is widely used in industrial and ar- try. Furthermore, the fundamental knowledge in tisanal production of fermented plant, food, and stress response mechanisms can also be used feed products such as sauerkraut, cucumber, to regulate the quality of products in industrial cabbage, olives, and silage (75). In addition, L. fermentations. plantarum is employed as a starter culture since

TABLE 1. Distribution of predicted open reading frames (ORFs) over functional classes (numbers based on the L. plantarum WCFS1 genome).

Main Cathegory Number of ORFs % Amino acid biosynthesis 82 2.69 Biosynthesis of cofactors, 67 2.20 prosthetic groups, and carriers Cell envelope 231 7.57 Cellular processes 115 3.77 Central intermediary metabolism 58 1.90 DNA metabolism 82 2.69 Energy metabolism 239 7.83 Fatty acid and phospholipid 61 2.00 metabolism Hypothetical proteins 932 30.54 Other cathegories 203 6.65 Protein fate 56 1.83 Protein synthesis 136 4.46 Purines, pyrimidines, nucleosides 89 2.92 and nucleotides Regulatory functions 262 8.58 Transcription 28 0.92 Transport and binding proteins 411 13.47 Total 3052 100 15 The Glutaredoxin and Thioredoxin sys- FIGURE 1. The redox cycles of the Thio- tems redoxin and Glutaredoxin systems. The ability to respond to reactive oxygen spe- Panel A depicts the redox cycle of the thiore- cies (ROS), i.e. adaptation to an aerobic envi- doxin system. Thioredoxin (TRX) is reduced by ronment, requires mechanisms to minimize the thioredoxin reductase (TR) in a NADPH depen- occurrence of thiol oxidation and to mitigate its dent reaction. Panel B depicts the redox cycle consequences. Under an oxygen regime, ROS of the glutaredoxin system. Glutathione (GSH) are formed as by-products of different oxygen is reduced by glutathione reductase (GR) in a consuming reactions catalyzed by for instance NADPH dependent reaction. The active forms of NADH oxidase, peroxidases, and pyruvate oxi- either TRX or GSH can then reduced proteins in dase. The reaction leading to the formation of the cytoplasm. Proteins are either represented at

ROS is known as the Fenton’s reaction and the their reduced SH2 or oxidized S2 state products are hydroxyl (OH.) hydrogen peroxide - A (H2O2), and superoxide ions ( O2 ) (42). The A B + metal cofactors like Fe2 react with H2O2 in a Fenton reaction to form HO. radicals that rap- NADPH + H+ NADP+ NADPH + H+ NADP+ idly oxidize an amino acid residue at or near the cation binding site of the protein (42, 49). ROS molecules can be neutralized through conver- TR GR sion into water by antioxidants like glutathione, ascorbate, pyruvate, flavonoids, carotenoids, TRX- S2 TRX- (SH)2 GSH- S2 GSH- (SH)2 and the action of peroxidases (12, 48). Thiols can serve as sensors to oxidative stress as well as PROTEIN-(SH) PROTEIN- S PROTEIN-(SH) PROTEIN- S antioxidants (15). Two of these ubiquitous thiols 2 2 2 2 are glutathione (GSH) and thioredoxin (TRX) (28, 66).A There is a significant overlap between the B functions of the glutathione and thioredoxin systems (Fig. 1)NADPH even though+ H+ theyNADP do not+ exchange NADPH + H+ NADP+ reductive equivalents (29, 32). The major function of both systems is to maintain a cellular reducing environment TR and regulate activity of GR enzymes (80).

TRX- S2 TRX- (SH)2 GSH- S2 GSH- (SH)2

PROTEIN-(SH)2 PROTEIN- S2 PROTEIN-(SH)2 PROTEIN- S2

16 1 Introduction and Outline of this Thesis

GSH is a tripeptide (L-g-glutamyl-L-cysteine-gly- molecular chaperone in protein folding and cine). Concentration of GSH is synthesized in two protein denaturation (34). This role was further sequential reactions catalyzed by g-glutamyl-L- studied in E. coli, where a proteome study asso- cysteine synthetase (gshA) and GSH synthetase ciated TRX with more than 80 proteins involved (gshB). GSH is found in the millimolar range in in 26 cellular processes (38). eukaryotes. The role of GSH in protecting cells against oxygen toxicity has been widely studied (15). In Gram-positive bacteria only some of The thioredoxin system in L. plantarum them can synthesize GSH or consume it from the WCFS1 growth medium (61). The latter is the case for The activated form of TRX is obtained by re- Lactococcus lactis and L. plantarum. Recently, duction of TRX by TR in a NADPH dependent a novel multidomain fusion protein has been reaction (Fig. 1). The annotated genome of L. found in Listeria monocytogenes , lmo2770, de- plantarum predicts that six genes are involved in nominated glutathione synthase fusion protein the thioredoxin system. Four genes predicted to (GSHF) (21). GSHF directs formation of gluta- encode TRX: trxA1 (lp_0236), trxA2 (lp_2270), thione in vitro in Li. monocytogenes. GSH is the trxA3 (lp_3637) and trxH (lp_2633) and two most important redox buffer inside the cell and genes predicted to encode TR: trxB1 (lp_0761) its system encompasses GSH in the oxidized and trxB2 (lp_2585). These genes are dispersed and reduced state and glutathione reductase throughout the chromosome (Fig. 2). From the (GR) (Fig.1). In a nutshell, GSH is involved in genes predicted to encode TRX, at the amino the reduction of disulfides and transduction of acid level TRXA2 is the closest homolog to the intracellular regulatory signals as well as in the well-documented TRX from E. coli (58) and Ba- redox regulation of transcriptional factors (61). cillus subtilis (59). Also, we found that between For example, in Escherichia coli, GSH is involved the two encoded TR proteins, TRXB1 is the closest in reduction of OxyR, a global regulator (81). homolog to TR from Lc. lactis (76). In addition, the amino acid sequence of TRXB1 and TRXB2 indicates that only TRXB1 has the catalytic site, Thioredoxins are small, heat stable, ubiqui- characteristic of the TRX proteins -CxxC-. Inac- tous proteins containing a -CxxC- active site. tivation of trxA2 in B. subtilis (59) and trxB in They were first isolated in-vitro as a hydrogen Staphylococcus aureus is lethal (71) for the cells. donor for ribonucleotide reductase (28). The This is not the case for Lc. lactis; disruption of thioredoxin system consists of TRX itself in the trxB1 is not lethal in this organism (76). Further- oxidized and reduced state and thioredoxin re- more, in Lc. lactis it was found that disruption of ductase (TR). TRX acts as an electron donor for trxB1 induced genes and proteins related to oxi- ribonucleotide reductase, 3-phoshoadenylsul- dative stress response, and in carbon and lipid fate reductase (PAPS), and methionine sulfoxide metabolism. Hence, these results suggests a role reductase (20) (6). In addition, TRX is involved of TR in the process of adaptation of Lc. lactis to in heat (31, 59) and oxidative stress (57, 80). Recent studies have suggested that TRX acts like 17 FIGURE 2. Genome Atlas of Lactobacillus plantarum WCFS1 and the Thioredoxin system (a color representation of this figure can be found in page 151). The atlas represents a circular view of the complete genome sequence of L. plantarum WCFS1. Seven circles were created using Microbial Genome Viewer (77). Circle 1, Innermost, GC% content. Circle 2, thioredoxin reductase (TR) encoding genes trxB1 red trxB2 green. Circle 3, thioredoxin (TRX) encoding genes trxA1 blue trxA2 orange trxA3 purple black. Circle 4, COG classification in antisense orientation. Circle 5, COG classification in sense orientation. Circle 6, anti-sense open reading frames (ORFs). Circle 7, Outermost, ORFs in sense orientation.

Lactobacillus plantarum WCFS1 Genome Atlas

General stress response in Gram-posi- death and cell lysis under limitation of nutrients tive bacteria (C, N, P, metal ions) and energy source (18, 24). Bacteria are constantly exposed to fluctuat- In an industrial fermentation process, the bacte- ing environmental conditions which affect their rium should be able to successfully counteract physiological state and consequently their various adverse conditions during processing growth rate. Periods of slow-growth, and non- such as acid, osmotic, temperature, and oxida- growing states are often followed by short peri- tive stress (72). In order to survive the changing ods of rapid growth. Most of the time, bacterial set of growth-restricting stimuli, bacteria have cells are in a non-growing state to escape cell evolved a very complex adaptation network. The

18 1 Introduction and Outline of this Thesis

stress response in bacteria is regulated at vari- towards varied stresses or starvation environ- ous levels including gene (transcription), protein ments is called a general stress response. In the (translation), and metabolite level (42). model organism B. subtilis the master switch in the general stress response has been identified Transcriptional regulation in Gram- as sigma B (sB). This sigma factor was detected positive bacteria by Haldenwang and coworkers (23) as a an al- In bacteria transcription starts when a complex ternative sigma factor of bacteria. The network of RNA polymerase (RNAP) and sigma factor (s) of sB includes more than 150 genes and con- binds to specific sites in the DNA denominated tains genes predicted to be involved in different promoters. Bacterial transcriptional regulation stress related conditions: heat, oxidative stress, can be mediated via two mechanisms: 1) via al- osmotic stress as well as in energy starvation, tering the binding capacity of RNA polymerase redox imbalance as well as in the regulation of (RNAP) to promoters and 2) via modulating the sB and cell envelope (25). In the annotated ge- activity of the RNAP. Sigma factors direct the nome of L. plantarum WCFS1 (36) no homolog binding of the RNAP core-enzyme to DNA and to the sB in B. subtilis has been found. In addition each one has its specific promoter sequence. to the general stress response mechanisms there The second mechanism of transcription regu- are specific regulons that are induced under a lation, modulating the RNAP, takes place via specific stress condition. These specific stress transcriptional factors (TFs). TFs interact with the regulons contain a set of specific-stress pro- different domains of the RNAP-complex or can teins that provide a protective function against a bind to specific sites in the proximity of the pro- single stimulus. The protective function might al- moter influencing transcription of downstream low the cell to neutralize the stress factor, adapt genes. Both sigma factor as well as TFs binding to its presence or repair damage caused by it. sites are highly conserved throughout different For example, under a heat stress stimulus three species and can be predicted using software regulons have been identified inB. subtilis: HrcA packages MEME (7) and MAST (8). controls class I genes; sB which controls class II genes; and CtsR controls class III (26).

Regulatory Networks The set of genes or regulon that is influenced Transcriptional regulation of oxidative by a s factor or by one of the TFs is called a stress in Gram-positive bacteria network. Adaptation networks exist as a single Since the time that oxygen began to accumu- regulon or they are organized in groups that late in the atmosphere (3.5 billion years ago), consist of more than one regulon. Moreover, a micro-organisms have evolved mechanisms single regulon can overlap with others, forming allowing them to maintain a reduced cytosolic complex modules and gene-expression networks environment (17). Bacteria have a limited toler- (26). A non-specific but essential stress response ance to oxygen. The balance between ROS pro- mechanism that provides a protective function duction and the antioxidant defences within the

19 cell determines the degree of oxidative stress. or as global stress response regulators under ROS molecules cause cellular damage at both oxidative stress. The PerR regulon includes perR, molecular and metabolic levels (42). Therefore, a ferric uptake regulator (fur) (45), a catalase the ability to respond appropriately to ROS is es- (katA), a DNA-binding protein (mrgA), alkyl hy- sential for survival and optimal performance of droperoxide reductase (ahpF), TRX peroxidase microbial cells. (tpx) (45) (27), and heme genes; the OhrR regu- Research on B. subtilis revealed that oxidative lon clusters among others glutathione reductase stress is mainly regulated by three transcription (gor), TRX (trxC), and sB is a global general factors: PerR, OhrR, and sB (27) (Fig. 3). These stress response gene (see section above). The factors are induced as a result of either specific OhrR in B. subtilis is homolog to well-character- stress environments: hydrogen peroxide (45), ized regulator OxyR from E. coli (81). organic acids, and heat shock (27) respectively

FIGURE 3. Induced transcriptional regulation in Bacillus subtilis under oxidative stress. Adapted from known studies and reviews on B. subtilis (26, 27, 39, 45, 59).

Ethanol Osmotic Cell envelope Oxidative Stress

Hydrogen peroxide Heat superoxide Disulfide

ıB

PerR Sulfur limitation OhrB CstR HrcA

SOS Spx furR

mrsA tpx recA lexA uvrA dinB metE cysK trxB mrgAOhrA perR fur zosA ahpC trxC katG gor katXkatB trxA clpP clpEclpC dnaK groEL OpuD OpuE gtaB ggaA uvrB uvrC

20 1 Introduction and Outline of this Thesis

Transcriptional regulation of oxidative lack of homology and the recovery of unknown stress in LAB proteins suggest that the response towards oxi- The response towards oxidative stress in Lc. lactis dative stress also includes strain-specific mecha- is less understood but it is suggested to include nisms. Our analysis suggests that the L. planta- sigma factors (rpoD, comX and s-x), metabolite rum WCFS1 genome encodes genes known to flux sensors, and two-component regulatory sys- be involved in general stress response systems tems (42). As for L. plantarum, only a few stud- (PerR) lp_3247; Spx lp_0836; Ohr lp_0889, ies have been done on regulatory mechanisms lp_1360) as well as stress-specific genes (trxA controlling the oxidative stress response in this lp_2770; grshR lp_1253 ; hrcA lp_2029). This LAB. It is known that the metabolism of L. plan- analysis is the first attempt into understanding tarum is influenced in the presence of oxygen oxidative stress response at the transcriptional (70) and that this bacterium can chemically de- level in L. plantarum. grade oxygen via catalase, NADH peroxidase, and pyruvate oxidase (46). On the other hand, Global transcriptional- analysis L. plantarum does not have a superoxide dis- The first published LAB genome was that of Lc. mutase (SOD) protein (22) and accumulates lactis IL1403 (10). In 2003, L. plantarum WCFS1

H2O2 when grown under oxygen (46). In addi- was the first fully sequenced and annotated lac- tion, it is also known that L. plantarum can ac- tobacillus species (36). Nowadays the genome cumulate high concentrations of manganese. In sequence database of LAB exceeds the 20 ge- the absence of chelators, dialyzed manganese nomes (53). Comparative genomics on the an- - acts as a stoichiometric scavenger of O2 pro- notated sequences show the capacity of LAB to + ducing MnO2 (5). Furthermore, recently efforts adapt to a variety of environmental niches (35) have been made to elucidate the regulatory net- through processes like horizontal gene transfer work in L. plantarum WCFS1 (79) using bioin- (43), sequence conservation, and metabolic sim- formatic tools. This in-silico analysis resulted in plification (40). The annotated genomes tell us a network containing eight clusters, where the what genes the bacterium possesses and what most prominent are carbon and nitrogen me- functionality they have. Nevertheless, it does not tabolism, the SOS regulon, and CstR. Using necessarily tell us which groups of genes are the information available for the model organ- actually expressed in specific environments. The ism, B. subtilis, and the annotated genome of total set of messenger RNA’s (mRNA) under a L. plantarum WCFS1 a first model of the oxida- specific condition -the transcriptome- serves as tive stress response mechanisms in L. plantarum a mirror of the biological state of the cell and WCFS1 can be defined on basis of homology the genes that play a role under that specific (Table 2). condition. Many of the transcriptional factors It should be noted that many genes characterized and two-component regulatory genes are signal in B. subtilis as being involved in oxidative stress transduction systems to environmental changes did not have a homolog in L. plantarum WCFS1 in nutrient, physico-chemical disturbances (65). and therefore were not included in here. The

21 -114 -118 -25 -41 -56 -274 -34 -31 -26 -24 -41 -114 -28 -46 -179 -26 -78 -36 -90 -37 score p 4.37 e 1.96 e 2.87 e 1.20 e 2.25 e 2.87 e 9.47 e 7.00 e 2.34 e 2.34 e 1.61 e 6.43 e 5.33 e 1.08 e 8.24 e 1.60 e 6.37 e 3.24 e 9.09 e 2.95 e using ERGO suite (47) WCFS1 main class Cellular processes Cellular processes Cellular processes Cellular processes Cellular processes Cellular processes Biosynthesis of cofactors, prosthetic groups, and carriers Cellular processes Biosynthesis of cofactors, prosthetic groups, and carriers Hypothetical proteins Cellular processes Cellular processes Biosynthesis of cofactors, prosthetic groups, and carriers Cellular processes Cellular processes Regulatory functions Regulatory functions Regulatory functions Regulatory functions Cellular processes Cellular processes Transport and binding proteins L. plantarum Lactobacillus plantarum and - Product chaperone protein DnaJ thioredoxin heat shock protein DnaK thioredoxin ATP-dependent Clp protease, ATP-binding subunit ClpC GroEL chaperonin GroES co-chaperonin heat shock protein GrpE glutathione reductase catalase heat-inducible transcription repressor HrcA stress induced DNA binding protein catalase catalase catalase catalase stress induced DNA binding protein transcription regulator transcription regulator transcription regulator glycine betaine/carnitine/choline transport protein ATPase of the PilT family ferric uptake regulator fur kat kat kat kat kat clpC hrcA grpE dnaJ Gene trxA2 dnaK trxA2 groEL groES gshR2 locus lp_2026 lp_0727 lp_2270 lp_3128 lp_2270 lp_1019 lp_1360 lp_1821 lp_0889 lp_3128 lp_2029 lp_2027 lp_3578 lp_3324 lp_0728 lp_1150 lp_2028 lp_3578 lp_3578 lp_1477 lp_3247 lp_3578 lp_1253 gor trxA Dps trxA heat katE katX katA katA hrcA grpE perR dnaJ ClpC ykuN dnaK OhrR OhrR OhrR mrgA BmrU OpuD groEL groES Osmotic oxidative Antibiotic 3 1 1,2 1 Bacillus subtilis B ı OhrR Fur Regulon PerR HrcA TABLE 2. Correlation between genomes of TABLE mechanisms. response stress on 31.08.2007 regarding

22 1 Introduction and Outline of this Thesis -228 -274 -107 -44 -62 -34 -79 -41 -10 -19 -25 -169 -118 -30 -162 1.11 e 2.25 e 1.17 e 2.18 e 8.09 e 2.34 e 7.00 e 6.41 e 1.83 e 1.97 e 1.43 e 1.24 e 3.04 e 3.24 e 9.09 e Cellular processes Hypothetical proteins Hypothetical proteins Energy metabolism Biosynthesis of cofactors, prosthetic groups, and carriers Hypothetical proteins Regulatory functions Regulatory functions Regulatory functions DNA metabolism DNA metabolism DNA metabolism Hypothetical proteins Hypothetical proteins Hypothetical proteins Central intermediary metabolism Cell envelope Hypothetical proteins DNA metabolism Hypothetical proteins DNA metabolism Cellular processes Transport and binding proteins Energy metabolism Hypothetical proteins Cellular processes Hypothetical proteins Cellular processes Regulatory functions Cellular processes (25, 27) B. subtilis exopolyphosphatase-related protein (putative) DNA-damage-inducible protein P transcription repressor of the SOS regulon ATP-dependent Clp protease, ATP-binding subunit ClpE thioredoxin ATP-dependent Clp protease, ATP-binding subunit ClpC regulatory protein Spx regulatory protein Spx pyruvate oxidase UV-damage repair protein ATP-dependent nuclease, subunit A ATP-dependent nuclease, subunit B unknown 6-phospho-beta-glucosidase DNA-directed DNA polymerase III, alpha chain unknown alpha-glucosidase endopeptidase Clp, proteolytic subunit catalase 1 segregation helicase (putative) stress induced DNA binding protein glycosyltransferase amino acid transport protein unknown unknown unknown integral membrane protein transcription regulator of multidrug-efflux transporter ATPase of the PilT family unknown C E 3 kat agl6 clcP lexA clpE dinP rhe clpC rexA rexB spx1 spx3 pox4 trxA2 dna umu cshA2 WCFS1 as described by Wels et al (79) lp_2693 lp_2694 lp_2279 lp_2280 lp_2063 lp_2270 lp_3128 lp_3627 lp_1269 lp_0786 lp_2071 lp_3141 lp_3142 lp_1543 lp_1899 lp_2290 lp_3278 lp_3578 lp_1150 lp_2817 lp_1611 lp_2778 lp_0145 lp_1019 lp_0836 lp_2228 lp_3587 lp_0607 lp_3022 lp_3023 E ytqI dps Spx Spx trxA lexA clpE clpP yqjH clpC uvrX yfhO katG yacL L. plantarum csbB ydeA ydaP dna addA addB ycdG yhdG bmrU bmrR 6 5 4 Regulon containing class III heat shock genes (27) Based on study performed by Hecker et al under strict anaerobe conditions (25) One of the main global regulators under oxidative stress response in Based on review from Helmann et al (26) SOS regulon in

Heat Stress CtsR SOS regulon

1 2 3 4, 5 6

23 There are different techniques available to study microarray contains a genome fragment. Some mRNA: northern blots, quantitative PCR (q-PCR), drawbacks of the clone-base microarray are and microarrays. Microarrays were first described the lack of specificity and identity of the spots. in 1995 (60) and allow a simultaneous com- The transcriptome studies included in this thesis parison of the expression levels of large sets of were preformed under the Agilent Platform us- genes in response to the conditions of interest. In ing gene-based microarrays containing specific this technique, RNA is isolated from cells grown probes synthesized by PCR and spotted on glass or incubated under conditions of interest. In the slides. The design of the slides can be accessed next step, the RNA in the sample is converted at the gene expression omnibus database (GEO) into cDNA by reverse transcription (73) and la- with accession number GPL4318 (52). belled with a fluorescent dye. The most common dyes used are cyanide dyes (cy3 and cy5). The The microarray technique generates reproduc- labelled cDNA is hybridized to the microarray. ible and large amounts of data (60) and custom The microarray is a small glass slide containing designed microarrays is becoming cheaper ev- thousands of DNA spots. The labelled cDNA will ery day. Transcriptome analysis is most applied hybridize to the equivalent spot on the micro- as a data-driven approach delivering datasets array and the total amount of signal on each of differentially expressed genes. Nevertheless, spot is a measure of the level at which the corre- much attention should be paid to the experimen- sponding genes were expressed in either studied tal design in order to obtain a dataset of genes condition. For some approaches, including the that represents only the genes that play a role in Affymetrix platform (1), labelled cDNA from a the physiological adaptation of the studied con- single sample is hybridized onto the microarray, dition. An experimental design for a microarray whereas other approaches such as the Agilent experiment should minimize the secondary ef- platform (2) rely on the co-hybridization of two fects. The secondary effects are caused by varia- samples (labelled with two dyes) onto the same tions in experimental setup, technical variance, slide. The chips used in both platforms contain and sample handling. These effects will only add prior synthesized oligos. These oligos are spot- complexity in the interpretation of results as well ted onto the chips. as false positive differentially expressed genes (50). Thus, in an ideal transcriptome analysis There are two kinds of microarrays that are of- the only difference between the samples should ten used: clone- and gene- based microarrays. be the condition of interest while cell density, Gene-based microarrays are usually applied growth phase, and growth rate should be identi- when the genome of the organism of interest has cal between the samples (64). RNA has a short been completely annotated. Then, each spot of half life and transcriptome response towards the microarray represents DNA fragments of the environmental changes occurs rapidly (78). For organism of interest. If the genome of the studied that matter, when working with microarrays RNA organism has not been annotated, clone-based quenching of the samples is essential. Quench- arrays are an option. In this case a spot in the ing strategies of culture samples including

24 1 Introduction and Outline of this Thesis

liquid nitrogen (51) or methanol at -40°C (16, (33), MetaCyc (41) or the commercial packages 51) have shown to lead to an immediate cell ar- such as SimphenyTM (68, 69) and GeneData rest and to decreased errors in harvesting. (19). The analytical methods together with the biological knowledge of the researcher allow ef- fective integration and interpretation of the in- Microarray data handling and analysis formation. The fluorescent signal of each spot and- back ground on the microarray are quantified by laser Outline of the thesis scanning. The signal intensities are then filtered The research presented in this thesis was initi- for homogeneity and signal-to-background ra- ated to gain more understanding in the oxida- tio in each spot. The filtered set of signals is then tive stress response in the lactic acid bacterium L. normalized. Normalization is conducted to cor- plantarum. Considerable attention was focused rect for the deviation between the observed sig- on thioredoxin (TRX), a ubiquitous protein in- nals and the actual amount of hybridized cDNA. volved in oxidative stress tolerance and redox The most commonly used and accepted nor- homeostasis in bacteria. In this study, we con- malization method is the locally weighted least structed a library of strains with alterations in the squares regression (LOESS or LOWESS) (54). thioredoxin system in L. plantarum WCFS1 which After filtering and normalization the datasets are is predicted to involve six genes in this bacterium. ready for analysis. The two major types of data Throughout the study, transcriptome analysis analysis are: significance and multivariable data was performed on the constructed strains using analysis (50). For finding whether a gene is dif- custom made DNA-microarrays. This genome- ferentially expressed under a certain condition wide analysis was subsequently validated using in a set of microarrays the most used tool is the different techniques including comparative ge- analysis of variance (ANOVA) (13). Neverthe- nomics, enzyme assays, and q-PCR. The leads less, nowadays different statistical analyses have obtained in this work can be used for strain im- developed including linear models such as LIM- provement in industrial fermentations. MA (63) and the false discovery rate (FDR) (55). Chapter 2 describes in depth the composition of the thioredoxin system in L. plantarum WCFS1 Biological Interpretation and its role in the oxidative stress response in After data handling and analysis the microarray this bacterium. Overexpression of gene trxB1 re- data is ready for interpretation. There are several sulted in a strain with 3-fold overproduction of tools available to handle the large sets of micro- thioredoxin reductase (TR) activity and higher re- array data. One method is clustering: hierarchi- sistance toward diamide and hydrogen peroxide cal, K-means, self-organizing maps, and princi- compared to the wild-type. At the transcriptome pal component analysis. Another approach uses level, the effect of trxB1-overexpression and hy- the existing knowledge and visualization tools drogen peroxide stress on L. plantarum was as- such as the online gene encyclopedia (KEGG) sociated with the induction of a set of 16 genes

25 including gapB, npr2, groEL, and genes predict- tion, we identified oxidative stress response pu- ed to be involved in purine metabolism. These tative regulatory networks in which among oth- results show that overproduction of TR results in ers, the transcription factor OhrR is activated. a mock-oxidative stress inside the cells. This observation offers an explanation for the observed Chapter 5 addresses the other major redox- improved tolerance towards oxidative stress of homeostasis system in the cytoplasm: the glu- the trxB1-overexpressing strain. taredoxin system. The glutaredoxin system was characterized in L. plantarum. Glutathione Chapter 3 addresses the impact of inactivation (GSH) was found to provide protection against of trxB1 on the oxidative stress response of L. exposure to hydrogen peroxide stress and is plantarum. Disruption of trxB1 resulted in a L. most likely functional as a scavenger of reactive plantarum strain with decreased TR activity (2.5 oxygen species. The gene responsible for glu- fold) and in a 19% lower growth rate under tathione production in Li. monocytogenes gshF aerobic conditions compared to the wild-type. has a homolog in L. plantarum WCFS1 namely The global transcriptome response of a trxB1- lp_2336. Disruption of lp_2336 affected survival disruption strain towards hydrogen peroxide re- of L. plantarum WCFS1 under hydrogen perox- vealed that 93 genes normally found induced ide stress. The impact of inactivation of lp_2336 under oxidative stress were significantly down- and hydrogen peroxide stress was studied using regulated in the trxB1::cam L. plantarum strain. DNA-microarrays and suggested an overlap be- These observations support the hypothesis pre- tween the thioredoxin and glutaredoxin system sented in Chapter 2, namely that TR encoded in L. plantarum. by the gene trxB1 in L. plantarum WCFS1 plays a role in oxidative stress response. In addition, Chapter 6 summarizes the main results ob- the global transcriptome analysis of the trxB1- tained in this study and includes a hypothetical disruption strain suggests that TRX is involved a model of the transcriptional regulation present many cellular processes including purine and in L. plantarum WCFS1 as a result of oxidative sugar metabolism. stress. The core or the presented model was based on studies of B. subtilis, L. plantarum, and Chapter 4 describes the role of the thioredoxin the data generate in this study. Finally, in this system in aerobic, anaerobic, and respiratory chapter concluding remarks and future perspec- growth of L. plantarum WCFS1. Analysis of the tives of the research described in this thesis are global transcriptome profiling data supported presented. conclusions from Chapter 2 and Chapter 3 regarding the role of trxB1 in oxidative stress and allowed us to hypothesize a role of trxB2 in heat stress response in L. plantarum WCFS1. In addi-

26 1 Introduction and Outline of this Thesis

REFERENCES

1. Affymetrix http://www.affimetrix.com/, posting date. Affymetrix. [Online.] 2. Agilent http://www.agilent.com/, posting date. Agilent. [Online.] 3. Ahrne, S., S. Nobaek, B. Jeppsson, I. Adlerberth, A. E. Wold, and G. Molin. 1998. The normal Lactobacillus flora of healthy human rectal and oral mucosa. J Appl Microbiol85 :88-94. 4. Annuk, H., J. Shchepetova, T. Kullisaar, E. Songisepp, M. Zilmer, and M. Mikelsaar. 2003. Char acterization of intestinal lactobacilli as putative probiotic candidates. J Appl Microbiol 94:403-12. 5. Archibald, F. S., and I. Fridovich. 1981. Manganese and defenses against oxygen toxicity in Lactobacil lus plantarum. J Bacteriol 145:442-51. 6. Arner, E. S., and A. Holmgren. 2000. Physiological functions of thioredoxin and thioredoxin reductase. Eur J Biochem 267:6102-9. 7. Bailey, T. L., and C. Elkan. 1994. Fitting a mixture model by expectation maximization to discover motifs in biopolymers. Proc Int Conf Intell Syst Mol Biol 2:28-36. 8. Bailey, T. L., and M. Gribskov. 1998. Combining evidence using p-values: application to sequence homology searches. Bioinformatics 14:48-54. 9. Beneduce, L., G. Spano, A. Vernile, D. Tarantino, and S. Massa. 2004. Molecular characterization of lactic acid populations associated with wine spoilage. J Basic Microbiol 44:10-6. 10. Bolotin, A., P. Wincker, S. Mauger, O. Jaillon, K. Malarme, J. Weissenbach, S. D. Ehrlich, and A. Sorokin. 2001. The complete genome sequence of the lactic acid bacterium Lactococcus lactis ssp. lactis IL1403. Genome Res 11:731-53. 11. Borch, E., M. L. Kant-Muermans, and Y. Blixt. 1996. Bacterial spoilage of meat and cured meat prod ucts. Int J Food Microbiol 33:103-20. 12. Condon, S. 1987. Responses of lactic acid bacteria to oxygen. FEMS Microbiology Reviews 46:269-280. 13. Cui, X., and G. A. Churchill. 2003. Statistical tests for differential expression in cDNA microarray experiments. Genome Biol 4:210. 14. De Vries, M. C., E. E. Vaughan, M. Kleerebezem, and W. M. de Vos. 2006. Lactobacillus plantarum - Survival, functional and potential probiotic properties in the human intestinal tract. Int. Dairy J. in press. 15. Fahey, R. C. 2001. Novel thiols of prokaryotes. Annu Rev Microbiol 55:333-56. 16. Faijes, M., A. E. Mars, and E. J. Smid. 2007. Comparison of quenching and extraction methodologies for metabolome analysis of Lactobacillus plantarum. Microb Cell Fact 6:27. 17. Fedoroff, N. 2006. Redox regulatory mechanisms in cellular stress responses. Ann Bot (Lond) 98:289-300. 18. Gallegos, M. T., R. Schleif, A. Bairoch, K. Hofmann, and J. L. Ramos. 1997. Arac/XylS family of transcriptional regulators. Microbiol Mol Biol Rev 61:393-410. 19. GeneData http://www.genedata.com/products/expressionist/, posting date. GeneData. [Online.] 20. Gonzalez Porque, P., A. Baldesten, and P. Reichard. 1970. The involvement of the thioredoxin system in the reduction of methionine sulfoxide and sulfate. J Biol Chem 245:2371-4. 21. Gopal, S., I. Borovok, A. Ofer, M. Yanku, G. Cohen, W. Goebel, J. Kreft, and Y. Aharonowitz. 2005. A multidomain fusion protein in Listeria monocytogenes catalyzes the two primary activities for glutathione biosynthesis. J Bacteriol 187:3839-47. 22. Gregory, E. M., and I. Fridovich. 1974. Oxygen metabolism in Lactobacillus plantarum. J Bacteriol 117:166-9. 23. Haldenwang, W. G., and R. Losick. 1980. Novel RNA polymerase sigma factor from Bacillus subtilis. Proc Natl Acad Sci U S A 77:7000-4. 24. Hantke, K. 2001. Iron and metal regulation in bacteria. Curr Opin Microbiol 4:172-7. 25. Hecker, M., J. Pane-Farre, and U. Volker. 2007. SigB-dependent general stress response in Bacillus subtilis and related gram-positive bacteria. Annu Rev Microbiol 61:215-36. 26. Hecker, M., and U. Volker. 2001. General stress response of Bacillus subtilis and other bacteria. Adv Microb Physiol 44:35-91. 27. Helmann, J. D., M. F. Wu, A. Gaballa, P. A. Kobel, M. M. Morshedi, P. Fawcett, and C. Paddon. 2003. The global transcriptional response of Bacillus subtilis to peroxide stress is coordinated by three transcription factors. J Bacteriol 185:243-53.

27 28. Holmgren, A. 1985. Thioredoxin. Annu Rev Biochem 54:237-71. 29. Holmgren, A. 1989. Thioredoxin and glutaredoxin systems. J Biol Chem 264:13963-6. 30. Hugenholtz, J., and M. Kleerebezem. 1999. Metabolic engineering of lactic acid bacteria: overview of the approaches and results of pathway rerouting involved in food fermentations. Curr Opin Biotechnol 10:492-7. 31. Jobin, M. P., D. Garmyn, C. Divies, and J. Guzzo. 1999. Expression of the Oenococcus oeni trxA gene is induced by hydrogen peroxide and heat shock. Microbiology 145 ( Pt 5):1245-51. 32. Kanzok, S. M., A. Fechner, H. Bauer, J. K. Ulschmid, H. M. Muller, J. Botella-Munoz, S. Schneuwly, R. Schirmer, and K. Becker. 2001. Substitution of the thioredoxin system for glutathione reductase in Drosophila melanogaster. Science 291:643-6. 33. KEGG http://www.genome.jp/kegg/, posting date. KEGG. [Online.] 34. Kern, R., A. Malki, A. Holmgren, and G. Richarme. 2003. Chaperone properties of Escherichia coli thioredoxin and thioredoxin reductase. Biochem J 371:965-72. 35. Klaenhammer, T., E. Altermann, F. Arigoni, A. Bolotin, F. Breidt, J. Broadbent, R. Cano, S. Chaillou, J. Deutscher, M. Gasson, M. van de Guchte, J. Guzzo, A. Hartke, T. Hawkins, P.Hols, R. Hutkins, M. Kleerebezem, J. Kok, O. Kuipers, M. Lubbers, E. Maguin, L. McK ay, D. Mills, A. Nauta, R. Overbeek, H. Pel, D. Pridmore, M. Saier, D. van Sinderen, A. -Sorokin, J. Steele, D. O’Sullivan, W. de Vos, B. Weimer, M. Zagorec, and R. Siezen. 2002. Discovering lactic acid bacteria by genomics. Antonie Van Leeuwenhoek 82:29-58. 36. Kleerebezem, M., J. Boekhorst, R. van Kranenburg, D. Molenaar, O. P. Kuipers, R. Leer, R. Tarchini, S. A. Peters, H. M. Sandbrink, M. W. Fiers, W. Stiekema, R. M. Lankhorst, P. A. Bron, S. M. Hoffer, M. N. Groot, R. Kerkhoven, M. de Vries, B. Ursing, W. M. de Vos, and R. J. Siezen. 2003. Complete genome sequence of Lactobacillus plantarum WCFS1. Proc Natl Acad Sci U S A 100:1990-5. 37. Kullisaar, T., M. Zilmer, M. Mikelsaar, T. Vihalemm, H. Annuk, C. Kairane, and A. Kilk. 2002. Two antioxidative lactobacilli strains as promising probiotics. Int J Food Microbiol 72:215-24. 38. Kumar, J. K., S. Tabor, and C. C. Richardson. 2004. Proteomic analysis of thioredoxin-targeted proteins in Escherichia coli. Proc Natl Acad Sci U S A 101:3759-64. 39. Leichert, L. I., C. Scharf, and M. Hecker. 2003. Global characterization of disulfide stress in Bacillus subtilis. J Bacteriol 185:1967-75. 40. Makarova, K. S., and E. V. Koonin. 2007. Evolutionary genomics of lactic acid bacteria. J Bacteriol 189:1199-208. 41. MetaCyc http://metacyc.org/, posting date. MetaCyc. [Online.] 42. Miyoshi, A., T. Rochat, J. J. Gratadoux, Y. Le Loir, S. C. Oliveira, P. Langella, and V. Azevedo. 2003. Oxidative stress in Lactococcus lactis. Genet Mol Res 2:348-59. 43. Molenaar, D., F. Bringel, F. H. Schuren, W. M. de Vos, R. J. Siezen, and M. Kleerebezem. 2005. Exploring Lactobacillus plantarum genome diversity by using microarrays. J Bacteriol 187:6119-27. 44. Molin, G. 2001. Probiotics in foods not containing milk or milk constituents, with special reference to Lacto bacillus plantarum 299v. Am J Clin Nutr 73:380S-385S. 45. Mostertz, J., C. Scharf, M. Hecker, and G. Homuth. 2004. Transcriptome and proteome analysis of Bacillus subtilis gene expression in response to superoxide and peroxide stress. Microbiology 150:497- 512. 46. Murphy, M. G., and S. Condon. 1984. Correlation of oxygen utilization and hydrogen peroxide accumula tion with oxygen induced enzymes in Lactobacillus plantarum cultures. Arch Microbiol 138:44-8. 47. Overbeek, R., N. Larsen, T. Walunas, M. D’Souza, G. Pusch, E. Selkov, Jr., K. Liolios, V. Joukov, D. Kaznadzey, I. Anderson, A. Bhattacharyya, H. Burd, W. Gardner, P. Hanke, V. Kapatral, N. Mikhailova, O. Vasieva, A. Osterman, V. Vonstein, M. Fonstein, N. Ivanova, and N. Kyrpides. 2003. The ERGO genome analysis and discovery system. Nucleic Acids Res 31:164-71. 48. Penninckx, M. 2000. A short review on the role of glutathione in the response of yeasts to nutritional, envi ron mental, and oxidative stresses. Enzyme Microb Technol 26:737-742.

28 1 Introduction and Outline of this Thesis

49. Pericone, C. D., S. Park, J. A. Imlay, and J. N. Weiser. 2003. Factors contributing to hydrogen perox ide resistance in Streptococcus pneumoniae include pyruvate oxidase (SpxB) and avoidance of the toxic effects of the fenton reaction. J Bacteriol 185:6815-25. 50. Pieterse, B. 2006. Transcriptome analysis of the lactic acid and NaCl-stress response in Lactobacillus plantarum. PhD thesis. Wageningen University. 51. Pieterse, B., R. H. Jellema, and M. J. van der Werf. 2006. Quenching of microbial samples for in creased reliability of microarray data. J Microbiol Methods 64:207-16. 52. PubMed http://www.ncbi.nlm.nih.gov/geo/, posting date. Gene Expression Omnibus (GEO). [Online.] 53. PUBMed http://www.ncbi.nlm.nih.gov/genomes/lproks.cgi, posting date. PUBMed. [Online.] 54. Quackenbush, J. 2002. Microarray data normalization and transformation. Nat Genet 32 Suppl:496-501. 55. Reiner, A., D. Yekutieli, and Y. Benjamini. 2003. Identifying differentially expressed genes using false discovery rate controlling procedures. Bioinformatics 19:368-75. 56. Renault, P., C. Gaillardin, and H. Heslot. 1988. Role of malolactic fermentation in lactic acid bacteria. Biochimie 70:375-9. 57. Ritz, D., H. Patel, B. Doan, M. Zheng, F. Aslund, G. Storz, and J. Beckwith. 2000. Thioredoxin 2 is involved in the oxidative stress response in Escherichia coli. J Biol Chem 275:2505-12. 58. Russel, M., P. Model, and A. Holmgren. 1990. Thioredoxin or glutaredoxin in Escherichia coli is essen tial for sulfate reduction but not for deoxyribonucleotide synthesis. J Bacteriol 172:1923-9. 59. Scharf, C., S. Riethdorf, H. Ernst, S. Engelmann, U. Volker, and M. Hecker. 1998. Thioredoxin is an essential protein induced by multiple stresses in Bacillus subtilis. J Bacteriol 180:1869-77. 60. Schena, M., D. Shalon, R. W. Davis, and P. O. Brown. 1995. Quantitative monitoring of gene expres sion patterns with a complementary DNA microarray. Science 270:467-70. 61. Smirnova, G. V., and O. N. Oktyabrsky. 2005. Glutathione in bacteria. Biochemistry (Mosc) 70:1199-211. 62. Smith, J. L., and S. A. Palumbo. 1983. Use of starter cultures in meats. J. Food Protection. 46:997-1006. 63. Smyth, G. K. 2005. Limma: linear models for microarray data. Springer,, New York. 64. Stevens, M. J. A. 2008. Transcriptome Response of Lactobacillus plantarum to Global Regulator Deficiency, Stress and other Environmental Conditions. PhD thesis. Wageningen University. 65. Sturme, M. 2005. Analysis of Quorum Sensing Regulatory Systems in the Human Isolate Lactobacillus plantarum. PhD thesis. Wageningen University. 66. Sundquist, A. R., and R. C. Fahey. 1989. Evolution of antioxidant mechanisms: thiol-dependent peroxi dases and thioltransferase among procaryotes. J Mol Evol 29:429-35. 67. Tannock, G. W. 2004. A special fondness for lactobacilli. Appl Environ Microbiol 70:3189-94. 68. Teusink, B., and E. J. Smid. 2006. Modelling strategies for the industrial exploitation of lactic acid bacte ria. Nat Rev Microbiol 4:46-56. 69. Teusink, B., A. Wiersma, D. Molenaar, C. Francke, W. M. de Vos, R. J. Siezen, and E. J. Smid. 2006. Analysis of growth of Lactobacillus plantarum WCFS1 on a complex medium using a genome-scale metabolic model. J Biol Chem 281:40041-8. 70. Tseng, C.-P., Thomas, J. Montville. 1992. Enzymatic Regulation of Glucose Catabolism by Lactobacillus plantarum in an Aerobic Chemostat. Biotechnol. Prog 8:125-131. 71. Uziel, O., I. Borovok, R. Schreiber, G. Cohen, and Y. Aharonowitz. 2004. Transcriptional regula tion of the Staphylococcus aureus thioredoxin and thioredoxin reductase genes in response to oxygen and disulfide stress. J Bacteriol186 :326-34. 72. van de Guchte, M., P. Serror, C. Chervaux, T. Smokvina, S. D. Ehrlich, and E. Maguin. 2002. Stress responses in lactic acid bacteria. Antonie Van Leeuwenhoek 82:187-216. 73. Van Gelder, R. N., M. E. von Zastrow, A. Yool, W. C. Dement, J. D. Barchas, and J. H. Eberwine. 1990. Amplified RNA synthesized from limited quantities of heterogeneous cDNA. Proc Natl Acad Sci U S A 87:1663-7. 74. Vesa, T., P. Pochart, and P. Marteau. 2000. Pharmacokinetics of Lactobacillus plantarum NCIMB 8826, Lactobacillus fermentum KLD, and Lactococcus lactis MG 1363 in the human gastrointestinal tract. Aliment Pharmacol Ther 14:823-8.

29 75. Vescovo, M., S. Torriani, F. Dellaglio, and V. Bottazzi. . 1993. Basis characteristics, ecology and ap plication of Lactobacillus plantarum: a review. Ann.Microbiol.Enzimol. 43:261-284. 76. Vido, K., H. Diemer, A. Van Dorsselaer, E. Leize, V. Juillard, A. Gruss, and P. Gaudu. 2005. Roles of thioredoxin reductase during the aerobic life of Lactococcus lactis. J Bacteriol 187:601-10. 77. Viewer, M. G. http://www.cmbi.ru.nl/genome/, posting date. Microbial Genome Viewer. [Online.] 78. Wang, Y., C. L. Liu, J. D. Storey, R. J. Tibshirani, D. Herschlag, and P. O. Brown. 2002. Precision and functional specificity in mRNA decay. Proc Natl Acad Sci U S A 99:5860-5. 79. Wels, M. 2008. Unraveling the regulatory network of Lactobacillus plantarum WCFS1. PhD thesis. Wagenin gen University. 80. Zeller, T., and G. Klug. 2006. Thioredoxins in bacteria: functions in oxidative stress response and regula tion of thioredoxin genes. Naturwissenschaften 93:259-66. 81. Zheng, M., and G. Storz. 2000. Redox sensing by prokaryotic transcription factors. Biochem Pharmacol 59:1-6.

30 2 Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1

L. Mariela Serrano, Douwe Molenaar, Michiel Wels, Bas Teusink, Peter A. Bron, Eddy J. Smid

A modified version was published in Microbial Cell Factories, Aug 2007(28) 31

31 32 2 Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1

ABSTRACT

Thioredoxin is a powerful electron donor that catalyzes a wide spectrum of redox reactions in the cell. The aim of this study is to elucidate the role of thioredoxin system in the oxidative stress response in Lactobacillus plantarum WCFS1. We have identified the trxB1-encoded thioredoxin reductase (TR) as a key enzyme in the oxidative stress response of Lactobacillus plantarum WCFS1. Overexpression of the trxB1 gene resulted in a 3-fold higher TR activity in comparison to the wild-type strain. Subsequently, higher TR activity was associated with an increased resistance towards oxidative stress. We further determined the global transcriptional response to hydrogen peroxide stress in the trxB1- overexpression and wild-type strains grown in continuous cultures. Hydrogen peroxide stress and overproduction of TR collectively resulted in the up-regulation of 267 genes. Additionally, gene expression profiling showed significant differential expression of27 genes in the trxB1-overexpression strain. Over expression of trxB1 was found to activate genes associated with DNA repair and stress mechanisms as well as genes associated with the activity of biosynthetic pathways for purine and sulfur-containing amino acids. A total of 16 genes showed a response to both TR overproduction and hydrogen peroxide stress. These genes are predicted to be involved in purine metabolism, energy metabolism (gapB) as well as in stress-response (groEL, npr2), and manganese transport (mntH2). The crossover between datasets may explain the phenotype of the trxB1-overexpression strain, which appears to be prepared for encountering oxidative stress. This latter property can be used for engineering robustness towards oxidative stress in industrial strains of L. plantarum.

3333 INTRODUCTION Thioredoxin (TRX) was first characterized as a The reducing power of the thioredoxin and the sole electron donor for ribonucleotide reduc- glutaredoxin system is essential for all organ- tase in Escherichia coli (17). The catalytic activity isms. Furthermore, these two systems are sug- of these oxido-reductases can be attributed to gested to be the only systems that maintain the the -CxxC- motif found in these proteins. At this cytoplasm of the cell in the reduced state (11). cysteine-rich site electrons are transferred from While glutathione is found in eukaryotes and the reduced TRX towards the substrate (proteins, Gram-negative bacteria, it has been reported disulfides, etc). The resulting oxidized TRX is re- that most Gram-positive bacteria lack the abil- generated via thioredoxin reductase (TR) using ity to synthesize glutathione but rather import it NADPH as a cofactor. Throughout the years, from the environment. This is the case for B. sub- studies on the effect of the ubiquitous and con- tilis, Listeria monocytogenes, and Lactobacillus served TRX in cellular metabolism have revealed plantarum (19). In the latter organism the gene that it plays a significant role in a variety of pro- gshB which is involved in the second step for the cesses, including oxidative stress, protein repair, synthesis of glutathione (glutathione synthase), and RNA biosynthesis (2, 23, 29). has not been identified in the genome sequence (16). Hence, it is believed that in L. plantarum Intensive research on the role of TRX and TR in- the thioredoxin system is the only active thiol- clude the use of transcriptomics and proteomics reducing system. approaches. Studies with Bacillus subtilis and Oenococcus oeni showed that gene trxA was Little is known about the thioredoxin system and induced under stress conditions like heat and its function in L. plantarum. This flexible and ver- hydrogen peroxide stress (13, 26). In addition, satile bacterium is a member of the human gut gene expression studies in E. coli elucidated that microbiota, and is commonly used in fermented under hydrogen peroxide stress, OxyR (transcrip- foods. The purpose of this study is to characterize tion regulator activated by hydrogen peroxide in the thioredoxin system in L. plantarum WCFS1 the cell) exerts transcription regulation on the pursuing a functional genomics approach us- thioredoxin system (23). Proteomic studies in mi- ing transcriptomics, enzyme activity assays, and cro-organisms have allowed the identification of bioinformatics studies. This investigation clarifies a range of TRX-targeted proteins via the use of a the suggested roles of TR in the response of L. Tandem Affinity Purification tag. In addition, the plantarum WCFS1 to oxidative stress. Finally, in-vivo TRX-interacting proteins in Saccaromyces this study unveils the role of TR or the thioredoxin cerevisiae were identified using yeast two-hybrid system as a redox sensor in the cell. systems (36). Both proteomic approaches revealed possible associations within the complex networks of redox regulation in micro-organ- MATERIAL AND METHODS isms and TRX and underlines the broad impact Bacterial strains, plasmids, media, and of the thioredoxin system in the metabolic and culture conditions. The bacterial strains used regulatory network of the cell. in this study are summarized in Table 1. E. coli

34 2 Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1

TABLE 1. Bacterial strains and plasmids used in this study.

Strain and Plasmids Characteristics Reference

Strains

E. coli DH5D

E. coli E10

Lc. lactis NZ9000 MG1363 pepn::nisRK (3, 8)

L. plantarum WCFS1 Sequenced wild-type strain single colony isolate of NCIMB8826 form human (15)

saliva.

NZ7100 L. plantarum WCFS1 derivative with chromosomal integration of pEMnisRK This work

plasmid.

NZ7606 CmR , L. plantarum NZ7100 derivative carrying the pNZ8150 plasmid. This work

NZ7601 CmR, L. plantarum NZ7100 derivative carrying the pMS011 plasmid. This work

NZ7607 CmR, L. plantarum WCFS1 derivative carrying the pNZ7021 plasmid. This work

NZ7602 CmR, L. plantarum WCFS1 derivative carrying the pMS040 plasmid. This work

Plasmids

pUC18Ery AmpR, Em R (31)

pACYC184 CmR, Tet R (4)

pNZ84 CmR, pACYC184 derivative with deletion of Tetracycline resistance gene (30)

and BamhI site.

pNZ9521 EmR, nisRK cloned in pIL253, expression of nisRK driven by rep read- (14)

through.

pNZ7130 AmpR, EmR pUC18EM derivative carrying 1.0 kb DNA fragments of both L. This work

plantarum WCFS1 lp_0075 and lp_0077 genes.

pNZ7131 AmpR, EmR, pNZ7130 derivative with extra cloning sites Nsa1 and NsiI. This work

ppNZ7132 CmR, Em R, pNZ84 derivative carrying 1.0 kb DNA fragments of L This work

.plantarum WCFS1 (lp_0075) and (lp_0077) genes and the EryR resistance

marker.

pNZ7133 CmR, Em R p7132 derivative carrying the whole genes nisRK from LC. lactis. This work

pNZ8150 CmR, pNZ8148 derivative lactococcal cloning and expression vector with (19)

nisA promoter upstream of a multiple cloning site and Sca1restriction site.

pMS011 CmR, pNZ8150 derivative carrying L. plantarum (lp_0761) trxB1 gene, This work

translation fused to the nisA promoter.

pNZ7021 CmR, pNZ8148 derivative carrying nisA::pepN. (35)

pMS040 CmR, pNZ7021 derivative carrying L. plantarum lp_0761 trxB1 gene, This work

translational fused to the pPEPN promoter.

35 strains were grown at 37°C in TY (25). Lacto- lp_0077-REV, shown in Table 2. The PCR ampli- coccus lactis was grown in M17 at 30°C and cons obtained were digested with EcorI-BamHI, L. plantarum WCFS1 was grown at 37°C in de and BamHI-XbaI (sites introduced in the primers Man-Rogosa-Sharp (MRS) or in Chemically De- used are underlined in Table 2) and cloned into fined Medium (CDM) (31). a EcoRI-XbaI digested pUC18Em (33). The ge- DNA Manipulations. All molecular biology netic organization of the resulting plasmid was techniques were performed following established verified by PCR and was designated pNZ7130. protocols by Sambrook (25). DNA was digested To introduce additional endonuclease restriction according to the conditions recommended by sites in pNZ7130 the multiple cloning sites con- the commercial suppliers of the restriction en- taining fragment was amplified by PCR using the zymes (Boehringer, Breda, The Netherlands). In primers EM1 and EM2 (Table 2). The resulting all cases, DNA was eluted from 0.7 % agarose amplicon (35 bp) was digested with BamHI and gels using the Purification Kits from Promega BglII and cloned in BamHI digested pNZ7130, (Leiden, The Netherlands). For a PCR reaction yielding pNZ7131. The 2.6 kb HindII fragment we used: 1 μl template DNA (10 to 100ng), 2 of pNZ7131 that contains the lp_0076 flank- μl of each primer combination (50ng/ml), 1 μl ing regions separated by the multiple cloning dNTP’s (100nM), 1 μl PWO-polymerase (5 U/μl) site and the erythromycin resistance encoding and 10 μl polymerase buffer (5x) (Roche, Wo- gene, was subcloned in vector pNZ84, yielding erden, The Netherlands). The reaction mixtures the lp_0076 replacement vector pNZ7132. The were adjusted to 50 μl with deionized H2O. lp_0076::nisRK replacement plasmid pNZ7133 Construction of strain L. plantarum was constructed by cloning of the 2.4 kb HpaII- NZ7100. Previously it has been established PstI fragment of pNZ9521 (15) into the NsaI- that the nisin controlled expression system can NsiI digested pNZ7132. The plasmid pNZ7133 be functionally implemented in L. plantarum, was introduced into L. plantarum WCFS1 and by chromosomal integration of the regulatory primary integrants were selected on basis of module encoding genes nisRK (21). However, erythromycin resistance. The anticipated con- the strain described by Pavan et al. harbors figuration of the pNZ7133 plasmid integration an erythromycin resistance marker. Therefore, in the lp_0076 locus was confirmed by PCR a chromosomal lp_0076::nisRK gene replace- and Southern blotting (data not shown). One ment without resistance marker was construct- of the integrants was cultured for 140 genera- ed. For this purpose, lp_0076 upstream and tions without antibiotic selection. Subsequently, downstream flanking regions were amplified erythromycin sensitive (EmS) colonies were iden- by proofreading PCR using genomic DNA of tified by plating on media without erythromycin L. plantarum WCFS1 as template. The primers followed by replication plating on erythromycin used to amplify the 5’- and 3’- regions of lp_ containing plates. In the EmS colonies identified, 0075 and lp_0077 respectively were lp_0075- a candidate lp_0076::nisRK replacement mu- FORW and lp_0075-REV; lp_0077-FORW and tant was identified by PCR using internal primers

36 2 Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1

in NisK and Orfx (data not shown). In a selected onies displaying CmR plasmid was isolated and candidate lp_0076::nisRK strain, the anticipat- checked by restriction and PCR analysis. Then, ed genetic organization of the lp_0076::nisRK purified plasmid pMS040 was inserted into L. locus was further confirmed by additional PCR plantarum NZ7100 (14). These transformants and Southern blot analyses (data not shown). contained the trxB1 gene translationally fused

The lp_0076::nisRK derivative of L. plantarum with the constitutive PpepN promoter (30) and one WCFS1 was designated NZ7100. of this colonies was denominated NZ7602. Construction of trxB1 over-expressing strains NZ7601 and NZ7602: Control strains L. plantarum NZ7606 and L. plantarum NZ7601. The gene trxB1 in L. NZ7607. In the experiments with the nisin in- plantarum WCFS1 was amplified by PCR us- ducible promoter we used as control strain ing genomic DNA from L. plantarum WCFS1 NZ7606, a L. plantarum NZ1700 strain con- as template and two primers: trxB1-FORW and taining the pNZ8150 vector. On the other hand, trxB1-REV with a SphI restriction site (site intro- for the chemostat studies strain NZ7607 was duced in the primer used; underlined Table 2). used as control. Strain NZ7607 is L. plantarum The amplified DNA fragment was cloned into WCFS1 containing the pNZ7021 vector. the linearized vector pNZ8150. The complete plasmid, designated pMS011, was cloned and Nisin Induction. Nisin induction was done as purified fromLc. lactis NZ900 cells (12) display- previously described by Pavan et al. (21) using ing chloramphenicol resistance (CmR) and intro- 50 ng/ml nisin for induction. duced into L. plantarum NZ7100 . The plasmid Quantitative PCR assays. Total RNA was iso- isolated form the L. plantarum colonies display- lated from exponentially growing L. plantarum ing CmR were checked by restriction and se- WCFS1 cultures with the High Pure RNA Isola- quencing analysis. One of these transformants tion Kit (Roche, Woerden, The Netherlands). To containing the L. plantarum trxB1 gene transla- eliminate genomic DNA contamination a 60 min tionally fused to the nisA promoter was denomi- DNAseI treatment was included in the isolation nated NZ7601. procedure. Furthermore, the quality and quantity L. plantarum NZ7602. The gene trxB1 in L. of the RNA was confirmed using the RNA 6000 plantarum WCFS1 was amplified by PCR us- Nano Assay (Agilent Technologies, Amstelveen, ing genomic DNA from L. plantarum WCFS1 The Netherlands) using the Agilent 2100 expert as template and two primers: trxB1-KPNFORW Bioanalyzer. Cultures of the bacterium were and trxB1-XBA1REV containing a Kpn1 and grown in anaerobic jars on CDM containing Xba1 site, respectively. The amplified fragment 5mM Diamide at 37°C, or CDM containing was purified from the gel and cloned into the di- 5mM DTT at 37°C and 30°C. In all cases 0.5% gested vector pNZ7021. The complete plasmid, (wt/vol) glucose was added as carbon source. pMS040, was cloned and isolated from Lc. lactis For the reverse transcription reaction 200 ng NZ9000 colonies displaying CmR. From the col- RNA was incubated at 65°C for 5 minutes with

37 TABLE 2. Oligonucleotides used in this study. Designed restriction sites in the primers are shown underlined in the table.

Oligonucleotides Sequence References (5’ –3’)

Lp_0075-FORW ACGTGAATTCCAGTTCAACTAGAACAAGC

Lp_0075 -REV ACGTGGATCCTACAATCCGTTTCATATTG

Lp_0077-FORW ACGTGGATCCAAGGCGTCATCAAAATAG

Lp-0077-REV ACGTTCTAGACGCAATTTTCTTCACATTAC

EM1 GGATCCGTTAACGAATTCGGCGCGCCGGCGCCA

EM2 AGATCTGGCGCCGGCGCGGCGAATTCGTTAACG

trxB1-FORW ATGGCAAAGAGTTACGACG

trxB1-REV GTTGCATGCTCAATTAA ATCAGCTACGC

trxB1- 300 TTC AGA ACC AGT CCC AAT GAC

trxB1-KPN1FORW CGCCCACGGTACCATGGCAAAGAGTTACGACG

trxB1-XBA1REV CGCGCCTCTAGACTCAATTAAATCAGCTACGC

trxA1-FORW ATGATCGAACCAGTCGATAAG

trxA1-REV TTAGGCAGGCTTC ACTTC

trxA2-FORW ATGGTCGCAGCAACTACTG

trxA2-REV TTATAGATA TTGAGCTAAAGTTTG

trxA3-FORW ATGATTAAAGAAATACATGACC

trxA3-REV TTAAGCAAGTGCTTTTTTGAGTTC

trxH-FORW ATGGCAACTGCA ACTTTAAG

trxH-REV CTACTTTTCAAGTGTATCTAAG

trxB2-FORW ATGAGTGCAGAATATGATTTAAC

trxB2-REV CGGGTACCCCGGTTTCTCGTCCTATTTGC

trxB2-truncFORW ATT CGG ATC CGT AGT CTC TGC AAG TTG C

trxB2-truncREV TTT GCG GTA CCA CAG AAT ATG ATT TAA CAA TTA

270 ng random nonamers, and 1 μl of 10mM were stored at -20°C until use. Quantitative PCR dNTP mixture. After 5 minutes on ice, the follow- amplification was performed in 96-well plate on ing was added: 5 μl 5xFirst strand Buffer, 2 μl a 7500 Fast System (Applied Biosystems), us- 0.1M DTT, 1 μl Superscript III (200 UNITS), 1 μl ing SYBR green for product detection. Each well RNASEout (40 UNITS) (all Invitrogen) and water contained 10 μl SYBR green Master Mix (Applied to a final volume of 20 μl. The reaction was in- Biosystems), 200nM of Reverse and forward cubated at 25°C for 5 min and then at 50°C for primers, and 1 μl of 10-fold or 100-fold diluted 60 min. The reaction was inactivated by heating RT product as template. The qPCR gene-specific at 70°C for 15 min. Generated cDNA samples probes of L. plantarum WCFS1 of trxA1 (0.3

38 2 Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1 kb), trxA2 (0.3 kb), trxA3 (0.3 kb), trxH (0.3 kb), To study oxidative stress response, either di- trxB1 (0.3 kb), and trxB2 (0.6 kb) were amplified amide (5mM) or hydrogen peroxide (3.5mM) using specific primer combinationstrxA1 -FORW was added to the medium. and trxA1-REV; trxA2-FORW and trxA2-REV; Growth-zone inhibition assays. Cultures were trxA3-FORW and trxA3-REV; trxH-FORW and grown at 37°C in MRS until OD600 of 0.4. At this trxH-REV; trxB1-FORW and trxB1-300; trxB2- point, 2.5 ml of culture was plated by mixing truncFORW and trxB2-truncREV. All primers are with 50 ml of 0.7 % Agarose MRS at 40°C. After specified in Table 2. Amplification was initiated the agar/culture plates hardened, perforations at 95°C for 10 min, followed by 40 cycles of of 6mm diameter were made. In each of these 95°C for 15 sec, 53°C for 30 sec and 60°C for apertures, 30 μl of solution (1M, 0.5 M, 0.25 M 60 sec for probes trxA1, trxA3 and, trxH. For the and 0.10 M) of diamide or (1M, 0.5 M) hydro- other probes (trxA2, trxB1, and trxB2) the step of gen peroxide was added. Plates were incubated 53°C for 30 sec was changed to 55°C for 30 sec. overnight at 37°C. The growth inhibition to- All samples were measured in duplicate. Control wards the oxidative stress agents was measured PCRs were included to detect background con- as the radius of growth inhibition expressed in tamination (no-template control) and remain- centimeters. ing chromosomal DNA (RT reactions in which Preparation of cell free extracts (CFE). Cells enzyme superscript III was not added). PCR of a growing culture OD600 of 1.0 (approx. 10 specificity and product detection were checked ml) were harvested by centrifugation at 12.000 post amplification by examining the dissocia- x g for 10 minutes at 4°C. The pellet was washed tion curves of the PCR products. These melting twice with 1 M Tris-HCl buffer pH=7.5. Next, curve profiles were generated by first heating the the cells were suspended in 1 ml of 1 M Tris- samples to 95°C and then cooling them to 60°C HCl buffer pH=7.5 and subsequently disrupted and slowly heating then at 2°C/min to 95°C for by Fastprep (Qbiogene Inc., Illkirch, France) in detection of SYBR green fluorescence. In each four treatments of 40 seconds at speed 4.0. The run, five standards of the gene of interest were suspension was centrifuged 2 min at maximum included with appropriate dilutions of the cDNA, speed and the supernatant or CFE was trans- to determine the cDNA concentration in the ferred to a new vial and used directly. Protein samples. All q-PCRs amplified a single product content of the CFE was determined using the as determined by the melting curve analysis and standard BCA Protein Assay kit (Pierce, Etten- the relative expression level is given as arbitrary Leur, The Netherlands). units (Au). Thioredoxin reductase enzyme assays. The Growth curves. For the growth experiments, reaction mixture (RM) was prepared by mixing 96-well plates were used. Each well was filled (on ice) 200 μl of 1M Tris-HCl, pH=7.5, 500 with 200 μl medium and 10 μl growing cells μl of insulin (10 μg/μl), 2.5 μl EDTA pH=7.5

(OD600 of 1.0). The 96-well plates were inocu- (0.2M), and 40 μl NADPH (40 mg/ml). For each lated at 37ºC and cell density was measured by enzymatic reaction 40 μl RM were used together detecting the turbidity of the cultures at 600 nm with 20 μl TRX (60μM) from E. coli (Sigma), and every 10 min.

39 80 μl CFE. Each reaction was incubated for 25 500 ml medium. A pH of 5.5 was maintained by minutes at 37°C; after this 500 μl of 6M Gua- the addition of 5 M NaOH and the stirrer nidine HCl / 0.2 M Tris-HCl, pH=7.5/ 1mM speed was set at 200 rotations per minute (rpm). DTNB was added. The final absorbance at 412 The headspace of the fermentors and medium nm was measured at 25°C. Note that the ex- vessel were full at all times with nitrogen gas at a tinction coefficient of DTNB reduction is 13.6 mM- flow rate of 520 ml · min-1. The cultures were kept 1 and that there are two molecules of TNB per at the dilution rate of 0.1 h-1 and steady state was molecule DTNB being reduced. The background assumed after 5 volume changes. At steady state signal in the assay was determined in the ab- or t0, samples were taken for transcriptomics, HPLC sence of TRX in one of the assays. TR activity was analysis, dry weight, and enzyme samples. In addi- expressed as nmol of DTNB · (min · mg protein)- tion, at steady state, 50 ml of hydrogen peroxide 1 and calculated as follows: Δabs412 · (0.620) · was added to the fermentors resulting in a final -1 -1 6 (protein content) · 2 · (e412DTNB) ·10 . concentration of 3.5mM in the fermentor. After 30

Glyceraldehyde-3-phosphate dehydroge- min t30, samples were taken for transcriptomics, nase enzyme assay. The assay mixture con- HPCL analysis, and enzymatic analysis. tained: triethanolamine-HCl buffer (pH=7.6), Microarrays.

100mM; ATP, 1mM; EDTA, 1mM; MgSO4, RNA isolation. A 40 ml L. plantarum WCFS1 1.5mM; NADH, 0.15mM; phosphoglycerate culture at steady state was added to 160 ml of kinase 22.5 U (Boehringer, Breda, The Nether- quenching buffer (22) (60% methanol, 66.7mM lands); and CFE. The reaction was started with HEPES; pH=6.5; −40°C). Following quenching, 5mM 3-phosphoglycerate. The absorption of the cells were immediately pelleted by centrifu- the assay mixture was monitor at 340 nm (E340 gation at 18.000 × g for 10 min at −20°C. The nm of reduced pyridine-dinucleotide cofactors is pellet was resuspended in a screw cap tube con- 6.3 mM-1). Enzyme activity is expressed as μM taining in 0.4 ml TE, 250 μl acidic phenol, 250 of amount of NADH converted per (min · mg μl chloroform, 30 μl 3 M sodium acetate pH 5.2, protein)-1. All assays were performed with two 30 μl 10% sodium dodecyl sulphate, and 1 gram concentrations of cell extract to confirm that re- glass beads. The tube was immediately frozen action rates were proportional to the amount of in liquid nitrogen and samples were stored at cell extract added. Protein content of the CFE -80ºC. The cells were disrupted with three sub- was determined through the standard BCA Pro- sequent 40 second treatments separated 1 min tein Assay kit. on ice incubations in a Fastprep (Qbiogene Inc., Continuous cultivations. Cultures were grown Illkirch, France). After disruption, 0.5 ml of the at 37°C in CDM (31) which was supplemented aqueous phase was used for RNA isolation with with 100mM glucose and 10 μg/ml of chloram- a High Pure kit, which included 1 h of treatment phenicol. A 1 Liter bioreactor (Applikon Depend- with DNaseI (Roche Diagnostics, Mannheim; able Instruments) was inoculated with cells from Germany). The isolated RNA was eluted in 50 an overnight culture to an initial OD600 of 0.1 in μl of elution supplied in the kit. cDNA synthesis

40 2 Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1

and purification. Before first-strand cDNA syn- Dried slides were scanned in the Scan Array thesis, the absence of genomic DNA and RNA Express (PerkinElmer Life Sciences; Packard Bio- degradation in the RNA samples was confirmed science) at 10 microns. Spot intensity data was using the RNA 6000 Nano Assay (Agilent Tech- quantified (average intensity) in ImaGene- ver nologies, Amstelveen, The Netherlands) using sion 5.0 (BioDiscovery, Inc., El Segundo, CA). the Agilent 2100 expert Bioanalyzer. First-strand Signal intensities of all probes were corrected cDNA synthesis was carried out by the CyScribe against background and normalized by fitting a

Post-Labelling and Purification kit (Amersham plot of M (M= 2log [cy5 intensity/cy3 intensity])

Biosciences, Buckinghamshire, UK) following against A (A= 0.5 · 2log [cy5 intensity · cy3 in- the manufacturer’s instructions with two modifi- tensity]) using the lowess algorithm in BASE (24). cations. The starting amount of mRNA used per The normalized data has been made available sample was 25 μg and the synthesis reactions at GEO with accession number GSE8348 (9). were incubated for 3 h at 42°C. The fold change (FC) is defined as 2M. For the Data acquisition and processing. Two inde- statistical analysis we used microarray analysis pendent chemostat cultivations were performed of variation (R/maanova) (5). In this maanova for both wild-type and trxB1 over-expression test we used three variables: fermentation, treat- strain. Three hybridization experiments, all with ment, and genotype. Moreover we tested the the same hybridization scheme (see below), were model taking into consideration the interaction performed with samples obtained from these between genotype and treatment. The maanova chemostats before and after treatment with hy- test resulted only in two sets of interesting data drogen peroxide. Two of these experiments used because the interaction effect did not reveal sig- samples obtained from the same fermentation. nificant changes in transcript levels. One dataset In each experiment three arrays were used. Per representing the genes affected as a result of the array two cDNA labeled targets were hybridized overproduction of TR and another set represent- on custom designed L. plantarum WCFS1 11K ing the genes affected due to oxidative stress. Agilent oligo microarrays published at the gene These two datasets were denominated genotype expression omnibus database with accession and hydrogen peroxide datasets respectively. number GPL4318 (9) using the Agilent 60-mer Significantly regulated genes within each data- oligo microarray processing protocol version set were defined as genes whose nominal pvalues 4.1. These microarrays contained an average were less than 1% and had a FC equal or higher of three probes per gene. The hybridization than 1.5. scheme contained the following cDNA com- Regulatory Motifs. Determination of regulatory parisons (1) wild-type with trxB1 over-expression motifs from a list of genes was based on pre- mutant; (2) wild-type with wild-type treated with dictions of upstream transcriptional units (TU’s) hydrogen peroxide; and (3) wild-type treated made using an in-house Python program, align- with hydrogen peroxide with trxB1 over-expres- ments of these TU’s using the MEME (7) and sion mutant treated with hydrogen peroxide. determining of nucleotide sequences using the

41 MAST software tools (10). The MEME software RESULTS determines through alignments small motifs In-silico analysis of the Thioredoxin which are present in the analyzed sequences system and assigns a significant value, valueE , and score The annotated genome of L. plantarum WCFS1 per position. This significant value represents (16) reveals that the thioredoxin system in this how well preserved the motif is, the number of organism is composed of six genes. Four genes TU’s that this motif has, and the conserved posi- are annotated as TRX encoding genes: trxA1 tion of the motif in the analyzed TU’s. The MAST (lp_0236), trxA2 (lp_2270), trxA3 (lp_3437), software allows searching for the motif in the and trxH (lp_2633). The other two genes trxB1 genome of interest. (lp_0761) and trxB2 (lp_2585) are annotated respectively as TR and nucleotide-disulphide oxidoreductase. These six genes are dispersed throughout the genome and are highly conserved within L. plantarum strains regardless of their ecological niche (19). Bioinformatics tools were used to investigate the evolutionary relationship of the TRX encoding genes in L. plantarum WCFS1. The translated gene sequences for trxA2 and trxB1 from L. plantarum WCFS1 showed the highest homology (pscores higher than e-152) to characterized TRX and TR respectively of different organisms such as Lc. lactis and B. subtilis. In L. plantarum TrxB1 and TrxB2 have a 28% similarity at the protein level. The orthologous relation of trxB1 from L. plantarum WCFS1 with trxB of B. subtilis and the similarity in the alignment with other trxB genes (including trxB from E. coli) suggests a molecular function of trxB1 as TR. Interest- ingly, the sequence of trxB2 does not possess an active center -CACV- which is an essential characteristic of the TR suggesting that this gene does not have the potential of reducing TRX. By phylogeny analysis it was determined that the TrxB2 protein of L. plantarum WFCS1 is orthologous to the B. subtilis protein YUMC suggesting that these two sequences have the same molecular

42 2 Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1

function namely to act as a ferredoxin NAD(P) tively higher expressed when the bacterium was reductase (27). cultivated at 37°C compared to the other four studied genes evaluated at the same incubation temperature. When comparing the effect of dif- In-vivo functionality of the thioredoxin ferent growth conditions within each probe, we system observed -with the exception of trxB2- that the The in-vivo functionality of the complete thiore- highest expression of each studied transcript is doxin system was studied by quantitative PCR (q- observed under diamide exposure compared to PCR) of RNA isolated from L. plantarum WCFS1 when grown with Diamide or different tempera- grown in CDM at 37°C under anaerobic and tures (37°C and 30°C). Our observations sug- oxidative stress conditions. The cultures were gest a role for both trxA2 and trxB1 genes in the exposed to different temperatures or redox en- response mechanism of this bacterium towards vironments created by the addition of diamide oxidative stress while the gene, trxB2 together or DTT to the medium. The relative expression with trxA2 could play a role during a heat shock expressed in arbitrary units (Au) of the six dif- or reductive stress. Based on the gene expres- ferent genes that compose the thioredoxin sys- sion data and the in-silico sequence analysis; we tem is presented in Table 3. We observed that propose that trxB1 is coding for the main TR in trxA2 and trxB1 were relatively higher expressed the thioredoxin system in L. plantarum WCFS1. under all studied conditions when compared to To study the role of this enzyme in more detail, the other trxA’s and trxB2 respectively. More- we constructed L. plantarum strains with elevated over, transcripts trxB2 and trxA2 were respec- expression levels of trxB1.

TABLE 3. Transcript levels of the thioredoxin system components in L. plantarum WCFS1 in response to different stresses. Relative expression levels of trxA1, trxA2, trxA3, trxH, trxB1, and trxB2 in Lactobacillus plantarum WCFS1 grown under different oxidative environments.

trxA1 trxA2 trxA3 trxH trxB1 trxB2 Sample AV 5 stdev AV stdev AV stdev AV stdev AV stdev AV stdev A1 115.95 67 186.78 28 128.83 33 94.80 32 197.93 12 151.40 27 B2 98.18 66 506.75 23 122.20 7 132.15 33 367.80 47 392.55 14 C3 156.50 33 357.43 68 152.20 41 138.13 18 180.75 55 304.17 69 D4 240.25 30 2104.90 124 870.00 8 380.05 81 1258.68 102 139.43 65

1 RNA isolated at OD600 =1.5 from L. plantarum WCFS1 grown at 30ºC on CDM with 0.5% glucose . 2 RNA isolated at OD600 =1.5 from L. plantarum WCFS1 grown at 37ºC on CDM with 0.5% glucose. 3 RNA isolated at OD600 =1.5 from L. plantarum WCFS1 grown at 37ºC on CDM with 0.5% glucose until OD600=1.0 when DTT was added to a final concentration of 5mM. 4 RNA isolated at OD600 =1.5 from L. plantarum WCFS1 grown at 37ºC on CDM with 0.5% glucose until OD600=1.0 when diamide was added to a final concentration of 5mM. 5 Average is given in Arbitrary units determined by a standard curve made with each probe.

43 Overproduction of TR to 300nmol DTNB · (min · mg prot)-1 when no Overproduction of TR was achieved using the nisin is added to the medium (Fig. 1). The TR NICE expression system (6). The transformant activity of NZ7601 without the addition of nisin (L. plantarum NZ7601), carries the gene trxB1 corresponded to the TR activity of the control under the control of the nisin promoter. Strain strain NZ7606 with values, both under aerobic NZ7601 displayed 3-fold higher TR activity af- and anaerobic growth conditions, of 300 and ter induction with 50 ng/ml nisinZ compared to 320nmol DTNB · (min · mg prot)-1 respectively. the wild type strain NZ7606 (data not shown). An elevated expression of the trxB1 transcript The strain NZ7601 is able to reduce more than was also detected in NZ7601 by northern blot 1000nmol DTNB · (min · mg prot)-1 compared analysis (data not shown).

FIGURE 1. Nisin-controlled overexpression of thioredoxin reductase (TR). The strain NZ7601 was grown (black bars) aerobically and (gray bars) anaerobically at 37°C and was induced with 0 or 50 ng/ml nisinZ. After 5 hours of induction cell free extracts were prepared and TR activity was measured. The data shown above is the result of three independent experiments. Enzyme activity is showed as nmolDTNB reduced per min per mg protein.

1200

1000

800

600

400

200 nmol / DTNB min mg protein

0 0 50 Nisin (ng/ml) 44 2 Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1

Next, we evaluated the effect of overproduction fected by the oxidative stress showing a μmax re- of the reductase in the oxidative stress response. duction to 0.17 h-1 and 0.15 h-1, respectively. For this we monitored the growth of the trxB1- However, later on the growth behavior of strain overexpressing strain NZ7601 and wild-type NZ7601 differed from that of the wild-type.

NZ7606 in the presence of 5mM diamide, a While strain NZ7606 after reaching an OD600 of known thiol oxidant (18) (Fig. 2). Both tested 0.5 stops growing and enters stationary phase, strains grew with a maximum specific growth strain NZ7601 continues to grow under oxida- rate (μmax) of 0.36 h-1 when no stress factor tive stress and reaches a cell density compara- was present in the medium. In the presence of ble to that observed in the absence of oxidative diamide, both strains were initially similarly af- stress.

FIGURE 2. Effect of TR overproduction in L. plantarum on response towards oxidative stress. Strain NZ7601 carrying the over expression of gene trxB1 was grown in MRS broth (white circles) and in MRS with 5mM diamide (black circles) and the wild type strain NZ7606 was grown in MRS broth (white triangles) and in MRS with 5mM diamide (black triangles). The growth was monitored by turbidity measurements at 600 nm.

0.5

-0.5 Ln OD600 Ln

-1

-1.5

-2 0 100 200 300 400 500 600 700 800 900 1000 Time (min)

45 This response under oxidative stress of strain drogen peroxide pulse. Because the NICE system NZ7601 was investigated in a quantitative is not suitable for continuous cultivations due to growth-zone inhibition assay (Fig. 3). In this as- the instability of nisin in the medium (data not say, we tested the inhibitory capacity of hydro- shown), we used a constitutive promoter, PpepN, gen peroxide and diamide on strains NZ7601 to drive expression of trxB1. Therefore, two new and NZ7606. We observed that strain NZ7601 strains were constructed: strain NZ7602 carry- at all studied oxidant concentrations showed a ing plasmid pMS040 where trxB1 is under the smaller inhibition zone compared to NZ7606. control of the a constitutive promoter from Lc.

lactis (30), PpepN, and strain NZ7607 as a control strain carrying the empty plasmid pNZ7021. Transcriptome analysis The continuous cultivations were performed in To analyze the correlation between TR over- biological triplicates (Table 4). In the chemo- production and oxidative stress we carried out stat cultivations glucose present in the medium transcriptome analysis of the constructed strains. (100mM) was consumed to undetectable levels The mutant and the wild-type strains were grown (<0.5mM) and the biomass produced was com- in chemostats at a dilution rate of 0.1 h-1, and parable in both strains. In addition, we found samples for transcriptome analysis were drawn no growth difference between both strains under at steady state both prior and 30 min after a hy- this cultivation condition.

FIGURE 3. Effect of TR overproduction on tolerance to hydrogen peroxide and diamide. The zone of growth inhibition towards hydrogen peroxide or diamide is showed in centimeters in the y-axis. The different strains are represented in the figure as such: L. plantarum NZ7606 (white bars) and L. plantarum NZ7601 (black bars).

1.2

1 Results of one experiment

0.8

0.6

0.4 Radius of inhibition (cm) inhibition of Radius

0.2

0 1M 0.5 M 1M 0.5 M 0.25 M 0.10 M

Hydrogen Peroxide Diamide

46 2 Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1

TABLE 4. Growth yield and GAPDH enzyme activity of L. plantarum strain NZ7602 and control strain NZ7607.

Strain Substrate Physiological parameters

a b c Glc Y glc-X GAPDH

Medium 100

NZ7607 u.d 0.12 0.62 ± 0.02

NZ7602 u.d 0.13 1.61 ± 0.12

NZ7607 + perox u.d 0.14 n.d

NZ7602 + perox u.d 0.15 n.d

a mmol of glucose present in the supernatant b Biomass yield on glucose (g of biomass/g glucose consumed). c Enzyme activity expressed as nmoles NADPH · (min · mg prot)-1 u.d. < 0.1 n.d Nt determined Mean ± SD of two separate chemostat steady states

The overexpression of trxB1 present in strain protein synthesis (tuf), protein fate (lp_1023), NZ7602 was in itself a good internal control in and cellular envelope (ica3). It should be not- the transcriptome analysis. The transcript of this ed that none of the mentioned stress-related gene was found to be 23-fold more abundant genes has a -CxxC- cysteine-rich active center. in the trxB1 over-expressing strain compared Furthermore, a gene related to cysteine amino to the control strain. This increase in transcript acid metabolism was also up-regulated in strain level resulted in a doubling of TR activity in NZ7602 compared to wild-type specifically strain NZ7602 compared to the wild-type (data the gene coding for serine O-acetyltransferase not shown). The ANOVA statistical test that sum- (cysE). We also observed a significant up-regu- marizes the significant effects due to overexpres- lation (1.7 fold) of the gene gapB predicted to sion of trxB1 showed that there were in total 27 encode glyceraldehyde-3-phosphate dehydro- significantly affected genes (pvalue <0.01 and FC genase. The GAPDH protein is involved in en- ≥1.5) in the trxB1 over-expressing strain when ergy metabolism of the cell. In the annotated compared to the wild-type (Table 5). We ob- genome of L. plantarum WCFS1 gapB is the served that 15% of them are predicted to encode only predicted glyceraldehyde-3-phosphate de- proteins involved in the purine and pyrimidine hydrogenase encoding gene. The in-vitro activity biosynthesis. Other affected genes are predicted of GAPDH in strain NZ7602 was analyzed and to encode proteins involved in stress response found to be approximately three-fold higher in

(groEL, npr2), transport and binding (mntH2), comparison with the wild-type (Table 4).

47 The transcriptome analysis of cells exposed to Regulatory networks a hydrogen peroxide pulse allowed us to ana- In the group of significantly affected genes due lyze the effect of hydrogen peroxide on both the to oxidative stress (Supplementary materials, Ta- trxB1- overexpressing and wild-type strains. Oxi- ble S1, see page 156) the largest group (22%) dative stress significantly affected a total of 267 corresponded to genes predicted to encode hy- genes with a pvalue <0.01 and FC ≥1.5 (Supple- pothetical proteins. In order to find a functional mentary materials, Table S1 see page 156). We correlation in the entire group of peroxide-af- observed that most of the up-regulated genes fected genes, we looked for regulatory motifs (159) are predicted to encode proteins associ- or binding sites in the upstream region of the ated with defense mechanisms against oxida- affected genes. As a result of this analysis, we tive challenge: hydrogen peroxide detoxification found two motifs (Supplementary materials, Ta- (npr2, kat, trxA2, pox3, pox5), exonucleases ble S1, see page 156) in the upstream region (rexA, rexB); stress response (asp1, asp2, groES, of a number of analyzed sequences. The first groEL); DNA repair (dinP, dnaE, recA); putative motif was found in the upstream region of 15 DNA helicases (lp_0910, lp_0308, lp_0432) of the affected genes and had the consensus se- and polymerases (umuC); transcription regula- quence of the LexA-DinR regulator in B. subtilis tor repressor of the SOS regulon (lexA), ferrous AGAACGTACGTTCG (Fig. 4) (38). Within the iron transport (feoA, feoB), as well as an un- group that contained this promoter sequence, characterized transcription regulator (lp_1360), we found well known stress-induced genes a manganese transporter (mntH2), and energy (ruvA, ruvB, lexA, rexA, rexB) (38). These genes metabolism (gapB). translate into proteins known from literature to be induced under conditions that cause DNA Our data showed that hydrogen peroxide stress damage or blockage of DNA replication. Also also resulted in significant down-regulation of genes predicted to encode proteins with hypo- the expression of 108 genes (Supplementary thetical functions (lp_0091, lp_0270, lp_0030, materials, Table S1, see page 156). The down- lp_2224, lp_2939, lp_3141, lp_0981, lp_3022, regulated genes are predicted to encode pro- and lp_1611) contained the LexA-DinR con- teins involved in glucose catabolic pathways: sensus promoter sequence. It is worth mention- the mannose PTS; energy metabolism (ndr, ing that these genes are highly induced under nar genes, adhE, fruk); fatty acid biosynthesis hydrogen peroxide stress. For example, gene (fab genes):nonribosomal peptide biosynthesis lp_1611 was induced 32-fold as compared to (nspA, nspB, and nspC); sugar uptake (sacR, the wild-type. ccpA). In addition, genes found down-regulated upon hydrogen peroxide stress included genes A second significant regulatory motif denoted predicted to encode prophages, cellular surface here as the Stress Response Element (SRE) and proteins, and acetyl-CoA carboxylases (accb2, found in 7 genes, had the consensus sequence accC2, accD2, and accA2). AACTAGCCGCGGTGGC (Fig. 4). The important exonucleases: (rexA, rexB), a DNA poly-

48 2 Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1 1 ) ) 4% ( 4% ( nthesis functions y y ulator Protein fate (4%) Cell envelope (7%) g Protein s DNA metabolism (4%) Energy metabolism (4%) Cellular processes (7%) Re Hypothetical proteins (19%) Main Functional Class (%) Amino acid biosynthesis (4%) Purines, pyrimidines, nucleosides and nucleotides (15%) ) . ) lase I y putative ) ( cos y ) l y NADPH g ( putative ( enz me y g L. plantarum putative ulator ( g ltransferase Product ladenine y y log(cy5/cy3) ratio. 2 Predicted gene names, function, fold change induction as well as main functional classes of the of classes functional main as well as induction change fold function, names, gene Predicted g ltransferase g y y ation factor Tu dro enase y g y cos clo-li ase y l y GroEL chaperonin NADH peroxidase extracellular protein, gamma-D- glutamate-meso-diaminopimelate muropeptidase DNA-3-meth type 4 prepilin-like proteins leader peptide processin serine O-acet g elon phosphoribosylglycinamide form ltransferase thioredoxin reductase glyceraldehyde 3-phosphate deh phosphoribosylformylglycinamidine c phosphoribosylaminoimidazole carbox lase, catal tic subunit unknown unknown unknown unknown unknown transcription re amino acid transport protein E 2 N s g urE ur tuf y urM apB , where is M the roEL ica3 ta npr2 p c p M p Gene trxB1 g g 2 1.6 1.6 1.6 1.5 1.6 2.1 2.3 3.9 1.8 1.7 1.6 0.6 1.5 1.6 1.8 1.5 1.7 1.7 1.6 FC 23.5 trx/wt < 0.01 & FC ≥ 1.5). ≥ FC & 0.01 < value Locus lp_2544 lp_3020 lp_3421 lp_1237 lp_1081 lp_1611 lp_1708 lp_1880 lp_1023 lp_2716 lp_0728 lp_2721 lp_2729 lp_0761 lp_1230 lp_3278 lp_0254 lp_0789 lp_2119 lp_2722 Values given in parenthesis correspond to the percentage of total amount genes (27) that belong each depicted functional class. Fold Change (FC) is express as 2 TABLE 5. Summary of significant affected genes (27) in the over expressing strain, NZ7602, when compared to the to compared when NZ7602, strain, expressing over the in (27) genes affected significant of Summary 5. TABLE (p wild-type in overrepresented found classes those are bold in presented classes functional Main columns. in displayed are genes affected significant this study when compared to the total genome of

1 2

49 FIGURE 4. Weblogo representation of conserved promoter regions in peroxide affected genes found using bioinformatics tools A) Regulatory motif LexA-DinR and B) uncharacterized regulatory motif referred in this study as the Stress Response Element (SRE). (a color representation of this figure can be found in page 151)

merase (DNA polymerase III), the transcription studied conditions (Table 6). The commonly af- regulator (rnhB), a glycolytic gene (gapB), a cell fected genes constitute 59% of the genes found envelope protein, and the hypothetical proteins affected by a trxB1-overexpression (Table 5) and lp_0145 and lp_1484 encoding genes con- 6% of the genes affected by hydrogen perox- tained this second promoter consensus. To the ide stress (Supplementary materials, Table S1, best of our knowledge, this motif has not been see page 156). From the 16 affected genes, 15 reported before. genes responded similarly to an over expression of trxB1 as well as to a hydrogen peroxide pulse; The same analytical approach was applied to these 15 genes were up regulated in both data- the data set containing the genes affected due to sets. The up-regulated genes (15) have already trxB1-overexpression. Yet, no significant regula- been described and correspond to genes pre- tory motif was found. It can be concluded that dicted to encode proteins involved in stress relat- genes predicted to encode the group of hypo- ed processes (npr2), energy metabolism (gapB), thetical proteins (19%) affected under trxB1- protein synthesis (tuf, rplB), manganese uptake overexpression do not share a clear single regu- (mntH2), purine biosynthesis (trxB1, and pur latory motif. genes), putative transcript regulators (lp_1360, lp_0126), and hypothetical proteins. The only gene found in both datasets that does Comparison between oxidative stress not share the same regulation is groEL. This gene response and the effect of trxB1 over- is 0.6-fold down-regulated as a result of the over production expression of trxB1 and 1.7-fold up-regulated in Finally, we explored the crossover between the response to a hydrogen peroxide pulse. In or- transcriptional response of TR over production der to analyze the effect at the metabolic level, with the transcriptional response obtained fol- the datasets were superimposed showing they lowing hydrogen peroxide treatment of both shared two major metabolic pathways: purine wild-type and NZ7602. This exercise resulted metabolism and cysteine biosynthesis. in a list of 16 genes which are affected in both

50 2 Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1 • conflicting Genes presented in • • • • • • • • • • • • • • • up Transcript Behavior Cell envelope DNA metabolism Protein synthesis Cellular processes Energy metabolism Purines, pyrimidines, Hypothetical proteins Main Functional Class nucleosides and nucleotides Transport and binding proteins ) t ) NADPH ( putative ( y Product g g ltransferase y y y anese transport protein ation factor Tu dro enase y g g y y cos cos lase I y y l l phosphoribosylformylglycinamidi ne c clo-li ase g NADH peroxidase DNA-3-methyladenine g elon phosphoribosylaminoimidazole carbox lase, catal tic subuni thioredoxin reductase man GroEL chaperonin phosphoribosylglycinamide form ltransferase glyceraldehyde 3-phosphate deh unknown unknown unknown cation transport protein unknown 2 2 E 3 f g2 ur urN tu urM apB roEL ica ta npr p p p Gene g trxB1 g mntH log (cy5/cy3) ratio. 2 0.5 4.2 2.9 1.8 0.7 1.7 0.7 3.4 0.6 1.8 3.1 1.8 5.0 1.8 1.8 32.9 1 treatment FC , where is M the M 1.6 2.1 2.3 3.9 1.7 1.6 0.6 1.6 1.6 1.8 2.6 1.7 1.5 1.6 1.7 23.5 genotype Locus lp_1081 lp_1611 lp_1708 lp_1880 lp_2119 lp_2716 lp_0728 lp_3020 lp_2722 lp_0761 lp_1087 lp_2992 lp_2721 lp_0789 lp_2544 lp_2729 TABLE 6. Crossover TABLE between genotype affected genes and hydrogen peroxide affected genes. this table are the common genes (16) between the two studied datasets: genotype and treatment. Further a summary of the regulation pattern for each of the depicted gene is given. Change (FC) is express as 2 1

51 DISCUSSION catalase, under conditions of hydrogen perox- In this study, we have characterized the role of ide stress. Furthermore, under hydrogen perox- TR in oxidative stress response of Lactobacillus ide stress pox genes (pox3, pox5) predicted to plantarum. We found that overproduction of TR encode pyruvate oxidase were found highly up- in L. plantarum WCFS1 improved tolerance of regulated compared to the wild-type. At first, this the strain towards an oxidative stress produced was an intriguing observation. Pyruvate oxidase by hydrogen peroxide or diamide. Global tran- catalyzes a reaction that utilizes pyruvate, oxy- scriptional analysis revealed a striking similar- gen, phosphate, and water and produces intra- ity in response towards the overproduction of cellular hydrogen peroxide, carbon dioxide, and the oxidoreductase TR and hydrogen peroxide acetyl phosphate. Yet, in this study there was no stress. Our observations suggest that overpro- oxygen added because the cultures were grown duction of TR triggers the induction of a specific anaerobically. However, genes predicted to en- set of 16 genes predicted to encode proteins as- code catalase and NADPH peroxidase were also sociated with oxidative stress response. This may found up regulated compared to the wild-type. If explain the phenotype of the trxB1-overexpres- these two latter enzymes were to be active, they sion strain, which appears to be prepared for would liberate oxygen and water respectively encountering oxidative stress. These genes are from hydrogen peroxide and therefore explain- predicted to encode proteins involved in purine ing the observed behavior of genes pox3 and metabolism, protein synthesis, as well as in cel- pox5 under hydrogen peroxide stress lular and energy metabolism. Hydrogen peroxide not only provokes up-regu- The global transcriptional analysis obtained lation of genes at the transcriptome level. We from cultures affected by hydrogen peroxide also observed down-regulation of genes pre- stress and in the absence of heme, offers com- dicted to encode proteins involved in main met- plementary information for the characteriza- abolic pathways: glycolysis, fatty acid biosyn- tion of the kat, and pox genes in L. plantarum thesis, non ribosomal peptide biosynthesis, and WCFS1. The catalase gene, kat, in L. plantarum amino acid metabolism. This observation sug- WCFS1 is an ortholog of the well characterized gests that in the presence of hydrogen peroxide manganese-dependent catalase in L. plantarum stress, the cell responds to it with a reduction CNRZ 1288 (1). Accordingly, in our dataset the in biomass formation. The transcriptome data genes kat and mntH2 predicted to encode cata- suggest that the pathways mentioned above are lase and a manganese transporter, were both being kept temporarily “on hold” until the cell significantly up-regulated three- and two-fold has detoxified from the oxidative stress induc- respectively under hydrogen peroxide. Hence, ing compounds. This reasoning was supported our transcriptome analysis only supports the fact by metabolite analysis data showing interrupted that the gene kat of L. plantarum WCFS1 may lactate production after exposure to hydrogen code a pseudocatalase or non-heme dependent peroxide (data not shown).

52 2 Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1

This study depicts the impact of overproduction (GAPDH) activity are significantly increased of TR at the transcriptome level in L. plantarum. 1.7-fold and 3-fold, respectively, upon overex- The overexpression of trxB1 affects the expres- pression of trxB1 and hydrogen peroxide stress. sion level of genes predicted to encode proteins GAPDH plays an important role and has been involved in stress related processes, DNA/RNA considered the key enzyme of glycolysis for its biosynthesis, and sulfur containing amino acid location in the pathway and the number of regu- biosynthesis. The resulting wide spectrum of latory interactions associated with this enzyme. pathways affected by TR has already been es- A correlation between the GAPDH and TR has tablished. Research performed by Vido et al. on already been reported in literature. Vido et al., aerobic growth of Lc. lactis (2005) revealed that (35) suggested that a disruption on the trxB1 in Lc. lactis, TR is involved not only in oxidative gene in Lc. lactis leads to induction of only one stress but also in carbon and lipid metabolism of the two genes that code for the GAPDH pro- (35). tein, specificallygapB . Furthermore, the induced GapB was only present in the reduced form. This Overproduction of TR in L. plantarum WCFS1 functionality prevents the formation of oxidized resulted in up-regulation of genes predicted to GapB and sustains formation of the reduced or encode heat shock and stress related proteins active form of GAPDH. Interestingly, in our case compared to the wild-type. For example, the the observed approximately three-fold higher gene groEL was found to be down regulated in GAPDH activity is observed as a result of over- strain NZ7602. This gene has been previously expression of trxB1 in contrast to a disruption in reported to be induced in B. subtilis (26) under trxB1 as reported by Vido et al (35) in Lc. lactis. stress conditions (heat, salt, and ethanol) to- Although overexpression of gapB in the strain gether with TRX encoding genes. In this study the NZ7602 results in higher GAPDH activity, the genes trxA1, trxA2, trxA3, and trxH predicted to elevated transcript levels under oxidative stress encode TRX were not found significantly affected were even more interesting. Studies from Van in strain NZ7602 at the transcript level. This re- Niel et al., (34) show that the GAPDH protein is sult suggests that the protein TRX is not affected an easy target for oxidative stress (cysteine resi- at the transcript level in strain NZ7602. Hence, dues) and together with glucokinase, fructose the genes regarded as significantly affected in 1-6 biphosphate, and aldolase has been found strain NZ7602 are more likely to be a response to be inhibited by 2.2mM hydrogen peroxide in to the direct manipulation of the TR level and/or Lc. lactis (34). Nevertheless, in our studies, we the protein load resulting from TR overproduc- observed elevated GAPDH activity upon expo- tion and not to the change in the TRX balance sure to 3.5mM hydrogen peroxide. Hence, we of the cell. can suggest that the functionality of GAPDH is organism dependent. We suggest that overpro- We have shown that gene levels of gapB and duction of TR in L. plantarum WCFS1 protects glyceraldehyde-3-phosphate dehydrogenase the GAPDH under conditions of oxidative stress

53 Using bioinformatics tools we were able to In summary, we have presented evidence that uncover new main stress response genes in L. TR is a factor in the oxidative stress response in plantarum WCFS1. For example the hypotheti- L. plantarum WCFS1. Our transcriptome data cal gene lp_1611 which contains the regulatory suggest that the thioredoxin system (trxA2 and motif LexA-dinR. The gene lp_1611 was found trxB1) is induced under hydrogen peroxide stress to be up-regulated 32-fold under hydrogen per- in L. plantarum WCFS1. Moreover, the discovery oxide stress. This gene is located adjacent to the of a crossover-group of genes with a common glycine ABC transporter operon just in the op- response under hydrogen peroxide stress, espe- posite direction of the opu genes. The position cially genes predicted to encode DNA-repairing and direction of the gene lp_1611 may suggest and stress proteins, including the pur genes, a role as a regulator of transcription; neverthe- npr2, and gapB, leads to the hypothesis that less, more analysis such as finding the DNA overproduction of TR results in a “mock-stress- binding region and activity in the protein need mode.” As a result under TR overproduction, 16 to be conducted. genes predicted to encode proteins involved in purine biosynthesis, cell wall biosynthesis, en- Another group of oxidative-stress-affected genes ergy metabolism, cellular envelope biosynthe- contains a second regulatory motif that we have sis, amino acid metabolism are activated. The termed SRE. In this group we found three genes activation of these genes provides an explana- that had already been defined as LexA-dinR tion of why strain L. plantarum NZ7602 is better regulated genes (rexA, rexB, recA). Moreover, in adapted to a challenge with hydrogen peroxide the group of genes containing the second mo- stress in comparison to the wild-type. The ob- tif, we found genes predicted to encode proteins served crossover between TR overproduction involved in primary energy metabolism (gapB) and hydrogen peroxide stress contributes es- and protein synthesis (rnhb) implying that there sential information to understand the oxidative is at least one more regulatory mechanism in- stress related signal transduction cascade in L. duced in L. plantarum WCFS1 upon oxidative plantarum. Nevertheless, this study clearly needs stress. We did not find a unique regulatory motif to be followed up with experiments showing the or SRE motif element in the group of genes af- effect, on the group of 16 genes resulting from fected by overproduction of TR. This reflects the an inactivation of trxB1 in L. plantarum. New complexity of the regulatory mechanisms associ- leads concerning catalase and the GAPDH pro- ated with TRX or TR. Perhaps, these up-regulated tein functionality in L. plantarum emerged from genes are governed by a complex network of our transcriptome dataset. These leads will be signal transduction events which is initiated by further studied. Using regulatory motif searches the thioredoxin system. Hence, we suggest the we identified a known stress regulatory element: set of 16 genes is part of the TR-specific defense LexA-dinR and also a new motif: AACTAGCC- mechanism of L. plantarum WCFS1 against oxi- GCGGTGGC (Fig. 4). Finally, transcription dative stress. profiling also revealed new actors such as the

54 2 Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1

uncharacterized gene lp_1611 which seems to play a role in the oxidative stress response in L. plantarum WCFS1. Current investigations aim at pinpointing the role of TR in the oxidative stress response.

55 REFERENCES

1. Abriouel, H., A. Herrmann, J. Starke, N. M. Yousif, A. Wijaya, B. Tauscher, W. Holzapfel, and C. M. Franz. 2004. Cloning and heterologous expression of hematin-dependent catalase produced by Lactobacillus plantarum CNRZ 1228. Appl Environ Microbiol 70:603-6. 2. Arner, E. S., and A. Holmgren. 2000. Physiological functions of thioredoxin and thioredoxin reductase. Eur J Biochem 267:6102-9. 3. Bolotin, A., P. Wincker, S. Mauger, O. Jaillon, K. Malarme, J. Weissenbach, S. D. Ehrlich, and A. Sorokin. 2001. The complete genome sequence of the lactic acid bacterium Lactococcus lactis ssp. lactis IL1403. Genome Res 11:731-53. 4. Chang, A. C., and S. N. Cohen. 1978. Construction and characterization of amplifiable multicopy DNA cloning vehicles derived from the P15A cryptic miniplasmid. J Bacteriol 134:1141-56. 5. Cui, X., and G. A. Churchill. 2003. Statistical tests for differential expression in cDNA microarray experi ments. Genome Biol 4:210. 6. de Ruyter, P. G., O. P. Kuipers, and W. M. de Vos. 1996. Controlled gene expression systems for Lactococcus lactis with the food-grade inducer nisin. Appl Environ Microbiol 62:3662-7. 7. Elkan, T. L. B. a. C. 1994. “Fitting a mixture model by expectation maximization to discover motifs in bio ploy mers”. Proceedings of the Second International Conference on Intelligent Systems for Molecular Biology:28-36. 8. Gasson, M. J. 1983. Plasmid complements of Streptococcus lactis NCDO 712 and other lactic streptococci after protoplast-induced curing. J Bacteriol 154:1-9. 9. GEO http://www.ncbi.nlm.nih.gov/geo/, posting date. Gene Expression Omnibus. [Online.] 10. Gribskov, T. L. B. a. M. 1998. “Combining evidence using p-values:application to sequence homology searches”. Bioinformatics 14:48-54. 11. Holmgren, A. 1985. Thioredoxin. Annu Rev Biochem 54:237-71. 12. Holo, H., and I. F. Nes. 1989. High-Frequency Transformation, by Electroporation, of Lactococcus lactis subsp. cremoris Grown with Glycine in Osmotically Stabilized Media. Appl Environ Microbiol 55:3119-3123 13. Jobin, M. P., D. Garmyn, C. Divies, and J. Guzzo. 1999. Expression of the Oenococcus oeni trxA gene is induced by hydrogen peroxide and heat shock. Microbiology 145 ( Pt 5):1245-51. 14. Josson, K., T. Scheirlinck, F. Michiels, C. Platteeuw, P. Stanssens, H. Joos, P. Dhaese, M. Za beau, and J. Mahillon. 1989. Characterization of a gram-positive broad-host-range plasmid isolated from Lactobacillus hilgardii. Plasmid 21:9-20. 15. Kleerebezem, M., M. M. Beerthuyzen, E. E. Vaughan, W. M. de Vos, and O. P. Kuipers. 1997. Controlled gene expression systems for lactic acid bacteria: transferable nisin-inducible expression cassettes for Lactococcus, Leuconostoc, and Lactobacillus spp. Appl Environ Microbiol 63:4581-4. 16. Kleerebezem, M., J. Boekhorst, R. van Kranenburg, D. Molenaar, O. P. Kuipers, R. Leer, R. Tarchini, S. A. Peters, H. M. Sandbrink, M. W. Fiers, W. Stiekema, R. M. Lankhorst, P. A. Bron, S. M. Hoffer, M. N. Groot, R. Kerkhoven, M. de Vries, B. Ursing, W. M. de Vos, and R. J. Siezen. 2003. Complete genome sequence of Lactobacillus plantarum WCFS1. Proc Natl Acad Sci U S A 100:1990-5. 17. Laurent, T. C., E. C. Moore, and P. Reichard. 1964. Enzymatic Synthesis of Deoxyribonucleotides. Iv. Isolation and Characterization of Thioredoxin, the Hydrogen Donor from Escherichia Coli B. J Biol Chem 239:3436-44. 18. Leichert, L. I., C. Scharf, and M. Hecker. 2003. Global characterization of disulfide stress in Bacillus subtilis. J Bacteriol 185:1967-75. 19. Li, Y., J. Hugenholtz, W. Sybesma, T. Abee, and D. Molenaar. 2005. Using Lactococcus lactis for glutathione overproduction. Appl Microbiol Biotechnol 67:83-90. 20. Mierau, I., and M. Kleerebezem. 2005. 10 years of the nisin-controlled gene expression system (NICE) in Lactococcus lactis. Appl Microbiol Biotechnol 68:705-17. 21. Pavan, S., P. Hols, J. Delcour, M. C. Geoffroy, C. Grangette, M. Kleerebezem, and A. Merce nier. 2000. Adaptation of the nisin-controlled expression system in Lactobacillus plantarum: a tool to study in vivo biological effects. Appl Environ Microbiol 66:4427-32.

56 2 Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1

22. Pieterse, B., R. H. Jellema, and M. J. van der Werf. 2006. Quenching of microbial samples for in creased reliability of microarray data. J Microbiol Methods 64:207-16. 23. Prieto-Alamo, M. J., J. Jurado, R. Gallardo-Madueno, F. Monje-Casas, A. Holmgren, and C. Pueyo. 2000. Transcriptional regulation of glutaredoxin and thioredoxin pathways and related enzymes in response to oxidative stress. J Biol Chem 275:13398-405. 24. Saal LH, T. C., Vallon-Christersson J, Gruvberger S, Borg A, Peterson C,. 2002. “BioArray Software Environment (BASE): a platform for comprehensive management and analysis of microarray data.” Genome Biol. 3:SOFTWARE0003. 25. Sambrook, J. F., E.F.; Maniatis T. 1989. Molecular Cloning: A Laboratory manual. , 2nd edition ed, vol. 1-3. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. 26. Scharf, C., S. Riethdorf, H. Ernst, S. Engelmann, U. Volker, and M. Hecker. 1998. Thioredoxin is an essential protein induced by multiple stresses in Bacillus subtilis. J Bacteriol 180:1869-77. 27. Seo, D., K. Kamino, K. Inoue, and H. Sakurai. 2004. Purification and characterization of ferredoxin- NADP+ reductase encoded by Bacillus subtilis yumC. Arch Microbiol 182:80-9. 28. Serrano, L. M., D. Molenaar, M. Wels, B. Teusink, P. A. Bron, W. M. de Vos, and E. J. Smid. 2007. Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1. Microb Cell Fact 6:29. 29. Stoyanovsky, D. A., Y. Y. Tyurina, V. A. Tyurin, D. Anand, D. N. Mandavia, D. Gius, J. Ivanova, B. Pitt, T. R. Billiar, and V. E. Kagan. 2005. Thioredoxin and lipoic acid catalyze the denitrosation of low molecular weight and protein S-nitrosothiols. J Am Chem Soc 127:15815-23. 30. Tan, P. S., I. J. van Alen-Boerrigter, B. Poolman, R. J. Siezen, W. M. de Vos, and W. N. Kon ings. 1992.Characterization of the Lactococcus lactis pepN gene encoding an aminopeptidase homologous to mamma lian aminopeptidase N. FEBS Lett 306:9-16. 31. Teusink, B., F. H. van Enckevort, C. Francke, A. Wiersma, A. Wegkamp, E. J. Smid, and R. J. Siezen. 2005. In silico reconstruction of the metabolic pathways of Lactobacillus plantarum: comparing predictions of nutrient requirements with those from growth experiments. Appl Environ Microbiol 71:7253-62. 32. van Alen-Boerrigter, I. J., R. Baankreis, and W. M. de Vos. 1991. Characterization and overexpres sion of the Lactococcus lactis pepN gene and localization of its product, aminopeptidase N. Appl Environ Microbiol 57:2555-61. 33. van Kranenburg, R., J. D. Marugg, S. van, II, N. J. Willem, and W. M. de Vos. 1997. Molecular characterization of the plasmid-encoded eps gene cluster essential for exopolysaccharide biosynthesis in Lactococcus lactis. Mol Microbiol 24:387-97. 34. van Niel, E. W., K. Hofvendahl, and B. Hahn-Hagerdal. 2002. Formation and conversion of oxygen metabolites by Lactococcus lactis subsp. lactis ATCC 19435 under different growth conditions. Appl Environ Microbiol 68:4350-6. 35. Vido, K., H. Diemer, A. Van Dorsselaer, E. Leize, V. Juillard, A. Gruss, and P. Gaudu. 2005. Roles of thioredoxin reductase during the aerobic life of Lactococcus lactis. J Bacteriol 187:601-10. 36. Vignols, F., C. Brehelin, Y. Surdin-Kerjan, D. Thomas, and Y. Meyer. 2005. A yeast two-hy brid knockout strain to explore thioredoxin-interacting proteins in vivo. Proc Natl Acad Sci U S A 102:16729-34. 37. Wegkamp, A., W. van Oorschot, W. M. de Vos, and E. J. Smid. 2007. Characterization of the role of para-aminobenzoic acid biosynthesis in folate production by Lactococcus lactis. Appl Environ Microbiol 73:2673-81. 38. Winterling, K. W., A. S. Levine, R. E. Yasbin, and R. Woodgate. 1997. Characterization of DinR, the Bacillus subtilis SOS repressor. J Bacteriol 179:1698-703.

57 58 3 Global transcriptional analysis reveals the specific role of thioredoxin reductase in oxidative stress response in Lactobacillus plantarum WCFS1

L. Mariela Serrano, Douwe Molenaar, Bas Teusink, Willem M. de Vos, Eddy J. Smid

Manuscript in preparation

59 60 3 Global transcriptional analysis reveals the specific role of thioredoxin reductase in oxidative stress response in Lactobacillus plantarum WCFS1

ABSTRACT

We have studied the role of thioredoxin reductase (trxB1) in the oxidative stress response of Lactobacillus plantarum by comparing the global transcriptional profile of the wild-type strain (L. plantarum WCFS1) with that of a trxB1::cam strain (L. plantarum NZ7608). Under aerobic conditions the specific growth rate of strain NZ7608 was slightly lower than that of the wild-type (0.47 ± 0.02 and 0.58 ± 0.01 h-1, respectively). We investigated this reproducible and significant difference in growth rate using enzyme assays, survival rates and global transcriptional analysis. Strain NZ7608 had a 2.5-fold decreased thioredoxin reductase activity compared to the wild-type. Furthermore, the observed growth rate reduction in strain NZ7608 was correlated to purine metabolism and sugar transport systems, which were affected at the level of transcription in strain NZ7608 compared to the wild-type. Moreover, even though strain NZ7608 had not completely lost the capability to respond towards hydrogen peroxide, the global transcriptional response of the strain NZ7608 under standard conditions showed a significant down regulation of 93 genes when compared to the wild-type that were directly or indirectly involved in the response of L. plantarum towards hydrogen peroxide stress. Finally, the transcriptome data also indicated that the response towards hydrogen peroxide in strain NZ7608 differed in the expression of 90 genes compared to the wild-type. Our results show that a fully functional thioredoxin system is important for optimal growth under aerobic conditions as well as for an adequate response of L. plantarum towards hydrogen peroxide stress. This latter property is of interest when engineering robustness towards oxidative stress in industrial strains of L. plantarum.

6161 INTRODUCTION dative stress conditions. Our results indicate that Thioredoxins (TRXs) are small 12 kDa sulfhydryl the trxB1 mutant has a reproducible growth im- electron donors which are activated (reduced) pairment compared to the wild-type when grown via thioredoxin reductase (TR) in a NADPH-de- aerobically. In addition, the trxB1 deletion strain pendent reaction otherwise known as the thio- did not affected the ability of L. plantarum to redoxin system. Zeller et al. (23) defines TRXs respond towards hydrogen peroxide. Neverthe- in bacteria as proteins with universal functions less, we have determined that the response of ranging from redox-homeostasis in the cyto- strain NZ7608 towards hydrogen peroxide is in- plasm to unique stress response functions which deed different from the wild-type response. Our depends upon the bacterium. Research on the data has revealed new leads into understanding thioredoxin system in Gram-positive bacteria oxidative stress response in this bacterium. such as Bacillus subtilis (13), Oenococcus oeni (5), and Lactococcus lactis (21) has shown a role for the thioredoxin system in oxidative stress re- MATERIALS AND METHODS sponse. The studies carried out on B. subtilis and Bacterial strains, plasmids, media, and O. oeni, reported induction of thioredoxin (TRX) culture conditions. The bacterial strains, plas- encoded by trxA under hydrogen peroxide stress. mids, and oligonucleotides used for this study are On the other hand, studies in Lc. lactis describe summarized in Table 1. Escherichia coli strains the induction of oxidative stress associated pro- were grown at 37°C in Tripton and yeast extract teins suggesting a role of TR encoded by trxB1 in (TY) (12). L. plantarum WCFS1 was grown at the oxidative stress response (21). Recently, we 37°C either in Man-Rogosa-Sharp broth (MRS) have shown that a trxB1-over expressing strain or in Chemically Defined Medium (CDM) (19). of Lactobacillus plantarum exhibits improved resistance against hydrogen peroxide and diamide DNA Manipulations. All molecular biology stress (14). This effect was attributed to the spe- techniques were performed following estab- cific up-regulation of 16 genes predicted to en- lished protocols by Sambrook et al. (12). DNA code proteins involved in the purine metabolism fragments were amplified using PCR. Digestion (pur genes), energy metabolism (gapB) as well of DNA fragments was done following the con- as in stress-response (groEL, npr2), and man- ditions recommended by the commercial sup- ganese transport (mntH2). The present study was pliers of the restriction enzymes (Boehringer, performed to explore the role of TR in L. planta- Breda, The Netherlands). Both DNA fragments rum under oxidative stress and to investigate the and used plasmids were eluted from 0.7% (wt/ association with the previously identified genes vol) agarose gels using the Purification Kits from (14). In order to do this, we constructed strain (Promega, Leiden, The Netherlands). Isolation L. plantarum NZ7608, a trxB1-deficient strain, of DNA was performed as explained elsewhere and analyzed the phenotype and transcriptome (14). response of this strain under standard and oxi-

62 3 Global transcriptional analysis reveals the specific role of thioredoxin reductase in oxidative stress response in Lactobacillus plantarum WCFS1

TABLE 1. Strains, plasmids, and oligonucleotides used in this study.

Strain Characteristics Reference

DH5Į E. coli

WCFS1 L. plantarum WCFS1. (7)

NZ7608 L. plantarum WCFS1 trxB1::cam derivative with chromosomal This work

integration of pMS019.

Plasmids

pNZ5318 CmR EmR; pNZ5317 derivative for multiple gene replacements (9)

containing las and pepN terminators

pMS019 CmR; pNZ518 derivative carrying 1.0 kb DNA fragments of L. This work

plantarum WCFS1 lp_0762 and lp_0763.

Oligonucleotides 5’-3’

trxB1-forwA GCCTGAGCTCTGAAGTAACGTTATTCCAAT

trxB1-revA AGGAGGATCCCACAACTGAAGTCAATTATAGC

trxB1-forwB TCCTGGATCCACACTTATATCACTGCTTTAGG

trxB1-revB AGGAGTCGACTCACAAGTGACACAAATTATGC

Construction of strain NZ7608, a trxB1 restriction sites. Chloramphenicol resistant (CmR) deficient L. plantarum. To construct a trxB1:: colonies were selected and checked by PCR. The cam replacement variant in L. plantarum WCFS1, resulting plasmid, containing 1 kb upstream plasmid pMS019 was constructed. This plasmid and downstream regions of lp_0761 separat- allowed the direct selection of the desired mu- ed by the cam gene was designated pMS019. tant by a double cross-over event. The upstream After purification and verification by sequence and downstream flanking regions of the gene analysis, plasmid pMS019 was integrated into L. trxB1 (lp_0761) were amplified by PCR using ge- plantarum WCFS1 (6) and plated in agar plates nomic DNA of L. plantarum WCFS1 as template containing MRS agar with 1 mM dithiothreitol and two primer combinations: trxB1_forwA with (DTT). Plates were incubated in anaerobic jars trxB1_revA and trxB1_forwB with trxB1_revB (Ta- overnight at 30ºC. Colonies displaying CmR and ble 1) The resulting amplified regions (approx. erythromycin sensitivity (Ems) were expected to 1 kB) were cloned into competent E. coli DH5 be the result of a one-step double-cross-over a- cells via a two-step ligation into the vector event on both flanking regions of thetrxB1 gene pNZ5318 using the Pme1 and Srf1 blunt ended and should possess the desired trxB1::cam re-

63 placement mutation. The resulting trxB1::cam 0.5% (wt/vol) glucose or 10 ml CDM containing variant of L. plantarum WCFS1 was designated 0.5% (wt/vol) glucose without guanine, adenine, NZ7608 and checked by restriction and South- uracil, xanthine, inosine, thymidine, and oro- ern blot analysis (data not shown). tate (GAUXITO-mix). Overnight cultures were Thioredoxin reductase enzyme assay. This washed twice in CDM with or without GAUXI- assay was performed as described previously by TO-mix respectively by 5 min centrifugation at Serrano et al. (14). In this study one unit (U) is 20.000 x g at room temperature. Washed cells defined as 1 nmol of TNB produced per min per were resuspended in 10ml CDM or 10 ml CDM milligram cell protein. without GAUXITO-mix containing 0.5% (wt/vol) Zone inhibition assays. These assays were glucose and incubated at 37ºC either under performed using L. plantarum strains WCFS1 aerobic (200 ml Erlenmeyer flask stirred at 120 and NZ7608 as described elsewhere (14). For rpm) or anaerobic (10 ml test tubes in anaerobe these assays potassium tellurite, diamide, hy- jars) conditions. drogen peroxide, and pyrogallol were used in Batch cultivations. Full grown cultures were concentrations ranges from 0.05 M to 1 M. diluted 1/50 in 500 ml CDM containing 0.5% Growth experiments. In all growth experi- (wt/vol) glucose. During growth pH was kept ments cells were pre-cultured overnight at 37ºC constant at pH=5.5, by the addition of 5M in 10 ml CDM containing 0.5% (wt/vol) glucose NaOH. Temperature was kept constant at 37ºC. under anaerobic conditions. Cell density was Filtered air was bubbled through the culture at measured by detecting the turbidity of the cul- a rate of 150 ml · min-1 while the culture was tures at 600 nm every 30 min. Specific growth stirred at a constant speed of 125 rotations per rate (μmax) was defined as ΔLnOD600 / Δtime at minute (rpm). the linear or exponential growth of each cul- Continuous cultivations. Three independent ture. chemostat cultivations were performed for each Growth on different carbon sources. Over- strain studied. A 1 Liter bioreactor (Applikon de- night cultures (10 ml) were washed twice in pendable instruments) was inoculated with cells CDM without C-source by 5 min centrifugation from an overnight culture of the strain in CDM at 20.000 x g at room temperature. The washed at 37°C to an initial OD600 of 0.1 in 500 ml me- cells were then resuspended in CDM (10 ml) dium (10 ml overnight culture). Cultures were containing 0.5% (wt/vol) of the sugar of interest grown at 37°C in 500 ml CDM (19) medium and incubated at 37ºC. Stock solutions of 20 supplemented with 2% (wt/vol) glucose. The pH % (wt/vol) of the following carbohydrates were was maintained at pH 5.5 by the automatic ad- prepared: glucose, sucrose, fructose, mannose, dition of 5 M NaOH and the stirrer speed was maltose, mannitol, sorbitol, trehalose, and cel- set at 200 rpm. The headspace of the fermen- lobiose. tors and medium vessel were full at all times with Growth with and without nucleosides nitrogen gas supplied at a flow rate of 520 ml · and nucleosbases. Cells were pre-cultured min-1. The cultures were kept at the dilution rate at 37ºC overnight in 10 ml CDM containing of 0.1 h-1 and steady state was assumed after 5

64 3 Global transcriptional analysis reveals the specific role of thioredoxin reductase in oxidative stress response in Lactobacillus plantarum WCFS1

volume changes. The continuous cultivation was omnibus database (GEO) with accession num- monitored by measuring the dry weight at dif- ber GSE8672 (2). Per array two cDNA labeled ferent intervals. At steady state or (t0), samples targets were hybridized on custom designed L. were taken for transcriptome, HPLC, and dry plantarum WCFS1 11K Agilent oligo microar- weight analysis. In addition, at steady state, 50 rays available at GEO with accession number ml of fresh hydrogen peroxide was added to the GPL4318 (2) using the Agilent 60-mer oligo fermentors to a final concentration of 3.5mM in microarray processing protocol version 4.1. The the fermentor. After 30 min (t30) following the hy- oligo-microarray contained an average of 3 drogen peroxide pulse, samples were taken for probes per transcript. Dried slides were scanned transcriptome, HPCL, and dry weight analysis. in the Scan Array Express (PerkinElmer Life Sci- HPLC analysis of fermentation products. ences; Packard Bioscience) for both Cy3 and Glucose, lactate, acetate, formate, ethanol, Cy5 at 10 microns and a PMT Gain between 40- acetoin, succinate, formic acid, and citrate were 50%. Spot intensity data were normalized and analyzed by HPLC, as described elsewhere quantified (average intensity) using a custom- (16). ized grid in ImaGene (BioDiscovery, Inc.). Signal Determination of culture dry weight. Dry intensities of all probes were corrected against weights of culture samples (10.0 ml) were deter- background in BioArray Software Environment mined using nitrocellulose filters (pore size 0.45 (BASE) (11). The ratio of each probe (M) was de-

μm HV; Durapore membrane filters, Millipore). fined as 2log of the (cy5/cy3) intensities where After removal of the medium by filtration, the fil- cy3 is the wild-type intensity. Fold change (FC) is ters were washed with demineralized water and defined as 2M. For the statistical analysis we used dried overnight at 55 ºC in a oven (Sharp Inc., Linear Models for Microarray Analysis (Limma) Japan). Triplicate samples varied by less than (R/limma) (15). In this test three variables were 1%. used: mutation, treatment, and interaction. The Microarrays. RNA isolation, cDNA synthesis model used takes into consideration the inter- and purification were done as described else- action between genotype and treatment. Signifi- where (14). For each studied culture (wild-type cant genes were defined as genes with a False and trxB1 mutant), independent chemostat culti- Discovery Rate (FDR) adjusted for the pvalues of vations were performed and used as a single ex- less than 1% and a fold change (FC) of higher or perimental design. For this experiment a hybrid- equal to 1.5. The statistical test resulted in three ization scheme was designed which consisted of sets of data. One dataset representing the genes six hybridizations: 1) wild-type versus NZ7608, affected as a result of the mutation of trxB1 (mu- 2) NZ7608 versus wild-type, 3) wild-type versus tation), another set representing the genes af- wild-type + peroxide, 4) wild-type + peroxide fected due to hydrogen peroxide stress and a versus NZ7608 + peroxide 5) NZ7608 + per- third representing the interaction or difference in oxide versus wild-type + peroxide and 6) wild- response towards peroxide between mutant and type + peroxide versus wild-type. The normal- wild-type (interaction). ized data can be found at the gene expression

65 RESULTS lower TR specific activity than the wild-type strain Construction of a trxB1 mutant (120 U and 300 U, respectively). Growth rates In previous work, the trxB1 gene, was found of strain NZ7608 in CDM containing 5mM di- to be involved in oxidative stress response of amide or 3.5mM peroxide did not differ from L. plantarum (14). In order to characterize the wild-type growth rates (data not shown). More- role of TR in L. plantarum in the oxidative stress over, both strain NZ7608 and the wild-type were response, a strain deficient in the gene trxB1 inhibited to the same extent by different oxida- was constructed. The trxB1::cam replacement tive agents as observed by zone inhibition assays was only obtained when the transformants of L. (data not shown). These analyses did not deliver plantarum were plated on agar containing a re- a clear phenotype of strain NZ7608. In order ducing agent (DTT) under strict anaerobic condi- to characterize strain NZ7608 further the global tions as described previously for Lc. lactis (21). transcriptional response of strain NZ7608 was The strain carrying the trxB1::cam replacement investigated. was designated L. plantarum NZ7608. Chemostat cultivation of L. plantarum Phenotypic analysis of the trxB1 mu- strains WCFS1 and NZ7608. tant For a detailed characterization of the transcrip- The TR activity and growth of strain NZ78608 tional response of the TR deficient L. plantarum was compared to that of the wild-type. This re- NZ7608, we used chemostat cultures. This setup vealed that strain NZ7608 showed a 2.5-fold allowed us to analyze gene-expression during

TABLE 2. Growth yield and lactate production in L. plantarum strains NZ7608 and wild-type.

Physiological Parameters a b c d L. plantarum strains Ysx Ysg Ysl qlact WCFS1 2.30 ± 0.2 0.10 ± 0.1 0.11 ± 0.1 7.20 ± 0.2 NZ7608 2.30 ± 0.1 0.10 ± 0.1 0.10 ± 0.1 7.40 ± 0.8 WCFS1 + peroxidee 1.90 ± 0.2 0.07 ± 0.1 0.15 ± 0.1 n.d.f e f NZ7608 + peroxide 1.90 ± 0.2 0.08 ± 0.1 0.11 ± 0.1 n.d. a Yield of biomass (g/liter) b Yield of biomass (g/g of glucose consumed). c Yield of biomass (g/g of lactated produced). d mmol of glucose consumed/g biomass/h e Culture exposed to 3.5mM hydrogen peroxide stress for 30 min. f Not determined.

66 3 Global transcriptional analysis reveals the specific role of thioredoxin reductase in oxidative stress response in Lactobacillus plantarum WCFS1

steady state under well defined anaerobic con- in anaerobic chemostat cultures at low dilution ditions and during exposure to sub-lethal hydro- rate is not affected. Hence, showing the chosen gen peroxide. All physiological parameters de- platform was indeed suitable for a global tran- termined from the chemostat cultivations were scriptional analysis of strains NZ7608 and wild- equal in both strains (Table 2). Strain WCFS1 type. (wild-type) and strain NZ7608 both consumed all glucose (96.5mM ± 5.00) and citrate (2.55mM ± 0.00) present in the media while producing Transcriptome profiling of chemostat the same amount of lactic acid. Exposure of both cultures of wild-type and trxB1-mutant. wild-type and strain NZ7608 to a hydrogen per- To analyze the effect of the trxB1 mutation both oxide stress for 30 minutes caused a reduction in the presence and absence of hydrogen perox- of the biomass of 17% in both strains. In addi- ide stress, total RNA was isolated from cultures of tion, the concentration of all end-products re- both wild-type and strain NZ7608 at steady state mained unchanged after exposure to hydrogen and 30 min after applying the hydrogen perox- peroxide (Table 3). The carbon recovery based ide pulse. Isolated RNA was used for performing on analysis of the fermentation products in the a global transcriptional analysis for which the cultures was 90% for both L. plantarum WCFS1 following variables were tested: mutation effect, and NZ7608; the remaining 10% of carbon is hydrogen peroxide effect, and interaction effect. attributed to the produced biomass. These ob- This analysis resulted in independent datasets of servations show that growth of the TR-mutant significantly affected genes with adjustedvalues p

TABLE 3. Extracellular metabolite concentrations (mM) in chemostat cultivations of strain L. plantarum WCFS1 and NZ7608.

Metabolite concentration (mM) a,b a,b Metabolite WCFS1 NZ7608 WCFS1+ peroxide NZ7608+ peroxide Glucose 0.00 ± 0.0 0.00 ± 0.0 0.00 ± 0.0 0.00 ± 0.0 Citric Acid 0.00 ± 0.0 0.11 ± 0.0 0.11± 0.0 0.11± 0.0 Pyruvate 1.70 ± 0.3 1.63 ± 0.3 2.22 ± 0.2 2.21 ± 0.4 Succinate 1.42 ± 0.1 1.50 ± 0.1 1.41 ± 0.1 1.53 ± 0.2 Lactic Acid 161.30 ± 10.3 170.60 ± 4.8 156.90 ± 7.8 169.80 ± 7.2 Formic Acid 4.71 ± 0.6 3.51 ± 0.8 4.32 ± 0.5 3.52 ± 0.8 Acetic Acid 16.71 ± 3.2 16.42 ± 2.8 16.41 ± 2.9 16.51 ± 2.3 Acetoine 0.62 ± 0.5 0.50 ± 0.5 0.72 ± 0.2 0.31 ± 0.5 Ethanol n.d.c n.d.c n.d.c n.d.c Butanodiol 0.11 ± 0.1 0.11± 0.2 0.11 ± 0.0 0.10 ± 0.2 a Cultures exposed to 3.5mM hydrogen peroxide for 30 min. b Values corrected by 20% for sample extraction. c Metabolite not detected by HPLC.

67 < 1% and FC ≥ 1.5 supplied as supplementary mutation-only genes. Most of these genes have material (Tables S1, S2, S3, and S4 see page not been functionally characterized. A functional 165). Disruption of trxB1 in L. plantarum WCFS1 thioredoxin system and therefore adequate in- caused significant differential expression in 106 tracellular control of the redox state is thought genes when compared to the wild-type strain to be relevant for the level of expression of these (Fig. 3 and Table S1 in the supplementary ma- genes. terial, see page 166). From these genes, 93 represent genes which are predicted to encode Effect of trxB1 disruption on aerobic proteins involved in energy metabolism, DNA growth performance. metabolism, purine metabolism, cell envelope, The global transcriptional profile of strains and molybdopterin biosynthesis. The remaining WCFS1 and NZ7608 under anaerobic condi- 13 genes included genes which are predicted to tions, suggested that strain NZ7608 was im- encode proteins involved in transport, cell enve- paired in DNA repair mechanisms and DNA/ lope and hypotheticals (Fig. 3 and Table S1 in RNA biosynthesis compared to wild-type (Table the supplementary material, see page 166). This 4). Therefore in order to validate the in-silico re- set of genes has been nominated in this study as sults we grew L. plantarum NZ7608 aerobically

TABLE 4. Genes encoding proteins involved in DNA repair and purine biosynthesis found significant affected (pvalue< 1% and FC ≥ 1.5) in strain NZ7608 compared to wild-type. Data extracted from Supplementary material Table S1, see page166.

Locus Gene M1 FC2 Product lp_2727 purC -0.67 0.63 phosphoribosylaminoimidazole-succinocarboxamide synthase lp_2729 purE -0.71 0.61 phosphoribosylaminoimidazole carboxylase, catalytic subunit lp_2723 purF -0.69 0.62 amidophosphoribosyltransferase precursor lp_2728 purK1 -0.61 0.65 phosphoribosylaminoimidazole carboxylase, ATPase subunit lp_2724 purL -0.73 0.60 phosphoribosylformylglycinamidine synthase II lp_2722 purM -0.72 0.61 phosphoribosylformylglycinamidine cyclo-ligase lp_2725 purQ -0.68 0.63 phosphoribosylformylglycinamidine synthase I lp_2726 purS -0.66 0.63 conserved purine biosynthesis cluster protein lp_3023 umuC -0.96 0.52 UV-damage repair protein lp_2280 dinP -0.77 0.59 DNA-damage-inducible protein P lp_2301 recA -0.61 0.66 recombinase A lp_2063 lexA -0.76 0.59 transcription repressor of the SOS regulon lp_1836 msrA3 -0.73 0.60 protein-methionine-S-oxide reductase

1 M is defined as the2 log (cy5/cy3) ratio. 2 FC is defined as Fold Change or M2

68 3 Global transcriptional analysis reveals the specific role of thioredoxin reductase in oxidative stress response in Lactobacillus plantarum WCFS1

in a pH-controlled batch culture and compared impairment in strain NZ7608 under aerobic its growth rate to that of the wild-type strain. The conditions could be a consequence of the estab- specific growth rates in aerated cultures of wild- lished down regulation of purine biosynthesis. type and strain NZ7608 were 0.58 ± 0.01 h-1 and 0.47 ± 0.02 h-1, respectively (19 % growth It is possible that the reduced growth rate of reduction, see Fig. 1). strain NZ7608 is linked with down-regulation of the biosynthesis pathways of purine nucleo- Furthermore, biomass production was 0.77 ± tides. Therefore, the growth effect of this trxB1 0.03 and 0.70 ± 0.01 grams dry weight / liter, mutant strain in the presence and absence of respectively, for wild-type and strain NZ7608. As nucleosides and nucleobases (GAUXITO-mix: mentioned before, at the transcriptome level pu- guanine, adenine, uracil, xanthine, inosine, rine biosynthesis was affected in strain NZ7608 thymidine, and orotate) (Table 5) was investi- compared with wild-type. The purC-purS genes gated. An aerobic culture of strain NZ7608 in a involved de novo purine biosynthesis were two- medium lacking the GAUXITO-mix shows 24% fold down regulated when compared to the growth impairment compared the wild-type. On wild-type (Fig. 2). Hence, the observed growth the other hand when strain NZ7608 was culti-

FIGURE 1. Exponential growth of L. plantarum strains WCFS1 (squares) and NZ7608 (triangles) on CDM containing 0.5% glucose at 37ºC. Strains were grown in controlled batch fermentors with the supply of air. Cell density was measured by detecting the turbidity of the cultures at 600 nm every 30 min.

1.5

1.0

0.5

0.0

-0.5

lnOD600 -1.0

-1.5

-2.0

-2.5

-3.0 0 1 2 3 4 5 6 7 8 Time (hours) 69 FIGURE 2. Representation of the biosynthesis of purines in L. plantarum WCFS1 in TM Simpheny obtained by projecting significantly affected genes (pvalue < 1% and FC ≥ 1.5) on the map of the metabolic network. Panel A represents the transcriptional response of the purine biosynthesis pathway in L. plantarum strain NZ7608 compared to wild-type. Panel B shows the transcriptional response of the purine biosynthesis in L. plantarum strains WCFS1 and NZ7608 upon a hydrogen peroxide treatment. Red colored genes (triangles) and reactions (arrows) represent an up-regulation, while green colored genes and reactions represent a down-regulation in the studied condition. (a color representation of this figure can be found in page 152)

70 3 Global transcriptional analysis reveals the specific role of thioredoxin reductase in oxidative stress response in Lactobacillus plantarum WCFS1

vated anaerobically without GAUXITO-mix only Sugar utilization 7% growth impairment in strain NZ7608 com- The global transcriptional analysis indicated that pared to wild-type was determined. These data the genes predicted to encode proteins involved support the notion that the impact on aerobic in a number of PTS-sugar transport systems growth of an impaired thioredoxin system can, were up-regulated in strain NZ7608 compared at least partially, be linked with down regulation to wild-type (Table 6). of the pathways for synthesis of DNA/RNA building blocks.

TABLE 5. Specific growth rate (μmax inh-1) of L. plantarum strains WCFS1 and NZ7608 under aerobe and anaerobe conditions in shake flasks containing CDM supplied with 0.5% glucose with or without nucleosides (GAUXITO-mix).

ȝmax (h-1)1 A2 B3 wt NZ7608 % wt NZ7608 % Aerobe 0.63 0.52 17 0.49 0.37 24 Anaerobe 0.61 0.55 10 0.41 0.38 7

TABLE 6. Significantly affected genes (pvalue< 1% and FC ≥ 1.5) encoding sugar transporters in strain NZ7608 compared to wild-type. Data extracted from Table S1 in the supplementary material, see page 166.

Transcriptome data Substrate Annotation 1 locus gene FC lp_1399 pts15B 0.64 beta-glucosides beta-glucosides PTS, EIIB lp_0175 malE 1.53 maltose maltose/maltodextrin ABC transporter lp_2969 pts22CBA 1.63 N-acetylglucosamine N-acetylglucosamine PTS, EIICBA lp_2780 pts20A 1.78 cellobiose cellobiose PTS, EIIA lp_0577 pts9D 1.86 mannose PTS, EIID lp_0576 pts9C 1.94 mannose PTS, EIIC mannose and glucose lp_0575 pts9AB 2.00 mannose PTS, EIIAB lp_0587 pts10B 1.65 mannose PTS, EIIB lp_1274 ptsI 1.65 unknown phosphoenolpyruvate-protein phosphotransferase lp_1273 ptsH 1.73 unknown phosphocarrier protein Hpr lp_1075 1.50 unknown ABC transporter, permease protein lp_2740 0.66 unknown ABC transporter, permease protein lp_0215 1.61 unknown ABC transporter, ATP-binding protein

1 M Fold change expressed as 2 , where M is defined as the 2log (cy5/cy3) ratio.

71 Specifically genes predicted to encode for trans- than wild-type. The growth disadvantage ob- porters of cellobiose (pts20A), maltose (malE) served for strain NZ7608 compared to wild-type and mannose/glucose (pts9ABCD) were found ranged from 10% to 40% for the different tested up-regulated in strain NZ7608 compared to carbohydrates. The difference in growth rate of wild-type. This suggests an association between strain NZ7608 compared to wild-type was 10% the function of TR and the regulatory network on trehalose and glucose; 17% on maltose, 25% for expression of sugar transporters. To link the on mannose and mannitol and even 40% when transcriptional response of the trxB1 mutation grown on fructose. This result confirms that an to an effect on sugar utilization in the mutant, impaired thioredoxin system affects the activity L. plantarum WCFS1 and strain NZ7608 were of the sugar utilization pathways in L. plantarum. cultivated on CDM supplemented with nine dif- However, it also shows that the physiological ferent sugars as the sole carbohydrate source response of a particular mutant can not always (Table 7). From this experiment it was observed be predicted straightforwardly from transcription that strain NZ7608 displayed two different re- data profiling. sponses when compared to the wild-type. It either grew at a similar rate (only 5% difference) compared to the wild-type (on sucrose and cellobiose), or it grew substantially slower

-1 TABLE 7. Anaerobic growth rate (μmax in h ) of L. plantarum strains WCFS1 and NZ7608 in CDM supplied with nine different carbon sources, each present at a concentration of 0.5% (wt/vol).

C-source µmax (h-1)1 2 Name Characteristic WCFS1 NZ7608 Difference (%) Sucrose disaccharide 0.56 0.53 5 Cellobiose disaccharide 0.47 0.44 6 Glucose monosaccharide 0.65 0.60 8 Trehalose disaccharide 0.53 0.47 11 Maltose disaccharide 0.48 0.40 17 Mannitol monosaccharide 0.44 0.33 25 Mannose monosaccharide 0.63 0.47 25 Fructose monosaccharide 0.27 0.16 41

1 Specific growth rate determined as ΔLnOD600/ Δtime 2 Growth of strain NZ7608 compared to wild-type and expressed as %.

72 3 Global transcriptional analysis reveals the specific role of thioredoxin reductase in oxidative stress response in Lactobacillus plantarum WCFS1

FIGURE 3. Summary of global transcriptional analysis performed in this study. Total number of significant affected genes (pvalue < 1% and FC ≥ 1.5) as a result of a trxB1 mutation compared to wild-type or as result of hydrogen peroxide stress on both strain NZ7608 and wild-type. 605b

13c 93d 512

106a a Mutation: Significant affected genes in strain NZ7608 compared to wild-type under anaerobic growth. b Treatment: Significant affected genes in strain NZ7608 compared to wild-type as a result of hydrogen peroxide stress c Mutation only: Significant affected genes in strain NZ7608 compared to wild-type. These genes were not found significantly af fected after hydrogen peroxide stress. d Crossover: Significant affected genes as a result of a trxB1 mutation as well as hydrogen peroxide stress.

Effect of exposure to sub-lethal dose of encode proteins involved in the stress response hydrogen peroxide mechanisms (Table 8). Upon hydrogen peroxide As was previously mentioned, comparison of stress in both studied strains, an up-regulation the global transcriptional response of the wild- of genes predicted to encode proteins involved type strain and the trxB1 disruption mutant upon in the SOS response mechanisms and stress pro- exposure to a sub-lethal dose of hydrogen per- teins was observed. Furthermore, genes predict- oxide was performed. A hydrogen peroxide ed to encode proteins involved in the thioredoxin pulse in cultures of both L. plantarum WCFS1 system (trxB1, trxB2, trxA2, trxA3), glutaredoxin and NZ7608 resulted in significant differential system (gpo, gshA1, grpE) as well as metabolic expression of a total of 605 genes (Fig. 3 and enzymes associated with oxidative stress (fum, Table S2 in the supplementary material, see mrsA3, gapB, pox3) were found affected. This page 169). shows that both wild-type strain and the TR impaired strain NZ7608 can respond to hydrogen This list was divided into two smaller datasets peroxide exposure, cope with oxidative stress for further analysis 1) 512 genes which were and most likely re-establish intracellular redox termed hydrogen peroxide-specific genes and homeostasis (Table 8). Stress response genes af-

2) 93 genes which were also found significantly fected (pvalue< 1% and FC ≥ 1.5) in both strains affected due to introduced mutation. The list of NZ7608 and wild-type as a result of hydrogen 512 genes encoded genes involved genes in peroxide stress. Data extracted from Supple- all functional categories. Hydrogen peroxide mentary material Table S2, see page 169. affected a large number of genes predicted to 73 Product A A A r TP-dependent nuclease, subunit TP-dependent nuclease, subunit B TP-dependent Clp protease, ATP-binding subunit ClpL A A A DNA-damage-inducible protein P holliday junction DNA helicase RuvB holliday junction DNA helicase Ruv regulatory protein Spx DNA-entry nuclease NADH dehydrogenase small heat shock protein catalase fumarate hydratase glutathione peroxidase glutamate--cysteine ligase nucleotide-disulphide oxidoreductase transcription repressor of the SOS regulon excinuclease ABC, subunit B excinuclease ABC, subunit protein-methionine-S-oxide reductase DNA-directed DNA polymerase III, alpha chain UV-damage repair protein heat shock protein GrpE thioredoxin thioredoxin recombinase A sigma54 activato fumarate reductase, flavoprotein subunit precursor flavodoxin fumarate reductase, flavoprotein subunit precursor fumarate reductase, flavoprotein subunit precursor,N-term truncated fumarate reductase, flavoprotein subunit precursor,N-term truncated DNA helicase (putative) DNA helicase (putative) DNA helicase (putative) < 1% and FC ≥ 1.5) in both strains NZ7608 and wild-type as a result result a as wild-type and NZ7608 strains both in 1.5) ≥ FC and 1% < FC2 5.09 1.88 1.84 0.45 3.65 2.69 2.68 4.42 0.22 0.34 3.34 0.48 1.62 1.50 0.42 3.55 1.78 6.85 0.41 1.53 2.32 1.65 3.37 2.60 3.33 0.60 0.52 0.60 0.36 0.46 0.46 1.53 2.00 1.60 value M1 2.35 0.91 0.88 1.74 0.69 0.59 1.87 1.43 1.42 2.15 1.83 0.83 2.78 0.61 1.21 0.72 1.75 1.38 1.74 0.61 1.00 0.68 -1.16 -2.19 -1.54 -1.06 -1.26 -1.29 -0.73 -0.94 -0.75 -1.47 -1.12 -1.12 3 2 2 C 2 2 E E kat fum gpo Data extracted from Supplementary material Table S2, see page 169. clpL lexA dinP ruvB ruvA uvrB recA rexA rexB grp spx4 hsp ndh Gene trxA endA dna trxA3 trxB uvrA1 umu gshA1 msrA Locus lp_2286 lp_2287 lp_0585 lp_2063 lp_0910 lp_0308 lp_0432 lp_0772 lp_0773 lp_2906 lp_3023 lp_1069 lp_2028 lp_3578 lp_1425 lp_1477 lp_0952 lp_1112 lp_1113 lp_3125 lp_0220 lp_2324 lp_2270 lp_1836 lp_1899 lp_2280 lp_2301 lp_3345 lp_2693 lp_2694 lp_3583 lp_2668 lp_3437 lp_2585 Stress response TABLE 8. Stress response genes affected (p affected genes response Stress 8. TABLE stress. peroxide of hydrogen

74 3 Global transcriptional analysis reveals the specific role of thioredoxin reductase in oxidative stress response in Lactobacillus plantarum WCFS1 bifunctional protein: phosphoribosylaminoimidazolecarboxamide formyltransferase; IMP cyclohydrolase anaerobic ribonucleotide reductase activator protein nucleoside-diphosphate kinase anaerobic ribonucleoside-triphosphate reductase bifunctional protein: alcohol dehydrogenase; acetaldehyde dehydrogenase glyceraldehyde 3-phosphate dehydrogenase GMP reductase pyruvate oxidase phosphoribosylformylglycinamidine synthase I carbamoyl-phosphate synthase, pyrimidine-specific, large chain L-lactate dehydrogenase pyruvate oxidase adenylosuccinate synthase adenylosuccinate lyase phosphoribosylformylglycinamidine synthase II conserved purine biosynthesis cluster protein dihydroorotase dihydroorotate oxidase orotidine-5'-phosphate decarboxylase thioredoxin reductase (NADPH) adenylate kinase UTP--glucose-1-phosphate uridylyltransferase phosphoribosylaminoimidazole carboxylase, catalytic subunit phosphoribosylformylglycinamidine cyclo-ligase phosphoribosylaminoimidazole-succinocarboxamide synthase phosphoribosylamine--glycine ligase amidophosphoribosyltransferase precursor phosphoribosylaminoimidazole carboxylase, ATPase subunit phosphoribosylglycinamide formyltransferase orotate phosphoribosyltransferase purine nucleosidase purine nucleosidase short-chain dehydrogenase/oxidoreductase purine/pyrimidine phosphoribosyltransferase (putative) 0.56 0.14 0.58 0.36 0.65 1.76 0.42 0.17 0.37 4.02 1.85 1.55 1.87 5.79 5.26 1.59 1.77 1.87 5.16 1.75 1.79 0.66 5.24 4.29 5.53 4.48 4.74 5.11 4.92 1.72 0.54 0.59 0.57 1.95 0.82 2.01 0.89 0.64 0.90 2.40 0.67 0.83 2.37 2.53 0.90 0.80 0.84 2.47 2.39 2.10 2.25 2.16 2.35 2.30 0.78 0.96 -0.82 -2.82 -0.80 -1.46 -0.63 -1.25 -2.56 -1.44 -0.60 -0.88 -0.76 -0.82 2 2 E urL yrF urF yrE urA urB ur urS yrC yrD urC urD urN ox1 ox urQ urM urK1 ndk adk yrAB p p p p galU p p p p p p nrdD p p p p p purH nrdG p p adhE gapB trxB1 guaC ldhL p p M lp_3270 lp_2932 lp_2931 lp_0849 lp_1101 lp_0242 lp_0852 lp_2726 lp_2702 lp_3662 lp_3269 lp_2727 lp_2700 lp_2699 lp_2697 lp_2722 lp_2591 lp_0363 lp_1289 lp_0060 lp_0789 lp_0761 lp_1058 lp_0757 lp_3271 lp_2723 lp_2721 lp_2719 lp_2729 lp_2724 lp_2725 lp_2720 lp_2728 lp_2698 log (cy5/cy3) ratio. 2 Purines Pyrimidines biosynthesis biosynthesis M is defined as Fold change is expressed as 2

1 2

75 Purine and pyrimidine biosynthesis the SimphenyTM modeling platform (18, 20) were Encountering exposure to hydrogen peroxide for used to visualize the expression data obtained both wild-type and strain NZ7608 resulted in for the mutation-affected genes and the hydro- up-regulation of genes predicted to encode pro- gen peroxide affected genes datasets (Tables teins involved in the purine biosynthesis and py- S1 and S2 in the supplementary material, see rimidine metabolism (Table 8). In addition, hy- page 166). This exercise showed that the disrup- drogen peroxide stress affected genes predicted tion of the gene trxB1 and hydrogen peroxide to encode purine nucleoside transporters. More stress response share five distinct metabolic ac- specifically, in the presence of hydrogen perox- tivities: purine biosynthesis and transport, sugar ide the genes predicted to encode the purine uptake systems, PMF generating metabolism, nucleoside transporters pns1, pns3, pns4, and molybdopterin metabolism, and transport of pns2 were found down regulated together with isoleucine (I), leucine (L) , and valine (V). Tran- deamidases, kinases, and reductases (Fig. 2). scriptome analysis shows that the trxB1 muta- On the other hand, the genes purC-purS, purA, tion and the hydrogen peroxide stress response A purB, adk, pucK, and guaC were differentially share 93 genes which are significantly affected. expressed by four-fold up regulation for both Interestingly, all of these 93 genes were found wild-type and TR mutant. Therefore, at the level to be inversely correlated (Fig 3 and Table S3 in of transcription upon a hydrogen peroxide stress the supplementary material, see page 187) be- the genes predicted to encode purine transport- tween the two datasets. This means that a gene pyruvate dehydrogenase complex, E2 component; dihydrolipoamide S- acetyltransferase pyruvate dehydrogenase complex, E3 component; dihydrolipoamide dehydrogenase catabolite control protein catabolite control protein B pyruvate dehydrogenase complex, E1 component, beta subunit pyruvate dehydrogenase complex, E1 component, alpha subunit pyruvate oxidase ers were inversely affected compared to the pur found down-regulated due to a trxB1 mutation genes. In the absence of hydrogen peroxide, was found up-regulated as a result of hydrogen 0.54 0.35 0.46 0.33 0.24 0.20 2.57 strain NZ7608 showed significant down-regula- peroxide treatment or vice- versa. There were tion of the purC-purS (Table 4) and up-regula- genes predicted to encode tress related proteins tion of the purine transporters compared to the shared in the two datasets namely: lexA, recA, 1.36 -0.88 -1.53 -1.12 -1.60 -2.07 -2.33 wild-type. Taking these observations together, it dinP, msrA3, and umuC. These stress-related en- is concluded that trxB1 or a fully functional thio- coding proteins were significantly down-regulat- ox3 dhB dhA redoxin system in L. plantarum does not play a ed in strain NZ7608 (Table 4) when compared p ccpA ccpB p p pdhC pdhD direct role in regulation of the oxidative stress to the wild-type and significantly up regulated response of the genes in the purine biosynthetic when both strains received a hydrogen peroxide and transport pathway. At the same time, this treatment. The metabolic genes predicted to en- lp_2153 lp_2256 lp_2152 lp_2154 lp_2602 lp_2629 lp_2151 analysis reveals the importance of the purine code pyruvate oxidase (pox2), 1-phosphofructo- transporters in oxidative stress response. kinase (fruk), and nitrate reductase (nar genes) were differentially expressed and showed up-

Glycolysis Crossover between trxB1 mutation and regulation due to the introduced mutation and Metabolism Carbohydrate hydrogen peroxide stress at the level of down-regulation when the strains received a hy- transcription drogen peroxide challenge. Furthermore, genes The maps of the reconstructed metabolic net- predicted to encode proteins involved in molyb- work of L. plantarum which were created using dopterin metabolism (moaE, moaB, moaD) were

76 3 Global transcriptional analysis reveals the specific role of thioredoxin reductase in oxidative stress response in Lactobacillus plantarum WCFS1 WCFS1 in the reconstructed metabolic < 1% and FC ≥ 1.5) as a result of the disruption value < 1% and FC ≥ 1.5) in response to a hydrogen peroxide value B L. plantarum Panel A shows the significant affected genes (p . WCFS1 and NZ7608. Red colored genes (triangles) and reactions (arrows) represent an up-regulation, TM . Panel B shows the significant affected genes (p L. plantarum L. plantarum in trxB1 FIGURE 4. Visualization of amino acid biosynthesis in maps using Simpheny of pulse on strains while green colored genes and reactions represent a down-regulation in the studied condition. (a color representation of this figure can be found in page 153) A

77 found up regulated in strain NZ7608 compared thesis, and 6% regulatory functions. Specifically to wild-type. Amino acid transport of the amino the relative transcriptional response of strain acids I, L and V displayed some interesting char- NZ7608 towards hydrogen peroxide differs from acteristics in our experiments (Fig 4). the wild-type in genes predicted to encode gly- cosyltranserfase (ica2); pyruvate oxidases (pox1, The gene (livE) predicted to encode the ABC pox, pox3, pox5); ribose, sugar transporters transporter for V, L, and I as well as other ABC (rbsU, pts8C, pts35C; malE, pts2C, pts2A, pts- transporters was significantly up-regulated as a 33BCA, pts30BCA); and proteins involved in py- response to a trxB1 mutation (Table S1 in the rimidine biosynthesis (pyrAB , pyrD, pyrC, pyrF, supplementary material, see page 166) while pyrE, pyrB); and amino acid biosynthesis (aroE, the genes predicted to encode amino acid trans- cysK, tyrA). To summarize, significant differ- aminases and other transporter mechanisms ences in transcriptional response upon exposure were found to be significantly down- regulated to hydrogen peroxide were found between wild- under hydrogen peroxide stress (Table S2 in type and strain NZ7608. These differences were the supplementary material, see page 169). associated with sugar transport systems, energy Moreover, upon hydrogen peroxide treatment metabolism, pyrimidine biosynthesis, and amino other amino acid transporters encoding genes acid biosynthesis. were affected. For example, we observed the up-regulation of lp_2818 predicted to encode a glutamate H+symporter and a serine transport encoding gene (sdaC).

Mechanisms used by strain NZ7608 and wild-type to cope with hydrogen peroxide stress The statistical analysis performed in this study also contains a variable of interaction. This variable represents the difference in response towards hydrogen peroxide of strain NZ7608 compared to wild-type at the transcriptional level. This analysis results in a list of 90 genes (Table S4 in the supplementary materials, see page 190). The three most prominent categories predicted to be represented by these genes are: 22% energy metabolism, 22% transport and binding proteins, 8% purine/pyrimidines biosyn-

78 3 Global transcriptional analysis reveals the specific role of thioredoxin reductase in oxidative stress response in Lactobacillus plantarum WCFS1

DISCUSSION established. The common denominator in both This study has characterized the role of the gene datasets showed the up-regulation of 16 genes trxB1 in the oxidative stress response of L. plan- predicted to encode proteins and enzymes in- tarum by comparing the global transcriptional volved in purine and energy metabolism as well analysis of the wild-type strain (L. plantarum as in the stress response and transport. Hence, WCFS1) with that of a trxB1 disrupted strain (L. the higher oxidative stress resistance observed in plantarum NZ7608). The results show that a a TR-overproduction strain was suggested to be functional thioredoxin system is important for the result of the mock-stress response that origi- optimal growth under aerobic conditions and nated from an altered thioredoxin system. The suggests a role for the thioredoxin system in an response of the thioredoxin system in B. subtilis adequate response of L. plantarum towards hy- under hydrogen peroxide stress has been de- drogen peroxide stress. A 19% growth impair- scribed by Mostertz et al. (10). These authors ment of strain N7608 compared to the wild-type demonstrated elevated transcription and protein under aerobic (non-respiring) conditions was production of TRXs and TR’s in B. subtilis under observed. Furthermore, the global transcription- hydrogen peroxide stress. The present study al profiling of strain NZ7608 under hydrogen shows that in L. plantarum, at the transcriptome peroxide stress showed a significant down regu- level, genes predicted to be involved in the redox lation of 93 genes which are directly or indirectly systems thioredoxin and the glutaredoxin system involved in the response of L. plantarum towards both respond to oxidative stress by a significant hydrogen peroxide stress. two-fold up-regulation of the genes trxB1, trxA2, trxA3, gor, and gop, whereas trxB2 is two-fold- The thioredoxin system in L. plantarum consists down-regulated. The latter observation supports of four TRX encoding genes trxA1, trxA2, txA3, the previously reported finding that trxB2 in L. trxH and two TR encoding genes by trxB1, trxB2. plantarum is not involved in the oxidative stress Serrano et al. (14), previously demonstrated the response of L. plantarum (14). The specific role role of trxB1 in the in oxidative stress response of trxB2 in L. plantarum is still under investiga- of L. plantarum. It is assumed that TR activity is tion. There is very sparse information available mainly affected by expression of the gene trxB1 concerning the glutathione system in Gram-pos- and not trxB2 as shown by studies in Lc. lactis (21) itive bacteria. A study by Fu et al. on Lc. lactis as well as in L. plantarum (14). In the latter study, NZ9000 has elucidated that glutathione produc- it was shown that overexpression of trxB1 result- tion can enhance hydrogen peroxide resistance ed in an overproduction of TR in L. plantarum. in Lc. lactis (1). Studies performed with Listeria A TR overproducing L. plantarum strain showed monocytogenes by Gopal and coworkers (4) higher resistance towards exposure to diamide have indicated the involvement of new enzymes and hydrogen peroxide. At the transcriptome in the synthesis of glutathione. level, a crossover between the overproduction of To pinpoint to actual role of TR encoded by TR and hydrogen peroxide stress response was

79 trxB1 in the oxidative stress response of L. plan- compared to the wild-type. As L. plantarum tarum, a comparison of the global transcrip- WCFS1 is prototrophic for all nucleobases, tion profile of strain NZ7608 with the wild-type nucleosides and nucleotides, aerobic growth was performed. The results obtained show 106 of strain NZ7608 and the wild-type on media significantly affected genes in strain NZ7608 with and without supplementation of a mixture compared to the wild-type. Affected genes are of nucleobases and nucleosides was performed. predicted to encode proteins involved in DNA The growth impairment of the mutant strain repair metabolism, purine biosynthesis, sugar was most profound (24%) under aerobic con- transport, molybdopterin biosynthesis, and ami- ditions in a medium lacking supplementation no acid biosynthesis. Furthermore, a 40% lower of the mixture of nucleobases and nucleosides. TR activity was measured in strain NZ7608 com- Transcriptome analysis predicted significant up- pared to the wild-type. Hence, based on these regulation of sugar PTS transporters in the trxB1- results it can be concluded that a functional TR in mutant strain. Growth on cellobiose and sucrose L. plantarum affects a wide spectrum of metabol- resulted in the loss of the phenotype (growth re- ic activities. A list of 13 genes was filtered from duction on glucose) of strain NZ7608 compared the transcriptome analysis comparing NZ7608 to wild-type. On the other hand, greater growth and wild-type. These genes were affected only impairment (10-40%) was found in the trxB1- as a result of a trxB1 mutation and not by oxida- mutant compared to the wild-type when grown tive stress. Consequently, the expression of these in other sugars such as mannose, mannitol and genes can be used as biomarkers to study be- fructose. This analysis demonstrates the effect of havior and alterations in the thioredoxin system. an impaired thioredoxin system on a diverse ar- The broad impact of thioredoxin on a variety of ray of sugar utilization pathways in L. plantarum. cellular processes has been studied in other mi- It also shows the limitations of making predic- cro-organisms (23). Kumar et al. (8), suggested tions of the physiological response of mutations the involvement of TR in over 26 distinct cellular solely on basis of global transcription profiling processes in E. coli. Thus, a change in the thiol- data. homeostasis of L. plantarum by a trxB1 disruption apparently affects the global metabolism of The observation that both wild-type strain and the the bacterium. TR impaired strain NZ7608 respond in a similar way to hydrogen peroxide exposure, suggests Global transcriptional analysis of strain NZ7608 that the thioredoxin system is most likely not the led us to discover a phenotype in strain NZ7608, only mechanism of the cell to cope with oxida- namely a 19% reduction in growth rate in an tive stress and to re-establish intracellular redox aerated culture of strain NZ7608 compared to homeostasis. Exposure to hydrogen peroxide af- the wild-type. In addition, our transcriptome fected 605 genes in both wild-type and the trxB1 data showed a two-fold down regulation in deficient strain NZ7608. These genes are pre- the purine biosynthesis (pur genes) in NZ7608 dicted to encode a high number of proteins in-

80 3 Global transcriptional analysis reveals the specific role of thioredoxin reductase in oxidative stress response in Lactobacillus plantarum WCFS1

volved in stress response mechanisms. The stress L. plantarum strain lp80 (3), show that the POX related genes recA, lexA, dinP, kat, mrsA3 were activity in L. plantarum is regulated by POXB and all found at least two-fold significantly up regu- POXF encoded by genes pox5 and pox3. The lated as a result of a hydrogen peroxide stress. results of this study clearly support the suggested Induction of these stress related proteins upon role of the pox2 encoding pyruvate oxidase in hydrogen peroxide stress have been previously L. plantarum during periods of exposure of the studied in B. subtilis (10). In addition, Wels and cells towards hydrogen peroxide. co-workers (22) have studied the SOS regulon in L. plantarum in relation to hydrogen peroxide In total, it has been demonstrated that 93 genes stress and have determined that the aforemen- were affected in their expression by both a trxB1 tioned genes belong to the hydrogen peroxide mutation as well as by hydrogen peroxide stress stress response mechanism of L. plantarum. in both strains NZ7608 and wild-type. Interest- The current study shows that upon a hydrogen ingly, the transcript levels of all of these genes peroxide pulse, L. plantarum down-regulates were found to be anti-correlated. The predicted expression of genes predicted to encode purine proteins encoded by these 93 genes are in- transporters, with the exception of the xanthine volved in energy metabolism (fruk, pmi, ndh2, transporter and as expected, up-regulates genes nar genes), purine biosynthesis (pur genes), predicted to encode proteins involved in the pu- stress response (recA, dinP, umuC, lexA, mrsA3), rine biosynthesis. Nevertheless, this observation as well as transport systems (pts genes). In addi- poses a new enigma that is whether a hydrogen tion, the response of NZ7608 towards hydrogen peroxide pulse in L. plantarum affects the purine peroxide differed in 90 genes when compared transporters or whether the stress mainly affects to the wild-type. Genes in this sub-category are purine biosynthesis. predicted to encode proteins involved in trans- Furthermore, genes that are predicted to encode port, pyrimidine metabolism, regulatory func- proteins involved in pyrimidine biosynthesis tions and hypothetical proteins. Hence, sug- (pyrE, pyrC, pyrB), pyruvate metabolism (pox2, gesting that a trxB1 deficient strain is equipped pox3, pox5), molybdopterin metabolism, and to cope with hydrogen peroxide stress through thiamine biosynthesis were found significantly induction of one or more back-up systems that up regulated in the studied strains as a result normally would not be the preference of the of hydrogen peroxide stress. Recent studies (17) bacterium. have shown that the gene pox5 predicted to In a previous study, in which the characteristics encode pyruvate oxidase in L. plantarum is up- of a TR-overproducer were studied, it was shown regulated under hydrogen peroxide stress and that elevated TR activity was associated with an that pox2 contains a regulatory motif shared by increased resistance towards oxidative stress hydrogen peroxide induced genes which suggest (14). Elevated levels of trxB1 caused activa- a role of pox2 in hydrogen peroxide scavenging tion of genes predicted to encode stress related (17). Studies on the five annotated pox genes in proteins and enzymes (DNA repair and stress

81 mechanisms) as well as genes associated with the activity of biosynthetic pathways for purine and sulfur-containing amino acids. The present study confirms this conclusion as a number of the genes up regulated as a result of overproduction of TR were found to be down-regulated in the trxB1 mutant. This is specifically found for the purine biosynthesis genes (purC-purS) and cysteine kinase (cysK). Hence, shedding more light into the complex role of the thioredoxin system of L. plantarum in regulatory networks associated with oxidative stress.

In summary, a trxB1 deficient strain of L. plantarum has been generated and characterized. In addition, TR has been found to play a major role in adaptation to aerobic growth as well as a role during exposure towards hydrogen peroxide. The presented transcriptional response analysis of strain NZ7608 compared to wild- type illustrates the global metabolic effect exerted by alterations in the redox-balance of L. plantarum. It has also been demonstrated that there is involvement of the thioredoxin system in regulation of sugar uptake systems as well as in the regulation of genes coding for purine and amino acid transporters. Finally, a confirmed role for the thioredoxin system in regulation of expression of genes involved in the signal transduction pathway associated with the oxidative response in L. plantarum has been shown.

82 3 Global transcriptional analysis reveals the specific role of thioredoxin reductase in oxidative stress response in Lactobacillus plantarum WCFS1

REFERENCES

1. Fu, R. Y., R. S. Bongers, S. van, II, J. Chen, D. Molenaar, M. Kleerebezem, J. Hugenholtz, and Y. Li. 2006. Introducing glutathione biosynthetic capability into Lactococcus lactis subsp. cremoris NZ9000 improves the oxidative-stress resistance of the host. Metab Eng 8:662-71. 2. GEO http://www.ncbi.nlm.nih.gov/geo/, posting date. Gene Expression Omnibus (GEO). [Online.] 3. Goffin, P., L. Muscariello, F. Lorquet, A. Stukkens, D. Prozzi, M. Sacco, M. Kleerebezem, and P. Hols. 2006. Involvement of pyruvate oxidase activity and acetate production in the survival of Lactobacil lus plantarum during the stationary phase of aerobic growth. Appl Environ Microbiol 72:7933-40. 4. Gopal, S., I. Borovok, A. Ofer, M. Yanku, G. Cohen, W. Goebel, J. Kreft, and Y. Aharonowitz. 2005. A multi domain fusion protein in Listeria monocytogenes catalyzes the two primary activities for glutathione biosynthesis. J Bacteriol 187:3839-47. 5. Jobin, M. P., D. Garmyn, C. Divies, and J. Guzzo. 1999. Expression of the Oenococcus oeni trxA gene is induced by hydrogen peroxide and heat shock. Microbiology 145 ( Pt 5):1245-51. 6. Josson, K., T. Scheirlinck, F. Michiels, C. Platteeuw, P. Stanssens, H. Joos, P. Dhaese, M. Za beau, and J. Mahillon. 1989. Characterization of a gram-positive broad-host-range plasmid isolated from Lactobacillus hilgardii. Plasmid 21:9-20. 7. Kleerebezem, M., J. Boekhorst, R. van Kranenburg, D. Molenaar, O. P. Kuipers, R. Leer, R. Tarchini, S. A. Peters, H. M. Sandbrink, M. W. Fiers, W. Stiekema, R. M. Lankhorst, P. A. Bron, S. M. Hoffer, M. N. Groot, R. Kerkhoven, M. de Vries, B. Urs,ing, W. M. de Vos, and R. J. Siezen. 2003. Complete genome sequence of Lactobacillus plantarum WCFS1. Proc Natl Acad Sci U S A 100:1990-5. 8. Kumar, J. K., S. Tabor, and C. C. Richardson. 2004. Proteomic analysis of thioredoxin-targeted proteins in Escherichia coli. Proc Natl Acad Sci U S A 101:3759-64. 9. Lambert, J. M., R. S. Bongers, and M. Kleerebezem. 2007. Cre-lox-Based System for Multiple Gene Deletions and Selectable-Marker Removal in Lactobacillus plantarum. Appl Environ Microbiol 73:1126-35. 10. Mostertz, J., C. Scharf, M. Hecker, and G. Homuth. 2004. Transcriptome and proteome analysis of Bacillus subtilis gene expression in response to superoxide and peroxide stress. Microbiology 150:497-512. 11. Saal LH, T. C., Vallon-Christersson J, Gruvberger S, Borg A, Peterson C,. 2002. “BioArray Software Environment (BASE): a platform for comprehensive management and analysis of microarray data.” Genome Biol. 3:SOFTWARE0003. 12. Sambrook, J. F., E.F.; Maniatis T. 1989. Molecular Cloning: A Laboratory manual. , 2nd edition ed, vol. 1-3. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. 13. Scharf, C., S. Riethdorf, H. Ernst, S. Engelmann, U. Volker, and M. Hecker. 1998. Thioredoxin is an essential protein induced by multiple stresses in Bacillus subtilis. J Bacteriol 180:1869-77. 14. Serrano, L. M., D. Molenaar, M. Wels, B. Teusink, P. A. Bron, W. M. de Vos, and E. J. Smid. 2007. Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1. Microb Cell Fact 6:29. 15. Smyth, G. K. 2005. Limma: linear models for microarray data. Springer, , New York. 16. Starrenburg, M. J., and J. Hugenholtz. 1991. Citrate Fermentation by Lactococcus and Leuconostoc spp. Appl Environ Microbiol 57:3535-3540. 17. Stevens, M. J. A. 2008. Transcriptome Response of Lactobacillus plantarum to Global Regulator Deficiency, Stress and other Environmental Conditions. PhD thesis. Wageningen University. 18. Teusink, B., and E. J. Smid. 2006. Modelling strategies for the industrial exploitation of lactic acid bacte ria. Nat Rev Microbiol 4:46-56. 19. Teusink, B., F. H. van Enckevort, C. Francke, A. Wiersma, A. Wegkamp, E. J. Smid, and R. J. Siezen. 2005. In silico reconstruction of the metabolic pathways of Lactobacillus plantarum: comparing predictions of nutrient requirements with those from growth experiments. Appl Environ Microbiol 71:7253-62.

83 20. Teusink, B., A. Wiersma, D. Molenaar, C. Francke, W. M. de Vos, R. J. Siezen, and E. J. Smid. 2006. Analysis of growth of Lactobacillus plantarum WCFS1 on a complex medium using a genome-scale metabolic model. J Biol Chem 281:40041-8. 21. Vido, K., H. Diemer, A. Van Dorsselaer, E. Leize, V. Juillard, A. Gruss, and P. Gaudu. 2005. Roles of thioredoxin reduc-tase during the aerobic life of Lactococcus lactis. J Bacteriol 187:601-10. 22. Wels, M. 2008. Unraveling the regulatory network of Lactobacillus plantarum WCFS1. PhD thesis. Wagenin gen University. 23. Zeller, T., and G. Klug. 2006. Thioredoxins in bacteria: functions in oxidative stress response and regula tion of thioredoxin genes. Naturwissenschaften 93:259-66.

84 4 The thioredoxin system plays an important role in adaptation of Lactobacillus plantarum WCFS1 to aerobic and respiratory growth.

L. Mariela Serrano, Christof Francke, Douwe Molenaar, Bas Teusink, Willem M. de Vos, Eddy J. Smid

Manuscript in preparation 85

85 86 4 The thioredoxin system plays an important role in adaptation of Lactobacillus plantarum WCFS1 to aerobic and respiratory growth.

ABSTRACT

In this study, a global transcriptional analysis was performed in order to further characterize the thioredoxin system in Lactobacillus plantarum and establish new insights in the role of the thioredoxin system at the transcript level under oxidative stress. Strains of L. plantarum WCFS1 containing alterations in the thioredoxin system were studied: NZ7608 (trxB1::cam), NZ7602 (trxB1 overexpression), and NZ7603 (trxB2::cam) under different oxygen regimes. The presented transcriptome analysis suggests that gene trxB1 is involved in oxidative stress response while the gene trxB2 is involved in heat stress response in L. plantarum. In addition, in this study the observed growth impairment of strain NZ7608 compared to the wild-type (Chapter 3), can be attributed at the transcriptome level, to the relative differential expression of four genes coding for: glyceraldehyde-3-phosphate dehydrogenase (gapB), catalase (kat), pyruvate oxidase (pox3), and a transcription regulator (lp_0889). Finally, this study offers further insights into understanding the regulation of oxidative stress response mechanisms present in L. plantarum. Adaptation to oxygen in L. plantarum WCFS1 is correlated to a set of nine genes including those coding for pyruvate oxidases pox5 and pox3 and the putative transcription regulators coded by Lp_0889 and Lp_1360. Using comparative genomics the latter genes were further characterized as homologs of the well-known oxidative regulator OhrR of Bacillus subtilis and their associated gene network predicted to include genes: gapB, pox3, kat, mrsA3, and others.

8787 INTRODUCTION code TR. Furthermore, genes spx1 and spx3 are Oxidative stress in bacteria results in cellular predicted to encode the SPX protein. damage due to the production of Reactive Oxy- gen Species (ROS) in the cell (19). In addition, The thioredoxin system and the glutaredoxin oxygen consumption by cells invokes an altered system are responsible for maintaining thiol- redox state which is manifested through meta- homeostasis in the cytoplasm of bacteria. While bolic changes such as a shift in the NADH/NAD+ most lactic acid bacteria do not have a functional ratio. Research on Bacillus subtilis showed that glutaredoxin system (17) the thioredoxin system oxidative stress in this Gram-positive bacterium is well established in these organisms (26). The is primarily regulated by three transcription fac- thioredoxin system includes TRX and TR. Studies tors (TFs): PerR, OhrR, and sB (11). These TFs are carried out in Lactococcus lactis by Vido et al. found to be induced as a result of either specific (34) have determined that in a TR (trxB1) disrupt- stresses such as hydrogen peroxide (20), organ- ed strain, genes encoding for proteins involved ic acids and heat shock (12), or as global stress in oxidative stress defence mechanism are found response regulators under conditions of oxida- induced both at the transcript and protein level. tive stress. Global transcriptional analysis in B. Hence, these results suggest a role for TR in the subtilis Mosterz et al. (20) revealed activation of aerobic life of Lc. lactis (34). Characterization an extended regulatory network involved in the of the gene trxB1 in L. plantarum WCFS1 using overall stress response in this microorganism. global transcriptional analysis provided addi- At the transcriptome level this network involved tional information on its role in aerobic cultiva- different regulators: PerR, OhrR, s-, CstR, and tion, carbon metabolism, and in the nucleotide the interaction with other regulation networks biosynthesis and amino acid transport (Chapter including the SOS response (an inducible DNA 3). Recent studies have demonstrated that trxB1 repair system that allows bacteria to survive sud- overexpression in L. plantarum results in a strain den increases in DNA damage), genes related with higher diamide and hydrogen peroxide re- to sulfur-limitation, and other genes like methio- sistance compared to the wild-type (5). This phe- nine sulfoxide reductase (mrsA), glutathione per- notype was attributed to the up-regulation of 16 oxidase (gpx), thiol peroxidase (tpx), thioredoxin genes, including gapB, groEL, npr2, and mntH2. (trxA), and thioredoxin reductase (trxB) (20). The The same 16 genes were shown earlier to be former two genes coding for thioredoxin (TRX) subject to up-regulation as a result of hydrogen and thioredoxin reductase (TR) have been char- peroxide stress (26). These studies in both Lc. acterized by Nakano et al. (21) as genes induced lactis and L. plantarum suggest that the thiore- under thiol-specific (disulfide) oxidative stress in doxin system plays an important role in oxidative B. subtilis and regulated by levels of the regula- response in lactic acid bacteria. In order to fur- tory SPX protein (21). In Lactobacillus plantarum ther test this, the global transcriptional response WCFS1 the gene trxA2 is predicted to encode of L. plantarum strains containing alterations in TRX while trxB1 and trxB2 are predicted to en- the thioredoxin system under different oxygen

88 4 The thioredoxin system plays an important role in adaptation of Lactobacillus plantarum WCFS1 to aerobic and respiratory growth.

regimes has been performed. By applying this MATERIAL AND METHODS global approach more insight into the thiore- Bacterial strains, plasmids, and media. doxin system in L. plantarum through character- The bacterial strains, plasmids, and oligonucle- ization of the genes trxB1 and trxB2 has been otides used for this study are summarized in gained. Increased understanding of the oxida- Table 1. Escherichia Coli strains were grown at tive stress response network in this bacterium 37°C in trypton and yeast extract (TY) (25). L. has also been achieved. plantarum WCFS1 was grown at 37°C, unless otherwise specified, in de Man-Rogosa-Sharp (MRS) or in Chemically Defined Medium (CDM) (30).

TABLE 1. Strains, plasmids, and oligonucleotides used in this study.

Characteristics Reference

Strains

E. coli

WCFS1 L. plantarum WCFS1 (16)

NZ7602 CmR, L. plantarum WCFS1 derivative carrying the pMS040 (26)

plasmid.

NZ7608 L. plantarum WCFS1 trxB1::cam derivative with chromosomal (Chapter 3)

integration of plasmid pMS019.

NZ7603 L. plantarum WCFS1 trxB2::cam derivative with chromosomal This work

integration of plasmid pMS004.

Plasmids

puc18Ery AmpR, Ery R (33)

pMS004 pUC18Ery derivative, AmpR, Ery R construct for D.C.O trxB2 with This work

1KB upstream (partial lp_2584) and 1KB downstream (partial

lp_2586) of trxB2.

Oligonucleotides 5’-3’

trxB2-FORWA GGC CGA ATT CTC CAG CTG ATG ACG AAT CG

trxB2-REVA GAA GGG TAC CCC ACT GCA TAG TAC GAC CG

trxB2-FORWB CTC CGG TAC CTA TTC TGC ACT CAT CGT TTC C

trxB2-REVB AGG GTC TAG AGA TAT GTG ATG TTT TAG GAT ACC

89 DNA manipulations. DNA manipulations ing trxB2::cam variant of L. plantarum WCFS1 were performed as described elsewhere (26). was designated NZ7602 and its integrity was Construction of L. plantarum strain NZ7603 checked by restriction and Southern blot analysis (trxB2::cam). To acquire a trxB2::cam replace- (data not shown). ment variant in L. plantarum WCFS1, plasmid Growth determination. In all growth experi- pMS004 was constructed. This plasmid allowed ments cell density was assessed by measuring the direct selection of the desired mutant when a the turbidity of the cultures at 600 nm every 30 double cross-over event occurred. The upstream min. Specific growth rate (μmax) was defined as and downstream flanking regions of the gene ΔLnOD600 / Δtime at the linear or exponential trxB2 (lp_2585) were amplified by PCR using growth phase of each culture. For the applica- genomic DNA of L. plantarum WCFS1 as tem- tion of aerobic and respiratory conditions, cul- plate and two primer combinations: trxB2_forwA tures were grown in the Innova 4340 Incubator with trxB2_revA and trxB2_forwB with trxB2_revB Shaker (New Brunswick Scientific). Temperature (Table 1). The resulting amplified regions -(ap was kept at 37ºC and cultures were shaken at a prox. 1 KB) with introduced restriction sites were constant rate of 125 rotations per minute (rpm). cloned via a two-step ligation into the vector Diamide growth cultivations. Cells were puc18ERY using EcorI and KpnI sticky restric- pre-cultured over night in 200-ml shake flasks tions sites for fragment A (lp_0284) and the KpnI containing 10 ml MRS and at 37º. Mature cells and Xba1 sticky restriction sites for fragment B were diluted 1:20 in 20 ml MRS and incubated (lp_2586). The resulting plasmid containing at three conditions: aerobic, anaerobic, or aero- the 1kb upstream and downstream regions of bic with 5mM diamide. lp_2585 (trxB2) was denominated pMS004 Shake Flask cultivations. Cells were pre-cul- was transformed into competent E. coli DH5 a tured over night at 37ºC in 200-ml shake flasks cells. After purification and verification of the containing 10 ml CDM with 0.5% (wt/vol) glu- sequences, plasmid pMS004 was introduced cose under either aerobic, anaerobic or respira- into L. plantarum WCFS. Transformants were tory (10 μg/ml Hemin, 40 μg/ml Vitamin k2). selected from agar plates containing 10 mg/ml Mature cells were diluted 1:20 in 20 ml CDM Erythromycin. Plates were incubated overnight containing 0.5% (wt/vol) glucose and main- at 37ºC. Colonies displaying Erythromycin resis- tained under the same oxygen regime. tance (EryR) should be as the result of a one-step HPLC. Extracellular metabolites present in the cross-over event on either flanking region of the supernatant of the different cultures were mea- trxB2 gene. Single EryR colonies were cultivated sured as described elsewhere (28). in MRS for over 100 generations. Then, cultures Microarrays. were plated on MRS agar. Colonies now dis- Batch cultivations. Cells were pre-cultured playing Erythromycin sensitivity (Erys) should be over night in 10 ml CDM with 0.5% (wt/vol) as the result of a double-cross over event on the glucose and at 37ºC in anaerobe jars. Cultures flanking regions of the trxB2 gene. The result- were diluted 1:50 in 500 ml CDM with 0.5% (wt/

90 4 The thioredoxin system plays an important role in adaptation of Lactobacillus plantarum WCFS1 to aerobic and respiratory growth.

vol) glucose and kept at 37ºC. During growth, effect of oxygen in each culture and (ii) the dif- the pH was kept at 5.5 by the addition of 5M ference in response toward aerobic and an- NaOH. The cultures were stirred at a constant aerobic conditions of each mutant compared to rate of 200 (rpm). the wild-type could be evaluated. The series of Anaerobic batch cultivations. Nitrogen gas normalized data obtained in this experiment are was supplied to the overhead of the fermentor available at gene expression omnibus (GEO) at 520 ml/min. with accession number GSE8743 (10). Aerobic batch cultivations. Filtered air was Comparison of respiratory growth be- bubbled through the culture at a rate of 150 tween L. plantarum strains using microar- ml/min. rays. Each culture (WCFS1, NZ7608, NZ7602, Respiratory cultivations. Cells were pre-cul- and NZ7603) was cultivated in a shake flask tured over night in 10 ml CDM with 0.5% (wt/ until OD600=1.0 and the following hybridization vol) glucose and at 37ºC. Cultures were diluted scheme (4 arrays) was applied: A) WCFS1 with 1:50 into 250-ml shake flasks containing 100 ml NZ7608 B) NZ7608 with NZ7602 C) NZ7602 CDM with 0.5% (wt/vol) glucose, 10 μg/ml He- with NZ7603 and D) NZ7603 with WCFS1. min, and 40 μg/ml Vitamin k2 until OD600=1.0. With this scheme the difference in response to- RNA isolation and cDNA synthesis. RNA wards respiratory cultivation of strains NZ7608, isolation and cDNA synthesis and purification NZ7602 and NZ7603 compared to wild-type as well as data processing was carried out as was evaluated. The normalized data of this described elsewhere (26). Significant differences experiment is available at GEO with accession in gene expression were defined as genes with a number GSE8744 (10).

False Discovery Rate (FDR) adjusted for the pvalue Functional annotation and transcription of less than 10% and FC ≥1.5 in the aerobe/ factor binding site prediction. All genomic anaerobe experiment or pvalue of less than 1% information was obtained from the ERGO ge- and FC ≥1.5 in the respiratory experiment. nome analysis and discovery system (23) (update Comparison of aerobic versus anaero- 7th of December 2007.) The information is also bic growth using microarrays. For each publicly accessible in the resources of NCBI (36). culture (WCFS1, NZ7608, and NZ7603) one The genome sequence of L. plantarum WCFS1 aerobic and one anaerobic batch cultivations and the functional annotation of its genes were were performed and the following hybridization taken from an in-house annotation database scheme (6 arrays) was applied: A) WCFS1_aer- (16). Potential MarR-family homologs were obe with NZ7608_aerobe B) NZ7608_aerobe collected from the database using the BLAST with NZ7603_aerobe C) NZ7603_aerobe with algorithm (1), with a typical cut-off of 10-10. WCFS1_aerobe D) WCFS1_anaerobe with A multiple sequence alignment was made with WCFS1_aerobe E) NZ7608_anaerobe with MUSCLE (7) (standard settings). The alignment NZ7608_aerobe and F) NZ7603_anaerobe was visually inspected and aberrant sequences with NZ7603_aerobe. With this scheme (i) the were removed (characterized by many gaps and

91 a different conservation pattern). ClustalW (31) RESULTS was used to create a bootstrapped neighbour- Phenotypic Characterization of L. plan- joining tree (with ‘correction for multiple substi- tarum strains tutions’ (15)). The resulting tree was analyzed The thioredoxin system plays an important role using LOFT, a tool that automatically divides the in the adaptation of L. plantarum strains to oxi- sequences into orthologous groups based on dative stress (26). In addition, previous results the hierarchy of the tree and the duplication and (26) showed the specific growth rate of a trxB1- speciation events implied by that hierarchy (32). deficient L. plantarum strain, NZ7608, was 19% Then, the orthologous groups were subdivided less compared to the wild-type when the strain into synteny groups (i.e. shared gene context) to was grown aerobically in a pH-controlled batch identify functional equivalents. Functional equiv- fermentation. Due to the suggested correlation alents are expected to share the same transcrip- between trxB1 and oxidative stress, it was de- tion factor (TF)-specific operator motif described cided to study the behavior of strain NZ7608 by Francke et al., unpublished results. In the case (trxB1::cam) when grown in a rich medium of the MarR homologs of interest, the upstream (MRS) and under different oxidative stress condi- regions of the TF encoding gene and neighbour- tions: aerobic (sparged with air), and disulfide ing genes were selected (indicated in Fig. 2) and stress (diamide addition to the media) (Fig. 1). potential regulatory elements (e.g. overrepresented DNA sequences) (200 bases) were iden- Under conditions of aerobic growth, strain tified automatically using MEME (2). The MEME NZ7608 showed reduction in specific growth output was screened manually, compared to the rate, thereby confirming the previous results. known binding site of B. subtilis OhrR (13) and In addition, the strains NZ7608 and wild-type specific operator motifs were defined. MAST (3) were severely affected when growth under con- was used to detect similar sites on other places ditions of disulfide stress (MRS + air + 5mM -5 of the genome (standard cut-off pvalue <10 ). diamide). Nevertheless, under conditions of disulfide stress, strain NZ7608 was affected to the same extent compared to wild-type. Further phenotypic characterization included monitoring the specific growth rate of L. plantarum strains WCFS1, NZ7608, and NZ7602 (trxB1 overexpression) under conditions that support aerobic respiration in Lc. lactis. To achieve this condition, the strains were cultivated aerobically in a

medium containing hemin, and vitamin k2. The specific growth rate in the trxB1-disrupted strain NZ7608 was 14% less when grown aerobically and 31% less under respiratory conditions compared to the wild-type (Table 2).

92 4 The thioredoxin system plays an important role in adaptation of Lactobacillus plantarum WCFS1 to aerobic and respiratory growth.

FIGURE 1. Growth of L. plantarum strains WCFS1 and NZ7608 under different oxygen regimes (A) anaerobic (B) aerobic and (C) aerobic with 5mM diamide in the medium.

Cell density (OD600) was measured every 30 min.

A A 6 BB 6

4 4 OD600 OD600

2 2

0 0 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 Time (hours) Time (hours)

C 6 C WCFS1 NZ7608 NZ7603

4 OD600

0 0 1 2 3 4 5 6 7 8 9 Time (hours)

TABLE 2. Specific growth rate depicted as μmax (h-1) of Lactobacillus plantarum WCFS1, NZ7608, and NZ7602 strains grown under anaerobic, aerobic, and respiratory conditions. Strains were grown in shake flasks anaerobically (anaerobe jars), aerobically (sup-

plied oxygen), and under respiratory (hemin, vitamin K2, and oxygen).

µmax 1 Strain Anaerobic Aerobic Respiratory WCFS 0.57 0.52 0.41 NZ7608 0.56 0.45 0.28 NZ7602 0.53 0.52 0.43

1 Specific growth rate (μmax) was defined as ΔLnOD600/Δtime at the linear or exponential growth of each culture

93 On the other hand, the trxB1-overexpressing Extracellular metabolite composition in strain NZ7602 did not show a significant dif- oxygen grown cultures ference in specific growth rate compared to the To study the effect of oxygen on the growth of wild-type under respiratory conditions. Hence, strains with altered TRX levels, further analysis of the disruption of gene trxB1 seems to have a strains WCFS1, NZ7608, NZ7602, and NZ7603 large impact on growth rate, particularly under (trxB2::cam) was performed. The end-product growth conditions that support oxidative respira- composition when strains WCFS1, NZ7608, tion in L. plantarum. Whereas the overexpres- and NZ7603 were grown in pH-controlled batch sion of trxB1 does not have a measurable effect, fermentors under anaerobic and aerobic con- the impact a trxB1 disruption is significant under ditions was assessed. Equal volume samples of aerobic and respiring conditions while these are all cultures were taken at the exponential growth not affected under anaerobic conditions. The ef- phase (OD600=1.0) for analysis of extracellular fect of oxygen on the metabolism of L. plantarum metabolites composition (Table 3). NZ7608 was further investigated by analysis of the metabolites produced in the fermentations.

TABLE 3. Effect of oxygen in extracellular metabolite concentrations at OD600=1.0 in L. plantarum strains WCFS1, NZ7608, and NZ7603 grown in controlled batch cultivations under aerobic and anaerobic conditions.

Anaerobe 3 Aerobe4 1 Metabolite Medium WCFS1 NZ7608 NZ7603 Medium WCFS1 NZ7608 NZ7603 Glucose 27.40 20.79 20.04 20.14 26.55 20.99 19.88 20.04 Citric Acid 7.65 7.55 7.28 7.33 7.49 7.78 3.98 7.49 Acetic Acid 2.21 2.25 2.16 2.14 2.20 2.08 0.99 2.07 Lactic Acid u.d 5 13.99 13.99 13.88 u.d 5 12.66 6.29 11.55 Succinate u.d 5 0.10 -0.15 0.06 u.d 5 u.d 5 u.d 5 u.d 5 2 DW 0.25 ± 0.01 0.28 ± 0.01 0.32 ± 0.02 0.25 ± 0.02 0.25 ± 0.01 0.25 ± 0.01

1 Metabolite values given in mM. All measurements were done on the supernant of the cultures when reached a cell density of

OD600=1.0. Acetoine, ethanol, formic acid, and pyruvate were under detection limits. 2 Dry weight is given as g biomass/liter. 3 Cultures grown in pH-contolled batch fermentors. Nitrogen gas was supplied to the overhead of the fermentor at a rate of 520 ml/ min while the culture was stirred at a constant speed of 200 rotations per minute (rpm). 4 Cultures grown in pH-contolled batch fermentors. Filtered air was sparged through the culture at a rate of 120 ml/ min while the culture was stirred at a constant speed of 200 rpm.

94 4 The thioredoxin system plays an important role in adaptation of Lactobacillus plantarum WCFS1 to aerobic and respiratory growth.

Metabolite analysis showed that the supply of a trxB2-disruption strain, NZ7603. The lack of air in these experiments triggered homo-lactic correlation in aerobic fermentations between fermentation in all fermentations. Metabolites lactic acid production and sugar consumption related to a mixed-fermentation (acetoine, ac- in strain NZ7608 suggests that oxygen affects etate, ethanol) were under detection limits and strain NZ7608 at the level of central carbon me- biomass formation of wild-type was equal to tabolism. Nevertheless, strain NZ7608 is able to 0.25 g biomass/liter both under anaerobic and grow with oxygen and produces equal amounts aerobic growth conditions. On the other hand of biomass compared to the wild-type and to under anaerobic growth conditions, both trxB- strain NZ7603. This result suggests activation of disrupted strains NZ7608 and NZ7603 pro- alternative processes in L. plantarum that ensure duced 0.28 and 0.32 g biomass/liter respec- redox homeostasis and take over functions of the tively. Therefore, the biomass production in the thioredoxin system. To identify these processes a trxB-disrupted strains was significantly higher global transcriptional analysis under anaerobic compared to 0.25 g biomass/liter produced in and aerobic growth conditions of L. plantarum the wild-type under anaerobic growth condi- wild-type and mutant strain was performed. tions. The total reduced lactic acid production in wild-type grown anaerobically and aerobically was 14mn and 13mM respectively. While Effect of TR mutations at the transcrip- the production of reduced lactic acid in strain tome level under anaerobic growth NZ7603 was comparable to that of the wild- conditions type in both growth conditions, strain NZ7608 As mentioned above, a trxB1-deficientL. planta- showed a different profile. Under aerobic condi- rum strain showed impaired growth compared tions, strain NZ7608 produced 6.30mM of re- to wild-type specifically under conditions of oxi- duced lactic acid in comparison to the amounts dative stress. To understand this behavior, the produced by wild-type (12.70mM) and to strain global transcriptional response of L. plantarum NZ7603 (11.60mM). On the other hand, when strain NZ7608 to wild-type grown anaerobi- strain NZ7608 was grown anaerobically, it pro- cally was compared. This resulted in a set of duced 14mM of lactic acid. This value is iden- significant differentially expressed genes with tical to the measured lactic acid production in pvalue <0.1 and FC ≥ 1.5 (Table 4). The genes wild-type and strain NZ7603 and represents a found up regulated in strain NZ7608 com- 50% increase to the observed lactic acid produc- pared to wild-type are genes predicted to en- tion in strain NZ7608 in the aerobic situation. code catalase (kat), a pyruvate oxidase (pox3), Under aerobic growth conditions strain NZ7608 glyceraldehyde-3-phosphate dehydrogenase consumed: 9mM glucose, 1.30mM citrate, and (gapB), methionine sulfoxide reductase (mrsA3), 4mM acetate. These values are 1.3, 10, and L-ribulokinase (araB), putative transcription 13 fold higher compared to the consumption regulators (lp_0889, lp_1360), proteins with of these metabolites in strains WCFS1 and in unknown function (lp_1880, lp_2113), and the

95 plantaricin operon (plnA-plnW) including the WCFS1 to the presence of air in the medium was regulator (plnA). Comparison of the global tran- shown to involve to nine genes. These genes are scriptional response of strain NZ7603 to strain predicted to encode pyruvate oxidases (pox3 NZ7608 under anaerobic cultivation was also and pox5), catalase (kat), a methionine sulfoxide performed. Such a comparison showed that the reductase (mrsA3), 1-deoxy-D-xylulose-5-phos- only difference between these two trxB-disrupted phate-synthase (dxs), a putative oxidoreductase strains is the up regulation of genes predicted (lp_1939), two transcription regulators (lp_0889 to encode for the biosynthesis of plantaricin in and lp_1360), and a protein with unknown strain NZ7608 compared to NZ7603. In sum- function (lp_2113). The predicted pyruvate mary, L. plantarum strain NZ7608 and NZ7603 oxidase encoding genes pox3 and pox5 were when grown anaerobically differ from wild-type up-regulated 4-fold and 8-fold, respectively in by modulating the expression of genes predicted the presence of air. In addition, the global tran- to encode proteins involved in stress response, scriptional of strains NZ7603 and NZ7608 dif- energy metabolism, regulatory functions, and fered significantly from wild-type for genes pre- plantaricin production. Hence, the global gene- dicted to encode plantaricin (pln genes, Table expression analysis suggests that there is already 4), and lp_3246 which is predicted to encode a a significant difference between the wild-type protein with unknown function. Moreover, gene and the trxB-deficient strains under anaerobic lp_0064, predicted to encode a hypothetical conditions protein, was found significantly up-regulated in strain NZ7608 compared to wild-type. In conclusion, in aerated cultures the differences at the global transcriptional level between a trxB1-dis- Effect of TR mutations at the transcrip- rupted L. plantarum strain and the wild-type are tome level under aerobic growth condi- rather few with the expression of the plantaricin tions cluster as main exception. Consequently, this To gain understanding on the role of the thio- analysis suggests these L. plantarum strains cope redoxin system in oxidative stress response of L. with oxygen similarly at the transcriptome level. plantarum the global transcriptional response of Although the response pattern does not seem to each trxB-mutant strain towards aerobic culti- be different, the level of regulation on the genes vation with the response of the wild-type strain involved does vary. It was found that in the pres- under the same condition was compared. The ence of oxygen the level of expression of genes global transcriptional analysis of L. plantarum in L. plantarum strains NZ7608 and NZ7603 strains WCFS1, NZ7608, and NZ7603 under varies compared to that of the wild-type. Strain aerobic cultivation resulted in several sets of sig- NZ7608 showed significant up-regulation of nificant differentially expressed genes with valuep the genes: kat, pox3, gapB, mrsA3, lp_1880, <0.01 and FC ≥ 1.5 (Table 4). The main adap- lp_2113, lp_0889, lp_1360, lp_3246, and the tation at the transcriptome level in L. plantarum plantaricin operon (pln genes). On the other

96 4 The thioredoxin system plays an important role in adaptation of Lactobacillus plantarum WCFS1 to aerobic and respiratory growth. strains L. plan L. plan strain WCFS1. Protein fate metabolism L. plantarum Cell envelope strain NZ7608 vs wild-type. strain NZ7603 vs wild-type. Cellular processes Cellular processes Energy metabolism Energy metabolism Central intermediary Regulatory functions Regulatory functions Hypothetical proteins Hypothetical proteins strain NZ7608 vs wild-type. strain NZ7603 vs wild-type. Main Functional Class L. plantarum Lactobacillus plantarum L. plantarum L. plantarum L. plantarum <0.1 and FC ≥1.5. value Product immunity protein PlnP, membrane-bound protease CAAX family immunity protein PlnI, membrane-bound protease CAAX family 1-deoxy-D-xylulose-5-phosphate synthase plantaricin biosynthesis protein PlnQ pyruvate oxidase L-ribulokinase (putative) glyceraldehyde 3-phosphate dehydrogenase protein-methionine-S-oxide reductase plantaricin biosynthesis protein PlnR immunity protein PlnM catalase plantaricin biosynthesis protein PlnO pyruvate oxidase unknown mannosyl-glycoprotein endo-beta-N-acetylglucosaminidase oxidoreductase bacteriocin precursor peptide PlnJ (putative) plantaricin A precursor peptide, induction factor bacteriocin precursor peptide PlnF (putative) bacteriocin precursor peptide PlnE (putative) plantaricin biosynthesis protein PlnS immunity protein PlnL histidine protein kinase PlnB; sensor unknown transcription regulator unknown transcription regulator unknown g 7 2.03 f 6 Response 1.58 1.47 2.11 1.51 1.39 1.73 1.59 2.29 1.87 2.39 1.88 2.08 2.55 1.67 1.46 1.61 1.82 1.63 e 5 1.62 2.05 1.85 2.85 2.48 3.70 2.94 3.58 3.30 5.56 3.05 3.47 5.38 2.00 d 4 2.29 2.65 2.41 3.77 3.45 3.97 3.71 4.67 3.43 3.60 5.24 1.64 1.58 2.02 Aerobe Significantly different expression of genes have a p c 3 -2.21 -2.27 -2.55 -4.32 -2.40 -2.22 -2.58 -2.59 -2.58 b 2 -3.75 -3.81 -5.06 -4.29 -5.59 -4.25 -4.55 -5.40 -3.68 a Anaerobe 1 3.36 7.23 3.28 5.87 4.34 6.63 8.54 7.11 9.45 7.18 7.76 4.11 5.11 3.23 3.12 3.61 3.58 3.18 4.80 10.34 log (cy5/cy3). This ratio depicts the significant affected genes under anaerobic conditions of log(cy5/cy3). This ratio depicts the significant affected genes under anaerobic conditions of log (cy5/cy3). This ratio depicts the significant affected genes in anaerobic vs aerobic growth conditions log (cy5/cy3). This ratio depicts the significant affected genes under aerobic conditions of log (cy5/cy3). This ratio depicts the significant affected genes under aerobic conditions of log (cy5/cy3). This ratio depicts the significant affected genes that show different response towards aerobic growth conditions in log (cy5/cy3). This ratio depicts the significant affected genes that show different response towards aerobic growth conditions in 2 2 2 2 2 2 2 3 F E lnJ lnL ln lnB lnA ln lnS lnR lnO lnQ lnM ox3 ox5 kat dxs plnI p p p p p p p plnP p p p araB p p p Gene gapB msrA Locus lp_2610 lp_0789 lp_0415 lp_0403 lp_1836 lp_3578 lp_0404 lp_0409 lp_3556 lp_0412 lp_0413 lp_0422 lp_0424a lp_0419 lp_0421 lp_0406 lp_3589 lp_0411 lp_2629 lp_1939 lp_2113 lp_0889 lp_1360 lp_0064 lp_1880 lp_3246 lp_0182 lp_0416 strain NZ7608 vs wild-type. strain NZ7603 vs wild-type. M ratio defined as M= M ratio defined as M= M ratio defined as M= tarum tarum M ratio defined as M= M ratio defined as M= M ratio defined as M= M ratio defined as M=

a b c d e f g

TABLE TABLE 4. Global transcriptional analysis comparing the effect of oxygen in WCFS1, NZ7608, and NZ7603.

97 hand, in L. plantarum NZ7603 only a single dif- sociation is conserved among the Bacilli (Fig. ferentially expressed gene was found in compar- 2). In the case of the three L. plantarum ohrR ison to the wild-type, namely pox3. These results orthologs, lp_1821 seems to be equivalent to a suggest a specific role of trxB1 at the regulatory specific ohrR ortholog found in Staphylococci as level in adaptation during aerobic growth of L. it shows a conserved genomic association with plantarum. glutathione reductase (lp_1822). In addition, the conserved association with a gene encoding Comparative genomics for transcrip- a putative exonuclease in various Lactobacilli tion factors (TFs) Lp_0889 and Lp_1360 suggests that lp_1360 encodes a Lactobacillus- To find the reason for the observed changes at specific OhrR equivalent. Finally, the absence of the level of the transcriptome under the various conservation with known genome sequences in conditions, two transcription factors (lp_0889 the case of lp_0889 might indicate, this ohrR-or- and lp_1360) that were found significantly up- tholog carries out an organism-specific role. regulated in the presence of air, were subject to In the upstream region of each of the three L. an independent comparative genomics analy- plantarum ohrR orthologs a clear regulatory sis (as described in the methods). A search for motif was identified (see Fig. 2 and methods). TF homologs in L. plantarum WCFS1 indicated Although, the motifs are very similar to the one that genes lp_0889, lp_1360 and a third gene, determined for B. subtilis OhrR (13), there are lp_1821, belong to a specific sub-family of the particularly conserved variations that allow the larger MarR protein family of transcription fac- attribution of individual motifs to each of the tors (PFAM: PF01047; (8)). A BLAST search for TFs. The L. plantarum genome was searched close homologs in all sequenced bacterial ge- with a consensus motif (Fig. 2) that was based nomes resulted in over 300 hits (not shown). on these three specific motifs, and the recovered The consecutive comparative sequence analysis sites were collected. Based on the leads from of homologs in species of the phylum Firmicutes bioinformatics (Supplementary material, Table – the analysis was based on alignment, tree S1, see page 194) and transcriptome analysis building and interpretation of that tree - showed (Table 4) of the OhrR regulon a model of how that the three L. plantarum sequences are closely TRX levels and oxygen presence could impose related (orthologous) to OhrR of B. subtilis. The regulation at the level of transcription in L. plan- gene ohrR has been shown to be involved in the tarum (Fig. 3) has been developed. Each studied transcriptional response toward peroxide stress situation will impose stress on the bacterium and (12). Moreover, it is associated with ykzA at the hence, adaptation will be required. These results DNA level, a gene that is part of the sB-depen- suggest that there are mechanisms that are used dent general stress regulon of B. subtilis (35). by the bacterium regardless of the stress (kat, The comparative analysis indicated that this as- pox3) and others are stress-specific (lp_0064, mrsA3). In addition, it is clear that the OhrR

98 4 The thioredoxin system plays an important role in adaptation of Lactobacillus plantarum WCFS1 to aerobic and respiratory growth.

FIGURE 2. Phylogeny and synteny of OhrR-like TFs for selected Firmicutes (a color version of this figure can be found in page 154). The sequences of transcriptional factors (TFs) orthologous to B. subtilis OhrR from a representative set of Firmicutes were aligned and a putative phylogeny was generated (bootstrap support indicated in red).

Orthologous genes are similary colored. The OhrR-like transcriptor factor is color-coded orange. The selected upstream regions to search for TF-specific operator sequences are indicated with small blue arrows. The retrieved operators are depicted on the right

regulon is not the only mechanism used by L. plantarum to cope with oxygen. This is because for the pox5 gene does not have an Ohr motif and it was found significantly up-regulated under aerobic growth.

99 FIGURE 3 Model for the transcriptional regulation displayed by OhrR of in L. plantarum towards oxidative stress response. Panel A depicts the genes suggested to be regulated by OhrR as a response to TR levels. Panel B shows the genes possibly regulated by OhrR in response to oxygen in the medium

gapB AA araB pox3 Thioredoxin Lp_0889 lp_2113 Reductase lp_1880 kat pln cluster

B ? pox5

pox3 lp_2113 Oxygen Lp_0889 lp_1880 kat pln cluster

mrsA3 Lp_1360 Lp_0064

Characterization of gene trxB1 in L. rum strain NZ7608 with the wild-type response plantarum WCFS1 towards respiratory cultivation was compared. Previous studies showed that trxB1 plays an es- This resulted in a set of 79 significant differen- sential role in oxidative stress response (26) and tially expressed genes summarized in Table 5 aerobic cultivation in L. plantarum (26). In order and within this set of 79 genes, 14 were sig- to identify whether the observed phenotype of nificantly up-regulated and 65 were down-regu- strain NZ7608 (trxB1::cam) is oxygen or stress lated in strain NZ7608 when compared to the specific the global transcriptional of L. planta- wild-type.

100 4 The thioredoxin system plays an important role in adaptation of Lactobacillus plantarum WCFS1 to aerobic and respiratory growth.

The up-regulated gene set contains genes pre- performed. This produced a dataset of eight dicted to encode a protein involved in the syn- significantly differentially regulated genes (Table thesis of a molybdenum cofactor (moeA), a 6). The only gene that was significantly up regu- metal ion transport protein; a putative Mn-like lated in strain NZ7602 compared to wild-type, transporter (mntH2), and a universal stress pro- was trxB1 encoding the TR protein. The other tein (uspA). Within the differentially 65 down- seven genes which were found down regulated regulated genes in strain NZ7608 are found in strain N7602 compared to wild-type are pre- genes predicted to encode for: a branch ami- dicted to encode for a TRX (trxA2), a pyruvate no acid transporter (livE), regulatory proteins kinase (pyk), a transcription terminator factor (spx1, spx3), heat shock proteins (hsp1, hsp2), (nusB), a protease (clpE) and ribosomal proteins an aldolase (deoC), ribosomal proteins (rpl op- (rpl operon). The trxB1 deficient strain is severely eron); as well as, proteins involved in cell enve- impaired in growth under conditions that al- lope structure (cps operon), energy metabolism low oxidative respiration and surprisingly the (araB, pox3), and amino acid biosynthesis (aro overproduction of the enzyme does not result in genes, his genes). An additionional, compari- phenotypic complementation. Apparently, over- son of the global transcriptional profiles of a production of the TR also brings about negative strain carrying the trxB1-overexpression plasmid effects during these specific growth conditions, (strain NZ7602) and that of the wild-type strain masking the effect of complementation. during respiratory cultivation conditions was

TABLE 6 Summary of global transcriptional analysis comparing the response to respiratory cultivation in Lactobacillus plantarum NZ7602 and WCFS1. Significantly different expression of genes have a pvalue <0.01 and FC ≥1.5.

1 2 Locus Gene M FC Product Main Functional Class Biosynthesis of cofactors, prosthetic lp_2270 trxA2 -0.66 0.6 thioredoxin groups, and carriers (13%) ATP-dependent Clp protease, ATP-binding subunit Cellular processes (13%) lp_1269 clpE -0.64 0.6 ClpE lp_1897 pyk -0.65 0.6 pyruvate kinase Energy metabolism (13%) lp_1594 rpmA -0.69 0.6 ribosomal protein L27 lp_1640 rplS -0.60 0.7 ribosomal protein L19 Protein synthesis (38%) lp_2125 rpsO -0.59 0.7 ribosomal protein S15 Purines, pyrimidines, nucleosides and lp_0761 trxB1 0.69 1.6 thioredoxin reductase (NADPH) nucleotides (13%) lp_1598 nusB -0.82 0.6 transcription termination factor NusB Transcription (13%)

1 M ratio defined as the ratio M=2log (cy5/cy3) of L. plantarum strain NZ7602 vs wild-type. 2 Fold Change (FC) is defined as M2 .

101 - Lac <0.01 and FC value (1%) Cell envelope (4%) Main Functional Class Cellular processes (4%) Energy metabolism (9%) Amino acid biosynthesis (16%) Biosynthesis of cofactors, prosthetic groups, and carriers Significantly different expression of genes have a p Product nucleotide-binding protein, universal stress protein UspA family imidazole glycerol phosphate synthase, amidotransferase sununit phospho-2-dehydro-3-deoxyheptonate aldolase / chorismate mutase polysaccharide biosynthesis protein small heat shock protein glycosyltransferase acyltransferase/acetyltransferase phosphoribosyl-AMP cyclohydrolase prephenate dehydrogenase shikimate 5-dehydrogenase shikimate kinase molybdopterin biosynthesis protein MoeA histidinol-phosphate aminotransferase histidinol dehydrogenase 3-dehydroquinate synthase 3-phosphoshikimate 1-carboxyvinyltransferase aspartate ammonia-lyase chorismate synthase cysteine synthase small heat shock protein L(+)-tartrate dehydratase, subunit B L-ribulokinase (putative) transketolase pyruvate oxidase 4-aminobutyrate aminotransferase fumarate reductase, flavoprotein subunit precursor acetoin utilization protein, C-terminal fragment (putative) acetoin utilization protein, N-terminal fragment (putative) extracellular protein, membrane-anchored (putative) short-chain dehydrogenase/oxidoreductase 2 FC 0.62 0.42 0.62 0.63 0.56 0.44 0.60 0.37 0.38 0.44 1.76 0.65 0.57 0.47 0.47 0.55 0.52 0.55 0.62 0.60 0.45 0.45 0.57 1.61 0.63 0.61 0.60 0.51 0.52 1.69 1 M 0.82 0.68 0.76 -0.68 -1.26 -0.69 -0.66 -0.84 -1.17 -0.75 -1.44 -1.39 -1.17 -0.62 -0.81 -1.08 -1.09 -0.85 -0.95 -0.87 -0.69 -0.73 -1.15 -1.16 -0.80 -0.68 -0.72 -0.74 -0.98 -0.95 NZ7608 and WCFS1 (79 genes). 2 E 2 ox3 hisI tkt aroI ttdB tyrA hisH hisC hisD aroF aroA aroB aro araB hsp1 hsp p cysK ansB Gene moeA aroD1 cps1B cps1E cps1G Locus lp_1701 lp_0129 lp_2668 lp_1178 lp_1183 lp_1085 lp_2553 lp_2034 lp_1084 lp_2033 lp_1494 lp_1181 lp_2557 lp_2551 lp_1086 lp_2035 lp_2830 lp_2037 lp_0256 lp_2559 lp_1721 lp_3491 lp_1089 lp_3556 lp_2629 lp_2979 lp_2980 lp_1357 lp_3045 lp_1083 TABLE TABLE 5 Summary of global transcriptional analysis comparing the response to respiratory cultivation in tobacillus plantarum ≥1.5.

102 4 The thioredoxin system plays an important role in adaptation of Lactobacillus plantarum WCFS1 to aerobic and respiratory growth. Protein fate (1%) Other categories (3%) Protein synthesis (6%) Regulatory functions (9%) Hypothetical proteins (28%) Transport and binding proteins (18%) Purines, pyrimidines, nucleosides and nucleotides (1%) strain NZ7608 vs wild-type. L. plantarum log (cy5/cy3) of 2 BC transporter, ATP-binding protein BC transporter, ATP-binding protein BC transporter, substrate binding protein A A A branched-chain amino acid ABC transporter, ATP-binding protein branched-chain amino acid ABC transporter, ATP-binding protein branched-chain amino acid ABC transporter, permease protein branched-chain amino acid ABC transporter, permease protein branched-chain amino acid ABC transporter, substrate binding protein Na(+)/H(+) antiporter glutamine ABC transporter, permease protein maltose/maltodextrin ABC transporter, permease protein galactitol PTS, EIIC manganese transport protein transport protein nucleotide-binding protein, universal stress protein UspA family deoxyribose-phosphate aldolase ribosomal protein L31 ribosomal protein S15 regulatory protein Spx regulatory protein Spx transcription antiterminator preprotein translocase, SecE subunit pseudouridylate synthase ribosomal protein L11 ribosomal protein L19 transcription regulator transcription regulator transcription regulator transcription regulator integral membrane protein integral membrane protein integral membrane protein (putative) unknown unknown unknown unknown unknown unknown unknown unknown unknown unknown unknown unknown unknown unknown unknown prophage P2b protein 20 prophage P2b protein 21 0.63 0.45 0.60 0.51 0.51 0.60 0.61 0.61 0.56 1.74 0.48 1.58 0.53 1.77 0.56 0.50 0.52 0.45 0.65 0.60 0.61 0.65 1.68 0.57 1.84 0.64 0.63 0.59 0.40 0.56 0.59 1.88 1.70 1.69 1.57 0.62 0.62 0.51 0.64 0.56 0.64 0.48 0.56 1.57 0.57 0.59 0.63 1.64 0.51 0.80 0.66 0.82 0.75 0.88 0.91 0.77 0.76 0.65 0.65 0.71 -0.67 -1.14 -0.75 -0.98 -0.98 -0.74 -0.70 -0.72 -0.83 -1.06 -0.91 -0.83 -1.01 -0.94 -1.14 -0.63 -0.75 -0.70 -0.61 -0.82 -0.65 -0.67 -0.76 -1.32 -0.83 -0.76 -0.68 -0.69 -0.96 -0.65 -0.85 -0.65 -1.06 -0.84 -0.80 -0.75 -0.68 -0.97 2 2 E K livE livB livA livD livC rpl rplS truB ts36C spx1 spx3 rpsO malG secE glnP rpm deoC bglG4 napA4 mntH p lp_2038 lp_2823 lp_0783 lp_2982 lp_2981 lp_2984 lp_2983 lp_2985 lp_3601 lp_0177 lp_2992 lp_1958 lp_2314 lp_3565 lp_2125 lp_0497 lp_0836 lp_2228 lp_3514 lp_1396 lp_2902 lp_2903 lp_3625 lp_0183 lp_1900 lp_3301 lp_1163 lp_0438 lp_0513 lp_0927 lp_0984 lp_0995 lp_1088 lp_1187 lp_1577 lp_1834 lp_2036 lp_2061 lp_2244 lp_2269 lp_2300 lp_2669 lp_2461 lp_2460 lp_0616 lp_2032 lp_0619 lp_1640 lp_0512 M ratio defined as the M= 1

103 Characterization of gene trxB2 in L. strain NZ77603 and that heat stress response plantarum WCFS1 is up regulated in strain NZ78603 compared to While gene trxB1 encoding for TR in L. plantarum wild-type under these conditions. Consequent- has been suggested to play a role in oxidative ly, it could be expected that under respiratory stress response, less is known about gene trxB2. conditions strain NZ7603 would have a lower Previous results showed that the transcript trxB2 growth rate and a higher heat stress tolerance was induced under heat stress conditions (26). compared to wild-type To further characterize gene trxB2 in L. plantarum the difference in response at the transcriptome level towards respiratory cultivation of L. plantarum strain NZ7603 compared to wild-type was investigated. A dataset of 370 significantly affected genes (Supplementary material Table S2, see page 195) was generated: 35% (128 genes) of these genes were up regulated and 65% (242 genes) were down regulated in strain NZ7603 compared to wild-type. The up-regulated genes in strain NZ7603 compared to wild- type include genes predicted to encode for proteins involved in cellular processes (dnaK, asp1, asp2, grpE ); DNA metabolism (uvrC, recQ2); protein fate (mrsA1); regulatory functions (HrcA), 18 transcription regulators, and purine and pyrimidine biosynthesis (nrdF). Within the set of genes which were found down regulated (65%) in strain NZ7603 compared to wild-type there were genes predicted to encode proteins involved in cofactor biosynthesis (trxA2), cell envelope proteins (mur); cellular processes (asp3, npr2, tpx); DNA metabolism (mutS, uvrA1); regulatory functions (spx1, cggR, fur); energy metabolism (gapB, pox3, pox5, ldhD, araA, araB); protein fate (peptidylpropil isomerases), and purines and pyrimidine biosynthesis (pyr genes, trxB1). Hence, transcriptome analysis suggests that principal cellular processes are more affected in

104 4 The thioredoxin system plays an important role in adaptation of Lactobacillus plantarum WCFS1 to aerobic and respiratory growth.

DISCUSSION a 19% reduction in growth rate (26). The global In this study the effect of alterations in the thiore- transcription profile of this disruption mutant doxin system at the transcriptome level was as- showed among other effects, reduced activity sessed In particular, the “mock stress” response of the purine biosynthesis (26). This study has shown in a trxB1-overexpressing L. plantarum shown that the global transcriptional response strain (26), the phenotype of a trxB1-disrupted of strain NZ7608 (trxB1::cam) under anaerobic strain (Chapter 3) and a trxB2-disrupted strain conditions is identical (with exception of the up- (this study) have been studied in more detail and regulation of the plantaricin operon) to that of under different growth conditions with different wild-type under aerobic cultivation. The pheno- oxygen regimes. By applying a global transcrip- type of NZ7608 under anaerobic conditions is tional analysis detailed characterization of the not significantly different to that of the wild-type. thioredoxin system in L. plantarum has been Further characterization showed that specific achieved and at the same time we have estab- growth rate in strain NZ7608 was 31% lower lished new insights in the regulatory network ex- compared to wild-type when grown under re- erted by TRX levels at the transcript level gained. spiratory conditions. Growth on a medium sup- A disruption of trxB1 is expected to prevent the plied with diamide did not result in a significant reduction of TRX and he would bring about a thi- growth impairment of strain NZ7608 compared ol imbalance in the cell manifested as a general to wild-type. At the transcriptome level the spe- stress response towards exposure to oxygen (kat, cific growth impairment of strain NZ7608 under pox3), even under anaerobic conditions. On the aerobic conditions correlated to six genes pre- other hand a trxB2 disruption under oxidative dicted to encode catalase (kat), pyruvate oxidase respiratory conditions resulted in the induction (pox3), glyceraldehyde-3-phosphate (gapB), of genes predicted to encode proteins correlated and methionine-S-oxide reductase (mrsA3) and to heat stress response in bacteria and down- two putative transcription regulators (lp_1360 regulation of general stress response genes (fur, and lp_0889). These six genes can be catego- spx) compared to the wild-type. Furthermore, rized into three different behavioral patterns. this study offers insights into understanding the The first behavioral pattern contains genes regulation of stress mechanisms present in L. (kat, gapB, and lp_0889). These three genes plantarum. These are that adaptation to oxy- are significantly up regulated in strain NZ7608 gen is correlated to nine genes in L. plantarum, compared to wild-type under anaerobic (8-fold) including pox5 and pox3 and the transcription and aerobic (2-fold) conditions while under re- regulators Lp_1360, and Lp_0889 predicted to spiratory conditions these genes do not show encode regulators of the MarR protein family of significant differences compared to wild-type. It transcription factors regulators (13). should be noted that these three genes were to When grown aerobically, disruption of the trxB1 found to up regulated (4- to 8- fold) in the wild- gene encoding for TR in L. plantarum, results in type under oxidative conditions underlining their a strain with a 2.5-fold decreased TR activity and important role in oxidative stress response in L.

105 plantarum. A second behavior pattern includes wild-type seems to be energy deficient needing genes mrsA3 and lp_1360. These two genes do regeneration of NADH (gapB) under anaero- not show any significant difference in expression bic and aerobic conditions. These observations when NZ7608 is grown anaerobically or under could be explained if glyceraldehydes-3-phos- respiratory conditions compared to wild-type. phate dehydrogenase encoded by gapB exerts Nevertheless, when strain NZ7608 is grown aer- a high degree of control on the glycolytic flux as obically and compared to wild-type both genes suggested for Lc. lactis by Poolman et al. (24). If are significantly up regulated compared to the so, the reduced expression of gapB would par- wild-type. The third behavioral pattern observed tially explain the observed growth impairment. in adaptation of strain NZ7608 to oxygen is rep- Alternatively, a study of Solem et al. (27) showed resented by the pox3 gene. This gene is signifi- that glyceraldehyde-3-phosphate dehydroge- cantly (eight-fold) up-regulated in strain NZ7608 nase has no control over the glycolytic flux inLc. compared to wild-type grown anaerobically, lactis. In addition the homofermentative behav- four-fold up regulated in NZ7608 compared to ior of strain NZ7608 in this study contradicts this wild-type grown aerobically and two-fold down assumed imbalance in the NADH/NAD+ ratio. regulated in NZ7608 compared to wild-type under respiratory conditions. To this behavioral It is known that aerobic cultivation leads to oxi- pattern could be added araB, with the excep- dation of proteins containing the -CxxC- motif, tion that this gene was not found significantly such as methionine sulfoxide reductase (MRS) up regulated in strain NZ7608 compare to wild- and TRX. Therefore, it is speculated that un- type when grown under aerobic conditions. der aerobic conditions both MRS (encoded by mrsA3) and TRX would be induced or activated The analysis of the global transcriptional re- to alleviate oxidative stress. Nevertheless, the sponse of strain NZ7608 compared to the trxB1 mutant strain, which is expected to have a wild-type under respiratory cultivation condi- reduced capacity to activate TRX, seems to adapt tions showed that the expression profile of strain by up regulating of expression of the gene mrsA3 NZ7608 differed significantly from wild-type un- to higher levels than the wild-type and in this der respiratory conditions in 79 genes. The re- way compensate for the reduced activity of the sponse of strain NZ7608 compared to wild-type thioredoxin system. On the other hand, reduced differs in the up regulation of genes predicted to activity of the thioredoxin system would hamper be involved in growth. On the other hand over- activation of MRS since the thioredoxin system expression of trxB1 under respiratory conditions is expected to function as a hydrogen donor in compared to wild-type showed a significant the reduction of methionine sulfoxide catalyzed down regulation of gene trxA2 and genes for ri- by MRS in E. coli (4). Since the trxB1 disruption bosomal proteases and obviously the trxB1 were mutant still possesses TR activity (Chapter 3), the up regulated. Thus, strain NZ7608 compared to latter effect is probably not relevant. Our study

106 4 The thioredoxin system plays an important role in adaptation of Lactobacillus plantarum WCFS1 to aerobic and respiratory growth.

shows that trxB1 disruption triggers a response non-active proteins and heat stress tolerance of the cell that leads hydrogen peroxide produc- and other stress related genes as well as the tion via pyruvate oxidase (pox3) and subsequent plantaricin operon were found to be up regu- consumption via catalase (kat), even under an- lated in strain NZ7603 compared to wild-type aerobic conditions. Moreover, these results sug- under respiratory conditions. In Gram-positive gest that mrsA3 is significantly induced in strain bacteria two different heat stress transcription NZ7608 under aerobic conditions suggesting repressors are known: HrcA and CstR (9). HrcA an altered thiol-homeostasis in the cytoplasm. is annotated as a heat-inducible transcription repressor in B. subtilis and regulates of dnak, and Characterization of trxB2, in previous stud- groELS gene expression while CstR regulates the ies showed that this transcript was induced in clp gene family. In L. plantarum the genes: hrcA, L. plantarum as a result of heat shock (26). In grpE, and dnaK, are located in one operon and addition, because TRXB2 only has a 28% ho- transcriptome analysis suggests that HrcA in mology with TRXB1 and does not contain an plantarum is an activator and not a repressor active -CxxC- center it was speculated that the under the studied conditions. Hence, these re- gene trxB2 is not an active TR. In this study it is sults show that under conditions that allow oxi- clear that a trxB2 deficient L. plantarum strain, dative respiration, strain NZ7603 induces genes NZ7603, when grown under anaerobe, aerobe, predicted to encode proteins correlated to heat or respiratory conditions does not have a clear stress response in bacteria and down regulation phenotype compared to wild-type. This obser- of general stress response genes compared to vation was confirmed at the transcriptome level the wild-type. where the plantaricin operon and the gene pox3 coding for pyruvate oxidase are the only sig- Adaptation of L. plantarum to an aerobic envi- nificant differences between NZ7603 and wild- ronment is reflected at the transcriptome level by type grown aerobically. Moreover, the global differential expression of nine genes predicted to transcriptional analysis of strain NZ7603 under encode: pyruvate oxidases (pox3, pox5), a cata- conditions that allow oxidative respiration, has lase (kat), a methionine-S-reductase (mrsA3), shown that a trxB2 disruption results in a major two putative oxidoreductases (lp_2113 and impact at the transcriptional level when com- lp_1939), two putative transcription regulators pared to respiring wild-type cultures. Disruption (lp_0889 and lp_1360), and a precursor of thia- of trxB2 was found to result in the down-regu- mine biosynthesis (dxs). Stevens et al. (29) have lation of genes encoding proteins involved in found up-regulation of pox3 coding for pyruvate ferric uptake system, thiol-homeostasis, osmotic oxidase in L. plantarum WCFS1 during aerobic stress, stress related genes, central metabolism, growth. Furthermore, pox5 or poxB and pox3 or and the putative transcription factor (lp_1360). poxF have been characterized in L. plantarum Conversely, genes involved in degradation of lp80 by Lorquet et al. (18) as the only contribu-

107 tors for pyruvate oxidase (POX) activity under (mrsA3 aerobe) was further investigated using aerobic conditions from the 5 present paralogs comparative genomics. The analysis revealed in L. plantarum. In total five paralogs of pox that both TFs are closely related to OhrR of B. genes have been identified in the genome of subtilis, which was shown to regulate the genes L. plantarum WCFS1 (16). The observed simul- coding for glutathione reductases (gor) and TRX taneous up-regulation of the gene coding for (trxA) (11). Moreover, a specific TF operator mo- catalase (kat) is functionally linked with pyruvate tif could be defined for both TFs, very similar to oxidase activity since the latter enzyme delivers that of OhrR. A genome-wide search for these the substrate (hydrogen peroxide) for catalase. motifs yielded putative minimal regulons which In addition, it is well known that L. plantarum included for the Lp_0889 regulon the genes: can accumulate manganese (Mn) up to micro- gapB, kat, pox3, lp_2113, lp_1880, and the molar concentrations and uses this metal to acti- pln cluster. On the other hand two genes where vate the L. plantarum non-heme Mncatalase (14, predicted to be under the regulation of Lp_0360 22). Exposure of L. plantarum cultures to oxygen regulon: lp_0064 and mrsA3. Furthermore, in during exponential growth was found to be as- the trxB1-disruption strain L. plantarum NZ7608 sociated with up regulation of a gene predicted the genes spx1, spx3 predicted to encode the to encode a well known redox damage repair SPX protein were found significantly down regu- protein, methionine sulfoxide reductase (mrsA3), lated compared to the wild-type. Moreover, in as well as up regulation of genes predicted to under respiratory conditions in strain NZ7608 encode putative oxidoreductases. This response the gene spx1 is down regulated compared to clearly reflects the oxidative stress the microbes the wild-type. From other studies it is known that are sensing during aerobic growth. the regulatory protein SPX plays an important role in stress survival and the degradation of Analysis of the global transcriptional profiles mis-folded proteins. The SPX protein contains presented in this study pinpoint leads into the a -CxxC- motif, thus is susceptible to oxidative network of response mechanisms activated in stress (21) and has been suggested to act as a the bacterium while dealing with stress. Earlier negative regulator of proteolysis homologue to studies have shown that the control of oxida- TrmA (9). In addition, disruption of TrmA in Lc. tive stress response in Lc. lactis is modulated lactis leads to a higher stress tolerance (temper- by sigma factors, two component systems, and ature and accumulation of non-native proteins) metabolite flux sensors (6). At the same time, while levels of SPX protein in B. subtilis have the mechanisms of stress response in Lc. lactis been shown to result in the induction of trxA and have been extensively studied and in include trxB1 (21). Other well-known regulators include the genes nox, npr, sod, recA, cydA (19). Es- PerR, and FurR. The PerR (homolog to lp_3247) pecially the potential role of the transcription has been found to regulate the vegetative kat factors Lp_0889 (kat aerobe) and Lp_1360 and tpx genes in B. subtilis, (20) (11); and,

108 4 The thioredoxin system plays an important role in adaptation of Lactobacillus plantarum WCFS1 to aerobic and respiratory growth.

the respiratory ferric uptake regulator (FurR) is known to be induced under oxidative stress (20) and OhrR (lp_1360) upon exposure to organic peroxides. In this study using bioinformatics suggests that the genes regulated by trxB1, lp_1360 and lp_0889 included the regulatory SPX protein (spx1), catalase (kat), and MRS (mrsA3) respectively. Nevertheless, more studies need to be conducted in L. plantarum to decipher the above mentioned regulons and their respective working networks.

In conclusion, the application of a global transcriptional analysis to further characterize the thioredoxin system in L. plantarum has been performed. The transcriptome analysis of strains NZ7608 and NZ7603 has revealed that trxB1 is involved in oxidative stress response while trxB2 is most likely involved in heat stress response in L. plantarum. Moreover, it has been shown that adaptation to oxygen is correlated to nine genes in L. plantarum including pox5 and pox3 and the transcription regulators Lp_0889 and Lp_1360. Both transcription regulators are homolog to the well known oxidative regulator OhrR of B. subtilis. New insights into the regulatory network exerted by TRX and by oxygen stress at the transcript level has been established.

109 REFERENCES

1. Altschul, S. F., T. L. Madden, A. A. Schaffer, J. Zhang, Z. Zhang, W. Miller, and D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25:3389-402. 2. Bailey, T. L., and C. Elkan. 1994. Fitting a mixture model by expectation maximization to discover motifs in biopolymers. Proc Int Conf Intell Syst Mol Biol 2:28-36. 3. Bailey, T. L., and M. Gribskov. 1998. Combining evidence using p-values: application to sequence homology searches. Bioinformatics 14:48-54. 4. Boschi-Muller, S., S. Azza, S. Sanglier-Cianferani, F. Talfournier, A. Van Dorsselear, and G. Branlant. 2000. A sulfenic acid enzyme intermediate is involved in the catalytic mechanism of peptide methionine sulfoxide reductase from Escherichia coli. J Biol Chem 275:35908-13. 5. Brooijmans, R. J., B. Poolman, G. K. Schuurman-Wolters, W. M. de Vos, and J. Hugenholtz. 2007. Generation of a membrane potential by Lactococcus lactis through aerobic electron transport. J Bacteriol 189:5203-9. 6. Duwat, P., B. Cesselin, S. Sourice, and A. Gruss. 2000. Lactococcus lactis, a bacterial model for stress re sponses and survival. Int J Food Microbiol 55:83-6. 7. Edgar, R. C. 2004. MUSCLE: a multiple sequence alignment method with reduced time and space complex ity. BMC Bioinformatics 5:113. 8. Finn, R. D., J. Mistry, B. Schuster-Bockler, S. Griffiths-Jones, V. Hollich, T. Lassmann, S. Moxon, M. Marshall, A. Khanna, R. Durbin, S. R. Eddy, E. L. Sonnhammer, and A. Bateman. 2006. Pfam: clans, web tools and services. Nucleic Acids Res 34:D247-51. 9. Frees, D., P. Varmanen, and H. Ingmer. 2001. Inactivation of a gene that is highly conserved in Gram- positive bacteria stimulates degradation of non-native proteins and concomitantly increases stress tolerance in Lactococcus lactis. Mol Microbiol 41:93-103. 10. GEO http://www.ncbi.nlm.nih.gov/geo/, posting date. Gene Expression Omnibus (GEO). [Online.] 11. Helmann, J. D., M. F. Wu, A. Gaballa, P. A. Kobel, M. M. Morshedi, P. Fawcett, and C. Paddon. 2003. The global transcriptional response of Bacillus subtilis to peroxide stress is coordinated by three transcription factors. J Bacteriol 185:243-53. 12. Helmann, J. D., M. F. Wu, P. A. Kobel, F. J. Gamo, M. Wilson, M. M. Morshedi, M. Navre, and C. Paddon. 2001. Global transcriptional response of Bacillus subtilis to heat shock. J Bacteriol 183:7318-28. 13. Hong, M., M. Fuangthong, J. D. Helmann, and R. G. Brennan. 2005. Structure of an OhrR-ohrA operator complex reveals the DNA binding mechanism of the MarR family. Mol Cell 20:131-41. 14. Igarashi, T., Y. Kono, and K. Tanaka. 1996. Molecular cloning of manganese catalase from Lactobacil lus plantarum. J Biol Chem 271:29521-4. 15. Kimura, M. 1981. Estimation of evolutionary distances between homologous nucleotide sequences. Proc Natl Acad Sci U S A 78:454-8. 16. Kleerebezem, M., J. Boekhorst, R. van Kranenburg, D. Molenaar, O. P. Kuipers, R. Leer, R. Tarchini, S. A. Peters, H. M. Sandbrink, M. W. Fiers, W. Stiekema, R. M. Lankhorst, P. A. Bron, S. M. Hoffer, M. N. Groot, R. Kerkhoven, M. de Vries, B. Ursing, W. M. de Vos, and R. J. Siezen. 2003. Complete genome sequence of Lactobacillus plantarum WCFS1. Proc Natl Acad Sci U S A 100:1990-5. 17. Li, Y., J. Hugenholtz, W. Sybesma, T. Abee, and D. Molenaar. 2005. Using Lactococcus lactis for glutathione overproduction. Appl Microbiol Biotechnol 67:83-90. 18. Lorquet, F., P. Goffin, L. Muscariello, J. B. Baudry, V. Ladero, M. Sacco, M. Kleerebezem, and P. Hols. 2004. Characterization and functional analysis of the poxB gene, which encodes pyruvate oxidase in Lactobacillus plantarum. J Bacteriol 186:3749-59. 19. Miyoshi, A., T. Rochat, J. J. Gratadoux, Y. Le Loir, S. C. Oliveira, P. Langella, and V. Azevedo. 2003. Oxidative stress in Lactococcus lactis. Genet Mol Res 2:348-59.

110 4 The thioredoxin system plays an important role in adaptation of Lactobacillus plantarum WCFS1 to aerobic and respiratory growth.

20. Mostertz, J., C. Scharf, M. Hecker, and G. Homuth. 2004. Transcriptome and proteome analysis of Bacillus subtilis gene expression in response to superoxide and peroxide stress. Microbiology 150:497-512. 21. Nakano, S., E. Kuster-Schock, A. D. Grossman, and P. Zuber. 2003. Spx-dependent global transcriptional control is induced by thiol-specific oxidative stress in Bacillus subtilis. Proc Natl Acad Sci U S A 100:13603-8. 22. Nierop Groot, M. N., and J. A. de Bont. 1999. Involvement of manganese in conversion of phenylala nine to benzaldehyde by lactic acid bacteria. Appl Environ Microbiol 65:5590-3. 23. Overbeek, R., N. Larsen, T. Walunas, M. D’Souza, G. Pusch, E. Selkov, Jr., K. Liolios, V. Joukov, D. Kaznadzey, I. Anderson, A. Bhattacharyya, H. Burd, W. Gardner, P. Hanke, V. Kapatral, N. Mikhailova, O. Vasieva, A. Osterman, V. Vonstein, M. Fonstein, N. Ivanova, and N. Kyrpides. 2003. The ERGO genome analysis and discovery system. Nucleic Acids Res 31:164-71. 24. Poolman, B., B. Bosman, J. Kiers, and W. N. Konings. 1987. Control of glycolysis by glyceraldehyde- 3-phosphate dehydrogenase in Streptococcus cremoris and Streptococcus lactis. J Bacteriol 169:5887-90. 25. Sambrook, J. F., E.F.; Maniatis T. 1989. Molecular Cloning: A Laboratory manual. , 2nd edition ed, vol. 1-3. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. 26. Serrano, L. M., D. Molenaar, M. Wels, B. Teusink, P. A. Bron, W. M. de Vos, and E. J. Smid. 2007. Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1. Microb Cell Fact 6:29. 27. Solem, C., B. J. Koebmann, and P. R. Jensen. 2003. Glyceraldehyde-3-phosphate dehydrogenase has no control over glycolytic flux in Lactococcus lactis MG1363. J Bacteriol 185:1564-71. 28. Starrenburg, M. J., and J. Hugenholtz. 1991. Citrate Fermentation by Lactococcus and Leuconostoc spp. Appl Environ Microbiol 57:3535-3540. 29. Stevens, M. J. A. 2008. Transcriptome Response of Lactobacillus plantarum to Global Regulator Deficiency, Stress and other Environmental Conditions. PhD thesis. Wageningen University. 30. Teusink, B., F. H. van Enckevort, C. Francke, A. Wiersma, A. Wegkamp, E. J. Smid, and R. J. Siezen. 2005. In silico reconstruction of the metabolic pathways of Lactobacillus plantarum: comparing predictions of nutrient requirements with those from growth experiments. Appl Environ Microbiol 71:7253-62. 31. Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin, and D. G. Higgins. 1997. The CLUST AL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 25:4876-82. 32. van der Heijden, R. T., B. Snel, V. van Noort, and M. A. Huynen. 2007. Orthology prediction at scalable resolution by phylogenetic tree analysis. BMC Bioinformatics 8:83. 33. van Kranenburg, R., J. D. Marugg, S. van, II, N. J. Willem, and W. M. de Vos. 1997. Molecular characterization of the plasmid-encoded eps gene cluster essential for exopolysaccharide biosynthesis in Lactococcus lactis. Mol Microbiol 24:387-97. 34. Vido, K., H. Diemer, A. Van Dorsselaer, E. Leize, V. Juillard, A. Gruss, and P. Gaudu. 2005. Roles of thiore doxin reductase during the aerobic life of Lactococcus lactis. J Bacteriol 187:601-10. 35. Voelker, L. L., and K. Dybvig. 1998. Characterization of the lysogenic bacteriophage MAV1 from Myco plasma arthritidis. J Bacteriol 180:5928-31. 36. Wheeler, D. L., T. Barrett, D. A. Benson, S. H. Bryant, K. Canese, V. Chetvernin, D. M. Church, M. Dicuccio, R. Edgar, S. Federhen, M. Feolo, L. Y. Geer, W. Helmberg, Y. Kapustin, O. Khovayko, D. Landsman, D. J. Lipman, T. L. Madden, D. R. Maglott, V. Miller, J. Ostell, K. D. Pruitt, G. D. Schuler, M. Shumway, E. Sequeira, S. T. Sherry, K. Sirotkin, A. Souvorov, G. Starchenko, R. L. Tatusov, T. A. Tatusova, L. Wagner, and E. Yaschenko. 2007. Database resources of the National Center for Biotechnology Information. Nucleic Acids Res.

111 112 5 Glutathione protects Lactobacillus plantarum WCFS1 against hydrogen peroxide stress

L. Mariela Serrano, Douwe Molenaar, Bas Teusink, Willem M. de Vos, Eddy J. Smid

Manuscript in preparation

113

113 114 5 Glutathione protects Lactobacillus plantarum WCFS1 against hydrogen peroxide stress.

ABSTRACT

Glutathione and thioredoxin are widely occurring disulfide reducing compounds that maintain intracellular redox homeostasis to protect against oxidative damage. In this study we analyzed the physiological effects of the controlled production of glutathione in Lactobacillus plantarum WCFS1. As this strain does not produce detectable amounts of glutathione, we introduced a plasmid carrying the Escherichia coli genes gshA and gshB cloned under the control of the inducible nisin promoter. The resulting L. plantarum strain NZ7609 synthesized high amounts of glutathione following nisin induction. Survival tests showed that this glutathione production protects L. plantarum WCFS1 against hydrogen peroxide stress in spite of the absence of glutathione production in the wild-type L. plantarum WCFS1. The genome of L. plantarum WCFS1 contains the gene lp_2336, predicted to encode a homolog of a glutathione synthase fusion protein detected in Listeria monocytogenes. Disruption of lp_2336 influenced the growth of L. plantarum under hydrogen peroxide stress and showed a longer lag phase compared to the wild-type. This phenotype was partially complemented when glutathione was added to the medium. However, under the studied conditions, over expression of lp_2336 did not result in detectable glutathione production in L. plantarum or in a higher resistance towards exposure to hydrogen peroxide. The role of lp_2336 in hydrogen peroxide stress tolerance was further characterized using a global transcriptional analysis. The response towards hydrogen peroxide in a lp_2336-disrupted strain compared to wild-type differed in the genes predicted to encode the thioredoxin system and stress related genes (trxA1, trxB2) and stress related genes (spx, clp, mrsA3). These results suggest an overlap of the thioredoxin and glutaredoxin system in L. plantarum WCFS1 and elucidate its complex network in response towards oxidative stress.

115115 INTRODUCTION Recently, studies have shown that GSH provides Thiol compounds such as thioredoxin (TRX) and protection against oxidative stress (18) and acid glutathione (GSH) are essential for many differ- stress (41) in Lactococcus lactis. In the genome ent organisms including Lactobacillus plantarum of L. plantarum WCFS1 (16) the genes lp_2324 (4). These cofactors have been shown to play a and lp_2336 have been annotated as gene role as antioxidant (34), in maintaining intracel- gshA coding for a glutamate-cysteine ligase lular thiol homeostasis (12), and oxidative stress while lp_0369, lp_1822, and lp_1253 are pre- response (10, 40). Bacteria are able to maintain dicted to encode GR. This annotation suggests an active GSH or TRX pool in the cell through that L. plantarum WCFS1 has the capacity to the glutaredoxin/thioredoxin system. The glu- synthesize the GSH intermediate g-glutamylcys- taredoxin system recycles GSH via glutathione teine. On the other hand, L. plantarum does not reductase (GR). On the other hand the TRX sys- seem to have the gene gshB coding for the en- tem recycles TRX by reducing TRX via thioredoxin zyme catalyzing the final step in the glutathione reductase (TR). Both systems use NADPH as a biosynthesis pathway. Nevertheless, a recent cofactor. A link between the glutaredoxin and study by Gopal et al. reported the presence of a thioredoxin systems has been reported in sev- fusion protein in Listeria monocytogenes, GSHF eral organisms (3, 13, 15, 27). Moreover stud- (lmo_2770) (8). This fusion protein consists of ies have shown both systems to be regulated by a g-glutamylcysteine ligase domain fused to an stress and environmental changes (9, 11, 32). ATP-grasp domain. Gopal and co-workers have shown that GSHF is responsible for the produc- The currently available L. plantarum WCFS1 tion, in vitro, of GSH. L. plantarum WCFS1 was genome sequence (16) shows the presence of found to contain a homolog of lmo_2770; the trxA and trxB genes coding for the production lp_2336 gene. of TRX and TR respectively. TRX is reported to be involved, as electron acceptor in a process This paper describes the characterization of the leading to protection against oxygen-damage glutaredoxin system using metabolic engineer- (29). We have recently showed that a func- ing techniques and functional analysis. We have tional TR is important for an adequate oxidative found that GSH protects L. plantarum towards stress response in L. plantarum (30). The bio- hydrogen peroxide stress. Furthermore, we have synthetic pathway of GSH, a tripeptide g-glu- determined that the gene product of the GSHF tamylcysteinylglycine (g-Glu-Cys-Gly) has been homolog lp_2336 plays a role in hydrogen per- well documented (20). This highly conserved oxide response but does not affect the GSH pool pathway includes two essential ATP dependent size in L. plantarum under the studied conditions. steps catalyzed by g-glutamylcysteine synthase The presented results add to our understanding (gshA) and glutathione synthase (gshB). GSH of the mechanisms present in L. plantarum to has been shown to act as an essential cofac- cope with oxidative stress and can be used by tor in amino acid transport, sulphur metabolism, industry to optimize strain robustness in indus- ion transport, and detoxification of the cells (26). trial fermentation processes.

116 5 Glutathione protects Lactobacillus plantarum WCFS1 against hydrogen peroxide stress.

MATERIALS AND METHODS (28). Lc. lactis was grown in M17 at 30°C and Bacterial strains, plasmids, media, and L. plantarum WCFS1 was grown at 37°C in de culture conditions. The bacterial strains used Man-Rogosa-Sharp (MRS) or in Chemically De- in this study are summarized in Table 1. Esch- fined Medium (CDM) (36). erichia coli strains were grown at 37°C in TY

TABLE 1. Strains and plasmids used in this study

Strain Characteristics Reference

DH5Į E. coli

NZ9000 Lc. lactis (1, 6)

WCFS1 L. plantarum WCFS1. (16)

NZ7100 L. plantarum WCFS1 derivative with chromosomal integration of pEMnisRK (30)

plasmid.

NZ7602 CmR, L. plantarum WCFS1 derivative carrying the pMS040 plasmid. (30)

L. plantarum WCFS1 trxB2::cam derivative with chromosomal integration of NZ7603 plasmid pMS004. (chapter 4)

NZ7606 CmR; L. plantarum NZ7100 derivative carrying the pNZ8150 plasmid. (30)

NZ7607 CmR; L. plantarum WCFS1 derivative carrying the pNZ7021 plasmid. (30)

L. plantarum WCFS1 trxB1::cam derivative with chromosomal integration of NZ7608 (chapter 4) plasmid pMS019.

NZ7618 L. plantarum WCFS1 gshf::cam derivative with chromosomal integration of This work

pMS046.

NZ7616 L. plantarum WCFS1 derivative carrying plasmid pMS047. This work

NZ7609 L. plantarum WCFS1 derivative carrying plasmid pNZ3203. This work

NZ7610 L. plantarum WCFS1 derivative carrying plasmid pNZ3204 This work

Plasmids

pNZ5318 CmR EmR; pNZ5317 derivative for multiple gene replacements containing las (17)

and pepN terminators

pMS046 CmR; pNZ5318 derivative carrying 1.0 kb DNA fragments of L. plantarum This work

WCFS1 lp_2335 and lp_2337.

pNZ7021 CmR; pNZ8148 derivative carrying nisA::pepN. (39)

pMS047 CmR; pNZ7021 derivative carrying L. plantarum lp_2336 gene, translational This work

fused to the pPEPN promoter.

pNZ3203 CmR; pNZ8148 derivative containing a functional E.coli gshA and gshB gene. (19)

pNZ3204 CmR; pNZ7021 derivative containing a functional E.coli gshA and gshB gene (19)

117 DNA Manipulations. All molecular biology Construction of L. plantarum strain techniques were performed following estab- NZ7618 (lp_2336::cam). A chromosomal lished protocols by Sambrook (28). DNA was di- lp_2336::cam gene replacement was construct- gested according to the conditions recommended in L. plantarum. For this purpose, upstream ed by the commercial suppliers of the restriction and downstream flanking regions of lp_2336 enzymes (Boehringer, Breda, The Netherlands). were amplified by proofreading PCR using ge- In all cases, DNA was eluted from 0.7 % (wt/ nomic DNA of L. plantarum WCFS1 as template. vol) agarose gels using the Purification Kits from The primers used to amplify the 1kb upstream Promega (Leiden, The Netherlands). For a PCR region were: 5’-CCGCTCGAGCGGTCG- reaction we used: 1 μl template DNA (10 to 100 TATCCAACACCATGAC-3’ and 5’- TCGAAA- ng), 2 μl of each primer combination (50 ng/ml), CAAATTTGCCGCGCGGCCGACTAGTTTT- 1 μl dNTP’s (100 nM), 1 μl PWO-polymerase (5 GCAAGGAG-3’. For the 1kb downstream U/μl) and 10 μl polymerase buffer (5x) (Roche, flanking region the primers used were: 5’- Woerden, The Netherlands). The reaction mix- TCCCCCGGGGGATAGCTCATAATTAAGTGTC tures were adjusted to 50 μl with demineralized -3’ and 5’- GGAAGATCTTCCGACTCATGATA- water. ACTAAGGC -3’ (restriction sites underlined). Construction of L. plantarum strain The PCR amplicons obtained were digested with NZ7616 (lp_2336 overexpression). The the restriction enzymes XhoI and PmeI, or with gene lp_2336 in L. plantarum WCFS1 was SmaI and BglII respectively and cloned in two amplified by proofreading PCR using genomic steps into plasmid pNZ5318 digested with the DNA from L. plantarum WCFS1 as template corresponding restriction enzymes. The genetic and two primers: 5’-GGCCCCGGGGTACCAT- organization of the resulting plasmid, desig- GGAATTAGATGCCGTTGG-3’ and 5’- GC- nated pMS046, was verified by PCR, restriction GCCTAGTCTAGAGAAATGGGTGTTTTCAAT- analysis and sequencing. Subsequently, plas- CATGACACTTA-3’ with KpnI and XbaI restriction mid pMS046 was introduced into L. plantarum sites (sites underlined in the primer used). The WCFS1 and primary integrants were selected on amplified DNA fragment was cloned into the basis of CmR. The anticipated configuration of linearized vector pNZ7021. The complete plas- the pMS046 plasmid integration in the lp_2336 mid, designated pMS047, was cloned and puri- locus was confirmed by PCR and Southern blot- fied from Lc. lactis NZ900 cells (14) displaying ting (data not shown). The lp_2336::cam deriv- chloramphenicol resistance (CmR). Further, plas- ative of L. plantarum WCFS1 was designated L. mid pMS047 was introduced into L. plantarum plantarum NZ7618. NZ7100 (30); colonies displaying CmR were Control strains. In the experiments with the ni- checked by restriction and sequencing analysis. sin inducible promoter we used as control strain One of these transformants containing the L. NZ7606: a L. plantarum NZ1700 strain contain- plantarum lp_2336 gene translationally fused to ing plasmid pNZ8150. L. plantarum NZ7607, the constitutive PpepN promoter (35) was denomi- carrying an empty plasmid, was used as con- nated L. plantarum NZ7616.

118 5 Glutathione protects Lactobacillus plantarum WCFS1 against hydrogen peroxide stress.

trol in growth studies analyzing the phenotype (1:20) and grown at 37ºC until OD600=1.0. of over-expression strains. For the experiments When reached the desired cell density we added including the disruption strains, the wild-type L. 10 μl of 8.8 M hydrogen peroxide solution to plantarum WCFS1 was used as control. each culture (final concentration 10mM). After Nisin induction. Nisin induction was done as 30 minutes of hydrogen peroxide challenge the previously described elsewhere (25) using 50 pellet was washed twice with phosphate buffer ng/ml nisin for induction. pH=7.5 by centrifugation to remove any trace of Growth curves. In all growth experiments cell hydrogen peroxide. Cell pellets were suspended density was measured by detecting the turbidity in fresh medium (10ml) and used for growth or of the cultures at 600 nm every 30 min. Also, spe- survival tests. cific growth rate (μmax) was defined as ΔLnOD600 Survival to hydrogen peroxide. To de- / Δtime during the exponential growth phase of termine survival rates of L. plantarum WCFS1, each culture. During incubation of the cultures, NZ7618, NZ7606, and NZ7609 we grew the temperature was kept at 37ºC and cultures were cultures overnight at on CDM containing 0.5% stirred before every measurement. Inoculum for (wt/vol) glucose. Overnight cultures were trans- grown experiments was prepared as follows: ferred to10 ml fresh medium in a ratio (1:20) pre-cultures were washed twice in phosphate and grown at 37ºC until OD600=1.0. At this point buffer pH=7.5 and resuspended in the same cells were either washed or challenged with hy- amount of fresh CDM containing 0.5% (wt/vol) drogen peroxide as described above. Cell pel- glucose. Growth experiments were carried out in lets of challenged or non-challenged cells were 96-well microtiter plates, 10 μl of washed-over- resuspended in 10 ml of fresh medium and a night culture was transferred to each well con- lethal dose (150 μl of 8.8 M) of hydrogen per- taining 240 μl of fresh medium. For growth ex- oxide solution was added to each culture (final periments carried out in 250-ml shake flasks 1 concentration 30mM). Cultures were plated ev- ml of an overnight culture was transferred to 50 ery 10 minutes on and MRS agar and incubated ml fresh medium. Shake flasks were constantly overnight at 37ºC. Survival was determined via stirred at 125 rotations per minute (rpm) in a colony forming units (CFU). temperature controlled (37ºC) incubator. These GSH enzyme assay. GSH (reduced form) was cultivations were carried out in GSH-free CDM assayed by enzymatic recycling based on the (36) containing 0.5% (wt/vol) glucose. Depend- procedure of Tiestze (37) and described else- ing on the experiment CDM was supplemented where (18). For these assays cell free extracts with GSH or hydrogen peroxide to final concen- were used and prepared as described elsewhere trations of 5mM and 10mM respectively. (30). Hydrogen peroxide challenge. Cultures HPLC total Glutathione (tGSH) determina- were grown overnight at 37ºC on CDM contain- tion. To determine tGSH, (reduced form GSH + ing 0.5% (wt/vol) glucose. Grown cultures were oxidized GSH), in the medium as well as inside transferred to 10 ml fresh medium in the ratio cells we used the mBBr fluorescent labelling de-

119 scribed by Mansoor et al.(21) with some modifi- Global transcriptional analysis: cations. To a 60 μl aliquot of cell free extract of Cultivations. L. plantarum strains WCFS1 and supernatant we added 60 μl of 2.0M NaBH4/0.5 NZ7618 were grown overnight at 37ºC on 10 ml M NaOH and 120 μl 20%sulfosalycylic acid with CDM containing 0.5% (wt/vol) glucose. Grown 100 μM DTT. This mix was left on ice for 30 min. cultures were transferred to 400 ml CDM con- After incubation on ice, to 120 μl of the mix we taining 0.5% (wt/vol) glucose in a ratio of (1:20). added 60 μl 1.4 M NaBH4/0.5M NaOH, 200 Cultures were grown until OD600=1.0 at 37ºC. μl solution B (65% DMSO, 35% water, 51 mM When cell density was reached, we sampled 100

NaCl and 140 mM HBr), 100 μl 1.0 M N-Ethyl- ml of each culture (t0) for RNA extraction. After maleimide and 20 μl 20 mM mBrB and 40 μl of sampling, a non-lethal dose (0.30 ml of 8.8 M) 1.06 M PCA. tGSH was detected by chromatog- of hydrogen peroxide was added to each culture raphy. Samples of 25 μl were injected into a 150 (final concentration 10 mM) and we took 100 x 4.6 mm, 3 μm PLRP-S column, equipped with ml samples after 10 min (t10) and 30 min (t30) for a PLRP-S guard column. The flow rate was 1 ml/ RNA extraction. min at 30°C. Elution solvent A was 0.1% trifluoric RNA isolation, cDNA synthesis, and labelling. acid (TFA) and 5% acetonitrile and solvent B was RNA extraction, cDNA synthesis and labelling of 0.1% TFA and 80% acetonitrile, both diluted with cDNA for each sample was done as described distilled water. The elution profile was as follows: elsewhere (30). 0-20 min, 0% B; 20-25 min, 16% B; 25-30min Hybridization scheme. A total of 16 hy- 50% B. The retention time of bimane derivatives bridizations were used to compare strains L. of GSH was 23 min. We used a Spectra Physics, plantarum WCFS1 and NZ7618 (Fig.1). This Spectra Systems FL2000 fluorometer detector hybridization scheme takes into consideration operating at excitation wavelength 394 nm and dye-swap, biological duplos, and is designed emission wavelength 480 nm. Plotting and in- in such a way that to determine mutation and tegration of peaks were performed by chrome- hydrogen peroxide significant changes on each leon version 6.6 from Dionex. To correlate peak studied strains at the different time points as well height with amount of GSH on the samples, a as to compare the response towards hydrogen standard curve was done with pure GSH. For peroxide between L. plantarum strains WCFS1 the standard curve, samples with known GSH and NZ7618. In each array two cDNA labeled concentrations (0-1mM) were prepared in water targets were hybridized on custom designed L. and treated other samples. The concentration of plantarum WCFS1 Agilent oligo microarrays GSH on the analyzed samples was obtained are available at Gene Expression Omnibus (GEO) expressed as μmol total GSH /ml supernatant or with accession number GPL5874 (7) using the μmol total GSH /g protein. Agilent 60-mer oligo microarray processing protocol version 4.1.

120 5 Glutathione protects Lactobacillus plantarum WCFS1 against hydrogen peroxide stress.

FIGURE 1. Hybridization scheme for the global transcriptional analysis of L. plantarum strains WCFS1 and NZ7618. Each arrow represents one array.

1 2 1W0 1W10 1W30 6 3

1ǻ 5 1ǻ 4 1ǻ 0 10 30 16 14 13 15 2ǻ 8 7 2ǻ 0 2ǻ30 10 9 12

2W0 2W30 2W10 10 11

Data acquisition and processing. Was done RESULTS as described elsewhere (Chapter 3) with some The glutaredoxin system in L. planta- modifications. The normalized data has been rum WCFS1 made available at GEO (7) with accession num- The biosynthesis of GSH includes two ATP-de- ber GSE9961. Significant genes were defined as pendent steps catalyzed by g-glutamylcysteine genes with a False Discovery Rate (FDR) adjusted synthase (gshA) and glutathione synthase (gshB). for the pvalues of less than 5%. The statistical test In general, Gram-positive bacteria are thought resulted in different sets of data. One dataset to be unable to synthesize GSH through this representing the genes affected as a result of the common route because most of them lack ei- mutation of lp_2336 compared to L. plantarum ther one or both genes coding for these enzymes WCFS1 (mutation), several datasets represent- in their genomes. Studies have shown that Lc. ing the genes affected due to hydrogen perox- lactis acquires GSH through uptake from the ide stress at different times (0, 10, 30) on each growth medium (20). This lactic acid bacterium studied culture, and finally datasets representing was found to accumulate GSH in response to different transcript response towards hydrogen different environmental stresses (18). Further- peroxide stress at different times of NZ7618 more, a recent study by Gopal et al, demon- compared to L. plantarum WCFS1. strated that the Gram-positive pathogen Listeria monocytogenes contains the lmo_2770 gene that encode the fusion protein GSHF. In vitro ex-

121 periments showed that GSHF is responsible for Glutathione uptake in L. plantarum the production of GSH (8). The pool of active WCFS1 GSH (reduced) inside the cell is kept constant by L. plantarum WCFS1 was found to actively take recycling of GSH by glutathione reductase (GR) up GSH. Following growth to OD600 of 1.0 in a using NADPH as donor of reducing equivalents. medium containing 5mM GSH, the GSH uptake This system is known as the glutaredoxin sys- amounted to 1.49 ± 0.15mM. When L. planta- tem. In the annotated genome of L. plantarum rum WCFS1 was grown in a medium contain- WCFS1 (16) g-glutamylcysteine synthase (gshA) ing 5mM GSH and 10mM hydrogen peroxide, and GR have been annotated while a homolog we found the same GSH uptake concentrations of gshB is absent. We found that in the genome (data not shown). These results were found re- of L. plantarum WCFS1 the gene lp_2336 is a gardless whether the L. plantarum cells were homolog of GSHF from Li. monocytogenes. We pre-cultured in a medium supplemented with therefore hypothesized that lp_2336 may play a GSH or not. In L. plantarum WCFS1, the intra- role in de novo biosynthesis of GSH in L. plan- cellular GSH pools were found to be below the tarum. Consequently, overexpression of gene detection limits of our enzyme assay. Because lp_2336 was expected to result in higher intra- of the enzyme assay is relatively insensitive, the cellular GSH levels. Based on earlier findings in total intracellular GSH pool in two different L. Lc. lactis a higher tolerance towards hydrogen plantarum strains was measured by HPCL peroxide is correlated to the intracellular GSH (Table 2). pool (5). In order to alter GSH levels in L. plantarum we constructed strains NZ7616, NZ7618 TABLE 2. Intracellular total GSH pools and NZ7609 (Table 1). L. plantarum NZ7616 (μmol GSH/ gr. protein) measured by contains a plasmid carrying the gene lp_2336 HPLC method in different L. plantarum- under the control of the constitutive PpepN de- strains. rived from Lc. lactis (35). Furthermore, L. plan- Amount Amount tarum NZ7618 carries an lp_2336 replacement [µmol/g prot] [µmol/g prot] (lp_2336::cam). Finally, L. plantarum strain Strain Glutathione Glutathione NZ7609 is a derivative of L. plantarum NZ7100 NZ7606 n.a. n.a. containing a plasmid carrying genes gshA and NZ7606* n.a. n.a. gshB from E. coli under the control of the nisin NZ7609 112.70 22.39 * regulated promoter Pnis (22). NZ7609 238.30 59.31 *Nisin induction (5 hours 50 ng/ml)

122 5 Glutathione protects Lactobacillus plantarum WCFS1 against hydrogen peroxide stress.

Again, in the wild-type strain (now with an empty ing for L. plantarum WCFS1 at a concentration control plasmid), we found no detectable GSH of 0.03 μmol/ g protein. The intracellular GSH pool while L. plantarum NZ7609 carrying the pool measured at different cell densities showed gshA and gshB genes from E. coli, was found that the pool was decreased with increasing to contain intracellular GSH in a concentration cell density. At a cell density of OD600=0.5 the of 59 μmol/ gr. protein. Even, in the absence intracellular GSH pool in all strains was mea- of the inducer nisin, we detected an intracel- sured to be above the 1. The pool values were lular GSH pool of 22 μmol of GSH/ g protein decreased to 0.14 μmol/ g protein or lower and in this strain. Because of the existing overlap to 0.12 μmol/ g protein or lower levels when between the glutaredoxin and thioredoxin sys- the GSH pool was measured at OD600=1.0 and tem in other organisms we measured the total OD600=2.0 respectively. GSH concentration in strains NZ7608, NZ7603, Exposure to a sub-lethal concentration of hy- corresponding to a TR trxB1 and trxB2 mutant drogen peroxide for 30 min at cell density of strains NZ7602 corresponding to a trxB1-over- OD600=1 .0 resulted in a lower intracellular expressing strain (Table 1). None of these three GSH pool in strains NZ7607, NZ7618, and strains produced detectable levels of GSH (data wild-type. Whereas hydrogen peroxide exposure not shown). Taking into consideration that pro- resulted in an increase in the intracellular GSH duction of GSH is stress-specific (20) we also pool in strains NZ7610 and NZ7616. Taken determined the amount of total GSH present into consideration that both strains NZ7610 in the cells after overnight cultivation, during and NZ7616 constitutively over-express the E. the exponential growth phase and in cells ex- coli genes gshA and gshB or the gene lp_2336, posed to 10mM hydrogen peroxide (Table 3). these results suggest that both strains NZ7610 In all different samples of L. plantarum cultures and NZ7616 affect the glutaredoxin system in L. traces of intracellular GSH were detected includ- plantarum WCFS1.

TABLE 3. Intracellular total-GSH concentration (μmol GSH/g protein) in different L. plantarum strains. Samples were taken at different cell density depicted as OD600 or from overnight cultures (O/N)

1 Total concentration of GSH expressed as μmol tGSH/gr. Protein. Where an assumption was made that 1ml of culture at

OD600=1 contains 155 ug protein. 2 Samples of cultures grown over night (O/N).

123 Hydrogen peroxide stress response in a similar response of the strains NZ7607 and L. plantarum NZ7616 towards hydrogen peroxide. On the To further characterize functionality of lp_2336 contrary, strain NZ7618 carrying an lp_2336:: in L. plantarum, we investigated the effect of hy- cam replacement, showed a significantly longer drogen peroxide on growth of strains NZ7607 lag phase compared to the other two strains (control), NZ7618 (lp_2336::cam), and NZ7616 upon hydrogen peroxide challenge (Fig. 2). (lp_2336 overexpression). The growth response of cultures that were challenged for 30 min In addition, we determined specific growth rates to a high dose of hydrogen peroxide (10mM) of a number of L. plantarum strains grown under prior to growth was investigated and compared varying concentrations of hydrogen peroxide that response to growth of untreated cultures. ranging from 0 to 10mM (Table 4). The spe- The resulting patterns (Fig. 2) show that strain cific growth rate of strain NZ7618 was reduced NZ7607, NZ7616 and NZ7618 grow equally in by 17% in a medium containing a sub-lethal the absence of a hydrogen peroxide challenge. concentration of 5mM hydrogen peroxide. The All cultures that were challenged with hydrogen reduction of the specific growth was brought peroxide prior to inoculation showed a signifi- down to 13% when NZ7618 was pre-cultured cantly longer lag phase. In addition, we found overnight in a medium supplemented with 5mM

FIGURE 2. Growth behaviour of L. plantarum strains grown on CDM containing 0.5% glucose (wt/vol). L. plantarum strains are represented as follows: NZ7607 (solid triangles), NZ7618 (solid squares), and NZ7616 (open squares). In addition all strains were either untreated (black symbols) or challenged with sub-lethal amounts of hydrogen peroxide for 10 minutes (gray symbols).

1.2

0.8

600 0.6 OD

0.4

0.2

0 0 2 4 6 8 10 12 14 Time (hours)

124 5 Glutathione protects Lactobacillus plantarum WCFS1 against hydrogen peroxide stress.

GSH, suggesting a protective effect of GSH. these strains have been incubated with GSH Furthermore, when strain NZ7618 was cultured overnight and subsequently grown on CDM. overnight on a medium containing GSH and The growth difference of cultures of the studied then cultured with 5mM or 10mM hydrogen per- strains was less pronounced in the presence of oxide the specific growth rate differs up to 19% GSH even if the cells had been challenged with compared to the control strain (WCFS1). Strain hydrogen peroxide (data not shown). Hence, the NZ7609 (GSH-producing) and strain NZ7610 amount of GSH in the medium seemed to in- (lp_2336 overexpression) show some different fluence growth rate of the L. plantarum cultures results. These strains seems to be able to cope under hydrogen peroxide stress. Furthermore, with hydrogen peroxide better than the control these results presented suggested that the sensi- strains NZ7606 and NZ7607 respectively even if tivity of L. plantarum WCFS1 is influenced by the the cells were challenged with hydrogen perox- presence of gene lp_2336. Therefore, we pro- ide prior to inoculation. On the other hand, the ceeded to check whether we could influence the specific growth rate of strains NZ7618, NZ7609, observed phenotype in strain NZ7618 by add- and NZ7610 is 13,1, and 4% less than the con- ing GSH in the medium. trols strains WCFS1 (0.49 h-1), NZ7606 (0.45 h-1) and NZ7607 (0.48 h-1) respectively when

TABLE 4. Specific growth rate of differentL. plantarum strains. Values shown in the table are shown as percentages and represent the difference in specific growth rate (μmax) of the strain compared to the specific growth rate of the control strain under the same growth conditions.

* µmax CDM CDM + 5mM CDM + 10mM A1 3 4 3 4 3 4 Strain n c n c n c NZ7618 13 19 3 17 13 12 NZ7609 5 6 -7 -1 -1 -3 NZ7610 7 5 -6 -3 -2 -1 B2 NZ7618 13 2 10 13 19 13 NZ7609 1 -3 4 10 6 9 NZ7610 4 -1 6 7 5 11

* Specific growth rates of strains WCFS1 (0.49 -1h ), NZ7606 (0.45 h-1), and NZ7607 (0.48 h-1) were taken as 100% for NZ7618, NZ7609 and NZ7610 respectively. Specific growth rate was determined as ΔLnOD/Δtime for each strain in triplicate at the linear or exponential growth. 1 L. plantarum strains grown overnight on CDM containing 0.5% (wt/vol). glucose 2 L. plantarum strains grown overnight on CDM containing 0.5% (wt/vol ) glucose and 5mM GSH. 3 Cultures grown until OD600=1.0 in CDM containing 0.5% (wt/vol) glucose. 4 Cultures grown until OD600=1.0 in CDM containing 0.5% (wt/vol) glucose and challenged with 10mM hydrogen peroxide.

125 Chemical complementation of observed or L. plantarum NZ7618 were challenged with phenotype. hydrogen peroxide prior to inoculation, no sig- Assuming that lp_2336 is involved in the produc- nificant differences between the different strains tion of GSH in L. plantarum WCFS1, we tested and medium compositions were found anymore whether supplementation of GSH would have an (Fig. 3). These results clearly show the protective effect on the growth behaviour of NZ7618 and effect of GSH during growth of L. plantarum in the wild-type in the presence of 10mM hydro- the presence of hydrogen peroxide. Because the gen peroxide (Fig. 3). Such an exercise showed lp_2366::cam replacement strain (L. plantarum that addition of GSH to the medium did shorten NZ7618) is more sensitive towards hydrogen the lag phase in strain NZ7618 compared to peroxide compared to the wild-type, we can also WCFS1 but not the full extent. Similar results conclude that the presence of an intact gene lp_ were found when using 5mM hydrogen peroxide 2366 is protecting the cells against the toxicity of in the medium (data not shown). Furthermore, hydrogen peroxide. when pre-cultures of either L. plantarum WCFS1

FIGURE 3. Growth behaviour of L. plantarum strains WCFS1 and NZ7618 grown on CDM (open symbols) or CDM + 5mM GSH (solid symbols). In both cases CDM contains 0.5% glucose (wt/vol) and 10mM hydrogen peroxide. L. plantarum strains are represented as follows: WCFS1 (triangles) and NZ7618 (squares). The inoculum of the cultures was either untreated (black symbols) or challenged

0.6

0.4 600 OD

0.2

0 0 5 10 15 20 25 Time (hours)

126 5 Glutathione protects Lactobacillus plantarum WCFS1 against hydrogen peroxide stress.

Survival upon exposure to lethal lev- plantarum WCFS1 (Fig. 4 Panel A) while sur- els of hydrogen peroxide. vival of strain NZ7609 was significantly higher To investigate the effect of glutathione on sur- than the control strain NZ7606 (Fig. 4 Panel B). vival of L. plantarum upon exposure to hydrogen Hence, these results suggest that intracellular peroxide, we analysed the behaviour of a gluta- production of GSH protects L. plantarum against thione producing strain, L. plantarum NZ7609 peroxide stress response and that lp_2336 is and a strain severely affected by hydrogen indeed involved in protection against hydrogen peroxide exposure, L. plantarum NZ7618 in a peroxide stress response in L. plantarum. To in- survival assay. For this assay, we used a dose crease our understanding of the physiological of 30mM of hydrogen peroxide and correlat- role of lp_2336 in the hydrogen peroxide stress ed survival with viable counts. This high dose response, we decided to perform a global tran- of hydrogen peroxide has proven to be lethal scriptional analysis of L. plantarum NZ7618 and for growing cultures of L. plantarum (data not L. plantarum WCFS1 under hydrogen peroxide shown). Survival of strain L. plantarum NZ7618 stress. was significantly lower than the control strain L.

FIGURE 4. Survival of L. plantarum strains WCFS1, NZ7618, NZ7606, and NZ7609 grown on CDM containing 0.5% glucose (wt/vol) towards 30mM hydrogen peroxide. All strains were either not-challenged (black) or challenged (gray) with hydrogen peroxide (10mM for 10 min). Panel A represents survival of L. plantarum strains WCFS1 (solid triangles) and NZ7618 (solid squares). Panel B represents survival of L. plantarum strains NZ7606 (solid triangles) and NZ7609 (solid squares).

AA BB 1.00E+09 1.00E+09

1.00E+08 1.00E+08

1.00E+07 1.00E+07

1.00E+06

1.00E+06 LOG cfu/ml LOGcfu/ml

1.00E+05 1.00E+05

1.00E+04 1.00E+04 0 10 20 30 0 10 20 30 Exposure time (min) Exposure time (min)

127 Global transcriptional analysis putative oxidoreductase (lp_0291); a putative The global transcriptional analysis of strains integral protein and (lp_2334); a transcription NZ7618 (lp_2336::cam) and WCFS1 was de- regulator (lp_2335) and an unknown (lp_1440) signed such (see Fig. 1) that it allowed us to de- compared to the wild-type. In addition, genes fine significantly the differentially affected genes predicted to be involved in energy metabolism responding to disruption of lp_2336 in L. plan- (aldH), cell envelope, transport systems (glycer- tarum as well as the genes responding towards ol) and unknown proteins were found up regu- hydrogen peroxide stress in each studied strain. lated in L. plantarum NZ7618 compared to the When grown under anaerobic conditions, L. wild-type. In summary, the consequences of a plantarum NZ7618 (disruption of lp_2336) dif- disruption of the gene lp_2336 in L. plantarum fered in 16 significant differentially expressed are specific to the operon where lp_2336 is lo- genes (Table 5) in comparison to the wild-type. cated (lp_2334 and lp_2335) and some cellu- We observed down-regulation of genes in L. lar processes such as transport and amino acid plantarum NZ7618 predicted to be involved in biosynthesis. amino acid biosynthesis (dapA2 and hisA); a

TABLE 5. Summary of significant affected genes (16) in L. plantarum NZ7618, when compared to the wild-type L. plantarum WCFS1 (pvalue < 0.05). Predicted gene names, function, fold change induction as well as main functional classes of the significant affected transcripts are displayed in columns.

Locus FC1 Gene Product Main Cathegory dapA lp_2685 3.15 2 dihydrodipicolinate synthase Amino acid biosynthesis phosphoribosylformimino-5-aminoimidazole lp_2556 2.37 hisA carboxamideribotide isomerase Biosynthesis of cofactors, prosthetic groups, and carriers lp_2771 -1.94 natC2 nicotinate phosphoribosyltransferase lp_3065 -2.26 cell surface protein precursor Cell envelope lp_0047 -3.44 aldH aldehyde dehydrogenase Energy metabolism lp_3631 -2.85 alpha-mannosidase (putative) lp_0104 2.26 unknown lp_0291 2.48 oxidoreductase lp_0960 -3.43 unknown Hypothetical proteins lp_1440 -2.86 unknown lp_2334 4.92 integral membrane protein lp_3524 -4.03 unknown lp_1191 4.82 transposase, fragment Other categories lp_2335 5.46 transcription regulator Regulatory functions lp_3650 -4.10 regulator glycerol-3-phosphate ABC transporter, permease Transport and binding proteins lp_1326 -11.02 protein (putative)

1 M Fold change (FC) calculated as 2 where M is the 2log (cy5/cy3) ratio of L. plantarum wild-type vs NZ7618.

128 5 Glutathione protects Lactobacillus plantarum WCFS1 against hydrogen peroxide stress.

The effect of exposure to 10mM hydrogen per- gen peroxide through up regulation of genes oxide on exponentially growing cultures of L. predicted to encode known stress response pro- plantarum NZ7618 and L. plantarum WCFS1 teins (trxA2, trxA3, trxH, gshR, npr2, kat, hsp2, was studied by analyzing the transcriptome at hsp3, fur, pox3, pox5, mrsA3, mrsA2) as well as different exposure times (0, 10, and 30 min). stress regulatory proteins (lp_1360, spx), pro- In L. plantarum WCFS1 a 10-min exposure to phages, proteins involved in copper transport hydrogen peroxide resulted in a significant dif- (copA, copB) and energy metabolism (pdhB, ferential expression of 41 genes (See supple- pdhC, xpk1, xpk2). In addition, hydrogen perox- mentary materials, Table S1, see page ---) ide exposure appeared to provoke a metabolic while a 30-min exposure resulted in a signifi- arrest in L. plantarum. This is suggested by the cant differential expression of 309 genes (See significant down regulation of genes predicted supplementary materials, Table S2, see page to encode proteins involved in central metabo- ---). After exposure to 10mM hydrogen perox- lism, energy metabolism, cell envelope, cell divi- ide for 10 min, the genes predicted to encode sion, cold shock proteins, DNA metabolism; as a glutathione reductase (gshR2); two pyruvate well as proteins involved in amino acid biosyn- oxidases (pox3 and pox5); two copper trans- thesis; protein synthesis, purine and pyrimide port proteins (cobA, cobB); a pyruvate dehy- biosynthesis, and EPS synthesis (See supple- drogenase (pdhB); a methionine-S-oxide reduc- mentary materials, Table S2, see page 208). In tase (mrsA3); two glutamine transport proteins summary, after a 30-min exposure to hydrogen (glnQ4, glnH2), and a regulatory protein (spx1), peroxide both strains seem to react by inducing among others were found to be up-regulated in stress response mechanisms (heat stress, oxida- L. plantarum. In addition, exposure to 10mM hy- tive stress, general stress) while the cells seem to drogen peroxide for 10 min was found to result be in growth arrest. This is concluded from the in the down-regulation in L. plantarum WCFS1 observed down regulation of cellular processes, of genes predicted to encode a TRX (trxA1); a DNA metabolism, and purine and pyrimidine ferrous iron transport protein (feoB); a putative biosynthesis at the transcriptome level when the glycerol-2-phosphate ABC transporter protein cells have been exposed to 10mM hydrogen (lp_1326) and a choloylglycine hydrolase (bsh1) peroxide for 30 min. among others (See supplementary materials, Table S1, see page 206). A 30-min exposure The response to exposure towards hydrogen to 10mM hydrogen peroxide resulted in a more peroxide after 30 min in L. plantarum WCFS1 extreme effect on the transcriptome of L. planta- compared to L. plantarum NZ7618 was ana- rum WCFS1 with 309 significantly differentially lyzed. Such an exercise resulted in 61 genes expressed genes (See supplementary materials, that responded differently in strain WCFS1 Table S2, see page ---). L. plantarum seemed compared to strain NZ7618 after a 30-min to respond to a longer exposure time to hydro- hydrogen peroxide exposure (See Supplemen-

129 tary materials, Table S3, see page 217). Out of DISCUSSION these 61 genes, the ones predicted to encode We have shown that GSH protects L. plantarum a TR (trxB2), a heat shock protein (hsp2), two against hydrogen peroxide stress. methionine-S-sulfoxides (mrsA2, mrsA3) and In this study a L. plantarum GSH producing two proteins involved in copper transport (cobA, strain, NZ7609, was made by cloning the E. coli cobB) were found to be up regulated in strain genes gshA and gshB under the control of the WCFS1 compared to NZ7618. To this pattern inducible nisin promoter in an expression plas- we could add genes predicted to encode pro- mid. L. plantarum strain NZ7609 produces GSH teins involved in fatty acid phospholipid me- and improves survival of L. plantarum towards tabolism as well as in protein synthesis. On the 30mM hydrogen peroxide stress. Characteriza- other hand genes predicted to encode a TRX tion of the glutaredoxin system in L. plantarum (trxA1), TRX peroxidase (tpx), a regulatory pro- included gene lp_2336. The gene lp_2336, is a tein (spx3), a choloylgycine hydrolase (bsh3), a homolog of a glutathione synthase fusion pro- phosphate starvation-inducible protein (phoH), tein present in Li. monocytogenes (8). Disrup- and a Clp protease (clpX) responded at higher tion of this gene negatively affected growth of levels in strain NZ7618 compared to wild-type the bacterium under hydrogen peroxide. Over- after a 30-min exposure to 10mM hydrogen expression of this gene did not result in higher peroxide. Hence, there is a significant difference tolerance towards hydrogen peroxide or in GSH in response towards 10mM hydrogen peroxide production. The role of gene lp_2336 in oxida- in strains WCFS1 and NZ7618. The difference tive stress response in L. plantarum was further includes the TRX and the stress-related proteins studied using a global transcriptional approach. (spx3, clpX, mrsA3) as well as copper transport This exercise revealed that the response towards and fatty acid metabolism. hydrogen peroxide stress in a lp_2336 disrupted strain compared to wild-type differed in the genes predicted to encode thioredoxin system and stress related genes (trxA1, trxB2, spx3, clpX, mrsA3) suggesting an overlap of the thioredoxin and glutaredoxin system in L. plantarum in response towards oxidative stress.

Exposure towards sub-lethal concentrations of hydrogen peroxide at the transcriptome level in L. plantarum has been previously described by Stevens et al. (33) and Serrano et al. (30). Both studies showed that hydrogen peroxide has a major impact at the transcriptome level affecting

130 5 Glutathione protects Lactobacillus plantarum WCFS1 against hydrogen peroxide stress.

transcription of genes predicted to encode pro- stress and production of ROS and the long-term teins involved in stress response and energy me- or adaptation response during which the bac- tabolism. The overall gene expression profiles in terium tries to survive in the new situation by these studies (28, 30) also suggest growth arrest optimal tuning of the expression pattern of its in the cells upon hydrogen peroxide. The results genetic potential among other mechanisms. presented here are in accordance with these ob- GSH, is a well known tripeptide known to protect servations. This study sheds more light on the cells against hydrogen peroxide stress (18, 32, differential gene expression upon short and long 41). In this study we have shown that a GSH term exposure towards sub-lethal concentra- producing L. plantarum strain, NZ7609, has tions of hydrogen peroxide. Genes predicted to an increased survival capacity towards lethal encode stress related proteins and copper and concentrations of hydrogen peroxide (30mM) glutamine transporters are significantly affected compared to the wild-type. This result suggests after a short-term (10-min) as well as after a that GSH protects L. plantarum towards hydro- long-term (30-min) exposure to hydrogen per- gen peroxide stress. In addition, we have shown oxide in L. plantarum. This is the case for genes that L. plantarum WCFS1 can take up GSH from predicted to encode two pyruvate oxidase (pox3, the environment. However, the protective effect pox5), copper and glutamine transporters (cobA, of GSH in L. plantarum raises the question of cobB, glnQ2, glnH4), a glutathione reductase why this bacterium as well as other Gram-posi- (gshR2), a methionine-S-sulfoxide (mrsA3), a tives do not have the capacity to synthesize this pyruvate dehydrogenase (pdhB) and a regula- tripeptide (20). A possible explanation is that tory protein (spx1). Growth arrest in the cells and L. plantarum has evolved in ecological niches the induction of some transcriptional factors and with an abundance of GSH, i.e. food and feed other stress response proteins is suggested from products (38). Especially vegetables and plants the transcriptome analysis only after a long-term are well known to be glutathione-rich niches. In exposure towards hydrogen peroxide. Some of general, adaptive evolution processes in these the genes that follow this pattern are genes pre- glutathione-rich niches may have favoured the dicted to be involved in the thioredoxin system, presence of GSH transport systems and not copper transport and energy metabolism as well genes coding for the GSH biosynthesis path- as stress related proteins (kat, fur, pox3, pox5, way. While the GSH-transports systems present mrsA2, mrsA3, gshR2, npr2). Another gene that in Gram-positive bacteria include uncharacter- follows this pattern is lp_1360, which has been ized energy-dependent mechanisms (31), re- suggested to encode an oxidative stress regu- cent studies have identified a GSH transporter lator in L. plantarum (Chapter 4). These results in yeast and rice denominated Hgt1p (2) and suggest that exposure towards hydrogen per- OxGT1 (42) respectively. oxide in L plantarum at the transcriptome level can be divided into two phases. The short-term Biosynthesis of GSH is carried out by two en- or immediate reaction towards dealing with the zymes gshA and gshB. The annotated genome

131 of L. plantarum WCFS1 (16) predicts that this study suggests that the response towards hydro- bacterium lacks one enzyme (gshB) in the bio- gen peroxide in L. plantarum is affected by a dis- synthetic pathway. While L. plantarum does not ruption of lp_2336. Specifically, exposure of L. produce significant levels of GSH (this study), plantarum NZ7618 (carrying an lp_2336::cam there are several genera of in the group of Fir- replacement) to hydrogen peroxide resulted in micutes including Streptococcus, Enterococcus, a significant down regulation of genes predicted and Li. monocytogenes that do synthesize GSH to encode proteins involved in protein synthesis, (24). In Li. monocytogenes it was shown that pyrimidine biosynthesis, thiol homeostasis, and there is a fusion protein, GSHF, that catalyzes copper transport. These proteins are expected both reactions normally carried out by gshA and to react towards oxidative stress. Furthermore, gshB and produces GSH in this organism (8). A genes predicted to encode proteins involved in homolog to GSHF was found in then genome the thioredoxin system (trxA1, trxB2, tpx) as well of L. plantarum WCFS: lp_2336. Disruption of as regulatory proteins (spx3) of the thioredoxin lp_2336 influenced the growth of L. plantarum system (23) are affected in strain NZ7618 com- under hydrogen peroxide stress. This phenotype pared to wild-type under hydrogen peroxide was partially repaired when GSH was added stress. This result shows that disruption of gene to the medium. Over-expression of lp_2336 lp_2336 affects the glutaredoxin system and that did not result in abundant GSH production in in this case, under hydrogen peroxide the thio- L. plantarum or in significant higher resistance redoxin system plays an essential role in strain. towards hydrogen peroxide stress. Nevertheless, An overlap between the thioredoxin and glu- upon hydrogen peroxide some GSH found in a taredoxin system has been previously reported L. plantarum strain carrying an over expression in several organisms (3, 9, 13, 15). Therefore, of lp_2336, NZ7616. This result suggests that our results suggest that this is also the case for L. the production of GSH in L. plantarum could be plantarum WCFS1. regulated by the environment. In conclusion, we have shown that GSH protects At the transcriptome level disruption of lp_2336 L. plantarum against oxidative stress. Further- in L. plantarum resulted in a set of 16 signifi- more, due to the glutathione-rich niches where cantly affected genes compared to the wild-type. this bacterium is found it seems that this organ- Among other, the affected genes were predicted ism has evolved and lost the capacity to produce to encode proteins involved in amino acid bio- GSH while still being able to efficiently transport synthesis or were genes located in the same op- this tripeptide. Characterization of the glutare- eron as lp_2336. Moreover, if wild-type behavior doxin system in L. plantarum included gene lp_ under hydrogen peroxide and if the mechanisms 2336. Disruption of this gene negatively affected that the wild-type uses are our reference to re- growth of the bacterium under hydrogen perox- flect tolerance towards hydrogen peroxide, then, ide. Overexpression of this gene did not result the transcriptome analysis presented in this in higher tolerance towards hydrogen peroxide

132 5 Glutathione protects Lactobacillus plantarum WCFS1 against hydrogen peroxide stress.

or in GSH production. The role of gene lp_2336 in oxidative stress response in L. plantarum was further studied using a global transcriptional approach. This exercise revealed an overlap between the thioredoxin and glutaredoxin system in L. plantarum WCFS1.

133 REFERENCES

1. Bolotin, A., P. Wincker, S. Mauger, O. Jaillon, K. Malarme, J. Weissenbach, S. D. Ehrlich, and A. Sorokin. 2001. The complete genome sequence of the lactic acid bacterium Lactococcus lactis ssp. lactis IL1403. Genome Res 11:731-53. 2. Bourbouloux, A., P. Shahi, A. Chakladar, S. Delrot, and A. K. Bachhawat. 2000. Hgt1p, a high affinity glutathione transporter from the yeastSaccharomyces cerevisiae. J Biol Chem 275:13259-65. 3. Casagrande, S., V. Bonetto, M. Fratelli, E. Gianazza, I. Eberini, T. Massignan, M. Salmona, G. Chang, A. Holmgren, and P. Ghezzi. 2002. Glutathionylation of human thioredoxin: a pos sible crosstalk between the glutathione and thioredoxin systems. Proc Natl Acad Sci U S A 99:9745-9. 4. Fahey, R. C. 2001. Novel thiols of prokaryotes. Annu Rev Microbiol 55:333-56. 5. Fu, R. Y., R. S. Bongers, S. van, II, J. Chen, D. Molenaar, M. Kleerebezem, J. Hugenholtz, and Y. Li. 2006. Introducing glutathione biosynthetic capability into Lactococcus lactis subsp. cremoris NZ9000 improves the oxidative-stress resistance of the host. Metab Eng 8:662-71. 6. Gasson, M. J. 1983. Plasmid complements of Streptococcus lactis NCDO 712 and other lactic streptococci after protoplast-induced curing. J Bacteriol 154:1-9. 7. GEO http://www.ncbi.nlm.nih.gov/geo/, posting date. Gene Expression Omnibus (GEO). [Online.] 8. Gopal, S., I. Borovok, A. Ofer, M. Yanku, G. Cohen, W. Goebel, J. Kreft, and Y. Aharonowitz. 2005. A multidomain fusion protein in Listeria monocytogenes catalyzes the two primary activities for glutathione biosynthesis. J Bacteriol 187:3839-47. 9. Grant, C. M. 2001. Role of the glutathione/glutaredoxin and thioredoxin systems in yeast growth and re sponse to stress conditions. Mol Microbiol 39:533-41. 10. Grant, C. M., L. P. Collinson, J. H. Roe, and I. W. Dawes. 1996. Yeast glutathione reductase is required for protection against oxidative stress and is a target gene for yAP-1 transcriptional regulation. Mol Microbiol 21:171-9. 11. Guzzo, J., M. P. Jobin, F. Delmas, L. C. Fortier, D. Garmyn, R. Tourdot-Marechal, B. Lee, and C. Divies. 2000. Regulation of stress response in Oenococcus oeni as a function of environmental changes and growth phase. Int J Food Microbiol 55:27-31. 12. Holmgren, A. 1979. Reduction of disulfides by thioredoxin. Exceptional reactivity of insulin and suggested func tions of thioredoxin in mechanism of hormone action. J Biol Chem 254:9113-9. 13. Holmgren, A. 1989. Thioredoxin and glutaredoxin systems. J Biol Chem 264:13963-6. 14. Holo, H., and I. F. Nes. 1989. High-Frequency Transformation, by Electroporation, of Lactococcus lactis subsp. cremoris Grown with Glycine in Osmotically Stabilized Media. Appl Environ Microbiol 55:3119-3123. 15. Kanzok, S. M., A. Fechner, H. Bauer, J. K. Ulschmid, H. M. Muller, J. Botella-Munoz, S. Schneu wly, R. Schirmer, and K. Becker. 2001. Substitution of the thioredoxin system for glutathione reductase in Drosophila melanogaster. Science 291:643-6. 16. Kleerebezem, M., J. Boekhorst, R. van Kranenburg, D. Molenaar, O. P. Kuipers, R. Leer, R. Tarchini, S. A. Peters, H. M. Sandbrink, M. W. Fiers, W. Stiekema, R. M. Lankhorst, P. A. Bron, S. M. Hoffer, M. N. Groot, R. Kerkhoven, M. de Vries, B. Urs ing, W. M. de Vos, and R. J. Siezen. 2003. Complete genome sequence of Lactobacillus plantarum WCFS1. Proc Natl Acad Sci U S A 100:1990-5. 17. Lambert, J. M., R. S. Bongers, and M. Kleerebezem. 2007. Cre-lox-Based System for Multiple Gene Deletions and lectable-Marker Removal in Lactobacillus plantarum. Appl Environ Microbiol 73:1126-35. 18. Li, Y., J. Hugenholtz, T. Abee, and D. Molenaar. 2003. Glutathione protects Lactococcus lactis against oxidative stress. Appl Environ Microbiol 69:5739-45. 19. Li, Y., J. Hugenholtz, W. Sybesma, T. Abee, and D. Molenaar. 2005. Using Lactococcus lactis for glutathione overproduction. Appl Microbiol Biotechnol 67:83-90. 20. Li, Y., G. Wei, and J. Chen. 2004. Glutathione: a review on biotechnological production. Appl Microbiol Biotechnol 66:233-42. 21. Mansoor, M. A., A. M. Svardal, and P. M. Ueland. 1992. Determination of the in vivo redox status of cysteine, cyste inylglycine, homocysteine, and glutathione in human plasma. Anal Biochem 200:218-29. 22. Mierau, I., and M. Kleerebezem. 2005. 10 years of the nisin-controlled gene expression system (NICE) in Lactococcus lactis. Appl Microbiol Biotechnol 68:705-17.

134 5 Glutathione protects Lactobacillus plantarum WCFS1 against hydrogen peroxide stress.

23. Nakano, S., E. Kuster-Schock, A. D. Grossman, and P. Zuber. 2003. Spx-dependent global transcriptional control is induced by thiol-specific oxidative stress in Bacillus subtilis. Proc Natl Acad Sci U S A 100:13603-8. 24. Newton, G. L., K. Arnold, M. S. Price, C. Sherrill, S. B. Delcardayre, Y. Aharonowitz, G. Cohen, J. Davies, R. C. Fahey, and C. Davis. 1996. Distribution of thiols in microorganisms: mycothiol is a major thiol in most actinomycetes. J Bacteriol 178:1990-5. 25. Pavan, S., P. Hols, J. Delcour, M. C. Geoffroy, C. Grangette, M. Kleerebezem, and A. Mercenier. 2000. Adaptation of the nisin-controlled expression system in Lactobacillus plantarum: a tool to study in vivo biological effects. Appl Environ Microbiol 66:4427-32. 26. Penninckx, M. J., and M. T. Elskens. 1993. Metabolism and functions of glutathione in micro-organisms. Adv Microb Physiol 34:239-301. 27. Prinz, W. A., F. Aslund, A. Holmgren, and J. Beckwith. 1997. The role of the thioredoxin and glutare doxin pathways in reducing protein disulfide bonds in theEscherichia coli cytoplasm. J Biol Chem 272:15661-7. 28. Sambrook, J. F., E.F.; Maniatis T. 1989. Molecular Cloning: A Laboratory manual. , 2nd edition ed, vol. 1-3. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. 29. Scharf, C., S. Riethdorf, H. Ernst, S. Engelmann, U. Volker, and M. Hecker. 1998. Thioredoxin is an essential protein induced by multiple stresses in Bacillus subtilis. J Bacteriol 180:1869-77. 30. Serrano, L. M., D. Molenaar, M. Wels, B. Teusink, P. A. Bron, W. M. de Vos, and E. J. Smid. 2007. Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1. Microb Cell Fact 6:29. 31. Sherrill, C., and R. C. Fahey. 1998. Import and metabolism of glutathione by Streptococcus mutans. J Bacte riol 180:1454-9. 32. Stephen, D. W., and D. J. Jamieson. 1996. Glutathione is an important antioxidant molecule in the yeast Saccharomyces cerevisiae. FEMS Microbiol Lett 141:207-12. 33. Stevens, M. J. A. 2008. Transcriptome Response of Lactobacillus plantarum to Global Regulator Deficiency, Stress and other Environmental Conditions. PhD thesis. Wageningen University. 34. Sundquist, A. R., and R. C. Fahey. 1989. Evolution of antioxidant mechanisms: thiol-dependent peroxi dases and thioltransferase among procaryotes. J Mol Evol 29:429-35. 35. Tan, P. S., I. J. van Alen-Boerrigter, B. Poolman, R. J. Siezen, W. M. de Vos, and W. N. Kon ings. 1992. Characterization of the Lactococcus lactis pepN gene encoding an aminopeptidase homologous to mammalian aminopeptidase N. FEBS Lett 306:9-16. 36. Teusink, B., F. H. van Enckevort, C. Francke, A. Wiersma, A. Wegkamp, E. J. Smid, and R. J. Siezen. 2005. In silico reconstruction of the metabolic pathways of Lactobacillus plantarum: comparing predictions of nutrient requirements with those from growth experiments. Appl Environ Microbiol 71:7253-62. 37. Tietze, F. 1969. Enzymic method for quantitative determination of nanogram amounts of total and oxi dized glutathione: applications to mammalian blood and other tissues. Anal Biochem 27:502-22. 38. Vescovo, M., S. Torriani, F. Dellaglio, and V. Bottazzi. 1993. Basis characteristics, ecology and ap plication of Lactobacillus plantarum: a review. Ann.Microbiol.Enzimol. 43:261-284. 39. Wegkamp, A., W. van Oorschot, W. M. de Vos, and E. J. Smid. 2007. Characterization of the role of paraaminobenzoic acid biosynthesis in folate production by Lactococcus lactis. Appl Environ Microbiol 73:2673-81. 40. Zeller, T., and G. Klug. 2006. Thioredoxins in bacteria: functions in oxidative stress response and regula tion of thioredoxin genes. Naturwissenschaften 93:259-66. 41. Zhang, J., R. Y. Fu, J. Hugenholtz, Y. Li, and J. Chen. 2007. Glutathione protects Lactococcus lactis against acid stress. Appl Environ Microbiol 73:5268-75. 42. Zhang, M. Y., A. Bourbouloux, O. Cagnac, C. V. Srikanth, D. Rentsch, A. K. Bachhawat, and S. Delrot. 2004. A novel family of transporters mediating the transport of glutathione derivatives in plants. Plant Physiol 134:482-91.

135 136 6 Summary, Concluding Remarks, and Future Perspectives

137

137 138 6 Summary, Concluding Remarks, and Future Perspectives

Lactobacillus plantarum is a bacterium traditionally used as starter culture to produce fermented foods. The reliability of starter cultures is essential in terms of quality and functional properties (acidification, flavour, and texture) but also in terms of growth- perfor mance and robustness (20). For example, robustness in probiotic cultures is an essential characteristic since the bacteria should survive passage through the gastrointestinal tract resist intestinal flora, persist in the intestine and express the expected function under un- favourable growth conditions. The quality of the products of industrial fermentations using bacteria depends on the evolved defence mechanisms present in the fermenting bacteria to resist adverse conditions encountered during processing (oxidation, oxygen, temperature, acid, salts). Hence, in order to secure the proper use and performance of bacteria in industrial fermentations as well as to control the quality of products, we need to understand the different adaptation mechanisms present in this organism. This thesis describes a study with focus on understanding oxidative stress in the model lactic acid bacterium L. plantarum WCFS1. There are two known systems involved in oxidative stress and redox homeostasis in bacteria: the thioredoxin and glutaredoxin systems. In this study, we constructed a collection of L. plantarum WCFS1 strains with alterations in these systems. Using the constructed strains under different oxidative stress conditions (hydrogen peroxide, diamide (thiol-stress), aerobic growth, and respiratory growth), global transcription al analysis was performed and subsequently validated using different techniques including comparative genomics, q-PCR, and enzyme assays. The comparative genomics and the global transcriptional analysis of the oxidative stress response of L. plantarum presented in this thesis can be used for optimizing industrial strains.

139139 Oxidative Stress oxide stress. In addition, in Chapter 2 we deter- Throughout an industrial fermentation process, the mined the transcription of the genes trxA2 and presence of oxygen is unavoidable. Since the time trxB1 under conditions of oxidative stress using that oxygen began to accumulate in the earth’s at- q-PCR techniques. The protective effect of GSH mosphere (3.5 billion years ago), (micro)organisms in L. plantarum towards oxidative stress is de- have evolved mechanisms allowing them to main- scribed in Chapter 5. The research presented in tain a reducing cytosolic environment (4). The this thesis shows that in L. plantarum both the toxicity of oxygen is attributed to reactive oxygen thioredoxin and the glutaredoxin systems play . species (ROS) like OH (hydroxyl radical), H2O2 (hy- an important role in oxidative stress response. - drogen peroxide) and O2 (superoxide) (14). Living cells have developed ways to cope with oxygen toxicity by either preventing the formation of ROS, The thioredoxin system eliminating them (enzymatic degradation and The thioredoxin system is a highly conserved scavenging), repairing the damage caused, or ren- mechanism in living organisms. This system can dering the most vulnerable targets (20). be found within Archaea as well as in Bacteria and Animalia (23). The members of this system The ability to respond to ROS requires mecha- (TRX, TR) are also highly conserved elements; nisms to minimize the occurrence of thiol oxida- nevertheless, TR has evolved. The evolution of tion and to mitigate its consequences. Thiols can TR from bacteria to mammals includes acquiring serve as sensors to oxidative stress as well as a broader substrate specificity and an extra re- antioxidants (3). Two of these ubiquious thiols dox catalytic site (1). The thioredoxin system in L. in nature are glutathione (GSH) (17) and thio- plantarum WCFS1 was extensively characterized redoxin (TRX) (9). TRX has been shown to play in Chapter 2 and Chapter 3. Using the com- a role in oxidative stress response in bacteria plete annotated genome sequence of L. plan- (15, 25) as well as GSH (13). These thiols are tarum WCFS1 (10) we predicted the members recycled inside the cell via their respective reduc- of the thioredoxin system which included four tase; glutathione reductase (GR) or thioredoxin genes predicted to encode TRX (trxA1, trxA2, reductase (TR). Thus the oxidized thiol, reduced trxA3, trxH) and two genes predicted to encode thiol, and reductase form a system denominat- TR (trxB1, trxB2). In Chapter 2 we showed that ed the thioredoxin or the glutaredoxin system. the gene trxB1 in L. plantarum WCFS1 encoded The thioredoxin system in L. plantarum WCFS1 TR. Overexpression of the trxB1 gene in L. plan- was studied in Chapter 2 and Chapter 3 where tarum resulted in a 3-fold increased TR activity we have identified the trxB1-encoding TR as a while a disruption of trxB1 resulted in a 2.5 fold key enzyme in the oxidative stress response of decrease in TR activity (Chapter 3) compared to L. plantarum WCFS1. Furthermore, in Chapter the wild type. 2 we showed that overexpression of trxB1 in L. plantarum WCFS1 resulted in strain with higher The thioredoxin system has been suggested to resistance towards diamide and hydrogen per- play an important role in aerobic life in Lac-

140 6 Summary, Concluding Remarks, and Future Perspectives

tococcus lactis (21). Adaptation to an aerobic larity to the well characterized TRX from Esch- environment is a known process for food-asso- erichia coli (9) and Bacillus subtilis (15). ciated lactic acid bacteria (LAB). Most LAB are facultative anaerobes but have been found to be The spectrum of reactions affected by TRX in aerotolerant and some are even able to respire; bacteria include reducing cytoplasmatic proteins this being the case of Lc. lactis upon the addition and scavenging hydrogen peroxide by providing of hemin and the presence of oxygen (2, 16). reducing equivalents to peroxiredoxin, as well as being an essential subunit of the bacteriophage Growth under aerobic conditions in a trxB1- T7 DNA polymerase and hydrogen donor to ri- disrupted L. plantarum strain was 19% affected bonucleotide reductase (25). In E. coli proteomic compared to the wild-type (Chapter 3). Aerobic studies revealed that TRX was associated with a and respiratory growth in L. plantarum is de- total of 80 proteins implicating the role of TRX in scribed in Chapter 4 where we present the es- more than 26 different cellular processes (11). sential role of a functional thioredoxin system in In L. plantarum WCFS1 transcriptome analyses adaptation of L. plantarum towards aerobic and presented in Chapters 2, 3, 4 suggest that the respiratory growth conditions. thioredoxin system is involve in several cellular processes in this bacterium. In Chapter 3 for A functional analysis of gene trxB2 was per- example, transcriptome analyses suggested that formed through q-PCR experiments, bioinfor- the thioredoxin system in L. plantarum WCFS1 matic tools (Chapter 2) and transcriptome data is involved in a wide range of processes includ- analysis of a trxB2-disrupted L. plantarum strain ing sugar and purine metabolism. In addition, (Chapter 4). These experiments showed that the in Chapter 2 and Chapter 4 alterations in the gene trxB2 did not contain a -CxxC- catalytic site thioredoxin system in L. plantarum WCFS1 re- (common trait of a TR); the gene trxB2 was in- sulted in a significant higher relative expression duced under heat stress, and that a disruption of of genes predicted to be involved in DNA repair gene trxB2 in L. plantarum resulted in a signifi- and stress mechanisms as well as genes associ- cant higher expression of genes predicted to en- ated with the activity of biosynthetic pathways for code heat stress proteins compared to the wild- purines and sulfur-containing amino acids. type. Hence, suggesting a role of gene trxB2 in heat stress response rather than oxidative stress response in L. plantarum WCFS1. Transcription- al analysis in Chapter 2 and Chapter 4 elucidate a possible role of genes trxA1, trxA2, trxA3, and trxH in response towards hydrogen peroxide. Homology studies, at the amino acid level, of the proteins encoded by these four genes suggest that the protein produced encoded by gene trxA2 L. plantarum WCFS1 has the highest simi-

141 Oxidative Stress response in L. planta- affected due to hydrogen peroxide stress as well rum WCFS1 as to overexpression of trxB1. The first promoter It is well established that LAB as well as other motif found in many of the hydrogen peroxide bacteria have evolved defence mechanisms to affected genes corresponded to the well-char- survive environmental changes and industrial acterized LexA-DinR regulon in B. subtilis known stresses (nutrient, cofactors, temperature, ROS, to induce proteins involved in DNA repair and etc) (19). Mechanisms involved in stress re- cell survival (24). In addition, a conserved mo- sponse in bacteria are known to be regulated at tif was found in the set of genes affected by an both transcription and metabolite level (14). In overexpression of trxB1. To our knowledge, this bacteria, transcription starts when a complex of is a novel regulon in L. plantarum which might RNA polymerase (RNAP) and a sigma factor (s) be TRX specific. Furthermore, in Chapter 4 we bind to specific sites in the DNA denominated used bioinformatic tools as well as promoter promoters. To our knowledge bacterial tran- sequence analysis to determine the functional scriptional regulation can be mediated via sig- role of lp_0889 and lp_1360. The latter genes ma factors (affect RNAP binding capacity) and lp_0889, lp_1360 were further characterized as by transcriptional factors (TFs) (affect activity of homologs of the well-known oxidative regulator RNAP). Even though in theory, every gene may OhrR of B. subtilis (8) and their associated gene have a regulator, organisms have developed network was predicted to include genes: gapB, networks to operate efficiently and minimize ge- pox3, kat, mrsA3. netic burden in the cell (20). The set of genes or regulon that is influenced by a sigma factor The transcriptome analyses presented in this or by one of the TFs is called a network. The thesis have also added information for under- genes within a network usually share a highly standing of adaptation processes following Redox stress conserved promoter motif. Identifying regulons oxidative stress in L. plantarum WCFS1 at the Oxygen and regulatory networks is essential to control, transcriptome level. In Chapter 4 we report that predict or engineer LAB behavior (20). Pioneer adaptation to oxygen in L. plantarum WCFS1 is Oxidative Stress work into unravelling the regulatory networks correlated to a set of nine genes including those present in L. plantarum using in-silico techniques coding for pyruvate oxidases pox5 and pox3 and

Hydrogen peroxide Heat (bioinformatics) and promoter analysis has been the putative transcription regulators coded by Thiol-imbalance done by Wels (22). Lp_0889 and Lp_1360. In Chapter 3 we determined that a fully functional thioredoxin system The study described in this thesis has made use of is essential for a sustained growth under aerobic ? comparative genomics, orthology predictions as and respiratory conditions. In Chapter 5 the re- well as bioinformatic tools and promoter analysis sponse towards hydrogen peroxide stress in an Lp_1360 Lp_3247 Lp_0889 CstR hrcA

Sulfur limitation to understand oxidative stress response and de- lp_2336 disrupted strain compared to wild-type SOS fine novel mechanisms in L. plantarum WCFS1. differed in the genes predicted to encode thio- FurR spx In Chapter 2 we identified two conserved motifs redoxin system and stress related genes (trxA1, using promoter analysis of the genes significantly trxB2, spx, clp, mrsA3). The gene lp_2336 is a

pox5 gpx tpx recA lexA uvrC dinB metEcysK fur trxB1 Lp_3128 kat ahpC mrsA3 lp_0064 pox3 gapB npr2 mntH2 gor trxA clpCclpP clpE dnaK groEL lp_3324 142 6 Summary, Concluding Remarks, and Future Perspectives

homolog of a glutathione synthase fusion pro- networks (Fig. 1). The major difference between tein, GSHF. This fusion protein is responsible for the well-studied transcriptome stress-response in production of GSH in vitro in Listeria monocyto- B. subtilis reviewed in the introduction of this the- genes (5). Therefore, these results suggest that sis and the model presented here (Fig. 1) is the there is a functional overlap of the thioredoxin absence of sB in L. plantarum WCFS1. The sigma and glutaredoxin system in L. plantarum in re- factor sB is known to control over 150 genes in B. sponse towards oxidative stress. subtilis (6). Oxidative stress response in this bacterium consists of a series of intertwined regulons and stress response mechanisms. Many genes CONCLUDING REMARKS have been omitted from this model to evade that The data-driven analyses of the oxidative stress the visual representation becomes a confusing response in L. plantarum WCFS1 described in spider web. In this way underlining as well the this thesis together with the literature on B. sub- necessity of new bioinformatic tools to support tilis (6-8, 12), and L. plantarum WCFS1 (18, 22) interpretation a vast amount of data generated and other bacteria (25) have given further in- by approaches like all the global transcriptional sight in the gene regulatory networks associated analysis used in this work. These tools should with this specific stress response and the role of include annotation, visualization, clustering, and the thioredoxin and glutaredoxin system in these libraries to assess biological function to genes.

FIGURE 1. Transcriptional networks activated in L. plantarum WCFS1 as a result of oxidative stress.

Redox stress

Oxygen

Oxidative Stress

Hydrogen peroxide Heat Thiol-imbalance

Lp_1360 Lp_3247 Lp_0889 CstR hrcA

Sulfur limitation SOS FurR

spx

pox5 gpx tpx recA lexA uvrC dinB metEcysK fur trxB1 Lp_3128 kat ahpC mrsA3 lp_0064 pox3 gapB npr2 mntH2 gor trxA clpCclpP clpE dnaK groEL lp_3324

143 FUTURE PERSPECTIVES rich supplements (plants, vegetables or encap- Control of starter cultures and probiotic cultures sulated GSH). Robustness of industrial strains is under industrial conditions is essential in order also relevant in the area of probiotic foods. The to provide a tasty, attractive, healthy, and safe functionality of certain probiotic bacteria relies product. Oxidative stress is one of the harsh on their survival of passage through the gastro- conditions that these fermentative microbes have intestinal tract (GI) and the ability to persist and managed to endure during their use in indus- sometimes colonize the intestinal flora. In addi- trial fermentation processes. The comparative tion, it may be important for probiotic activity genomics and global transcriptional analysis of that the microbes deliver under these severe con- the on oxidative stress response of L. plantarum ditions in the GI tract the factors that trigger the presented in this thesis can be used for optimiz- probiotic effect. Hence, engineering robustness ing industrial strains. in the probiotic strains could have be the key for success in achieving the desired functionality. A The results presented in this study show that number of leads obtained in this study towards overproduction of TR results in a “mock“-oxida- improving robustness of L. plantarum against tive stress response inside the cells and conse- oxidative stress conditions could also be used quently in a strain with better resistance towards for improving robustness in other (probiotic) lac- diamide and hydrogen peroxide (Chapter 2). tic acid bacteria. Instead of pursuing the trait of This conclusion implies that we can trigger a robustness we could also use the knowledge to higher resistance towards oxidative stress in an obtain strains that easily lyse. This characteristic industrial strain by influencing the amount of TR is important for adjunct cultures in a variety of present inside the cells. This strategy could be fermentations where these microbes determine applied in partially-anaerobic industrial pro- the quality of the end-product in terms of taste cesses like in the preparation of fresh cheeses and texture. and sausages. The intracellular TR concentration could be directed using metabolic engineering The broad applicability of the research on the techniques (this thesis) but theoretically also by thioredoxin system performed in this thesis relies influencing the amounts of TRX by adjusting the on the fact that this system is highly conserved in concentration of oxygen in the fermentation. An- organisms of diverse taxonomic groups. There- other interesting and applicable result obtained fore, lessons obtained form research in bacte- in this research is that GSH protects L. plantarum ria can lead to possible applications in eukary- against hydrogen peroxide (Chapter 5). Thus, the otes including humans. In cancer research, it is optimized growth environment that would ren- known that cancer cells have a high content of der protection of L. plantarum against oxidative TRX and that high reduced TRX levels can cause stress will be obtained in a GSH-rich medium. cancer-prone disease (25). Therefore, knowl- Changing the media composition to obtain edge on the thioredoxin system can increase our a GSH-rich medium can be done by supple- knowledge on unravelling the mechanisms in- menting GSH directly, engineering the strains side cancer cells. to produce GSH (this thesis) or by using GSH-

144 6 Summary, Concluding Remarks, and Future Perspectives

The work described in this thesis indicates that linked to sugar and purine metabolism. Conse- the oxidative stress response is an intertwined quently, when we cultured the same strain on network of single and complex regulons. While another sugar (cellobiose) as only carbon source we have observed a relation between aeration, we were able influence the phenotype and im- hydrogen peroxide, disulfide, sulfur limitation, prove growth rate of the mutant under aerobic and heat stress, the hierarchy of stress response culture conditions. However, it was not possible -if any exists- is less understood. The overlap be- to perform a functional analysis of each lead or tween stress responses has been well established validate every significant affected gene obtained in bacteria. For example, it is known that in B. in our microarrays. Therefore, undoubtedly subtilis oxidative stress affects proteins which are this high data-driven approach together with also induced under heat stress (15). In this the- data interpretation, used statistical cut-offs, and sis we show that the thioredoxin system plays a presence of false positives will contain flaws. In crucial role in oxidative stress response (Chap- addition the regulatory networks present in L. ter 3) and adaptation to aerobic and respiratory plantarum and operative under oxidative stress growth conditions (Chapter 4); yet, the thiore- described above only include regulation at the doxin system is also induced under heat stress transcription level. It is known that stress re- especially gene trxB2 (Chapter 2). While a link sponse in for instance Lc. lactis is regulated at between oxidative stress and heat stress response the transcriptional level (s factors), at the protein is easily drawn, there are other stress-networks (two component regulatory systems), and at the harder to visualize as the relation between a TR metabolite level (metabolite flux sensors) (14). overproduction and proteins normally induced Therefore, the next step is to use a systems biol- under sulfur limitation. ogy approach and complement the work in this thesis with proteome and metabolome data on In this thesis pioneering work has been present- oxidative stress response in L. plantarum. ed into understanding the response mechanisms present at the transcriptome level in L. plantarum as a result of oxidative stress have been presented in this thesis. The purpose of this work is to serve as a tool as well as a building block for further studies. The conclusions drawn in relation to stress response in L. plantarum have been based on well-thought experimental design, objective statistical analysis of the data and in many cases additional functional analysis. For example the reduced growth rate observed upon aeration in a trxB1-disruption L plantarum strain was investigated using microarrays (Chapter 3). This analysis suggested that the observed phenotype was

145 REFERENCES

1. Arner, E. S., and A. Holmgren. 2000. Physiological functions of thioredoxin and thioredoxin reductase. Eur J Biochem 267:6102-9. 2. Duwat, P., S. Sourice, B. Cesselin, G. Lamberet, K. Vido, P. Gaudu, Y. Le Loir, F. Violet, P. Lou biere, and A. Gruss. 2001. Respiration capacity of the fermenting bacterium Lactococcus lactis and its positive effects on growth and survival. J Bacteriol 183:4509-16. 3. Fahey, R. C. 2001. Novel thiols of prokaryotes. Annu Rev Microbiol 55:333-56. 4. Fedoroff, N. 2006. Redox regulatory mechanisms in cellular stress responses. Ann Bot (Lond) 98:289-300. 5. Gopal, S., I. Borovok, A. Ofer, M. Yanku, G. Cohen, W. Goebel, J. Kreft, and Y. Aharonowitz. 2005. A multidomain fusion protein in Listeria monocytogenes catalyzes the two primary activities for glutathione biosynthesis. J Bacteriol 187:3839-47. 6. Hecker, M., J. Pane-Farre, and U. Volker. 2007. SigB-dependent general stress response in Bacillus subtilis and related gram-positive bacteria. Annu Rev Microbiol 61:215-36. 7. Hecker, M., and U. Volker. 2001. General stress response of Bacillus subtilis and other bacteria. Adv Microb Physiol 44:35-91. 8. Helmann, J. D., M. F. Wu, A. Gaballa, P. A. Kobel, M. M. Morshedi, P. Fawcett, and C. Paddon. 2003. The global transcriptional response of Bacillus subtilis to peroxide stress is coordinated by three transcription factors. J Bacteriol 185:243-53. 9. Holmgren, A. 1985. Thioredoxin. Annu Rev Biochem 54:237-71. 10. Kleerebezem, M., J. Boekhorst, R. van Kranenburg, D. Molenaar, O. P. Kuipers, R. Leer, R. Tarchini, S. A. Peters, H. M. Sandbrink, M. W. Fiers, W. Stiekema, R. M. Lankhorst, P. A. Bron, S. M. Hoffer, M. N. Groot, R. Kerkhoven, M. de Vries, B. Ursing, W. M. de Vos, and R. J. Siezen. 2003. Complete genome sequence of Lactobacillus plantarum WCFS1. Proc Natl Acad Sci U S A 100:1990-5. 11. Kumar, J. K., S. Tabor, and C. C. Richardson. 2004. Proteomic analysis of thioredoxin-targeted proteins in Escherichia coli. Proc Natl Acad Sci U S A 101:3759-64. 12. Leichert, L. I., C. Scharf, and M. Hecker. 2003. Global characterization of disulfide stress in Bacillus subtilis. J Bacteriol 185:1967-75. 13. Li, Y., J. Hugenholtz, T. Abee, and D. Molenaar. 2003. Glutathione protects Lactococcus lactis against oxidative stress. Appl Environ Microbiol 69:5739-45. 14. Miyoshi, A., T. Rochat, J. J. Gratadoux, Y. Le Loir, S. C. Oliveira, P. Langella, and V. Azevedo. 2003. Oxidative stress in Lactococcus lactis. Genet Mol Res 2:348-59. 15. Scharf, C., S. Riethdorf, H. Ernst, S. Engelmann, U. Volker, and M. Hecker. 1998. Thioredoxin is an essential protein induced by multiple stresses in Bacillus subtilis. J Bacteriol 180:1869-77. 16. Sijpesteijn, A. K. 1970. Induction of cytochrome formation and stimulation of oxidative dissimilation by hemin in Streptococcus lactis and Leuconostoc mesenteroides. Antonie Van Leeuwenhoek 36:335-48. 17. Smirnova, G. V., and O. N. Oktyabrsky. 2005. Glutathione in bacteria. Biochemistry (Mosc) 70:1199-211. 18. Stevens, M. J. A. 2008. Transcriptome Response of Lactobacillus plantarum to Global Regulator Deficiency, Stress and other Environmental Conditions. PhD thesis. Wageningen University. 19. Stortz, G., and R. Hengge-Aronis. 2000. Bacterial Stress Responses, Washington DC. 20. van de Guchte, M., P. Serror, C. Chervaux, T. Smokvina, S. D. Ehrlich, and E. Maguin. 2002. Stress responses in lactic acid bacteria. Antonie Van Leeuwenhoek 82:187-216. 21. Vido, K., H. Diemer, A. Van Dorsselaer, E. Leize, V. Juillard, A. Gruss, and P. Gaudu. 2005. Roles of thioredoxin reductase during the aerobic life of Lactococcus lactis. J Bacteriol 187:601-10. 22. Wels, M. 2008. Unraveling the regulatory network of Lactobacillus plantarum WCFS1. PhD thesis. Wagenin gen University. 23. Williams, C. H., Jr. 2000. Thioredoxin-thioredoxin reductase--a system that has come of age. Eur J Biochem 267:6101.

146 6 Summary, Concluding Remarks, and Future Perspectives

24. Winterling, K. W., A. S. Levine, R. E. Yasbin, and R. Woodgate. 1997. Characterization of DinR, the Bacillus subtilis SOS repressor. J Bacteriol 179:1698-703. 25. Zeller, T., and G. Klug. 2006. Thioredoxins in bacteria: functions in oxidative stress response and regula tion of thioredoxin genes. Naturwissenschaften 93:259-66.

147 148 Appendix: Color figures and supplementary material

149

149 150 Appendix: Color figures and supplementary material

CHAPTER 1, FIGURE 2. Genome Atlas of Lactobacillus plantarum WCFS1 and the Thioredoxin system. The atlas represents a circular view of the complete genome sequence of L. plantarum WCFS1. Seven circles were created using Microbial Genome Viewer (77). Circle 1, Inner- most, GC% content. Circle 2, thioredoxin reductase (TR) encoding genes trxB1 red trxB2 green. Circle 3, thioredoxin (TRX) encoding genes trxA1 blue trxA2 orange trxA3 purple trxH black. Circle 4, COG classification in antisense orientation. Circle 5, COG classification in sense orientation. Circle 6, antisense open reading frames (ORFs). Circle 7, Outermost, ORFs in sense orientation.

Lactobacillus plantarum WCFS1 Genome Atlas

CHAPTER 2 FIGURE 4. Weblogo representation of conserved promoter regions in peroxide affected genes found using bioinformatics tools A) Regulatory motif LexA- DinR and B)

A B A B

151151 CHAPTER 3 FIGURE 2. Representation of the biosynthesis of purines in L. plantarum TM WCFS1 in Simpheny obtained by projecting significantly affected genes (pvalue < 1% and FC ≥ 1.5) on the map of the metabolic network. Panel A represents the transcriptional response of the purine biosynthesis pathway in L. plantarum strain NZ7608 compared to wild-type. Panel B shows the transcriptional response of the purine biosynthesis in L. plantarum strains WCFS1 and NZ7608 upon a hydrogen peroxide treatment. Red colored genes (triangles) and reactions (arrows) represent an up-regulation, while green colored genes and reactions represent a down-regulation in the studied condition.

152 Appendix: Color figures and supplementary material - WCFS1 in the reconstructed < 1% and FC ≥ 1.5) as a result of the value < 1% and FC ≥ 1.5) in response to a hydrogen L. plantarum value B Panel A shows the significant affected genes (p . TM WCFS1 and NZ7608. Red colored genes (triangles) and reactions (arrows) represent an up-regula an represent (arrows) reactions and (triangles) genes colored Red NZ7608. and WCFS1 . Panel B shows the significant affected genes (p L. plantarum L. L. plantarum in trxB1 CHAPTER 3 FIGURE 4. Visualization of amino acid biosynthesis in metabolic maps using Simpheny disruption of strains on pulse peroxide tion, while green colored genes and reactions represent a down-regulation in the studied condition. A

153 - . The sequences of transcrip CHAPTER 4 FIGURE 2. Phylogeny and synteny of OhrR-like TFs for selected Firmicutes Orthologous genes are similary colored. The OhrR-like transcriptor factor is color-coded orange. The selected upstream regions to search for TF-specific operator sequences are indicated with small blue arrows. retrieved operators depicted on the right tional factors (TFs) orthologous to B. subtilis OhrR from a representative set of Firmicutes were aligned and a putative phylogeny was generated (bootstrap support indicated in red).

154 Appendix: Color figures and supplementary material

CHAPTER 2 SUPPLEMENTARY MATERIAL

155 1 L. plantarum carriers (1%) Cell envelope (4%) DNA metabolism (7%) Cellular processes (3%) Main Functional Class (%) Amino acid biosynthesis (1%) Central intermediary metabolism (1%) Biosynthesis of cofactors, prosthetic groups, and e C ) I ) ) e Product e e e Global transcriptome response towards hydrogen peroxide stress in WCFS1. TP-dependent DNA helicase RecQ cystathionine beta-lyas molybdopterin precursor synthase Moa protoporphyrinogen oxidase (putative aspartate kinase A cysteine synthas cysteine desulfurase thioredoxin muramidase, C-terminal fragment phytoene synthas DNA-3-methyladenine glycosylase excinuclease ABC, subunit C autoinducer production protein GroEL chaperonin prephenate dehydrogenas glycosyltransferase (putative GroES co-chaperonin adherence protein alkaline shock protein alkaline shock protein catalase DNA-entry nuclease NADH peroxidase 1-deoxy-D-xylulose-5-phosphate synthas extracellular protein glycosyltransferase (putative extracellular protein extracellular protein cell surface protein precursor cell surface hydrolase (putative) cell surface protein precursor DNA helicase (putative) stress induced DNA binding protein L. plantarum kat dxs tyrA ica3 crtM tag2 luxS npr2 csd1 uvrC cysK asp2 asp1 Gene endA trxA2 thrA1 hemK moaC groEL recQ2 groES metC1 acm3-C 2 FC 1.89 1.86 1.99 0.66 1.51 1.95 1.62 1.69 1.53 0.66 1.52 1.52 1.54 1.57 2.06 0.52 0.57 0.62 0.63 0.65 0.65 0.66 1.57 1.62 1.66 1.85 1.93 2.28 2.47 3.25 3.68 5.01 1.51 treatment 3 <0.01 & FC ≥1.5). Predicted gene names, function, fold change induction as well as main class of the genes are displayed in Locus value lp_0255 lp_0256 lp_1492 lp_2377 lp_1470 lp_0304 lp_2658 lp_1793 lp_3263 lp_3020 lp_2109 lp_0774 lp_0979 lp_2034 lp_2270 lp_3154 lp_3679 lp_3676 lp_3677 lp_1165 lp_0197 lp_2716 lp_0910 lp_1885 lp_0728 lp_3128 lp_0727 lp_0930 lp_0929 lp_3578 lp_2906 lp_2544 lp_2610 Supplementary materials file 1,TABLE S1. strains NZ7607 and NZ7602 Significantly affected genes (267) due to hydrogen peroxide stress both in strains NZ7607 and NZ7602 (p when study this in overrepresented found classes those are bold in presented classes functional Main respectively. three and one column compared to the total genome of

156 Appendix: Color figures and supplementary material DNA metabolism (7%) Energy metabolism (10%) t t e e e A A e beta-galactosidase, small subunit holliday junction DNA helicase RuvB holliday junction DNA helicase RuvA topoisomerase IV, subunit B ATP-dependent nuclease, subunit A ATP-dependent nuclease, subunit B excinuclease ABC, subunit B recombinase A topoisomerase IV, subunit myo-inositol 2-dehydrogenas L-serine dehydratase, beta subuni myo-inositol 2-dehydrogenas formate acetyltransferase activating enzyme inositol catabolism protein IolE nitrate reductase, alpha chain branched-chain amino acid aminotransferase DNA-directed DNA polymerase III, alpha chain excinuclease ABC, subunit DNA-damage-inducible protein P UV-damage repair protein formate C-acetyltransferase citrate lyase, alpha chain 1-phosphofructokinase bifunctional protein: alcohol dehydrogenase; acetaldehyde dehydrogenase aspartate 1-decarboxylase aromatic amino acid specific aminotransferase myo-inositol 2-dehydrogenas nitrate reductase, beta chain pyruvate,water dikinas beta-galactosidase, large subuni DNA-directed DNA polymerase III, epsilon chain (putative) DNA helicase (putative) DNA helicase (putative) citF pps iolE fruK lacL dinP ruvB ruvA rexA rexB uvrB recA lacM parE parC narH narG bcaT sdhB pflA2 pflB2 iolG1 iolG2 dnaE adhE iolG3 panD uvrA1 araT2 umuC 1.77 1.78 2.60 2.69 4.11 1.69 2.35 0.56 0.56 0.57 0.58 0.58 0.61 0.64 1.86 1.93 2.77 4.35 0.58 0.59 0.62 0.65 2.12 2.14 2.21 4.62 0.45 0.60 0.64 0.45 0.50 0.55 0.64 2 2 2,3 2,3 2,3 lp_0811 lp_2286 lp_1839 lp_2693 lp_3605 lp_0505 lp_3606 lp_3484 lp_3314 lp_1497 lp_2390 lp_2287 lp_1840 lp_1899 lp_2694 lp_0772 lp_2301 lp_2280 lp_3607 lp_3313 lp_1109 lp_2096 lp_0432 lp_0308 lp_0773 lp_3023 lp_3662 lp_3608 lp_1498 lp_0579 lp_2684 lp_1912 lp_3483

157 Regulatory functions (4%) Energy metabolism (10%) Fatty acid and phospholipid metabolism (5%) n t ) e e A r r r r e e 3-oxoacyl-[acyl-carrier protein] synthase II acetyl-CoA carboxylase, biotin carboxylase subuni 3-oxoacyl-[acyl-carrier protein] synthase III glycerol-3-phosphate dehydrogenas glyceraldehyde 3-phosphate dehydrogenase ribokinase transcription repressor of the SOS regulon p-nitrobenzoate reductase pyruvate oxidas acyl carrier protein (3R)-hydroxymyristoyl-[acyl carrier protein] dehydratase [acyl-carrier protein] S-malonyltransferase acetyl-CoA carboxylase, biotin carboxyl carrier protei acetyl-CoA carboxylase, carboxyl transferase subunit alpha acetyl-CoA carboxylase, carboxyl transferase subunit beta transcription regulator of fructose operon catabolite control protein pyruvate oxidas glutamate decarboxylase 3-oxoacyl-[acyl-carrier protein] reductas enoyl-[acyl-carrier protein] reductase (NADH sorbitol operon activato fumarate hydratase glycine cleavage system, H protein fumarate reductase, flavoprotein subunit precursor phosphopantetheinyltransferase transcription regulato transcription regulato transcription regulator (putative) stress-responsive transcription regulator (putative) transcription regulator (putative) transcription regulato short-chain dehydrogenase/oxidoreductase pnb fabI fum fruR lexA fabF glpD fabD pox5 pox3 ccpA gapB gadB srlM2 fabZ1 rbsK1 fabH2 fabG1 accB2 accA2 acpA2 accC2 accD2 gcsH1 1.83 4.84 0.65 1.52 9.39 0.28 0.33 0.34 0.35 0.36 0.37 0.48 2.01 0.61 0.64 0.65 0.66 1.54 1.57 1.60 3.31 1.85 1.86 5.02 1.60 0.32 0.32 0.33 0.38 0.42 0.63 1.63 1.65 3 2 lp_0952 lp_0050 lp_3589 lp_1672 lp_1675 lp_1676 lp_1678 lp_1671 lp_1680 lp_1682 lp_0371 lp_2095 lp_2651 lp_2256 lp_3234 lp_0825 lp_0126 lp_0294 lp_1360 lp_2063 lp_3045 lp_0500 lp_2629 lp_3420 lp_1670 lp_1673 lp_1674 lp_1679 lp_1681 lp_3655 lp_1112 lp_0305 lp_0789

158 Appendix: Color figures and supplementary material Transcription (1%) Hypothetical proteins (22%) y TP-dependent RNA helicase ribonuclease HII nucleotide-binding protein, universal stress protein UspA family A 1 segregation helicase (putative) unknown unknown unknown unknown HD superfamily hydrolase unknown unknown unknown integral membrane protein unknown unknown unknown unknown unknown unknown unknown unknown unknown unknown unknown unknown integral membrane protein hydrolase, HAD superfamily, Cof famil unknown lipase/esterase (putative) acetyltransferase (putative) unknown integral membrane protein unknown unknown rhe3 rnhB cshA3 1.75 0.53 0.59 1.56 1.72 0.49 0.52 1.57 1.59 1.59 1.60 1.61 1.98 0.57 0.58 0.61 0.62 0.66 0.66 0.66 1.51 1.51 1.51 1.51 1.52 1.53 1.57 1.78 1.78 1.62 1.63 1.65 1.66 1.71 3 2 3 2 2 lp_0580 lp_1081 lp_2939 lp_1853 lp_3438 lp_2937 lp_1704 lp_0990 lp_2337 lp_2278 lp_1486 lp_1485 lp_1484 lp_1872 lp_2669 lp_0063 lp_0899 lp_0155 lp_1637 lp_2573 lp_1163 lp_1439 lp_0053 lp_3359 lp_0030 lp_1723 lp_1886 lp_2631 lp_1292 lp_3448 lp_1290 lp_2513 lp_0837 lp_2224

159 Other categories (13%) Hypothetical proteins (22%) ) integral membrane protein unknown unknown unknown unknown unknown unknown unknown integral membrane protein unknown unknown unknown oxidoreductase unknown unknown unknown 1 segregation helicase (putative) unknown unknown acetyltransferase (putative) unknown exopolyphosphatase-related protein (putative unknown unknown unknown unknown unknown prophage P2a protein 15 prophage P1 protein 27 prophage P2a protein 25 prophage P2a protein 24 prophage P1 protein 20 prophage P1 protein 30 prophage P2a protein 27 prophage P2a protein 26 1.89 2.09 2.15 2.73 3.03 5.30 1.82 1.86 2.25 2.32 2.37 2.42 2.49 2.55 2.66 2.69 1.93 1.97 1.98 2.08 2.23 2.84 2.87 4.19 4.71 9.18 0.47 0.48 0.49 0.49 0.50 0.50 0.51 0.51 32.85 2 2 2 3 2 2 2 lp_2718 lp_0533 lp_0981 lp_2212 lp_0089 lp_2342 lp_1939 lp_0928 lp_0307 lp_0499 lp_1543 lp_0145 lp_0927 lp_1703 lp_3356 lp_0306 lp_0207 lp_0091 lp_2279 lp_2113 lp_1880 lp_3141 lp_1708 lp_3142 lp_3022 lp_0960 lp_1611 lp_2442 lp_0650 lp_2432 lp_2433 lp_0643 lp_0653 lp_2430 lp_2431

160 Appendix: Color figures and supplementary material Protein fate (1%) Protein synthesis (6%) Other categories (13%) A prolyl aminopeptidase dipeptidase protein-methionine-S-oxide reductase 4'-phosphopantetheinyl transferase non-ribosomal peptide synthetase NpsB ribosomal protein S21 non-ribosomal peptide synthetase Nps ribosomal protein S14 prophage P1 protein 25 prophage P1 protein 24 prophage P1 protein 18 prophage P1 protein 17 prophage P2a protein 29 prophage P1 protein 26 prophage P1 protein 31 prophage P2a protein 16 prophage P2a protein 30 prophage P1 protein 15 prophage P2a protein 28 prophage P2a protein 20 prophage P1 protein 19 prophage P2a protein 17 prophage P1 protein 32 prophage P1 protein 56 prophage P1 protein 14 prophage P1 protein 29 prophage P2a protein 13 prophage P1 protein 13 prophage P2a protein 11 transposase, fragment (putative) prophage P1 protein 12 prophage P2a protein 7 prophage P2a protein 12 GTPase prophage P2a protein 2, integrase prophage P1 protein 1, integrase rpsU rpsN npsB npsA npsC msrA3 pepR2 pepD3 0.52 0.53 0.53 0.53 0.54 0.55 0.55 0.55 0.56 0.56 0.58 0.58 0.58 0.60 0.61 0.61 0.61 0.61 0.63 0.63 0.63 0.63 0.65 0.65 0.65 1.54 2.65 2.80 0.66 1.85 3.34 0.47 0.57 0.64 0.53 1.51 lp_0648 lp_0647 lp_0641 lp_0640 lp_2428 lp_0649 lp_0654 lp_2441 lp_2427 lp_0638 lp_2429 lp_2437 lp_0642 lp_2440 lp_0655 lp_0679 lp_0637 lp_0652 lp_2444 lp_0636 lp_2446 lp_1192 lp_0635 lp_2450 lp_2445 lp_1854 lp_2455 lp_0624 lp_2919 lp_0959 lp_1836 lp_0581 lp_0582 lp_1973 lp_0578 lp_1048

161 (6%) Protein synthesis (6%) Transport and binding proteinsm (13%) Purines, pyrimidines, nucleosides and nucleotides ) e I I e e e A elongation factor Tu phosphoribosylaminoimidazole carboxylase, ATPase subunit amidophosphoribosyltransferase precursor phosphoribosylformylglycinamidine synthase I deoxyribose-phosphate aldolas conserved purine biosynthesis cluster protein mannose PTS, EIID serine transporter mannose PTS, EIIC cation transport protein ribosomal protein L4 ribosomal protein L13 phosphoribosylaminoimidazole-succinocarboxamide synthase phosphoribosylamine--glycine ligas phosphoribosylglycinamide formyltransferas bifunctional protein: phosphoribosylaminoimidazolecarboxamide formyltransferase; IMP cyclohydrolase peptide chain release factor 3 ribosomal protein S6 ribosomal protein L2 anaerobic ribonucleoside-triphosphate reductase thioredoxin reductase (NADPH) phosphoribosylaminoimidazole carboxylase, catalytic subunit phosphoribosylformylglycinamidine synthase phosphoribosylformylglycinamidine cyclo-ligas ribosomal protein S14-2 ribosomal protein L32 ribosomal protein L23 S-adenosylmethionine tRNA ribosyltransferase-isomerase purine/pyrimidine phosphoribosyltransferase (putative ribosomal protein S30E tuf rplB rplD prfC rplM purL rpsF purF rplW purS purE purC purD purN purH nrdD purQ purM queA sdaC trxB1 rpmF deoC pts9D pts9C purK1 rpsN2 lp_1087 1.53 1.82 1.90 1.91 2.18 2.32 2.33 2.38 2.42 2.91 3.06 3.14 0.53 0.55 0.58 0.60 1.55 0.61 2.35 3.44 1.55 1.56 1.62 1.77 1.80 1.89 1.76 1.80 1.74 1.80 lp_1034 lp_2119 lp_1077 lp_2728 lp_2723 lp_2724 lp_2727 lp_0497 lp_2726 lp_2719 lp_2721 lp_2720 lp_0577 lp_0502 lp_0576 lp_1087 lp_1255 lp_2932 lp_2725 lp_2722 lp_0009 lp_1036 lp_2216 lp_0761 lp_2729 lp_1289 lp_1535a lp_1035 lp_2285 lp_0737

162 Appendix: Color figures and supplementary material Plasmids (1%) WCFS1 Transport and binding proteinsm (13%) L. plantarum BC transporter, ATP-binding protein BC transporter, ATP-binding protein BC transporter component, iron regulated (putative) BC transporter, permease protein BC transporter, ATP-binding protein BC transporter component (putative) manganese transport protein ferrous iron transport protein B glycine betaine/carnitine/choline ABC transporter, permease protein ferrous iron transport protein A N-acetylgalactosamine PTS, EIIB N-acetylgalactosamine PTS, EIID mannose PTS, EIIAB N-acetylgalactosamine PTS, EIIC iron chelatin ABC transporter, ATP-binding protein A glycine betaine/carnitine/choline ABC transporter, substrate binding and permease protein iron chelatin ABC transporter, substrate binding protein (putative) A potassium uptake protein A A A A fructose PTS, EIIABC quaternary ammonium compound-resistance protein mannitol PTS, EIICB multidrug transport protein transport protein cation efflux protein sugar transport protein fucose transport protein myo-inositol transport protein transport protein amino acid transport protein multidrug transport protein fecE fecB feoB feoA kup2 choS qacH opuD mntH2 pts19B pts19D pts19C pts9AB pts2CB pts16ABC 1.54 1.55 1.65 1.71 1.76 1.78 1.94 1.96 1.98 2.01 2.03 2.17 2.46 2.91 3.17 0.60 0.62 0.64 0.64 0.65 0.65 0.65 1.59 1.61 1.62 1.66 1.51 1.52 1.52 1.62 1.64 0.61 0.66 Values given in parenthesis correspond to the percentage of total amount genes (267) that belong each depicted functional class. Upstream region of gene contains the LexA-dinR regulatory motif: AGAACTCATGTTCG. Upstream region of gene contains an uncharacterized regulatory motif: AGCTAATAGCATCGGC. Gray cathegories depict functional classes which are over-represented when compared to the total genome of lp_2836 lp_0367 lp_3279 lp_2822 lp_2038 lp_2992 lp_1466 lp_1468 lp_1469 lp_1610 lp_3288 lp_0331 lp_1467 lp_0498 lp_p3_05 lp_p1_01 lp_RNA02 lp_2650 lp_3604 lp_2648 lp_0575 lp_2649 lp_1475 lp_2097 lp_1473 lp_0729 lp_0230 lp_1472 lp_0982 lp_3285 lp_1335 lp_0991 lp_2823

* 1 2 3

163 164 Appendix: Color figures and supplementary material

CHAPTER 3 SUPPLEMENTARY MATERIAL

165 3 mutant strain, trxB1 Cell envelope (9%) DNA metabolism (3%) Cellular processes (2%) groups, and carriers (3%) Energy metabolism (10%) Main Functional Class (%) Amino acid biosynthesis (1%) Central intermediary metabolism (2%) Biosynthesis of cofactors, prosthetic Product pyruvate dehydrogenase complex, E2 component; dihydrolipoamide S-acetyltransferase nitrate reductase, alpha chain recombinase A phosphoribosyl-AMP cyclohydrolase molybdopterin biosynthesis protein MoaB molybdopterin biosynthesis protein, D chain 6-phospho-beta-glucosidase nitrate reductase, beta chain nitrate reductase, gamma chain nitrate reductase, delta chain molybdopterin biosynthesis protein, E chain small heat shock protein alpha-glucosidase galactoside O-acetyltransferase 1-phosphofructokinase NADH dehydrogenase pyruvate oxidase succinate-semialdehyde dehydrogenase (NAD(P)+) mannose-6-phosphate isomerase UV-damage repair protein DNA-damage-inducible protein P DNA-entry nuclease extracellular protein glycosyltransferase lipoprotein precursor cell surface protein precursor extracellular protein lipoprotein precursor extracellular protein cell surface protein precursor extracellular protein extracellular protein 2 FC 1.50 1.53 0.66 1.60 1.58 1.59 1.73 1.62 1.53 1.55 1.53 1.59 1.62 1.65 1.67 1.67 1.79 1.90 2.01 0.52 0.59 0.60 1.51 1.55 1.57 1.57 1.65 1.69 1.76 2.09 2.31 2.73 1 M 0.59 0.61 0.68 0.66 0.67 0.79 0.69 0.61 0.63 0.62 0.67 0.69 0.73 0.74 0.74 0.84 0.92 1.01 0.60 0.63 0.65 0.65 0.72 0.76 0.81 1.07 1.21 1.45 -0.61 -0.96 -0.77 -0.74 E C i 2 2 2 K m ox bg9 hisI p narI fru agl1 narJ dinP recA narH hsp p narG p ndh Gene endA pdhC gabD moaB moa thgA1 moaD umu Locus lp_1497 lp_2152 lp_3525 lp_2280 lp_2301 lp_2553 lp_1495 lp_1479 lp_1478 lp_2847 lp_1763 lp_2098 lp_0800 lp_1449 lp_1070 lp_0304 lp_3413 lp_3412 lp_3414 lp_2906 lp_2668 lp_0174 lp_0393 lp_3023 lp_1498 lp_1500 lp_1499 lp_2096 lp_1069 lp_0852 lp_3092 lp_2384 Supplementary material File 1, TABLE S1. Significantly affected genes (106) found in the to the wild-type. NZ7608 when compared

166 Appendix: Color figures and supplementary material Hypothetical proteins (30%) y y protein containing diguanylate cyclase/phosphodiesterase domain 2 (EAL) 1 segregation helicase (putative) integral membrane protein lipase/esterase (putative) unknown integral membrane protein integral membrane protein unknown unknown unknown unknown oxidoreductase unknown integral membrane protein hydrolase, HAD superfamily, Cof famil integral membrane protein adenylyl transferase (putative) oxidoreductase hydrolase, HAD superfamily hydrolase, HAD superfamily, Cof famil unknown integral membrane protein (putative) acetyltransferase (putative) unknown unknown unknown unknown unknown unknown unknown unknown unknown 0.64 1.51 1.51 1.51 1.52 1.52 1.53 1.54 1.58 1.59 1.60 1.61 1.63 1.66 1.66 1.69 1.71 1.71 1.75 1.78 1.80 1.84 0.40 0.48 0.50 0.55 0.58 0.60 0.65 0.65 0.66 0.66 0.59 0.60 0.60 0.60 0.61 0.61 0.63 0.66 0.67 0.68 0.69 0.70 0.73 0.74 0.76 0.77 0.78 0.81 0.83 0.85 0.88 -0.64 -0.62 -0.61 -0.60 -0.59 -1.32 -1.05 -1.00 -0.87 -0.79 -0.73 2 cshA lp_3356 lp_0091 lp_0145 lp_2113 lp_2900 lp_3561 lp_0199 lp_3016 lp_3640 lp_1484 lp_1486 lp_3243 lp_0320 lp_3244 lp_1168 lp_0472 lp_1876 lp_0444 lp_0960 lp_3022 lp_1611 lp_3142 lp_3141 lp_1543 lp_0778 lp_0098 lp_0823 lp_3318 lp_3078 lp_1353 lp_1485 lp_2755

167 Protein fate (2%) nucleotides (8%) Other categories (5%) Regulatory functions (8%) Transport and binding proteins (18%) Purines, pyrimidines, nucleosides and <1% and FC ≥1.5). value A BC transporter, permease protein BC transporter, ATP-binding protein BC transporter, permease protein BC transporter, permease protein BC transporter, ATP-binding protein A A A A A phosphocarrier protein Hpr branched-chain amino acid ABC transporter, ATP-binding protein iron chelatin ABC transporter, substrate binding protein N-acetylglucosamine PTS, EIICBA maltose/maltodextrin ABC transporter, substrate binding protein nitrite extrusion protein mannose PTS, EIIB cellobiose PTS, EII mannose PTS, EIIAB phosphoenolpyruvate-protein phosphotransferase mannose PTS, EIID mannose PTS, EIIC beta-glucosides PTS, EIIB amino acid transport protein (putative) transcription antiterminator transcription regulator of fructose operon protein-tyrosine phosphatase transcription regulator transcription antiterminator response regulator phosphoribosylformylglycinamidine synthase II conserved purine biosynthesis cluster protein phosphoribosylformylglycinamidine cyclo-ligase phosphoribosylaminoimidazole-succinocarboxamide synthase phosphoribosylaminoimidazole carboxylase, ATPase subunit transcription repressor of the SOS regulon protein-methionine-S-oxide reductase phosphoribosylaminoimidazole carboxylase, catalytic subunit amidophosphoribosyltransferase precursor phosphoribosylformylglycinamidine synthase I transcription regulator of gluconeogenic genes transcription regulator GTPase transposase, fragment prophage P2a protein 3 prophage P2a protein 2, integrase prophage P1 protein 1, integrase M 1.73 1.80 1.52 1.63 1.53 1.62 1.65 1.78 1.86 1.94 2.00 1.65 0.64 1.80 1.50 1.61 0.60 0.66 0.65 1.51 1.56 1.59 1.62 1.64 1.64 0.60 0.63 0.61 0.60 0.61 0.62 0.63 0.65 0.59 0.63 1.52 1.61 1.60 1.82 0.26 0.59 0.61 0.79 0.60 0.70 0.85 0.62 0.70 1.00 0.73 0.73 0.83 0.90 0.96 0.59 0.85 0.69 0.60 0.65 0.67 0.70 0.71 0.71 0.61 0.69 0.67 0.86 -0.64 -0.73 -0.59 -0.61 -0.73 -0.66 -0.72 -0.67 -0.61 -0.76 -0.73 -0.71 -0.69 -0.68 -1.96 -0.77 -0.71 log (cy5/cy3) ratio 2 3 E 2 K E R tsI A tp urL tsH urF ur urS urC urQ urM ts9D ts9C urK1 p livE rrp4 fruR p lexA ts15B ts10B ts20A p p p fhuD nar ts9AB p p gnt p p mal p p p bglG4 p bglG5 ts22CB msrA p p p p p M is defined as the FC is defined as Fold Change or 2 Values given as percentage of the total significant affected transcripts (p lp_2780 lp_2077 lp_2981 lp_0861 lp_0577 lp_1399 lp_2740 lp_1075 lp_3103 lp_0175 lp_1481 lp_0587 lp_2076 lp_0215 lp_2969 lp_1274 lp_0576 lp_0575 lp_1273 lp_2722 lp_2728 lp_2729 lp_2063 lp_3514 lp_1974 lp_2095 lp_3234 lp_2782 lp_2725 lp_2726 lp_2454 lp_2455 lp_0624 lp_1687 lp_0860 lp_1836 lp_3272 lp_2723 lp_2724 lp_2727 lp_3529 lp_1487

1 2 3

168 Appendix: Color figures and supplementary material 3 (3%) Biosynthesis of cofactors, Main Functional Class (%) Amino acid biosynthesis (15%) prosthetic groups, and carriers Protein fate (2%) nucleotides (8%) Other categories (5%) Regulatory functions (8%) Transport and binding proteins (18%) Purines, pyrimidines, nucleosides and A Product <1% and FC ≥1.5). value A BC transporter, permease protein BC transporter, ATP-binding protein BC transporter, permease protein BC transporter, permease protein BC transporter, ATP-binding protein A A A A A phosphocarrier protein Hpr branched-chain amino acid ABC transporter, ATP-binding protein iron chelatin ABC transporter, substrate binding protein N-acetylglucosamine PTS, EIICBA maltose/maltodextrin ABC transporter, substrate binding protein nitrite extrusion protein mannose PTS, EIIB cellobiose PTS, EII mannose PTS, EIIAB phosphoenolpyruvate-protein phosphotransferase mannose PTS, EIID mannose PTS, EIIC beta-glucosides PTS, EIIB amino acid transport protein (putative) transcription antiterminator transcription regulator of fructose operon protein-tyrosine phosphatase transcription regulator transcription antiterminator response regulator phosphoribosylformylglycinamidine synthase II conserved purine biosynthesis cluster protein phosphoribosylformylglycinamidine cyclo-ligase phosphoribosylaminoimidazole-succinocarboxamide synthase phosphoribosylaminoimidazole carboxylase, ATPase subunit transcription repressor of the SOS regulon protein-methionine-S-oxide reductase phosphoribosylaminoimidazole carboxylase, catalytic subunit amidophosphoribosyltransferase precursor phosphoribosylformylglycinamidine synthase I transcription regulator of gluconeogenic genes transcription regulator GTPase transposase, fragment prophage P2a protein 3 prophage P2a protein 2, integrase prophage P1 protein 1, integrase molybdopterin-guanine dinucleotide biosynthesis protein MobB molybdopterin precursor synthase MoaA serine O-acetyltransferase chorismate synthase molybdopterin biosynthesis protein MoaB molybdopterin-guanine dinucleotide biosynthesis protein MobA (putative) prephenate dehydrogenase trans-hexaprenyltranstransferase, component II GTP cyclohydrolase II glutamate--cysteine ligase cysteine synthase pantothenate kinase aspartate kinase shikimate 5-dehydrogenase cystathionine beta-lyase molybdopterin biosynthesis protein MoeB molybdopterin biosynthesis protein, D chain molybdopterin biosynthesis protein Moe thiamin biosynthesis lipoprotein ApbE 3-phosphoshikimate 1-carboxyvinyltransferase glycine hydroxymethyltransferase aspartate racemase riboflavin synthase, alpha chain, N-terminally truncated aspartate kinase molybdopterin biosynthesis protein, E chain molybdopterin precursor synthase MoaC riboflavin synthase, beta chain thiamine-phosphate pyrophosphorylase dihydrodipicolinate reductase shikimate 5-dehydrogenase phosphoribosyl-AMP cyclohydrolase 3-dehydroquinate dehydratase 2 FC 0.41 0.50 1.84 1.89 0.35 0.49 1.76 0.59 0.49 1.50 1.74 0.55 1.51 1.54 1.65 0.46 0.48 0.53 0.61 0.64 1.66 1.67 0.55 0.65 0.43 0.51 0.51 0.59 0.61 0.61 0.61 0.62 M 1.73 1.80 1.52 1.63 1.53 1.62 1.65 1.78 1.86 1.94 2.00 1.65 0.64 1.80 1.50 1.61 0.60 0.66 0.65 1.51 1.56 1.59 1.62 1.64 1.64 0.60 0.63 0.61 0.60 0.61 0.62 0.63 0.65 0.59 0.63 1.52 1.61 1.60 1.82 0.26 0.59 0.61 1 M 0.88 0.92 0.82 0.59 0.80 0.60 0.62 0.72 0.73 0.74 -1.29 -1.00 -1.52 -1.02 -0.76 -1.04 -0.87 -1.12 -1.07 -0.91 -0.72 -0.63 -0.88 -0.62 -1.22 -0.96 -0.96 -0.76 -0.71 -0.70 -0.70 -0.68 0.79 0.60 0.70 0.85 0.62 0.70 1.00 0.73 0.73 0.83 0.90 0.96 0.59 0.85 0.69 0.60 0.65 0.67 0.70 0.71 0.71 0.61 0.69 0.67 0.86 -0.64 -0.73 -0.59 -0.61 -0.73 -0.66 -0.72 -0.67 -0.61 -0.76 -0.73 -0.71 -0.69 -0.68 -1.96 -0.77 -0.71 log (cy5/cy3) ratio 2 E 2 C E E 3 E 2 K E hisI R thi ribA tyrA ribB ribH glyA aroF aro racD cysE cysK coaA Gene dapB thrA thrA1 moa mobB moaA moaB mobA moeB moeA moa moaD tsI aroD1 aroD2 aroC1 gshA1 A hepB1 apbE1 metC1 tp urL tsH urF ur urS urC urQ urM ts9D ts9C urK1 p livE rrp4 fruR p lexA ts15B ts10B ts20A p p p fhuD nar ts9AB p p gnt p p mal p p p bglG4 p bglG5 ts22CB msrA p p p p p Locus M is defined as the FC is defined as Fold Change or 2 Values given as percentage of the total significant affected transcripts (p lp_1478 lp_1492 lp_1493 lp_1480 lp_2037 lp_1495 lp_0115 lp_1491 lp_0254 lp_1066 lp_1479 lp_1437 lp_2324 lp_2034 lp_2308 lp_0913 lp_0979 lp_1084 lp_0255 lp_2035 lp_1496 lp_1438 lp_1494 lp_1072 lp_2375 lp_0256 lp_1874 lp_3494 lp_2553 lp_2798 lp_1523 lp_1436 lp_2780 lp_2077 lp_2981 lp_0861 lp_0577 lp_1399 lp_2740 lp_1075 lp_3103 lp_0175 lp_1481 lp_0587 lp_2076 lp_0215 lp_2969 lp_1274 lp_0576 lp_0575 lp_1273 lp_2722 lp_2728 lp_2729 lp_2063 lp_3514 lp_1974 lp_2095 lp_3234 lp_2782 lp_2725 lp_2726 lp_2454 lp_2455 lp_0624 lp_1687 lp_0860 lp_1836 lp_3272 lp_2723 lp_2724 lp_2727 lp_3529 lp_1487

Supplementary material File 2, TABLE S2. Significantly affected genes (605) in both strains NZ7608 and wild-type and NZ7608 strains both in (605) genes affected Significantly S2. TABLE 2, File material Supplementary pulse. peroxide of a hydrogen as a result 1 2 3

169 (3%) Cell envelope (5%) Biosynthesis of cofactors, prosthetic groups, and carriers A extracellular protein, gamma-D-glutamate-meso-diaminopimelate muropeptidase (putative) D-alanine activating enzyme Dlt D-alanyl transfer protein DltD beta-lactamase lipopolysaccharide biosynthesis protein LicD protoporphyrinogen oxidase (putative) thioredoxin thioredoxin beta-lactamase (putative) poly(glycerol-phosphate) alpha-glucosyltransferase D-alanyl transfer protein DltB teichoic acid glycosylation protein extracellular protein adherence protein lipoprotein precursor (putative) glycosyltransferase (putative) lipoprotein precursor prenyltransferase lipoprotein precursor glycosyltransferase extracellular protein cell surface protein precursor lipoprotein precursor extracellular protein extracellular protein extracellular protein cell surface protein precursor extracellular protein lipoprotein precursor (putative) extracellular protein (putative) cell surface protein precursor, GY family cell surface protein precursor cell surface protein precursor 1.58 1.73 0.62 0.64 1.64 1.65 2.32 1.58 1.63 1.72 0.62 0.65 1.50 1.51 1.54 1.78 0.10 0.34 0.34 0.41 0.43 0.47 0.47 0.47 0.48 0.49 0.50 0.53 0.54 0.59 0.60 1.63 1.51 0.66 0.79 0.72 0.72 1.21 0.66 0.70 0.78 0.59 0.60 0.63 0.83 0.71 0.59 -0.70 -0.65 -0.70 -0.62 -3.32 -1.56 -1.56 -1.29 -1.21 -1.09 -1.09 -1.08 -1.06 -1.02 -0.99 -0.90 -0.90 -0.77 -0.74 5 2 K 2 2 licD dltA dltB dltD bla1 bla trxA3 trxA gtcA hem tagE lp_0844 lp_3116 lp_1793 lp_1812 lp_2658 lp_2019 lp_2393 lp_0469 lp_3437 lp_2270 lp_1070 lp_1763 lp_0304 lp_0800 lp_2098 lp_1446 lp_2847 lp_3421 lp_1447 lp_1449 lp_0473 lp_3077 lp_0946 lp_2162 lp_1643 lp_2747 lp_2843 lp_3059 lp_2018 lp_2016 lp_1715 lp_2377 lp_2341

170 Appendix: Color figures and supplementary material (1%) Cellular processes (2%) Energy metabolism (12%) Central intermediary metabolism TP-dependent Clp protease, ATP-binding subunit ClpL A galactoside O-acetyltransferase cell division protein FtsL (putative) alpha-glucosidase nisin resistance protein (putative) copper homeostasis protein phytoene synthase squalene synthase heat shock protein GrpE competence-damage protein cell division protein FtsK glutathione peroxidase catalase DNA-entry nuclease alpha-glucosidase phosphoenolpyruvate carboxykinase (ATP) cell division protein SufI phosphoglucomutase alpha, alpha-phosphotrehalase pyruvate dehydrogenase complex, E1 component, alpha subunit pyruvate dehydrogenase complex, E1 component, beta subunit small heat shock protein alpha-amylase formate C-acetyltransferase L-lactate dehydrogenase myo-inositol 2-dehydrogenase myo-inositol 2-dehydrogenase NADH dehydrogenase formate acetyltransferase activating enzyme 1-phosphofructokinase myo-inositol 2-dehydrogenase mannose-6-phosphate isomerase myo-inositol 2-dehydrogenase inositol catabolism protein IolE 2-nitropropane dioxygenase glucan 1,4-alpha-maltohydrolase 0.31 0.66 0.29 0.52 0.59 1.65 1.84 1.53 1.53 1.57 1.62 3.34 4.42 0.13 0.14 0.47 0.40 0.44 0.20 0.24 0.34 0.41 0.61 0.16 0.17 0.18 0.19 0.22 0.24 0.25 0.25 0.26 0.27 0.28 0.50 0.46 0.72 0.88 0.61 0.62 0.65 0.69 1.74 2.15 -1.71 -0.60 -1.76 -0.93 -0.77 -2.97 -1.10 -2.79 -1.31 -1.20 -2.33 -2.07 -1.54 -1.29 -0.71 -2.67 -2.56 -2.45 -2.43 -2.19 -2.05 -2.01 -2.00 -1.97 -1.90 -1.86 -1.01 -1.11 2 2 3 2 i 2 2 2 2 2 E K E ck m gm flB flA dhA dhB kat nsr p p ftsL sufI iol gpo treA fru clpL crtN agl1 agl3 p crtM cinA cutC grp hsp ndh p p ftsK iolG iolG4 iolG1 iolG endA p p amy ldhL thgA1 lp_0764 lp_2028 lp_0393 lp_2220 lp_2201 lp_0174 lp_0912 lp_1420 lp_3313 lp_0263 lp_2757 lp_0179 lp_1069 lp_3314 lp_3583 lp_3263 lp_3262 lp_3418 lp_1101 lp_3608 lp_3612 lp_2154 lp_2153 lp_2096 lp_3605 lp_2384 lp_3606 lp_3607 lp_2302 lp_2210 lp_0220 lp_3578 lp_2906 lp_0193 lp_2668 lp_0355

171 Energy metabolism (12%) pyruvate dehydrogenase complex, E3 component; dihydrolipoamide dehydrogenase pyruvate dehydrogenase complex, E2 component; dihydrolipoamide S- acetyltransferase beta-galactosidase, small subunit fructokinase L-iditol 2-dehydrogenase succinate-semialdehyde dehydrogenase (NAD(P)+) nitrate reductase, beta chain L-ribulokinase (putative) [citrate (pro-3S)-lyase] ligase mannitol-1-phosphate 5-dehydrogenase L-serine dehydratase, beta subunit 6-phospho-beta-glucosidase aldose 1-epimerase beta-galactosidase, large subunit bifunctional protein: alcohol dehydrogenase; acetaldehyde dehydrogenase pyruvate oxidase nitrate reductase, alpha chain nitrate reductase, gamma chain citrate lyase, beta chain citrate lyase, alpha chain pyruvate,water dikinase nitrate reductase, delta chain fumarate hydratase malic enzyme, NAD-dependent transaldolase aspartate 1-decarboxylase pyruvate oxidase fumarate reductase, flavoprotein subunit precursor glucokinase alpha-1,2-mannosidase (putative) fumarate reductase, flavoprotein subunit precursor,N-term truncated fumarate reductase, flavoprotein subunit precursor,N-term truncated fumarate reductase, flavoprotein subunit precursor N-acetyl mannosamide kinase 0.33 0.35 0.36 0.32 0.34 0.29 0.37 0.49 0.49 0.49 0.44 0.29 0.29 0.31 0.36 0.37 0.38 0.39 0.40 0.41 0.44 0.48 0.48 0.48 0.51 0.44 0.42 0.36 0.50 0.50 0.46 0.46 0.52 0.42 -1.60 -1.53 -1.49 -1.79 -1.66 -1.55 -1.44 -1.04 -1.03 -1.03 -1.20 -1.71 -1.81 -1.81 -1.46 -1.44 -1.38 -1.34 -1.32 -1.28 -1.19 -1.07 -1.06 -1.04 -0.96 -1.20 -1.25 -1.47 -1.00 -0.99 -1.12 -1.12 -0.94 -1.24 3 2 E 2 F E ps ox ox1 bg9 anD tal cit p cit fum citC narI lacL mae narJ mtlD gutB lacM araB narH p p narG p sdhB adh pdhD pdhC gabD p galM sacK1 lp_0952 lp_3662 lp_3525 lp_2151 lp_3545 lp_2152 lp_3484 lp_0852 lp_1106 lp_0233 lp_0441 lp_3634 lp_3539 lp_0579 lp_0184 lp_3483 lp_3092 lp_1498 lp_1497 lp_1500 lp_1108 lp_1109 lp_0849 lp_1113 lp_3125 lp_1499 lp_1112 lp_1105 lp_3556 lp_1425 lp_1912 lp_3487 lp_3567 lp_0505

172 Appendix: Color figures and supplementary material DNA metabolism (3%) Energy metabolism (12%) A glycine cleavage system, H protein acetolactate synthase ribokinase L-arabinose isomerase glyceraldehyde 3-phosphate dehydrogenase cytochrome D ubiquinol oxidase, subunit II L-serine dehydratase, alpha subunit 2-keto-3-deoxygluconate kinase L-2-hydroxyisocaproate dehydrogenase p-nitrobenzoate reductase pyruvate oxidase DNA mismatch repair protein MutS2 DNA processing protein DNA-directed DNA polymerase III, alpha chain holliday junction DNA helicase Ruv holliday junction DNA helicase RuvB aldose 1-epimerase L-2-hydroxyisocaproate dehydrogenase uracil-DNA glycosylase DNA-binding protein II phosphate acetyltransferase ribose 5-phosphate epimerase branched-chain amino acid aminotransferase acetaldehyde dehydrogenase citrate lyase, acyl carrier protein cytochrome D ubiquinol oxidase, subunit I flavodoxin sugar-phosphate aldolase oxidoreductase N-methylpurine-DNA glycosylase (putative) short-chain dehydrogenase/oxidoreductase DNA helicase (putative) DNA helicase (putative) N-acetylglucosamine kinase (putative) 4-hydroxyphenylacetate-3-hydroxylase, C-terminus DNA helicase (putative) 1.62 0.66 0.66 0.64 0.65 0.66 0.58 1.71 1.77 2.24 2.57 1.52 1.56 1.78 1.84 1.88 0.60 0.63 0.54 0.63 0.56 0.56 0.56 0.58 0.52 0.53 0.60 0.62 0.62 1.51 0.57 1.53 1.60 0.53 0.55 2.00 0.70 0.77 0.82 1.16 1.36 0.61 0.64 0.83 0.88 0.91 0.60 0.61 0.68 1.00 -0.60 -0.60 -0.64 -0.63 -0.61 -0.78 -0.73 -0.66 -0.90 -0.67 -0.85 -0.84 -0.83 -0.80 -0.93 -0.92 -0.75 -0.70 -0.69 -0.82 -0.91 -0.85 2 S E K g ta nb ox3 als p p un citD ruvA ruvB araA dprA p cydB bcaT cydA kdg sdhA mut gapB rpiA1 dna hbsU acdH hicD2 hicD1 rbsK1 galM gcsH1 lp_0500 lp_0305 lp_0253 lp_0789 lp_1126 lp_1005 lp_1477 lp_1731 lp_3603 lp_3355 lp_1245 lp_3554 lp_1879 lp_1991 lp_2271 lp_0602 lp_2390 lp_0060 lp_0329 lp_0506 lp_1107 lp_1125 lp_3509 lp_0070 lp_0807 lp_0350 lp_0050 lp_2629 lp_0806 lp_0910 lp_1852 lp_0432 lp_1899 lp_2287 lp_2286 lp_0308

173 metabolism (2%) Other categories (3%) DNA metabolism (3%) Fatty acid and phospholipid A A A TP-dependent nuclease, subunit TP-dependent nuclease, subunit B A A acetyl-CoA carboxylase, biotin carboxylase subunit glycerol-3-phosphate dehydrogenase (NAD(P)+) choloylglycine hydrolase topoisomerase IV, subunit B topoisomerase IV, subunit UV-damage repair protein 1,3-propanediol dehydrogenase acetyl-CoA carboxylase, carboxyl transferase subunit alpha excinuclease ABC, subunit excinuclease ABC, subunit B recombinase A DNA-damage-inducible protein P acetylesterase dihydroxyacetone phosphotransferase, binding sub-unit glycerol-3-phosphate dehydrogenase glycerone kinase acyl carrier protein phosphodiesterase transposase, fragment prophage P1 protein 24 prophage P1 protein 31 prophage P1 protein 30 prophage P1 protein 27 prophage P1 protein 19 prophage P2a protein 30 prophage P1 protein 14 prophage P1 protein 20 prophage P2a protein 15 prophage P1 protein 17 transposase, fragment prophage P2a protein 1 integrase, fragment prophage P1 protein 1, integrase prophage P2a protein 3 prophage P2a protein 2, integrase 1.52 0.62 0.55 2.37 2.57 2.60 2.68 2.69 3.33 6.85 0.25 1.52 3.37 5.09 0.42 0.49 0.54 0.55 1.51 0.22 0.56 0.57 0.59 0.60 0.61 0.65 0.65 0.66 0.66 0.66 1.73 1.78 1.96 3.40 3.72 4.11 0.61 1.24 1.36 1.38 1.42 1.43 1.74 2.78 0.61 1.75 2.35 0.59 0.79 0.83 0.97 1.77 1.90 2.04 -0.68 -0.85 -2.02 -1.25 -1.01 -0.88 -0.86 -2.20 -0.83 -0.82 -0.77 -0.73 -0.71 -0.63 -0.62 -0.60 -0.60 -0.59 2 C 2 2 E ar arC est dinP glpD rexA uvrB rexB recA p dak bsh4 p gspA dhaT uvrA1 umu accA dak1B accC2 lp_0955 lp_1678 lp_1680 lp_0756 lp_3505 lp_0169 lp_0371 lp_0168 lp_2572 lp_1839 lp_0860 lp_0647 lp_0654 lp_0653 lp_0650 lp_0642 lp_2427 lp_0637 lp_0643 lp_2442 lp_0640 lp_3168 lp_2456 lp_1997 lp_0624 lp_2454 lp_2455 lp_3051 lp_1840 lp_2693 lp_0773 lp_0772 lp_2694 lp_2301 lp_2280 lp_3023

174 Appendix: Color figures and supplementary material (5%) Protein fate (1%) metabolism (2%) Protein synthesis (3%) Purines, pyrimidines, Fatty acid and phospholipid nucleosides and nucleotides A A ribosomal protein S10 ribosomal protein S3 ribosomal protein S8 non-ribosomal peptide synthetase NpsB cyclopropane-fatty-acyl-phospholipid synthase peptidylprolyl isomerase protein-methionine-S-oxide reductase non-ribosomal peptide synthetase NpsA tRNA (5-methylaminomethyl-2-thiouridylate)-methyltransferase ribosomal protein L6 ribosomal protein L29 peptide chain release factor 3 GTP-binding protein Typ ribosomal protein L24 S-adenosylmethionine tRNA ribosyltransferase-isomerase queuine tRNA-ribosyltransferase nucleoside-diphosphate kinase acetyl-CoA carboxylase, carboxyl transferase subunit beta enoyl-[acyl-carrier protein] reductase (NADH) dipeptidase prolyl aminopeptidase dipeptidase protein-tyrosine phosphatase dipeptidase prolyl aminopeptidase elongation factor Tu serine--tRNA ligase nucleotide-disulphide oxidoreductase anaerobic ribonucleotide reductase activator protein tripeptidase 4'-phosphopantetheinyl transferase pseudouridylate synthase ribosomal protein S30E purine nucleosidase phosphopantetheinyltransferase 1.52 1.53 1.53 0.65 1.61 1.65 1.53 1.53 1.60 1.57 3.55 0.30 1.55 1.59 1.60 1.66 1.77 1.77 0.14 0.62 0.63 2.02 0.45 0.55 0.59 0.62 0.65 0.42 0.56 0.57 0.49 0.56 0.31 0.54 1.53 0.61 0.61 0.62 0.69 0.73 0.61 0.62 0.67 0.65 1.83 0.63 0.67 0.67 0.74 0.82 0.83 1.01 0.61 -0.61 -1.73 -2.82 -0.70 -0.67 -1.15 -0.87 -0.77 -0.69 -0.62 -1.26 -0.82 -0.81 -1.04 -0.85 -1.68 -0.88 2 2 3 2 2 C 2 2 U f X rfC tp tgt tu epT rtM ndk fabI rplF rpl truB epD epR epD1 epR1 epD3 cfa p rpsJ p typA rpsC rpsH trm nrdG npsB npsA p trxB queA npsC rpm p serS1 accD2 msrA p p p p p lp_1032 lp_1040 lp_1050 lp_2178 lp_0737 lp_2585 lp_0853 lp_3193 lp_0334 lp_1904 lp_0228 lp_0501 lp_0581 lp_2591 lp_2931 lp_3272 lp_2919 lp_0959 lp_0578 lp_1051 lp_1043 lp_1255 lp_2146 lp_2285 lp_2282 lp_0242 lp_1679 lp_1682 lp_1681 lp_3174 lp_0582 lp_2032 lp_2119 lp_1046 lp_1836

175 (5%) Transcription (1%) Purines, pyrimidines, nucleosides and nucleotides TP-dependent RNA helicase bifunctional protein: phosphoribosylaminoimidazolecarboxamide formyltransferase; IMP cyclohydrolase A adenylosuccinate synthase dihydroorotase dihydroorotate oxidase adenylate kinase carbamoyl-phosphate synthase, pyrimidine-specific, large chain adenylosuccinate lyase phosphoribosylformylglycinamidine synthase I conserved purine biosynthesis cluster protein transcription-repair coupling factor ribonuclease HII transcription termination-antitermination factor NusA DNA-directed RNA polymerase, beta subunit orotidine-5'-phosphate decarboxylase phosphoribosylformylglycinamidine synthase II phosphoribosylamine--glycine ligase phosphoribosylglycinamide formyltransferase phosphoribosylaminoimidazole carboxylase, catalytic subunit phosphoribosylaminoimidazole-succinocarboxamide synthase phosphoribosylformylglycinamidine cyclo-ligase anaerobic ribonucleoside-triphosphate reductase UTP--glucose-1-phosphate uridylyltransferase phosphoribosylaminoimidazole carboxylase, ATPase subunit orotate phosphoribosyltransferase thioredoxin reductase (NADPH) GMP reductase amidophosphoribosyltransferase precursor polynucleotide adenylyltransferase ribonuclease H (putative) purine nucleosidase purine/pyrimidine phosphoribosyltransferase (putative) 4.92 1.85 4.02 1.52 1.57 1.62 1.98 2.05 1.55 1.59 1.77 1.79 1.87 1.87 5.26 5.79 5.53 0.58 0.66 4.29 4.48 4.74 5.16 5.24 1.72 1.75 1.76 5.11 0.56 0.65 0.59 1.95 2.30 0.89 2.01 0.60 0.65 0.70 0.99 1.03 0.64 0.67 0.83 0.84 0.90 2.40 0.90 2.53 2.47 2.10 2.16 2.25 2.37 2.39 0.78 0.80 0.82 2.35 0.96 -0.80 -0.60 -0.84 -0.63 -0.76 E yrF urL yrE urF urA yrC yrD urB ur urS urD urN urC urQ apL urM rnh urK1 adk yrAB mfd rhe3 p p galU p p rnhB rpoB p p p p p p nrdD purH p p p p p nusA p trxB1 guaC p p lp_0363 lp_0757 lp_3270 lp_2723 lp_2702 lp_2697 lp_1058 lp_2700 lp_2698 lp_1289 lp_2725 lp_2719 lp_2722 lp_2593 lp_0539 lp_1853 lp_2043 lp_1021 lp_2278 lp_3269 lp_1873 lp_2721 lp_2728 lp_2727 lp_2726 lp_2724 lp_2932 lp_2720 lp_0761 lp_3271 lp_2699 lp_2729

176 Appendix: Color figures and supplementary material Hypothetical proteins (24%) y oxidoreductase oxidoreductase unknown integral membrane protein integral membrane protein (putative) unknown unknown oxidoreductase unknown unknown unknown hydrolase, HAD superfamily, Cof famil unknown acetyltransferase (putative) unknown unknown protein containing diguanylate cyclase/phosphodiesterase domain 2 (EAL) unknown oxidoreductase unknown oxidoreductase unknown unknown hydrolase, HAD superfamily unknown unknown unknown unknown unknown unknown oxidoreductase unknown unknown unknown integral membrane protein 0.16 0.18 0.18 0.19 0.19 0.20 0.22 0.22 0.23 0.23 0.23 0.25 0.26 0.26 0.27 0.27 0.28 0.28 0.28 0.30 0.31 0.31 0.31 0.31 0.32 0.32 0.33 0.34 0.34 0.35 0.36 0.37 0.37 0.37 0.38 -2.64 -2.48 -2.45 -2.38 -2.38 -2.34 -2.18 -2.15 -2.13 -2.11 -2.11 -1.99 -1.97 -1.95 -1.90 -1.89 -1.85 -1.85 -1.81 -1.76 -1.71 -1.71 -1.70 -1.70 -1.65 -1.63 -1.61 -1.56 -1.55 -1.53 -1.48 -1.44 -1.43 -1.42 -1.39 lp_3318 lp_2732 lp_0240 lp_2900 lp_2755 lp_2766 lp_0063 lp_3614 lp_0058 lp_2813 lp_3613 lp_1353 lp_1168 lp_2767 lp_0320 lp_0199 lp_0823 lp_0096 lp_2733 lp_0155 lp_3489 lp_1872 lp_1485 lp_3078 lp_0156 lp_3243 lp_3250 lp_0139 lp_0154 lp_2669 lp_3244 lp_3566 lp_1486 lp_0875 lp_0472

177 Hypothetical proteins (24%) y y y y y dihydroxyacetone phosphotransferase, phosphoryl donor protein unknown lipase/esterase (putative) unknown unknown unknown unknown integral membrane protein hydrolase, HAD superfamily, Cof famil hydrolase, HAD superfamily, Cof famil adenylyl transferase (putative) unknown HD superfamily hydrolase oxidoreductase integral membrane protein integral membrane protein unknown unknown unknown unknown unknown hydrolase, HAD superfamily, Cof famil unknown unknown unknown unknown unknown hydrolase, HAD superfamily, Cof famil unknown acetyltransferase (putative) unknown unknown unknown lipase/esterase (putative) hydrolase, HAD superfamily, Cof famil unknown 0.52 0.52 0.52 0.53 0.53 0.53 0.53 0.53 0.53 0.41 0.41 0.42 0.43 0.43 0.44 0.44 0.44 0.45 0.45 0.46 0.47 0.47 0.47 0.48 0.48 0.49 0.49 0.50 0.50 0.54 0.55 0.55 0.55 0.55 0.56 0.56 -0.96 -0.95 -0.94 -0.93 -0.93 -0.92 -0.92 -0.91 -0.90 -1.29 -1.28 -1.26 -1.22 -1.21 -1.20 -1.20 -1.17 -1.15 -1.15 -1.11 -1.10 -1.09 -1.08 -1.07 -1.05 -1.04 -1.02 -1.01 -0.99 -0.89 -0.87 -0.86 -0.85 -0.85 -0.85 -0.84 dak3 lp_1484 lp_1002 lp_1880 lp_2616 lp_2219 lp_0438 lp_1067 lp_2601 lp_1876 lp_0098 lp_3602 lp_0580 lp_1424 lp_3640 lp_0778 lp_0899 lp_1762 lp_1490 lp_3232 lp_0574 lp_0395 lp_2093 lp_0846 lp_3246 lp_0542 lp_1576 lp_2787 lp_1856 lp_0170 lp_0603 lp_0507 lp_1502 lp_1566 lp_2923 lp_3537 lp_0061

178 Appendix: Color figures and supplementary material Hypothetical proteins (24%) y y unknown unknown unknown unknown extracellular protein integral membrane protein of the dedA famil unknown oxidoreductase (putative) unknown unknown unknown unknown unknown unknown unknown unknown unknown unknown unknown acetyltransferase (putative) unknown unknown unknown unknown hydrolase, HAD superfamily, Cof famil integral membrane protein (putative) unknown integral membrane protein integral membrane protein integral membrane protein integral membrane protein integral membrane protein integral membrane protein unknown unknown polyribonucleotide nucleotidyltransferase (putative) 0.57 0.57 0.57 0.58 0.58 0.59 0.59 0.60 0.60 0.60 0.61 0.61 0.61 0.62 0.62 0.62 0.62 0.62 0.62 0.63 0.63 0.65 0.65 0.66 0.66 0.66 1.51 1.51 1.51 1.51 1.52 1.53 1.53 1.54 1.56 0.64 0.59 0.59 0.59 0.59 0.60 0.61 0.61 0.62 0.64 -0.82 -0.82 -0.81 -0.79 -0.78 -0.77 -0.76 -0.74 -0.74 -0.73 -0.72 -0.71 -0.71 -0.70 -0.69 -0.69 -0.68 -0.68 -0.68 -0.67 -0.66 -0.63 -0.62 -0.60 -0.60 -0.59 -0.63 lp_1068 lp_3291 lp_2748 lp_0543 lp_2173 lp_0907 lp_1518 lp_3236 lp_1503 lp_2809 lp_3501 lp_0072 lp_2483 lp_1906 lp_1572 lp_0583 lp_0865 lp_3645 lp_2994 lp_1003 lp_2260 lp_2231b lp_0499 lp_1905 lp_0509 lp_0824 lp_3504 lp_2085 lp_1809 lp_1702 lp_0365 lp_0183 lp_2806 lp_3359 lp_2515 lp_1412

179 Hypothetical proteins (24%) 1 segregation helicase (putative) unknown oxidoreductase unknown integral membrane protein unknown unknown integral membrane protein unknown unknown unknown unknown integral membrane protein unknown oxidoreductase integral membrane protein unknown lipase/esterase (putative) unknown SUA5 family translation factor (putative) acetyltransferase (putative) integral membrane protein acetyltransferase (putative) exopolyphosphatase-related protein (putative) unknown integral membrane protein integral membrane protein unknown unknown unknown unknown unknown integral membrane protein (putative) unknown unknown unknown 3.23 3.41 1.56 1.57 1.59 1.59 1.60 1.60 1.62 1.62 1.62 1.64 1.64 1.64 1.71 1.72 1.73 1.73 1.74 1.76 1.76 1.80 1.83 1.84 1.92 2.02 2.06 2.11 2.11 2.22 2.23 2.28 2.46 2.49 2.60 2.67 1.69 1.77 0.64 0.65 0.67 0.67 0.68 0.68 0.69 0.69 0.70 0.71 0.71 0.72 0.78 0.79 0.79 0.79 0.80 0.81 0.82 0.85 0.87 0.88 0.94 1.02 1.04 1.08 1.08 1.15 1.16 1.19 1.30 1.31 1.38 1.42 2 cshA lp_0244 lp_1703 lp_2696 lp_0445 lp_0151 lp_0150 lp_2212 lp_2718 lp_3268 lp_0924 lp_0213 lp_3394 lp_1939 lp_0214 lp_2036 lp_2631 lp_0444 lp_2376 lp_3356 lp_1290 lp_2526 lp_2279 lp_0837 lp_0533 lp_1260 lp_2513 lp_0089 lp_0306 lp_0091 lp_0307 lp_2954 lp_2342 lp_2113 lp_0990 lp_1543 lp_3141

180 Appendix: Color figures and supplementary material Regulatory functions (9%) Hypothetical proteins (24%) A transcription regulator of beta-galactosidase gene catabolite control protein regulatory protein Spx catabolite control protein B transcription antiterminator response regulator transcription regulator transcription regulator of fructose operon transcription regulator sorbitol operon activator transcription antiterminator response regulator transcription regulator transcription regulator transcription regulator transcription regulator transcription regulator transcription regulator transcription regulator transcription regulator (putative) transcription regulator glucokinase regulatory protein transcription regulator transcription regulator transcription regulator of gluconeogenic genes unknown unknown unknown unknown unknown transcription regulator transcription regulator transcription regulator transcription regulator transcription regulator transcription regulator 0.49 0.54 0.45 0.46 0.48 0.41 0.38 0.37 0.35 0.37 0.25 0.55 0.49 0.50 0.50 0.52 0.45 0.48 0.48 0.48 0.48 0.49 0.44 0.44 0.37 3.57 4.17 5.09 5.26 8.12 0.29 0.34 0.56 0.56 0.57 0.58 1.84 2.06 2.35 2.40 3.02 -1.04 -0.88 -1.16 -1.12 -1.06 -1.30 -1.39 -1.42 -1.52 -1.44 -1.98 -0.86 -1.04 -1.01 -1.00 -0.96 -1.15 -1.07 -1.06 -1.05 -1.05 -1.04 -1.20 -1.18 -1.42 -1.77 -1.56 -0.84 -0.82 -0.81 -0.79 2 R rrp4 treR fruR lacR spx4 gnt ccpA ccpB rrp12 srlM bglG4 bglG5 lp_0396 lp_0285 lp_0173 lp_3006 lp_2256 lp_3476 lp_2602 lp_0172 lp_3514 lp_3124 lp_1004 lp_0435 lp_3508 lp_3470 lp_3597 lp_3549 lp_3345 lp_1487 lp_1974 lp_0262 lp_3655 lp_2095 lp_0145 lp_3142 lp_3022 lp_1611 lp_0960 lp_3529 lp_3234 lp_2651 lp_2782 lp_3638 lp_3506 lp_2903 lp_0593 lp_2902

181 proteins (21%) Transport and binding Regulatory functions (9%) A A r response regulator histidine protein kinase; sensor transcription repressor of the SOS regulon N-acetylglucosamine PTS, EIICBA phosphocarrier protein Hpr mannose PTS, EIIAB mannose PTS, EIIC mannose PTS, EIID cellobiose PTS, EIIC histidine protein kinase; sensor cellobiose PTS, EII galactitol PTS, EIIB N-acetylgalactosamine PTS, EIIB beta-glucosides PTS, EII transcription antiterminator histidine protein kinase; sensor sorbitol operon transcription regulator N-acetylgalactosamine PTS, EIID transcription regulator (putative) transcription regulator transcription regulator transcription regulator (putative) transcription regulator transcription regulator transcription regulator transcription regulator transcription regulator transcription regulator transcription regulator transcription regulator transcription regulator sigma54 activato transcription regulator transcription regulator cation transport protein 1.53 1.53 3.65 0.33 0.30 0.65 0.05 0.05 0.07 0.23 0.15 0.28 0.30 0.30 0.58 0.63 0.64 0.32 1.55 1.56 1.59 1.76 1.87 2.07 0.66 1.52 0.59 0.59 0.60 0.60 0.60 0.60 0.62 0.63 0.22 0.61 0.62 1.87 0.63 0.64 0.67 0.81 0.90 1.05 0.60 -1.59 -1.73 -4.29 -3.79 -2.13 -0.62 -4.35 -2.73 -1.84 -1.75 -1.72 -1.65 -0.79 -0.66 -0.64 -0.60 -0.77 -0.77 -0.74 -0.74 -0.74 -0.73 -0.68 -0.67 -2.18 C tsH ts9C ts9D ts8C rrp8 lexA ts20A ts35B ts19B ts15A p ts19D ts9AB lam srlR2 p p p bglG1 hpk11 hpk12 p p p p p p pts22CBA lp_3190 lp_0443 lp_3415 lp_3287 lp_0294 lp_0889 lp_1360 lp_2063 lp_0575 lp_1398 lp_0319 lp_1685 lp_2665 lp_0576 lp_0577 lp_2780 lp_3547 lp_2650 lp_1273 lp_0885 lp_3495 lp_2599 lp_3649 lp_3013 lp_0951 lp_0585 lp_3646 lp_3290 lp_3639 lp_3656 lp_3581 lp_1102 lp_0439 lp_2648 lp_2969

182 Appendix: Color figures and supplementary material proteins (21%) Transport and binding A A A iron chelatin ABC transporter, substrate binding protein beta-glucosides PTS, EIIBCA ribose transport protein fructose PTS, EIIABC maltose/maltodextrin ABC transporter subunit (putative) ribose transport protein, membrane-associated protein sucrose PTS, EIIBC beta-glucosides PTS, EIIBC iron chelatin ABC transporter, permease protein cellobiose PTS, EIIC phosphoenolpyruvate-protein phosphotransferase maltose/maltodextrin ABC transporter, substrate binding protein galacitol PTS, EII nitrite extrusion protein PTS, EIIA (putative) mannose PTS, EIIB N-acetylglucosamine/galactosamine PTS, EII iron chelatin ABC transporter, ATP-binding protein bifunctional protein: HPr kinase; P-ser-HPr phosphatase galactitol PTS, EIIB maltose/maltodextrin ABC transporter, permease protein iron chelatin ABC transporter, permease protein PTS system, trehalose-specific IIBC component cellobiose PTS, EIIC cellobiose PTS, EIIC N-acetylgalactosamine PTS, EIIC galacitol PTS, EIIC cellobiose PTS, EIIC transport protein sugar ABC transporter, substrate binding protein sugar ABC transporter, permease protein myo-inositol transport protein fucose transport protein 0.43 0.46 0.53 0.34 0.36 0.48 0.50 0.53 0.53 0.36 0.37 0.43 0.47 0.47 0.50 0.52 0.52 0.54 0.40 0.44 0.45 0.46 0.51 0.33 0.36 0.38 0.47 0.48 0.48 0.37 0.45 0.51 0.42 -1.21 -1.12 -0.92 -1.05 -1.56 -1.49 -0.99 -0.90 -0.90 -1.08 -1.48 -1.43 -1.32 -1.23 -1.17 -1.15 -1.11 -1.09 -0.99 -0.98 -0.94 -0.93 -0.90 -1.48 -1.40 -1.09 -1.06 -1.58 -1.05 -1.43 -1.17 -0.97 -1.24 C F E K K C tsI ts7C p ts35A ts11A ts10B ts19A ts36B fhuB ts29C ts19C ts35C ts23C fhuD rbsU nar fhu rbsD hpr fhuG mal malA mal pts6C p ts11BC ts1BCA ts5AB p p p p p p p p p p p p pts24BCA pts16ABC lp_0884 lp_0886 lp_2647 lp_0178 lp_1274 lp_0394 lp_0185 lp_3105 lp_0265 lp_3507 lp_3642 lp_2649 lp_3103 lp_0436 lp_3010 lp_3600 lp_3106 lp_3133 lp_3643 lp_0587 lp_3658 lp_3604 lp_0176 lp_0286 lp_2097 lp_0175 lp_3548 lp_0498 lp_1481 lp_3546 lp_3104 lp_3659 lp_0754

183 proteins (21%) Transport and binding A A A A A A BC transporter, permease protein (putative) BC transporter, permease protein BC transporter, ATP-binding protein BC transporter, permease protein A A A A sucrose PTS, EIIBCA beta-glucosides PTS, EIIBCA maltose/maltodextrin ABC transporter, permease protein glycerol uptake facilitator protein mannitol PTS, EII beta-glucosides PTS, EIIBCA quaternary ammonium compound-resistance protein teichoic acid ABC transporter, ATP-binding protein lactose transport protein mannose PTS, EII PTS, EIIB PTS, EII galactitol PTS, EII malate transport protein (putative) iron chelatin ABC transporter, permease protein iron chelatin ABC transporter, ATP-binding protein gluconate transport protein serine transporter galactitol PTS, EIIC dihydroxyacetone transport protein (putative) bifunctional protein: transcriptional regulator; PTS, EII amino acid transport protein transport protein amino acid transport protein sugar transport protein sugar transport protein bifunctional protein: transcriptional regulator; PTS, EII cadmium-/zinc-/cobalt- transporting ATPase 0.64 0.63 0.62 0.62 0.59 0.65 1.51 1.51 0.54 0.61 0.61 0.66 0.67 0.55 0.60 0.60 0.61 0.54 0.55 0.58 0.65 0.62 0.59 0.59 0.60 1.50 0.56 0.60 0.61 0.56 0.57 1.52 0.59 0.59 0.59 0.60 -0.64 -0.67 -0.69 -0.70 -0.62 -0.77 -0.88 -0.86 -0.74 -0.74 -0.72 -0.72 -0.70 -0.61 -0.59 -0.86 -0.89 -0.77 -0.63 -0.68 -0.77 -0.75 -0.75 -0.84 -0.73 -0.72 -0.83 -0.82 3 2 E ts2A fec ts10A ts34B ts21A ts36A fecD gntP ts36C tagH dhaP sdaC qacC malG glpF5 lacS p mleP p p p p p pts26BCA pts30BCA pts27BCA lp_3543 lp_3219 lp_2920 lp_3513 lp_0177 lp_3503 lp_2799 lp_3533 lp_3541 lp_3601 lp_1476 lp_3436 lp_1249 lp_2743 lp_2927 lp_0502 lp_3486 lp_0586 lp_2856 lp_3611 lp_3599 lp_3138 lp_2829 lp_0232 lp_1475 lp_3327 lp_1075 lp_0171 lp_3229 lp_3284 lp_0344 lp_2744

184 Appendix: Color figures and supplementary material proteins (21%) Transport and binding BC transporter, ATP-binding and permease protein BC transporter, permease protein BC transporter, ATP-binding protein BC transporter, ATP-binding protein BC transporter, permease protein BC transporter, permease protein BC transporter, permease protein BC transporter, ATP-binding protein BC transporter, ATP-binding protein BC transporter, ATP-binding and permease protein BC transporter, permease protein BC transporter, ATP-binding and permease protein BC transporter, ATP-binding protein A A A A A A A A A A A A A glycine betaine/carnitine/choline ABC transporter, substrate binding and permease protein potassium uptake protein glutamine ABC transporter, substrate binding and permease protein malate transport protein phosphate ABC transporter, substrate binding protein manganese ABC transporter, permease protein teichoic acid ABC transporter, permease protein transport protein multidrug transport protein, major facilitator superfamily sugar transport protein glycine betaine/carnitine/choline transport protein proton/sodium-glutamate symport protein transport protein amino acid transport protein amino acid transport protein cobalt transport protein transport protein transport protein, C-terminal fragment amino acid transport protein amino acid transport protein multidrug transport protein 1.66 1.53 1.53 1.72 1.54 1.58 1.63 1.82 1.67 1.67 1.53 1.53 1.53 1.53 1.73 1.52 1.55 1.55 1.55 1.56 1.56 1.57 1.59 1.59 1.68 1.68 1.70 1.71 1.73 1.75 1.77 1.78 1.78 1.79 0.73 0.61 0.62 0.78 0.62 0.66 0.71 0.86 0.74 0.74 0.61 0.61 0.62 0.62 0.79 0.60 0.63 0.63 0.64 0.64 0.64 0.65 0.67 0.67 0.75 0.75 0.77 0.77 0.79 0.81 0.83 0.83 0.83 0.84 2 2 E st p tagG kup1 ecsB mtsB choS mleP glnPH lp_1259 lp_0217 lp_2675 lp_0331 lp_1119 lp_3324 lp_2394 lp_1336 lp_2818 lp_2111 lp_1096 lp_2038 lp_0148 lp_3225 lp_0149 lp_0888 lp_1334 lp_1456 lp_0982 lp_0101 lp_0343 lp_0525 lp_1759 lp_2897 lp_1120 lp_2912 lp_0746 lp_1811 lp_0770 lp_2893 lp_2822 lp_2894 lp_1335 lp_0367

185 proteins (21%) Transport and binding <1% and FC ≥1.5). value BC transporter, ATP-binding and permease protein BC transporter, ATP-binding and permease protein BC transporter, ATP-binding protein BC transporter, substrate binding protein A A A A phosphate ABC transporter, ATP-binding protein glycine betaine/carnitine/choline ABC transporter, ATP-binding protein copper transporting ATPase phosphate ABC transporter, permease protein phosphate ABC transporter, permease protein N-acetylglucosamine and glucose PTS, EIICBA cation transport protein amino acid transport protein transport protein transport protein transport protein transport protein purine transport protein multidrug transport protein M 1.90 1.83 1.84 1.88 2.18 2.32 1.94 1.83 1.84 1.86 1.97 1.97 2.00 2.02 2.05 2.05 2.09 2.98 0.93 0.87 0.88 0.91 1.13 1.22 0.96 0.87 0.88 0.90 1.58 0.98 0.98 1.00 1.01 1.04 1.04 1.06 log (cy5/cy3) ratio 2 stB stD stC p p p copB choQ pts18CBA M is defined as the FC is defined as Fold Change or 2 Values given as percentage of the total significant affected transcripts (p 1 2 3 lp_2395 lp_0558 lp_0368 lp_3363 lp_3278 lp_3000 lp_0747 lp_0749 lp_2525 lp_3358 lp_2509 lp_0783 lp_2514 lp_0295 lp_2710 lp_0748 lp_2531 lp_0991

186 Appendix: Color figures and supplementary material 3 trxB1 (2%) (3%) Cell envelope (8%) envelope Cell DNA metabolism (3%) Cellular processes (2%) Biosynthesis of cofactors, Energy metabolism (12%) Main Functional Class (%) Amino acid biosynthesis (1%) prosthetic groups, and carriers Central intermediary metabolism Product nitrate reductase, beta chain galactoside O-acetyltransferase recombinase A DNA-damage-inducible protein P pyruvate dehydrogenase complex, E2 component; dihydrolipoamide S-acetyltransferase molybdopterin biosynthesis protein MoaB alpha-glucosidase succinate-semialdehyde dehydrogenase (NAD(P)+) nitrate reductase, gamma chain nitrate reductase, delta chain phosphoribosyl-AMP cyclohydrolase molybdopterin biosynthesis protein, E chain molybdopterin biosynthesis protein, D chain lipoprotein precursor glycosyltransferase extracellular protein cell surface protein precursor lipoprotein precursor extracellular protein extracellular protein small heat shock protein DNA-entry nuclease UV-damage repair protein NADH dehydrogenase 1-phosphofructokinase mannose-6-phosphate isomerase 6-phospho-beta-glucosidase pyruvate oxidase nitrate reductase, alpha chain E C i 2 2 2 K m ox bg9 hisI p narI fru agl1 narJ dinP recA gene narH hsp p ndh narG p endA pdhC gabD moa thgA1 moaB umu moaD • • • • peroxide Ĺ • • • • • • • • • • • • • • • • • • • • • • • • • mutation Transcript response Ĺ t 2 men t FC 0.37 3.37 5.09 6.85 0.37 0.43 0.31 0.31 0.48 0.35 0.61 0.35 0.48 0.10 0.34 0.34 0.41 0.43 0.47 0.49 0.34 4.42 0.29 0.22 0.25 0.26 0.29 0.38 0.39 rea t on 2 ti a t FC 1.79 0.66 0.59 0.52 1.73 1.55 1.53 1.59 1.65 1.53 1.60 1.58 1.59 1.69 1.55 1.76 1.57 1.57 1.51 1.65 1.62 0.60 1.53 1.67 1.67 2.01 1.90 1.50 1.62 mu Locus lp_2301 lp_2280 lp_3023 lp_0852 lp_1478 lp_1498 lp_0393 lp_3525 lp_1499 lp_2553 lp_1495 lp_1479 lp_1070 lp_1763 lp_0304 lp_0800 lp_2098 lp_2847 lp_1449 lp_2668 lp_2906 lp_0174 lp_1069 lp_2096 lp_2384 lp_3092 lp_2152 lp_1497 lp_1500 Supplementary material File3, TABLE S3. Supplementary Significantlymaterial affected genesFile3, (93)TABLE which are found as aresult of a mutation as well by oxidative stress.

187 (9%) (13%) Protein fate (2%) Other categories (4%) Purines, pyrimidines, Regulatory functions (9%) Hypothetical proteins (32%) nucleosides and nucleotides Transport and binding proteins y y A A BC transporter, permease protein response regulator transcription antiterminator transcription repressor of the SOS regulon transcription regulator of fructose operon transcription regulator of gluconeogenic genes iron chelatin ABC transporter, substrate binding protein phosphoribosylaminoimidazole carboxylase, ATPase amidophosphoribosyltransferase precursor phosphocarrier protein Hpr protein-methionine-S-oxide reductase phosphoribosylformylglycinamidine synthase I phosphoribosylaminoimidazole carboxylase, catalytic conserved purine biosynthesis cluster protein transcription regulator mannose PTS, EIIAB mannose PTS, EIIC mannose PTS, EIID cellobiose PTS, EII phosphoenolpyruvate-protein phosphotransferase maltose/maltodextrin ABC transporter, substrate nitrite extrusion protein mannose PTS, EIIB A phosphoribosylformylglycinamidine synthase II transcription antiterminator transcription regulator oxidoreductase integral membrane protein integral membrane protein (putative) hydrolase, HAD superfamily, Cof famil unknown unknown unknown protein containing diguanylate cyclase/phosphodiesterase domain 2 (EAL) unknown hydrolase, HAD superfamily unknown oxidoreductase unknown integral membrane protein hydrolase, HAD superfamily, Cof famil adenylyl transferase (putative) integral membrane protein integral membrane protein unknown unknown acetyltransferase (putative) unknown unknown 1 segregation helicase (putative) unknown unknown unknown unknown unknown unknown transposase, fragment prophage P1 protein 1, integrase prophage P2a protein 3 prophage P2a protein 2, integrase protein-tyrosine phosphatase phosphoribosylformylglycinamidine cyclo-ligase N-acetylglucosamine PTS, EIICB phosphoribosylaminoimidazole-succinocarboxamide 3 E E K R tsI urL urF tsH ur urS urC urQ urM ts9C ts9D urK1 p rrp4 fruR lexA ptp2 ts20A ts10B p p p ts9AB gnt nar fhuD p p p p mal p p p bglG4 p bglG5 msrA p p p ts22CBA lp_1075 p • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • 0.48 3.65 0.05 0.37 0.41 0.45 5.11 5.16 0.33 4.74 4.02 5.24 5.53 0.37 0.05 0.07 0.15 0.37 0.43 0.43 0.57 0.25 0.29 0.35 0.16 0.19 0.19 0.25 0.26 0.27 0.27 0.28 0.31 0.31 0.32 0.36 0.37 0.38 0.41 0.41 0.44 0.44 0.52 1.74 1.76 2.23 2.60 3.23 3.41 3.57 4.17 5.09 5.26 8.12 0.22 3.40 3.72 4.11 0.45 3.55 5.79 0.30 0.36 5.26 1.51 0.59 2.00 1.52 1.64 1.65 0.62 0.61 1.63 0.65 0.61 1.94 1.86 1.78 0.63 0.63 1.64 1.61 1.62 1.56 1.53 1.62 1.52 1.50 1.71 1.51 1.84 1.78 1.61 1.59 1.51 1.71 1.80 1.75 1.58 1.60 1.54 1.63 1.66 1.69 1.52 1.66 1.53 0.40 0.65 0.65 0.66 0.64 0.60 0.66 0.58 0.50 0.55 0.48 1.82 0.61 0.26 0.59 1.59 0.60 1.73 1.65 0.63 0.60 lp_3514 lp_2063 lp_0575 lp_1974 lp_1487 lp_0587 lp_2723 lp_2729 lp_2969 lp_2725 lp_2728 lp_2727 lp_2722 lp_2095 lp_0576 lp_0577 lp_2780 lp_1481 lp_3103 lp_3529 lp_3234 lp_2782 lp_0175 lp_1075 lp_3318 lp_2900 lp_2755 lp_1353 lp_1168 lp_0320 lp_0199 lp_0823 lp_1485 lp_3078 lp_3243 lp_3244 lp_1486 lp_0472 lp_1876 lp_0098 lp_3640 lp_0778 lp_1484 lp_0444 lp_3356 lp_0091 lp_2113 lp_1543 lp_3141 lp_0145 lp_3142 lp_3022 lp_1611 lp_0960 lp_0860 lp_0624 lp_2454 lp_2455 lp_3272 lp_1836 lp_1273 lp_1274 lp_2726 lp_2724

188 Appendix: Color figures and supplementary material (9%) (13%) Protein fate (2%) Other categories (4%) Purines, pyrimidines, Regulatory functions (9%) nucleosides and nucleotides Transport and binding proteins A A <1% and FC ≥1.5). BC transporter, permease protein response regulator transcription antiterminator transcription repressor of the SOS regulon transcription regulator of fructose operon transcription regulator of gluconeogenic genes iron chelatin ABC transporter, substrate binding protein phosphoribosylaminoimidazole carboxylase, ATPase amidophosphoribosyltransferase precursor phosphocarrier protein Hpr protein-methionine-S-oxide reductase phosphoribosylformylglycinamidine synthase I phosphoribosylaminoimidazole carboxylase, catalytic conserved purine biosynthesis cluster protein transcription regulator mannose PTS, EIIAB mannose PTS, EIIC mannose PTS, EIID cellobiose PTS, EII phosphoenolpyruvate-protein phosphotransferase maltose/maltodextrin ABC transporter, substrate nitrite extrusion protein mannose PTS, EIIB A phosphoribosylformylglycinamidine synthase II transcription antiterminator transcription regulator transposase, fragment prophage P1 protein 1, integrase prophage P2a protein 3 prophage P2a protein 2, integrase protein-tyrosine phosphatase phosphoribosylformylglycinamidine cyclo-ligase N-acetylglucosamine PTS, EIICB phosphoribosylaminoimidazole-succinocarboxamide value 3 E E K R tsI urL urF tsH ur urS urC urQ urM ts9C ts9D urK1 p rrp4 fruR lexA ptp2 ts20A ts10B p p p ts9AB gnt nar fhuD p p p p mal p p p bglG4 p bglG5 msrA p p p ts22CBA lp_1075 p • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • M 0.48 3.65 0.05 0.37 0.41 0.45 5.11 5.16 0.33 4.74 4.02 5.24 5.53 0.37 0.05 0.07 0.15 0.37 0.43 0.43 0.57 0.25 0.29 0.35 0.22 3.40 3.72 4.11 0.45 3.55 5.79 0.30 0.36 5.26 log (cy5/cy3) ratio. 2 1.51 0.59 2.00 1.52 1.64 1.65 0.62 0.61 1.63 0.65 0.61 1.94 1.86 1.78 0.63 0.63 1.64 1.61 1.62 1.56 1.53 1.62 1.52 1.50 1.82 0.61 0.26 0.59 1.59 0.60 1.73 1.65 0.63 0.60 M is defined as the FC is defined as Fold Change or 2 Values given as percentage of the total significant affected transcripts (p lp_3514 lp_2063 lp_0575 lp_1974 lp_1487 lp_0587 lp_2723 lp_2729 lp_2969 lp_2725 lp_2728 lp_2727 lp_2722 lp_2095 lp_0576 lp_0577 lp_2780 lp_1481 lp_3103 lp_3529 lp_3234 lp_2782 lp_0175 lp_1075 lp_0860 lp_0624 lp_2454 lp_2455 lp_3272 lp_1836 lp_1273 lp_1274 lp_2726 lp_2724 1 2 3

189 3 (2%) metabolism (1%) Cell envelope (2%) DNA metabolism (1%) Cellular processes (2%) Fatty acid and phospholipid Energy metabolism (22%) Main Functional Class (%) Amino acid biosynthesis (3%) Central intermediary metabolism Product TP-dependent Clp protease, ATP-binding subunit ClpL A acetaldehyde dehydrogenase fumarate hydratase mannitol-1-phosphate 5-dehydrogenase alpha, alpha-phosphotrehalase DNA-(apurinic or apyrimidinic site) lyase pyruvate,water dikinase phosphoglycerate mutase cell division protein SufI formate C-acetyltransferase aldose 1-epimerase L-lactate dehydrogenase pyruvate oxidase pyruvate oxidase pyruvate oxidase L-ribulokinase (putative) dihydroxyacetone phosphotransferase, binding sub-unit phosphoenolpyruvate carboxykinase (ATP) bifunctional protein: alcohol dehydrogenase; acetaldehyde dehydrogenase glycosyltransferase (putative) phosphoglucomutase pyruvate dehydrogenase complex, E1 component, beta subunit pyruvate dehydrogenase complex, E1 component, alpha subunit pyruvate oxidase L-iditol 2-dehydrogenase 3-phosphoshikimate 1-carboxyvinyltransferase cysteine synthase prephenate dehydrogenase lipoprotein precursor short-chain dehydrogenase/oxidoreductase glucokinase 2 FC 0.11 0.31 0.36 0.60 6.73 0.54 0.38 0.60 1.77 0.40 0.52 0.60 1.81 1.91 2.45 2.48 1.75 0.43 0.45 1.72 1.74 0.53 0.57 1.95 2.01 0.56 0.58 0.59 0.45 1.70 1.73 1 M 2.75 0.82 0.86 0.94 1.29 1.31 0.81 0.78 0.80 0.96 1.01 0.76 0.79 -3.24 -1.70 -1.49 -0.73 -0.90 -1.40 -0.74 -1.32 -0.96 -0.73 -1.23 -1.15 -0.91 -0.80 -0.82 -0.79 -0.76 -1.16 3 2 2 2 2 ck ps gm ox1 ox5 ox ox3 flB dhB dhA mg9 p sufI p fum tyrA treA clpL ica nth1 p mtlD gutB araB aroE dak2 p p p p cysK p Gene acdH p p adhE ldhL p galM Locus lp_0329 lp_1112 lp_0233 lp_1912 lp_0355 lp_0263 lp_0764 lp_3589 lp_0852 lp_3556 lp_0169 lp_3583 lp_2860 lp_3545 lp_2629 lp_2035 lp_3313 lp_1101 lp_3418 lp_2153 lp_0060 lp_0441 lp_0849 lp_3662 lp_3487 lp_2154 lp_3170 lp_0256 lp_2034 lp_1070 lp_2142 Supplementary material File 4, TABLE S4. Significantly affected genes (90) that show a different response towards towards response different a show that (90) genes affected Significantly S4. TABLE 4, File material Supplementary to wild-type in strain NZ7608 compared peroxide, hydrogen

190 Appendix: Color figures and supplementary material (8%) Protein fate (1%) Other categories (1%) Protein synthesis (2%) Purines, pyrimidines, Hypothetical proteins (24%) nucleosides and nucleotides y y ribosomal protein S15 ribosomal protein S21 nucleoside-diphosphate kinase orotate phosphoribosyltransferase aspartate carbamoyltransferase protein-methionine-S-oxide reductase carbamoyl-phosphate synthase, pyrimidine-specific, large chain dihydroorotate oxidase dihydroorotase orotidine-5'-phosphate decarboxylase unknown unknown unknown unknown unknown unknown unknown unknown unknown integral membrane protein unknown hydrolase, HAD superfamily, Cof famil hydrolase, HAD superfamily, Cof famil integral membrane protein integral membrane protein integral membrane protein integral membrane protein hydrolase, HAD superfamily integral membrane protein integral membrane protein unknown unknown prophage P2a protein 3 0.49 0.58 0.46 1.80 2.11 1.70 1.74 1.75 1.71 1.75 0.18 0.48 0.50 0.53 0.55 0.55 0.56 0.59 0.59 1.67 1.68 1.70 1.73 1.75 1.79 1.80 1.80 1.86 1.93 1.98 2.02 2.98 3.34 0.85 1.08 0.76 0.78 0.80 0.81 0.81 0.74 0.75 0.77 0.79 0.81 0.84 0.85 0.85 0.89 0.95 0.99 1.02 1.57 1.74 -1.04 -0.79 -1.11 -2.45 -1.06 -0.99 -0.90 -0.86 -0.85 -0.84 -0.75 -0.75 3 yrF yrE yrB yrD yrC yrAB ndk p p p rpsU p p rpsO p msrA lp_2703 lp_2125 lp_1973 lp_0242 lp_2700 lp_1345 lp_1708 lp_3566 lp_2513 lp_2160 lp_2616 lp_3057 lp_0058 lp_2036 lp_3360 lp_2766 lp_3537 lp_1353 lp_2696 lp_0472 lp_3359 lp_0778 lp_3078 lp_1358 lp_0365 lp_0061 lp_0444 lp_2454 lp_1836 lp_2699 lp_2702 lp_2698 lp_2697

191 (22%) Transcription (1%) Regulatory functions (6%) Transport and binding proteins A A <1% and FC ≥1.5). value A BC transporter, permease protein (putative) BC transporter, permease protein A A mannose PTS, EIIAB beta-glucosides PTS, EIIBC ferrous iron transport protein A beta-glucosides PTS, EIIBC transcription regulator, mannitol operon mannitol PTS, EIICB maltose/maltodextrin ABC transporter, substrate binding protein ribose transport protein galactitol PTS, EIIC cellobiose PTS, EIIC maltose/maltodextrin ABC transporter, permease protein galacitol PTS, EIIC mannitol PTS, EII Na(+)/H(+) antiporter phosphoenolpyruvate-protein phosphotransferase galactitol PTS, EIIB cellobiose PTS, EIIC transcription regulator DNA-directed RNA polymerase, beta subunit transport protein multidrug transport protein transcription regulator transcription regulator transcription regulator 0.57 0.40 0.47 0.49 0.35 0.32 0.57 1.67 1.83 1.87 1.92 1.97 0.43 1.73 2.18 0.54 1.72 0.47 2.25 0.58 1.68 1.71 2.59 1.69 1.69 1.74 M 0.74 0.87 0.90 0.94 0.98 0.79 0.79 1.12 1.17 0.75 0.78 1.37 0.76 0.76 0.80 -1.08 -0.81 -1.33 -1.04 -1.62 -1.50 -0.80 -1.21 -0.88 -1.07 -0.78 log (cy5/cy3) ratio 2 E R tsI ts2A ts8C p treR ts35B mtl feoA ts36C ts23C ts35C rpoB ts9AB rbsU ts2CB mal malG p p napA4 p p p p p p ts33BCA ts30BCA p p M is defined as the FC is defined as Fold Change or 2 Values given as percentage of the total significant affected transcripts (p 1 2 3 lp_3303 lp_2856 lp_1274 lp_3010 lp_0177 lp_3546 lp_0439 lp_0175 lp_3601 lp_0231 lp_0232 lp_1467 lp_3513 lp_3565 lp_3547 lp_2613 lp_0262 lp_3527 lp_2038 lp_3658 lp_0575 lp_3598 lp_0442 lp_3234 lp_1021 lp_0230

192 Appendix: Color figures and supplementary material

CHAPTER 4 SUPPLEMENTARY MATERIAL

193 Product genome containing the L. plantarum hypothetical protein methionine sulfoxide reductase B transcription regulator transcription regulator transcription regulator integral membrane protein NADH dehydrogenase carbamoyl phosphate synthase small subunit N-acetyl-gamma-glutamyl-phosphate reductase GTP pyrophosphokinase (putative) methionyl aminopeptidase integral membrane protein biotin--[acetyl-CoA-carboxylase] ligase and biotin operon repressor NADH peroxidase ferrichrome ABC transporter, substrate binding lipo protein ferrichrome ABC transporter, ATP-binding protein glycerol kinase (3R)-hydroxymyristoyl-[acyl carrier protein] dehydratase pyruvate oxidase integral membrane protein DNA topoisomerase I plantaricin biosynthesis protein PlnQ ferrichrome ABC transporter, substrate binding lipo protein ferrichrome ABC transporter, ATP-binding protein aldehyde dehydrogenase plantaricin A precursor peptide, induction factor response regulator PlnD, repressor response regulator L-arabinose isomerase glutathione reductase response regulator aryl-alcohol dehydrogenase oxidoreductase maltose phosphorylase glutathione reductase pyruvate oxidase Motif sequence CTCAAGTGCTTGCAATCAAT TTGAGTTGTGTACAATCTAA ATTAATTGCGTACAATGTAC ATCGATTGTGCACAATCATT ATCAATCGCACGCGATTGAA ATCGATTGTGCACAATCATT ATTAATTGTGAACAATTAAA ATCGATCGCATCCAATTATA ATCGATCGCATCCAATTATA ATTGATTGAGCACAACAAAT ATTAATTGTAAACAATTCAA ATTAATTGTAAACAATTCAA ATGAATTGTGTACAATCGGT ATAAATTGTTCACAACTAAA ATCAATTGAACACTAATTAT ATCAATTGAACACTAATTAT TTCAATTGTGCAAAATTTAA CTCAATTGCAAACAATGGTT TTCAATTGCAAACAATCATT TTCAATTGCAAACAATCATT ATTGATTGCACCCAACTGAA ATAAATAGTATACAATTTTA TTTAGTTGTGCACAATCTTA TTTAGTTGTGCACAATCTTA ATTGGTTGCGCACGGTCATA ATTAATTGCAATCTATTTAT ATAAATTGGAAACAATCATT ATAAATTGCGCACGGATGAA GTCAATTGCGCAGTATGATA TTCAATCGCGTGCGACTGAA AATGATTGTAAGCAATTTAA ATTAATTGTAAGCGGTTTAC TTGAATTGAGCACAATTAAA AACGATTGCGAGCCATGAAC ATACATTGTACACAACTTAA ATAAATCGCTTTCAATTATA 2E-05 6E-08 4E-07 4E-07 1E-06 1E-06 1E-05 1E-05 1E-05 1E-06 2E-06 2E-06 3E-06 3E-06 4E-06 4E-06 7E-06 1E-05 1E-05 2E-05 2E-05 7E-06 8E-06 2E-05 2E-05 2E-05 2E-05 2E-09 5E-09 1E-08 5E-09 1E-06 2E-06 8E-06 1E-05 1E-05 pvalue Motif Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr Ohr rrp8 rrp8 glpK plnA npr2 carA topA aldH plnD fhuD fhuC plnQ fhuD fhuC araA pox3 pox2 ndh1 birA2 pepM map2 fabZ2 Gene argC2 msrA3 gshR3 gshR2 Locus lp_0450 lp_1836 lp_1360 lp_0889 lp_1821 lp_0891 lp_0313 lp_0527 lp_0528 lp_0257 lp_2544 lp_0259 lp_0854 lp_0370 lp_3103 lp_3104 lp_0415 lp_1677 lp_2629 lp_2630 lp_1850 lp_0413 lp_3103 lp_3104 lp_0047 lp_1822 lp_0418 lp_2665 lp_3554 lp_0181 lp_2665 lp_3054 lp_1939 lp_0293 lp_1253 lp_0852 Ohr-like consensus. Ohr-like Supplementary material S1. File Genes 1, extracted TABLE from the

194 Appendix: Color figures and supplementary material <0.01 value (1%) Cell envelope (7%) Main Functional Class Biosynthesis of cofactors, Amino acid biosynthesis (1%) prosthetic groups, and carriers Significant genes have a p Product NZ7603 and WCFS1 (370 affected genes). UDP-N-acetylmuramoylalanyl-D-glutamyl-2,6-diaminopimelate--D-alanyl-D- alanine ligase extracellular protein, gamma-D-glutamate-meso-diaminopimelate muropeptidase (putative) aspartate kinase diaminopimelate epimerase 2,3,4,5-tetrahydropyridine-2-carboxylate N-succinyltransferase glycine hydroxymethyltransferase phosphoglycerate dehydrogenase penicillin binding protein 2B phospho-N-acetylmuramoyl-pentapeptide-transferase UDP-N-acetylmuramate--alanine ligase penicillin binding protein 2B serine-type D-Ala-D-Ala carboxypeptidase N-acetylmuramoyl-L-alanine amidase thioredoxin cobyric acid synthase (putative) D-alanyl transfer protein DltD lysozyme (putative) UDP-N-acetylmuramoylalanyl-D-glutamate--2,6-diaminopimelate ligase serine-type D-Ala-D-Ala carboxypeptidase glycosyltransferase acyltransferase (putative) mannosyl-glycoprotein endo-beta-N-acetylglucosaminidase cell surface protein precursor cell surface protein precursor lipoprotein precursor cell surface protein precursor (putative) cell surface protein precursor adherence protein cell surface hydrolase, membrane-bound (putative) pyrazinamidase/nicotinamidase glycosyltransferase extracellular protein 2 FC 0.57 0.64 0.66 1.52 1.73 0.64 0.58 0.58 0.63 0.67 1.50 1.59 0.37 0.61 0.37 0.50 0.62 0.51 0.54 0.67 0.59 1.54 1.55 1.57 1.58 1.60 1.64 1.80 0.65 0.44 0.63 0.55 1 0.60 0.79 0.59 0.67 0.62 0.63 0.65 0.66 0.68 0.71 0.85 -0.80 -0.65 -0.60 -0.65 -0.80 -0.79 -0.66 -0.59 -1.43 -0.72 -1.42 -1.00 -0.68 -0.96 -0.88 -0.59 -0.76 -0.62 -1.18 -0.68 -0.87 M 2 2 Y C Lactobacillus plantarum lys clp\ lytH dltD glyA dapF dacB Gene murF bp2B1 bp2B mra thrA1 dapD cobQ mur serA dacA1 cps1D murE1 p p Locus lp_0979 lp_2185 lp_2264 lp_2375 lp_0785 lp_2270 lp_0856 lp_2200 lp_0518 lp_0182 lp_0977 lp_1568 lp_1010 lp_3451 lp_3074 lp_2098 lp_3001 lp_1982 lp_0923 lp_1793 lp_2586 lp_2382 lp_2612 lp_2016 lp_2162 lp_1158 lp_3189 lp_2783 lp_1462 lp_1180 lp_0304 lp_2199 Supplementary material File 2, S2. TABLE Global transcriptional analysis comparing the response to respiratory cultivation in and FC ≥1.5.

195 metabolism (2%) Cell envelope (7%) Central intermediary DNA metabolism (3%) Cellular processes (5%) A A TP-dependent Clp protease, ATP-binding subunit ClpC TP-dependent Clp protease, ATP-binding subunit ClpX TP-dependent Clp protease, ATP-binding subunit ClpB TP-dependent DNA helicase RecQ TP-dependent DNA helicase PcrA A A A A A chaperone protein DnaJ DNA-entry nuclease cell division protein FtsH, ATP-dependent zinc metallopeptidase alkaline shock protein alkaline shock protein alkaline shock protein heat shock protein GrpE heat shock protein DnaK DNA replication protein DnaD excinuclease ABC, subunit C phosphoglucomutase phosphoglycolate phosphatase (putative) alpha-glucosidase rod-shape determining protein cell division initiation protein DivIV NADH peroxidase thiol peroxidase methionine adenosyltransferase glutamine-fructose-6-phosphate transaminase (isomerizing) DNA mismatch repair protein MutS2 excinuclease ABC, subunit autoinducer production protein phosphoglucosamine mutase DNA-binding protein II DNA helicase (putative) cell surface protein precursor cell surface protein precursor nucleotide-binding protein, universal stress protein UspA family DNA-directed DNA polymerase III, epsilon chain (putative) NTP pyrophosphohydrolase 0.61 0.61 0.62 0.39 0.60 1.50 1.59 1.64 1.69 1.73 0.62 1.64 1.66 1.67 0.66 1.51 1.61 0.64 0.66 0.66 0.67 0.54 0.63 0.58 0.61 0.63 0.66 0.34 0.66 1.83 3.09 0.55 0.64 1.62 1.69 0.59 0.67 0.71 0.75 0.79 0.71 0.73 0.60 0.69 0.74 0.76 0.87 1.63 0.69 -0.71 -0.70 -0.69 -1.35 -0.73 -0.69 -0.64 -0.60 -0.59 -0.59 -0.90 -0.66 -0.59 -0.78 -0.70 -0.67 -0.60 -1.55 -0.60 -0.87 -0.64 2 2 K S K 2 E x gm crA tp ftsH agl3 p luxS clpX clpB npr clpC p uvrC grp dnaJ asp3 asp1 asp gph1 met mut endA dna hbsU dnaD glmM uvrA1 rodA recQ2 glmS1 divIVA lp_2026 lp_2906 lp_0547 lp_0774 lp_2544 lp_2027 lp_1144 lp_0432 lp_2925 lp_1229 lp_1597 lp_1322 lp_1019 lp_2116 lp_0929 lp_2028 lp_1903 lp_1301 lp_1741 lp_0811 lp_0119 lp_2109 lp_1885 lp_0872 lp_0193 lp_1879 lp_2189 lp_2323 lp_0930 lp_0822 lp_0820 lp_0773 lp_2354 lp_2271 lp_0764

196 Appendix: Color figures and supplementary material metabolism (2%) DNA metabolism (3%) Energy metabolism (9%) Fatty acid and phospholipid L-arabinose isomerase phosphoglycerate mutase phosphate acetyltransferase L-ribulokinase (putative) mannitol-1-phosphate 5-dehydrogenase acetolactate synthase glyceraldehyde 3-phosphate dehydrogenase phosphogluconate dehydrogenase (decarboxylating) pyruvate oxidase L-2-hydroxyisocaproate dehydrogenase mannose-6-phosphate isomerase D-lactate dehydrogenase phosphopyruvate hydratase H(+)-transporting two-sector ATPase, beta subunit glycine cleavage system, H protein Putative N-acetyldiaminopimelate deacetylase fructose-bisphosphate aldolase pyruvate dehydrogenase complex, E2 component; dihydrolipoamide S- acetyltransferase pyruvate oxidase glucose-6-phosphate isomerase L-2-hydroxyisocaproate dehydrogenase H(+)-transporting two-sector ATPase, epsilon subunit phosphoglycerate mutase aldose 1-epimerase aminotransferase with N-terminal regulator domain aminotransferase cyclopropane-fatty-acyl-phospholipid synthase pyruvate kinase p-nitrobenzoate reductase p-coumaric acid decarboxylase aldose 1-epimerase oxidoreductase aryl-alcohol dehydrogenase phosphopantetheinyltransferase DNA-3-methyladenine glycosylase I 0.66 0.65 0.64 0.64 0.64 0.60 0.60 0.62 0.63 0.63 0.59 0.59 0.59 0.55 0.56 0.54 0.46 0.47 0.49 0.50 0.45 0.39 0.67 1.59 1.60 1.67 0.60 0.36 0.58 1.52 1.56 1.93 2.26 0.55 1.99 0.99 0.67 0.68 0.74 0.60 0.65 0.95 1.18 -0.59 -0.62 -0.64 -0.65 -0.64 -0.75 -0.73 -0.70 -0.67 -0.66 -0.75 -0.75 -0.77 -0.86 -0.84 -0.89 -1.12 -1.09 -1.04 -1.01 -1.15 -1.35 -0.59 -0.73 -1.46 -0.78 -0.85 als pgi pta fba pyk pmi pnb cfa2 tag1 mtlD ldhD araA araB atpD atpC pox5 pox3 gnd2 gapB padA pdhC pgm2 pmg9 hicD2 hicD1 galM2 galM1 gcsH2 enoA1 lp_1731 lp_3554 lp_0597 lp_0233 lp_0807 lp_0789 lp_1541 lp_3589 lp_1245 lp_3556 lp_2057 lp_1005 lp_2384 lp_2353 lp_2263 lp_0792 lp_2364 lp_2629 lp_2502 lp_0350 lp_0330 lp_2152 lp_2363 lp_2683 lp_3207 lp_0050 lp_0296 lp_1897 lp_3170 lp_3665 lp_0826 lp_3355 lp_3054 lp_1682 lp_3174

197 Protein fate (3%) metabolism (2%) Other categories (2%) Protein synthesis (11%) Fatty acid and phospholipid ribosomal protein L27 ribosomal protein S11 ribosomal protein L19 glycine--tRNA ligase, beta chain aspartate--tRNA ligase ribosomal protein L12/L7 SSRA RNA binding protein isoleucine--tRNA ligase peptidylprolyl isomerase peptidylprolyl isomerase ribosomal protein L17 ribosomal protein S15 ribosome-binding factor A ribosomal protein S9 ribosomal protein S4 diphosphomevalonate decarboxylase acyl carrier protein phosphodiesterase carboxy-terminal processing proteinase formylmethionine deformylase oligoendopeptidase F preprotein translocase, SecE subunit Xaa-Pro dipeptidase peptidylprolyl isomerase serine/threonine protein kinase protein-methionine-S-oxide reductase elongation factor G choloylglycine hydrolase phosphomevalonate kinase prophage P2a protein 14 lysin prophage P1 protein 5 prophage P2b protein 20 prophage P2b protein 7 transposase, fragment prophage P1 protein 66, lipoprotein precursor prophage P1 protein 13 0.45 0.45 0.47 0.41 0.43 0.49 0.51 0.52 0.53 1.51 0.65 0.66 1.63 1.64 0.36 0.38 0.47 0.47 1.53 1.62 0.51 0.57 0.57 0.61 0.63 0.64 0.33 1.52 0.48 0.55 0.66 1.54 1.54 1.55 1.58 1.74 0.59 0.70 0.72 0.62 0.69 0.60 0.62 0.63 0.64 0.66 0.80 -1.15 -1.14 -1.09 -1.28 -1.23 -1.03 -0.96 -0.93 -0.91 -0.63 -0.61 -1.49 -1.40 -1.09 -1.09 -0.97 -0.82 -0.82 -0.72 -0.66 -0.64 -1.61 -1.05 -0.87 -0.60 rplL ileS rpsI rplS rbfA rplQ glyS def1 ppiB ctpA rpsK rpsD pkn2 rpsO bsh3 secE aspS rpmA pepQ smpB prtM1 prtM2 fusA2 mvaD pepF1 mrsA1 mvaK2 lp_1061 lp_2039 lp_1964 lp_1594 lp_0622 lp_0799 lp_2187 lp_1452 lp_3176 lp_2125 lp_1640 lp_1078 lp_2331 lp_1980 lp_1734 lp_0955 lp_3362 lp_2155 lp_2225 lp_0616 lp_2258 lp_3193 lp_2231c lp_1339 lp_1027 lp_1063 lp_1733 lp_2443 lp_1767 lp_0628 lp_2461 lp_2474 lp_3165 lp_0689 lp_0636 lp_1864

198 Appendix: Color figures and supplementary material Protein synthesis (11%) Hypothetical proteins (28%) A ribosomal protein L16 ribosomal protein L22 translation initiation factor IF-1 threonine--tRNA ligase 1 tyrosine--tRNA ligase ribosomal protein L20 elongation factor TS ribosomal protein L33 tRNA (5-methylaminomethyl-2-thiouridylate)-methyltransferase proline--tRNA ligase ribosomal protein (putative) ribosomal protein L36 ribosomal protein L21 tryptophan--tRNA ligase elongation factor Tu valine--tRNA ligase ribosomal protein S2 elongation factor P ribosomal protein L31 ribosomal protein L10 ribosomal protein S13 ribosomal protein S16 methionine--tRNA ligase lysine--tRNA ligase GTP-binding translation elongation factor Lep ribosomal protein S18 ribosomal protein L9 glutamate--tRNA ligase oxidoreductase esterase (putative) methylase (putative) integral membrane protein unknown unknown unknown unknown 0.67 1.51 0.66 0.65 0.66 0.64 0.64 0.65 0.61 0.63 0.59 0.59 0.60 0.61 0.58 0.58 0.59 0.58 0.55 0.55 0.55 0.55 0.56 0.54 1.51 1.60 0.53 0.59 0.28 0.34 0.43 0.44 0.48 0.49 0.50 0.52 0.59 0.59 0.68 -0.59 -0.59 -0.61 -0.61 -0.65 -0.65 -0.63 -0.71 -0.68 -0.77 -0.76 -0.73 -0.72 -0.79 -0.79 -0.77 -0.79 -0.87 -0.87 -0.86 -0.85 -0.83 -0.89 -0.91 -0.76 -1.83 -1.58 -1.21 -1.18 -1.06 -1.04 -0.99 -0.95 tsf tuf efp rplI rplJ infA gltX rplT rplP rplV tyrS thrS rplU trpS lysS valS rpsB rpsP proS rpsR trmU rpmJ rpsM metS rpmE rpmG lepA2 lp_3120 lp_1041 lp_1514 lp_1039 lp_1517 lp_2054 lp_1059 lp_2178 lp_2807 lp_1593 lp_1059a lp_1592 lp_0434 lp_0615 lp_2322 lp_2055 lp_2048 lp_2119 lp_0621 lp_1060 lp_1636 lp_0454 lp_1596 lp_0512 lp_0011 lp_0550 lp_0013 lp_0609 lp_2677 lp_3312 lp_1754 lp_0183 lp_0543 lp_2253 lp_3459 lp_2061

199 Hypothetical proteins (28%) unknown unknown unknown unknown unknown unknown unknown integral membrane protein unknown integral membrane protein (putative) unknown unknown unknown unknown unknown unknown unknown GTP-binding protein unknown unknown integral membrane protein unknown extracellular protein, membrane-anchored (putative) unknown unknown unknown integral membrane protein unknown unknown unknown unknown integral membrane protein unknown acetyltransferase (putative) integral membrane protein 0.52 0.52 0.54 0.54 0.55 0.56 0.56 0.57 0.57 0.57 0.57 0.57 0.58 0.58 0.58 0.59 0.59 0.60 0.60 0.60 0.61 0.61 0.61 0.61 0.61 0.61 0.62 0.62 0.62 0.62 0.63 0.63 0.63 0.63 0.63 -0.95 -0.94 -0.89 -0.89 -0.86 -0.83 -0.83 -0.82 -0.82 -0.81 -0.81 -0.80 -0.80 -0.79 -0.78 -0.76 -0.76 -0.75 -0.74 -0.73 -0.72 -0.71 -0.71 -0.71 -0.71 -0.71 -0.70 -0.70 -0.69 -0.69 -0.67 -0.67 -0.67 -0.66 -0.66 lp_1150 lp_1354 lp_0583 lp_0524 lp_1577 lp_2357 lp_1953 lp_2205 lp_2160 lp_2954 lp_0402 lp_2636 lp_3373 lp_2342 lp_1535 lp_1637 lp_2753 lp_1881 lp_2809 lp_1585 lp_2901 lp_0899 lp_1357 lp_0725 lp_3348 lp_3169 lp_2864 lp_1834 lp_0032 lp_0139 lp_0568 lp_1702 lp_1415 lp_3668 lp_2824

200 Appendix: Color figures and supplementary material Hypothetical proteins (28%) unknown unknown unknown unknown oxidoreductase unknown unknown integral membrane protein integral membrane protein unknown integral membrane protein unknown integral membrane protein integral membrane protein unknown unknown unknown oxidoreductase integral membrane protein unknown unknown oxidoreductase unknown unknown unknown unknown unknown unknown unknown unknown integral membrane protein (putative) unknown unknown unknown unknown unknown 0.64 0.64 0.64 0.64 0.64 0.64 0.64 0.64 0.64 0.64 0.65 0.65 0.65 0.65 0.65 0.65 0.66 0.66 0.66 0.66 0.66 1.50 1.51 1.51 1.51 1.51 1.52 1.52 1.52 1.52 1.53 1.53 1.53 1.53 1.53 1.53 0.59 0.59 0.60 0.60 0.60 0.60 0.61 0.61 0.61 0.61 0.61 0.61 0.61 0.61 0.62 -0.66 -0.66 -0.65 -0.64 -0.64 -0.64 -0.64 -0.64 -0.64 -0.64 -0.63 -0.63 -0.62 -0.62 -0.62 -0.61 -0.61 -0.61 -0.60 -0.59 -0.59 lp_1459 lp_3433 lp_2813 lp_0701 lp_3244 lp_2261 lp_0438 lp_1842 lp_3180 lp_1708 lp_0351 lp_0073 lp_2234 lp_3195 lp_1313 lp_2300 lp_1586 lp_0206 lp_3577 lp_0381 lp_0984 lp_0291 lp_3432 lp_0287 lp_0928 lp_0320 lp_1583 lp_0307 lp_3048 lp_3047 lp_3217 lp_1789 lp_1561 lp_2690 lp_3397 lp_1833

201 (4%) Purines, pyrimidines, Hypothetical proteins (28%) nucleosides and nucleotides dTDP-4-dehydrorhamnose reductase GTP pyrophosphokinase (putative) carbamoyl-phosphate synthase (glutamine-hydrolysing), large chain, truncated orotate phosphoribosyltransferase phosphoribosylaminoimidazole carboxylase, ATPase subunit thioredoxin reductase (NADPH) UTP--glucose-1-phosphate uridylyltransferase uridine kinase ribonucleotide reductase protein NrdI purine nucleosidase unknown unknown unknown unknown unknown unknown 1 segregation helicase (putative) unknown unknown integral membrane protein unknown unknown 1 segregation helicase (putative) unknown unknown integral membrane protein integral membrane protein oxidoreductase unknown unknown oxidoreductase lipase/esterase (putative) acetyltransferase (putative) unknown deoxyguanosine kinase 0.67 0.66 0.66 1.59 1.65 0.48 0.49 0.55 0.58 0.63 0.46 1.50 0.65 1.53 1.54 1.54 1.56 1.58 1.58 1.60 1.60 1.61 1.63 1.63 1.67 1.67 1.72 1.74 1.78 1.78 1.78 1.80 1.83 1.86 1.91 0.67 0.72 0.59 0.62 0.62 0.62 0.64 0.66 0.66 0.67 0.68 0.68 0.70 0.71 0.74 0.74 0.78 0.80 0.83 0.83 0.84 0.85 0.87 0.90 0.94 -0.59 -0.59 -0.59 -1.05 -1.02 -0.85 -0.79 -0.66 -1.13 -0.62 udk nrdI rfbD pyrE galU dgk2 trxB1 purK1 cshA3 cshA1 pyrAB2 lp_0293 lp_2697 lp_1190 lp_0761 lp_0757 lp_1562 lp_2673 lp_2591 lp_1784 lp_0509 lp_1788 lp_0927 lp_0828 lp_2085 lp_0960 lp_2337 lp_2485 lp_3448 lp_3080 lp_0554 lp_0990 lp_1519 lp_2883 lp_1960 lp_0533 lp_2004 lp_0146 lp_2666 lp_3394 lp_0127 lp_2631 lp_3356 lp_2584 lp_1329 lp_2728

202 Appendix: Color figures and supplementary material Purines, pyrimidines, Regulatory functions (11%) nucleosides and nucleotides aspartate carbamoyltransferase transcription antiterminator transcription regulator (putative) transcription regulator transcription regulator (putative) transcription regulator (putative) transcription regulator transcription regulator (putative) transcription regulator citrate lyase regulator histidine protein kinase PlnB; sensor transcription regulator regulator of phenolic acid metabolism PadR transcription regulator transcription regulator, N-terminal fragment (putative) transcription regulator (putative) transcription regulator transcription regulator (putative) transcription regulator transcription regulator transcription regulator ribonucleoside-diphosphate reductase, beta chain transcription regulator regulatory protein Spx central glycolytic genes regulator ferric uptake regulator histidine protein kinase; sensor transcription regulator transcription regulator (putative) histidine protein kinase; sensor (putative) response regulator transcription regulator (putative) transcription regulator (putative) transcription regulator (putative) transcription regulator (putative) transcription regulator 1.59 1.51 1.51 1.52 1.64 0.61 1.56 1.56 0.52 0.52 0.54 0.55 0.61 0.64 0.64 0.64 0.66 0.66 1.50 1.51 1.52 1.52 1.53 1.54 1.54 1.55 1.55 1.55 0.40 0.56 0.60 1.56 1.57 1.57 1.63 1.64 0.67 0.59 0.60 0.60 0.72 0.64 0.64 0.59 0.60 0.61 0.61 0.61 0.62 0.62 0.63 0.63 0.63 0.65 0.65 0.65 0.71 0.72 -0.72 -0.95 -0.94 -0.89 -0.87 -0.71 -0.65 -0.65 -0.65 -0.60 -0.59 -1.33 -0.85 -0.73 fur citR pltR plnB pyrB nrdF spx1 hpk4 hpk3 cggR padR bglG5 lp_2703 lp_0692 lp_1153 lp_1757 lp_2964 lp_3164 lp_0563 lp_3416 lp_3597 lp_1103 lp_0416 lp_2842 lp_3664 lp_1857 lp_1693 lp_0921 lp_1243 lp_0128 lp_0482 lp_3060 lp_1914 lp_1488 lp_1360 lp_0836 lp_0788 lp_3247 lp_0744 lp_3625 lp_1557 lp_3529 lp_1356 lp_3079 lp_1922 lp_3409 lp_3417 lp_2804

203 Transport and binding proteins (12%) strain NZ7603 vs wild-type. L. plantarum transport protein teichoic acid ABC transporter, permease protein amino acid transport protein ABC transporter, ATP-binding protein proton/sodium-glutamate symport protein N-acetylgalactosamine PTS, EIID cation efflux protein transport protein ABC transporter, ATP-binding protein amino acid transport protein cation transporting P-type ATPase pyrimidine nucleoside transport protein iron chelatin ABC transporter, permease protein transport protein, C-terminal fragment multidrug transport protein cation efflux protein ABC transporter, ATP-binding protein cation efflux protein (putative) spermidine/putrescine ABC transporter, permease protein multidrug transport protein ABC transporter, ATP-binding protein glycerol uptake facilitator protein ABC transporter, permease protein ammonium transport protein transport protein, C-terminal fragment 0.61 0.64 0.66 0.67 1.50 1.58 1.62 1.65 1.79 0.61 0.62 0.62 0.64 0.64 0.65 0.66 0.66 1.50 1.51 1.51 1.52 1.52 1.62 1.62 1.82 . M log (cy5/cy3) of 2 0.59 0.66 0.70 0.73 0.84 0.59 0.59 0.60 0.60 0.60 0.69 0.69 0.87 -0.71 -0.64 -0.59 -0.59 -0.71 -0.69 -0.69 -0.65 -0.64 -0.61 -0.60 -0.59 potC tagG fhuG ecsB amtB nupC glpF6 pacL2 pts19D M ratio defined as the M= Fold Change (FC) is defined as 2 lp_0595 lp_0343 lp_1722 lp_0299 lp_2818 lp_2648 lp_3280 lp_0729 lp_1771 lp_0120 lp_0567 lp_3204 lp_3106 lp_0831 lp_2836 lp_3288 lp_0723 lp_1386 lp_0316 lp_0991 lp_1336 lp_3463 lp_1456 lp_0349 lp_2897 1 2

204 Appendix: Color figures and supplementary material

CHAPTER 5 SUPPLEMENTARY MATERIAL

205 strain WCFS1 as Protein fate Main Cathegory DNA metabolism Protein synthesis Cellular processes Energy metabolism groups, and carriers Regulatory functions Hypothetical proteins Amino acid biosynthesis Biosynthesis of cofactors, prosthetic Fatty acid and phospholipid metabolism Predicted gene names, function, fold L. plantarum <0.05). value A plantaricin biosynthesis protein PlnY (putative) type I site-specific deoxyribonuclease, HsdR subunit Product argininosuccinate synthase shikimate 5-dehydrogenase thioredoxin nicotinate phosphoribosyltransferase glutathione reductase pyruvate oxidase pyruvate oxidase choloylglycine hydrolase regulatory protein Spx 2-amino-4-hydroxy-6-hydroxymethyldihydropteridine pyrophosphokinase phosphoketolase protein-methionine-S-oxide reductase signal peptidase I glutamate--tRNA ligase GTP-binding protein Typ pyruvate dehydrogenase complex, E1 component, beta subunit unknown unknown unknown unknown unknown unknown unknown secreted protein (putative) unknown unknown unknown DNA helicase (putative) transcription regulator unknown 3 K X lnY ox3 ox5 pk1 dhB fol glt sip1 typA p x spx1 bsh1 p p argG trxA1 p hsdR aroD3 msrA gshR2 nadC1 Gene 1 FC 3.09 7.78 8.25 3.02 2.73 2.54 2.84 4.77 3.87 2.80 3.37 2.83 4.29 -4.22 -3.97 -2.53 -5.07 -3.62 -4.45 -3.02 -3.07 -3.58 -2.85 -3.47 -3.80 -2.62 -3.68 -6.09 -4.10 -3.16 -3.59 18.81 lp_1234 lp_1427 lp_1535 lp_1572 lp_1766 lp_2113 lp_2306 lp_3057 lp_3058 lp_3238 lp_3429 lp_3442 lp_1836 lp_0432 lp_0938 lp_2153 lp_0775 lp_3498 lp_0236 lp_0565 lp_1253 lp_3298 lp_2659 lp_3536 Locus lp_0836 lp_3506 lp_0431 lp_3589 lp_2862 lp_0609 lp_2146 lp_2629 Supplementary File 1, TABLE S1. Summary of significant affected genes (41) in a consequence of a 10-min exposure to 10mM hydrogen peroxide (p change induction as well main functional classes of the significant affected transcripts are displayed in columns.

206 Appendix: Color figures and supplementary material Transport and binding proteins WCFS1 exposed to 10mM hydrogen peroxide L. plantarum (cy5/cy3) of a

log 2 where M = maltose/maltodextrin ABC transporter, substrate binding protein glutamine ABC transporter, substrate binding protein teichoic acid ABC transporter, permease protein ferrous iron transport protein B glutamine ABC transporter, ATP-binding protein copper transporting ATPase copper transporting ATPase ribose transport protein, membrane-associated protein glycerol-3-phosphate ABC transporter, permease protein (putative) M 2 E feoB rbsD tagG mal copA copB glnH glnQ4 3.83 2.96 2.80 6.14 5.64 3.74 -3.59 -5.66 -20.33 Fold Change (FC) is defined as 2 1 lp_0175 lp_0343 lp_2313 lp_1326 lp_1466 lp_2312 lp_3055 lp_3363 lp_3659

207 strain WCFS1 as Cell envelope Main Cathegory Amino acid biosynthesis Biosynthesis of cofactors, prosthetic groups, and carriers L. plantarum Predicted gene names, function, fold <0.05). value exopolysaccharide biosynthesis protein; chain length determinator serine-type D-Ala-D-Ala carboxypeptidase cell surface protein precursor cell surface hydrolase, membrane-bound (putative) cell surface hydrolase, membrane-bound (putative) lipoprotein precursor lipoprotein precursor (putative) cell surface hydrolase, membrane-bound (putative) glycosyltransferase cell surface protein precursor polysaccharide biosynthesis protein exopolysaccharide biosynthesis protein repeat unit transporter pantothenate kinase formylTHFpolyglutamate synthase / folylpolyglutamate dihydrofolate synthase 2-amino-4-hydroxy-6-hydroxymethyldihydropteridine pyrophosphokinase 5-formyltetrahydrofolate cyclo-ligase glutathione reductase prenyltransferase nicotinate phosphoribosyltransferase 2-dehydropantoate 2-reductase thioredoxin thioredoxin thioredoxin H-type diaminopimelate decarboxylase anthranilate synthase, component II biotin--[acetyl-CoA-carboxylase] ligase and biotin operon repressor aspartate--ammonia ligase phosphoribosylformimino-5-aminoimidazole carboxamideribotide isomerase phosphoribosyl-ATP pyrophosphatase argininosuccinate lyase shikimate 5-dehydrogenase Product argininosuccinate synthase folK fthC trxH lysA trpG hisA hisE argH argG coaA asnA folC2 trxA2 trxA3 birA2 cps2I cps1F aroD3 cps4B cps1A dacA2 gshR2 panE1 nadC1 Gene 1 FC 2.99 2.47 2.13 2.59 2.76 4.14 2.76 3.80 2.83 3.04 2.60 2.13 4.74 2.26 2.91 -3.86 -2.17 -3.25 -3.80 -2.13 -2.20 -5.96 -2.14 -2.75 -3.03 -2.22 -3.65 -2.60 -4.61 -3.23 -3.66 -2.19 -2.67 lp_1278 lp_0197 lp_0461 lp_0618 lp_0988 lp_1812 lp_2737 lp_2783 lp_2795 lp_1182 lp_1205 lp_2107 lp_3296 lp_3298 lp_1570 lp_1253 lp_1715 lp_0565 lp_2532 lp_2270 lp_3437 lp_2633 lp_1177 lp_1653 lp_0854 lp_0913 lp_2556 lp_2552 lp_1713 lp_3498 lp_0957 lp_0775 lp_0776 Locus Supplementary File 2, S2. TABLE Summary of significant affected genes (309) in a consequence of a 30-min exposure to 10mM hydrogen peroxide (p change induction as well main functional classes of the significant affected transcripts are displayed in columns.

208 Appendix: Color figures and supplementary material Cell envelope Cellular processes Central intermediary metabolism A glucan 1,6-alpha-glucosidase starch (bacterial glycogen) synthase cold shock protein CspC cell filamentation protein Fic cell division protein Fts small heat shock protein catalase nucleotide-binding protein, universal stress protein UspA family nucleotide-binding protein, universal stress protein UspA family nucleotide-binding protein, universal stress protein UspA family stress induced DNA binding protein septum site-determining protein MinC cell shape determining protein MreB NADH peroxidase plantaricin biosynthesis protein PlnY (putative) cell division protein Smc cell division protein SufI carbonate dehydratase teichoic acid biosynthesis protein di-trans-poly-cis-decaprenylcistransferase small heat shock protein glycerol-3-phosphate cytidylyltransferase poly(glycerol-phosphate) alpha-glucosyltransferase cell division protein FtsK cell division protein FtsW cell division protein GidB alpha-hemolysin homolog cell surface protein (putative) cell surface protein (putative) cell surface protein precursor extracellular protein, gamma-D-glutamate-meso-diaminopimelate muropeptidase (putative) extracellular protein cell surface protein (putative UDP-N-acetylglucosamine 1-carboxyvinyltransferase UDP-N-acetylmuramoylalanine--D-glutamate ligase fic kat sufI cah ftsA smc hlyA glgA plnY ftsW gidB npr2 hsp3 hsp2 ftsK1 dexB cspC minC uppS murD tagF1 tagE1 tagD1 mreB1 murA2 2.40 3.18 2.05 2.62 2.53 2.99 2.21 3.03 2.03 3.16 2.12 2.34 2.80 1.97 4.87 4.60 -2.27 -2.36 -2.41 -4.36 -4.06 -3.29 -1.95 -2.17 -2.74 -2.10 -2.17 -3.56 -2.79 -2.91 -2.11 -2.59 -2.34 -2.33 15.51 lp_0023 lp_3477 lp_2194 lp_3578 lp_1322 lp_2340 lp_2745 lp_3128 lp_2316 lp_2319 lp_2544 lp_0431 lp_1632 lp_0355 lp_2736 lp_3641 lp_2051 lp_0997 lp_3352 lp_1299 lp_0268 lp_2137 lp_3201 lp_2358 lp_2668 lp_3075 lp_3117 lp_3413 lp_3421 lp_3452 lp_3454 lp_2361 lp_2197 lp_0267 lp_1461

209 DNA metabolism Energy metabolism Energy metabolism Central intermediary metabolism TP-dependent helicase DinG TP-dependent DNA helicase RecQ pyruvate oxidase ribokinase phosphoketolase phosphoketolase 4-oxalocrotonate tautomerase pyruvate oxidase NifU-like protein pyruvate dehydrogenase complex, E1 component, beta subunit pyruvate dehydrogenase complex, E2 component; dihydrolipoamide S- acetyltransferase maltose O-acetyltransferase NADH dehydrogenase fructose-bisphosphate aldolase aldose 1-epimerase beta-galactosidase, large subunit 4-aminobutyrate aminotransferase UDP-glucose hydrolase (5'-nucleotidase) aminotransferase short-chain dehydrogenase/oxidoreductase malate dehydrogenase (putative) aminotransferase oxidoreductase alpha-1,2-mannosidase (putative) 2-(5''-triphosphoribosyl)-3'-dephosphocoenzyme-A synthase DNA topoisomerase aromatic amino acid specific aminotransferase aminotransferase branched-chain amino acid aminotransferase beta-glucosidase A DNA-directed DNA polymerase III, gamma/tau subunit DNA-binding protein II type I site-specific deoxyribonuclease, HsdR subunit DNA helicase (putative) A glutamine-fructose-6-phosphate transaminase (isomerizing) bgl fba nifU citG lacL xylH arcT topA xpk1 xpk2 dinG pox5 pox3 ndh1 bcaT pdhB dnaX hbsU hsdR pdhC rbsK3 araT2 galM1 recQ2 glmS1 mrt-03 2.50 3.25 2.94 2.78 2.09 2.39 4.13 1.71 1.88 2.84 2.07 2.64 2.99 4.09 2.28 2.82 3.11 2.36 2.59 2.36 -1.77 -3.97 -2.71 -2.22 -3.91 -4.39 -3.77 -4.59 -3.37 -1.89 -3.23 -2.29 -2.71 -2.95 -3.57 lp_3660 lp_2659 lp_3551 lp_1712 lp_3589 lp_2153 lp_2152 lp_2629 lp_0313 lp_1471 lp_0826 lp_3483 lp_1721 lp_2237 lp_2751 lp_2764 lp_3150 lp_3207 lp_3355 lp_3634 lp_3553 lp_0330 lp_2684 lp_1410 lp_2390 lp_3629 lp_1093 lp_0698 lp_1879 lp_0938 lp_0432 lp_1885 lp_1850 lp_0822 lp_1737

210 Appendix: Color figures and supplementary material metabolism Hypothetical proteins Fatty acid and phospholipid dihydroxyacetone phosphotransferase, phosphoryl donor protein unknown unknown unknown oxidoreductase unknown integral membrane protein unknown integral membrane protein unknown unknown unknown extracellular protein unknown nucleotide-binding protein, universal stress protein UspA family unknown unknown unknown unknown alkaline phosphatase superfamily protein methyltransferase (putative) phosphatase (putative) integral membrane protein unknown unknown acetyltransferase (putative) unknown unknown dihydroxyacetone phosphotransferase, binding sub-unit diphosphomevalonate decarboxylase choloylglycine hydrolase phosphatidate cytidylyltransferase dihydroxyacetone phosphotransferase, binding sub-unit glycerone kinase choloylglycine hydrolase holo-[acyl-carrier protein] synthase dak3 dak2 bsh3 bsh1 cdsA acpS mvaD dak1A dak1B 2.34 2.17 2.18 2.77 2.79 1.95 2.04 2.28 2.64 2.84 2.55 2.29 2.60 2.53 3.52 -3.54 -2.46 -2.63 -2.48 -1.85 -2.30 -3.33 -3.84 -2.74 -4.17 -2.66 -3.72 -2.78 -2.31 -2.90 -4.62 -3.35 -2.56 -3.21 -4.07 14.82 lp_0099 lp_0104 lp_0138 lp_0244 lp_0260 lp_0309 lp_0335 lp_0357 lp_0457 lp_0568 lp_0864 lp_0869 lp_0899 lp_1163 lp_1187 lp_1213 lp_1234 lp_1281 lp_1283 lp_1314 lp_1315 lp_1317 lp_1371 lp_1388 lp_1390 lp_1426 lp_1427 lp_1734 lp_0170 lp_2050 lp_0166 lp_0168 lp_0169 lp_3362 lp_3536 lp_0522

211 Hypothetical proteins y integral membrane protein unknown unknown integral membrane protein unknown unknown integral membrane protein unknown unknown unknown integral membrane protein oxidoreductase unknown unknown acetyltransferase (putative) unknown hydrolase, HAD superfamily unknown unknown unknown unknown unknown integral membrane protein unknown unknown polyribonucleotide nucleotidyltransferase (putative) integral membrane protein of the dedA famil phosphoesterase (putative) unknown unknown unknown unknown protein containing diguanylate cyclase/phosphodiesterase domain 1 (GGDEF) acetyltransferase (putative) unknown unknown 2.04 2.53 2.82 2.23 1.86 1.99 2.09 1.96 4.62 2.31 2.33 2.95 1.96 -2.22 -3.78 -2.75 -3.35 -3.10 -3.58 -6.02 -3.38 -3.40 -6.48 -2.91 -3.14 -3.03 -3.85 -2.74 -5.12 -3.55 -5.36 -2.89 -1.85 -4.14 -2.42 -4.96 lp_1431 lp_1535 lp_1561 lp_1571 lp_1572 lp_1583 lp_1690 lp_1708 lp_1766 lp_1833 lp_1842 lp_1860 lp_1872 lp_1884 lp_1947 lp_2000 lp_2056 lp_2061 lp_2062 lp_2083 lp_2093 lp_2113 lp_2141 lp_2230 lp_2231a lp_2231b lp_2232 lp_2238 lp_2292 lp_2299 lp_2306 lp_2347 lp_2524 lp_2579 lp_2616 lp_2636

212 Appendix: Color figures and supplementary material Other categories Hypothetical proteins unknown integral membrane protein (putative) unknown unknown unknown unknown secreted protein (putative) metal-dependent hydrolase (putative) unknown unknown unknown unknown integral membrane protein (putative) unknown unknown unknown unknown unknown unknown chloramphenicol O-acetyltransferase prophage P1 protein 20 prophage P1 protein 23 prophage P1 protein 43 lysin GTPase prophage P2a protein 56, lysin prophage P2a protein 39 prophage P2a protein 14 prophage P2a protein 13 prophage P2a protein 7 prophage P2b protein 20 prophage P2b protein 10 lysin prophage P3 protein 2, CI-like repressor 3-octaprenyl-4-hydroxybenzoate carboxy-lyase cat vdcB 2.08 2.40 4.45 7.70 7.37 3.50 2.50 2.05 2.84 2.36 3.75 2.23 3.44 3.95 3.02 2.09 3.03 2.50 2.05 3.09 3.20 3.39 3.12 -2.29 -2.18 -3.31 -1.87 -2.53 -2.88 -3.44 -2.40 -2.95 -3.23 -4.36 -3.85 lp_2669 lp_2755 lp_2786 lp_2937 lp_2987 lp_3057 lp_3058 lp_3228 lp_3238 lp_3316 lp_3323 lp_3346 lp_3410 lp_3429 lp_3433 lp_3438 lp_3442 lp_3501 lp_3566 lp_1787 lp_0643 lp_0646 lp_0666 lp_1767 lp_1854 lp_2401 lp_2418 lp_2443 lp_2444 lp_2450 lp_2461 lp_2471 lp_2810 lp_3389 lp_0271

213 Protein fate and nucleotides Protein synthesis Purines, pyrimidines, nucleosides A A glycine--tRNA ligase, alpha chain leucine--tRNA ligase ribosomal protein S30E rRNA methyltransferase glutamate--tRNA ligase preprotein translocase, YidC subunit (putative) alanine--tRNA ligase peptidylprolyl isomerase signal peptidase I dTDP-4-dehydrorhamnose 3,5-epimerase uridine kinase tripeptidase carbamoyl-phosphate synthase (glutamine-hydrolysing), small chain protein-methionine-S-oxide reductase endopeptidase PepO UDP-glucose 4-epimerase UDP-galactopyranose mutase GMP synthase (glutamine-hydrolysing) nucleotide-disulphide oxidoreductase purine nucleosidase protein-methionine-S-oxide reductase ribosomal protein L28 ribosomal protein S12 pseudouridylate synthase GTP-binding protein Typ adenine phosphoribosyltransferase cytidylate kinase formylmethionine deformylase signal peptidase II rRNA methylase (putative) ribosomal protein L11 methyltransferase (putative) unknown methionine--tRNA ligase 4'-phosphopantetheinyl transferase phenylalanine--tRNA ligase, alpha chain pseudouridylate synthase apt udk glf1 gltX truB cmk sip1 rfbC rpsL def1 lspA leuS alaS typA glyQ cspR pepT metS guaA trxB2 pheS npsC rpmB rluD1 pepO yidC1 prtM1 galE2 msrA3 msrA2 pyrAA2 1.92 1.89 1.80 4.71 2.76 4.07 2.10 -2.15 -3.09 -2.34 -2.09 -6.58 -3.12 -1.93 -2.53 -2.24 -2.58 -3.58 -2.54 -3.59 -3.50 -3.03 -3.05 -5.30 -2.70 -2.21 -3.60 -4.29 -2.06 -2.82 -2.85 -2.70 -2.56 -1.74 -2.72 -2.36 lp_1316 lp_0737 lp_0609 lp_1965 lp_2277 lp_2211 lp_2862 lp_1553 lp_1562 lp_1452 lp_1188 lp_3445 lp_1904 lp_1176 lp_0914 lp_2585 lp_2825 lp_1783 lp_1836 lp_1025 lp_2032 lp_2146 lp_1883 lp_1200 lp_2155 lp_1780 lp_1835 lp_1724 lp_1990 lp_2044 lp_0454 lp_0582 lp_1558 lp_1781 lp_1624 lp_2086

214 Appendix: Color figures and supplementary material Transcription Regulatory functions Transport and binding proteins TP-dependent RNA helicase BC transporter, ATP-binding protein BC transporter, substrate binding proteins transcription termination-antitermination factor NusA transcription antitermination protein NusG pyrimidine operon regulator regulatory protein Spx regulatory protein Spx response regulator; accesory gene regulator protein A transcription regulator transcription regulator (putative) transcription regulator transcription regulator transcription regulator transcription regulator transcription regulator transcription regulator transcription regulator transcription regulator (putative) transcription regulator transcription regulator copper transporting ATPase dihydroxyacetone transport protein (putative) glutamine ABC transporter, ATP-binding protein glycerol uptake facilitator protein ferric uptake regulator copper transporting ATPase ferrous iron transport protein B glutamine ABC transporter, substrate binding protein transcription antiterminator transcription regulator A A iron chelatin ABC transporter, substrate binding protein (putative) A transport protein polar amino acid ABC transporter, substrate binding and permease protein amino acid transport protein (putative) amino acid transport protein fur fecB rhe2 feoB spx1 spx5 ecsA lamA nusA copB copA dhaP copR nusG glpF4 glnH2 glnQ4 bglG1 pyrR1 1.92 3.86 1.97 1.99 2.64 3.24 3.87 2.43 2.09 5.71 2.20 2.08 2.47 2.97 7.14 2.56 4.31 -3.08 -1.91 -2.31 -1.79 -3.48 -2.78 -2.28 -3.97 -2.66 -3.97 -3.54 -2.58 -2.22 -2.25 -5.32 -4.11 -2.70 -2.83 -2.68 lp_0617 lp_1162 lp_0836 lp_3579 lp_2043 lp_0319 lp_0358 lp_0442 lp_0593 lp_1258 lp_1360 lp_2742 lp_2871 lp_3444 lp_3458 lp_3478 lp_3506 lp_2704 lp_0171 lp_1455 lp_1175 lp_0201 lp_3580 lp_3363 lp_2312 lp_2313 lp_3365 lp_3247 lp_3055 lp_1473 lp_1466 lp_0559 lp_0802 lp_0861 lp_0982 lp_0885

215 Transport and binding proteins WCFS1 exposed to 10mM hydrogen t=0 L. plantarum vs WT t=30 A A log (cy5/cy3) of a 2 BC transporter, ATP-binding protein BC transporter, permease protein BC transporter, ATP-binding protein BC transporter, ATP-binding protein BC transporter, ATP-binding and permease protein WCFS1. Thus WT phosphate ABC transporter, permease protein PTS, EII maltose/maltodextrin ABC transporter, substrate binding protein malate transport protein Na(+)/H(+) antiporter oligopeptide ABC transporter, substrate binding protein oligopeptide ABC transporter, permease protein teichoic acid ABC transporter, permease protein teichoic acid ABC transporter, ATP-binding protein A amino acid transport protein sugar transport protein Mg2+ and Co2+ transport protein transport protein A A A transport protein A glycerol-3-phosphate ABC transporter, permease protein (putative) transport protein amino acid transport protein transport protein sorbitol PTS, EII PTS system, trehalose-specific IIBC component ribose transport protein, membrane-associated protein ribose transport protein where M = M pstC tagH rbsD rbsU tagG nha1 malE oppA oppB mleP1 pts34A pts37A pts5ABC L. plantarum 2.26 2.79 2.62 3.24 3.22 2.88 3.31 2.20 3.89 3.39 -4.60 -3.58 -3.30 -2.99 -1.84 -4.32 -3.02 -4.03 -2.75 -2.37 -3.83 -2.41 -2.55 -3.87 -3.00 -3.23 -20.18 peroxide for 30 minutes vs Fold Change (FC) is defined as 2 lp_3542 lp_0594 lp_0192 lp_1261 lp_1262 lp_0748 lp_0344 lp_2240 lp_2503 lp_2565 lp_2712 lp_2740 lp_2743 lp_2774 lp_2789 lp_3000 lp_0175 lp_1326 lp_1689 lp_1811 lp_1863 lp_1958 lp_3618 lp_0265 lp_3659 lp_3658 lp_0343

216 Appendix: Color figures and supplementary material

- value strain WCFS1 com metabolism Cell envelope Main Cathegory DNA metabolism Cellular processes Energy metabolism Hypothetical proteins Biosynthesis of cofactors, Fatty acid and phospholipid prosthetic groups, and carriers L. plantarum WCFS1 exposed to 10mM hydrogen t=0 L. plantarum vs WT t=30 log (cy5/cy3) of a 2 TP-dependent Clp protease, ATP-binding subunit ClpX dihydroxyacetone phosphotransferase, binding sub-unit Product thioredoxin extracellular protein, gamma-D-glutamate-meso-diaminopimelate muropeptidase (putative) A cell division protein FtsH, ATP-dependent zinc metallopeptidase small heat shock protein thiol peroxidase DNA topoisomerase acetate kinase H(+)-transporting two-sector ATPase, epsilon subunit acetyl-CoA carboxylase, biotin carboxyl carrier protein choloylglycine hydrolase cyclopropane-fatty-acyl-phospholipid synthase cobyric acid synthase (putative) phosphate starvation-inducible protein unknown unknown unknown unknown unknown unknown integral membrane protein unknown unknown unknown unknown unknown integral membrane protein unknown nucleotide-binding protein, universal stress protein UspA family FMN-binding protein extracellular protein cell surface protein (putative) WCFS1. Thus WT 2 C where M = x M hoH tp strain NZ7618 as a consequence of a 30-min exposure to 10mM hydrogen peroxide (p ftsH cfa1 clpX topA atp ack3 hsp bsh3 trxA1 p cobQ accB3 dak1A Gene L. plantarum 1 FC 2.88 2.45 5.49 2.63 2.32 2.60 2.86 2.95 1.86 2.79 -2.48 -2.29 -2.22 -4.01 -3.33 -3.45 -2.57 -3.01 -3.62 -2.62 -2.44 -2.90 -2.89 -2.18 -2.47 -2.34 -2.54 -1.91 -2.11 -2.53 -2.17 -22.88 -24.89 L. plantarum Predicted gene names, function, fold change induction as well as main functional classes of the significant affected transcripts peroxide for 30 minutes vs Fold Change (FC) is defined as 2 lp_0158 lp_0383 lp_0479 lp_0837 lp_0995 lp_1354 lp_1507 lp_1556 lp_1992 lp_2113 lp_2332 lp_2333 lp_2334 lp_2342 lp_1449 lp_2162 lp_3117 lp_2116 lp_2652 lp_1970 lp_1850 lp_2242 lp_2363 lp_3490 lp_0362 lp_3362 lp_1696 lp_0166 Locus lp_0547 lp_2668 lp_2382 lp_0236 lp_2323

Supplementary File 3, TABLE S3. Summary of significant genes affected (61) Supplementary in File 3, TABLE pared to <0.05). are displayed in columns.

217 Protein fate and nucleotides Other categories Protein synthesis Regulatory functions Hypothetical proteins Transport and binding proteins Purines, pyrimidines, nucleosides WCFS1 vs NZ7618 as a result of exposure to 10mM hydrogen peroxide for30 min. A L. plantarum log (cy5/cy3) of a 2 central glycolytic genes regulator transcription regulator of beta-galactosidase gene regulatory protein Spx glycine betaine/carnitine/choline ABC transporter, substrate binding and permease protein copper transporting ATPase copper transporting ATPase teichoic acid ABC transporter, ATP-binding protein nucleotide-disulphide oxidoreductase phenylalanine--tRNA ligase, beta chain GTP-binding protein Typ protein-methionine-S-oxide reductase protein-methionine-S-oxide reductase preprotein translocase, SecA subunit alanine--tRNA ligase methionine--tRNA ligase stress-responsive transcription regulator (putative) transcription regulator (putative) transcription regulator (putative) transcription regulator transcription regulator (putative) sugar transport protein ribosomal protein acetylating enzyme unknown unknown unknown unknown prophage P1 protein 63 prophage P1 protein 64 2 3 where M = S M heT lacR typA alaS tagH spx3 secA choS copA copB p met cggR trxB2 msrA msrA 5.48 3.10 2.51 1.97 2.22 2.34 2.71 2.87 1.94 2.26 2.09 2.16 2.95 -2.11 -1.86 -2.13 -7.97 -3.05 -2.25 -3.02 -2.70 -3.49 -2.56 -2.43 -2.48 -2.93 -2.93 -16.70 Fold Change (FC) is defined as 2 lp_3470 lp_0126 lp_1557 lp_1911 lp_2335 lp_2800 lp_2228 lp_0367 lp_3055 lp_3363 lp_3611 lp_0344 lp_0788 lp_2146 lp_1836 lp_0739 lp_2277 lp_2360 lp_0454 lp_1559 lp_2669 lp_2770 lp_3283 lp_3322 lp_0686 lp_0687 lp_1835 lp_2585 1

218 Nederlandse samenvatting

Nederlandse Samenvatting

De bacterie Lactobacillus plantarum behoort tot doxine systeem in L. plantarum bij blootstelling de groep van de melkzuurbacteriën. Deze aan relevante kweekcondities. Bacteriën uit deze fylogenetisch heterogene Door analyse van gen-expressie profielen met groep hebben als gemeenschappelijke ken- behulp van DNA-chips is inzicht verkregen in merk dat ze suikers snel omzetten in melkzuur. de metabole- en regulatieprocessen in L. planta- L. plantarum wordt wereldwijd gebruikt in voed- rum die samenhangen met een functioneel selfermenaties en wordt daarnaast ook aanget- thioredoxine systeem. Deze gegevens en inzich- roffen in het maagdarmstelsel van zoogdieren ten kunnen worden gebruikt voor het opti- inclusief dat van de mens. Er zijn L. plantarum maliseren van industriële stammen en fermen- stammen beschreven die een positieve invloed taties. op de gezondheid van de gastheer kan hebben, het zogenaamde probiotische effect. Het thioredoxine systeem is een geconserveerd enzymsysteem dat voorkomt in alle hoofdlijnen De controle van activiteit en functionaliteit van van de fylogenetische boom (Archaea, Eubac- bacteriecultures tijdens industriële- en voedself- teria en Eukaryotes). Met deze informatie kon ermentaties is essentieel voor het verkrijgen van op basis van sequentievergelijkingen de aan- smaakvolle, gezonde en veilige producten voor wezigheid van een aantal componenten van dit een concurrerende prijs. Gedurende industriële systeem in het genoom van L. plantarum wor- fermentaties moeten bacteriën een groot aan- den voorspeld. Deze voorspelling is bevestigd tal stringente condities overleven (zogenaamde door het onderzoek dat wordt beschreven in stress condities) waaronder stress veroorzaakt hoofdstuk 2 en 3. In deze hoofdstukken wordt door de aanwezigheid van zuurstof en zuurstof beschreven dat het tot het thioredoxine systeem radicalen (oxidatieve stress). Het onderzoek bes- behorende enzym thioredoxin reductase (TRXB1) chreven in dit proefschrift handelt over de reac- een belangrijke rol speelt in de oxidatieve stress tie op en overleving van oxidative stress door de respons van L. plantarum. Tevens is aangetoond industrieel relevante melkzuurbacterie L. planta- dat een variant van Lactobacillus plantarum die rum. extra veel van dit enzym (TRXB1) aanmaakt, in vergelijking met een wildtype stam, een ver- Er zijn twee mechanismen bekend die bacteriën hoogde resistentie vertoont tegen de schadelijke toepassen bij de handhaving van redoxhomeo- effecten van zuurstof en zuurstof radicalen. stase en de reactie op oxidatieve stress: (i) het thioredoxine systeem en (ii) het glutaredoxine De meeste melkzuurbacteriën zijn facultatief systeem. Dit proefschrift beschrijft de rol en anaërobe organismen, hoewel ze aerotolerant werking van het thioredoxine - en het glutare- zijn en sommige soorten zelfs een oxidatieve

219 ademhalingsketen bezitten. Stammen van L. transcriptie analyses (zie hoofdstukken 2, 3 en 4) plantarum met een defect in het thioredoxine laten zien dat ook het thioredoxine systeem van systeem bleken slechter te groeien in de aan- L. plantarum is betrokken bij een groot aantal wezigheid van zuurstof (hoofdstuk 3). Aërobe cellulaire en metabole processen zoals het sui- en respiratoire groei van L. plantarum werd kermetabolisme, purine metabolisme, DNA-re- daarom in meer detail bestudeerd in hoofstuk paratie, synthese van zwavelhoudende amino- 4. In deze studie kon de belangrijke rol van het zuren en stress respons. thioredoxine systeem in de bescherming tegen oxidatieve stress worden bevestigend. Naast het thioredoxine systeem is er ook nog Hoofdstuk 2 beschrijft verder een functionele een ander systeem waarvan bekend is dat het analyse van het gen trxB2. De voorspelling was een rol speelt bij de instandhouding van de intra- dat dit gen codeert voor een thioredoxine redu cellulaire redox balans en de oxidatieve stress tase. Deze studie heeft laten zien dat de functi- response: het glutaredoxine systeem. Dit systeem onaliteit van TrxB2- verschilt van andere thiore- bestaat uit de enzymen GSH1 en GSH2 die doxine reductases. TrxB2 komt in L. plantarum gezamenlijk verantwoordelijk zijn voor de bio- tot verhoogde expressie als reactie op een tijde- synthese van het tripeptide glutathion. lijke verhoging van de temperatuur (hitte-shock) Gereduceerd glutathion reageert makkelijk met terwijl de expressie van TrxB1 wordt verhoogd sterk geoxideerde stoffen en neutraliseert daar- na blootstelling aan waterstofperoxide. Deletie mee deze oxidanten. De rol van het glutare- van het gen trxB2 leidde tot een verhoogde ex- doxine systeem wordt beschreven in hoofdstuk 5 pressie van genen met een voorspelde functie in en ook hier is een rol in de bescherming tegen de hitte shock respons, hetgeen lijkt te duiden oxidatieve stress aangetoond. In eerste instantie om een regulatoire rol van TrxB2 in de respons kon de aanwezigheid van slechts één van twee op hitte- het tripeptide glutathion die wordt ge- genen van het glutaredoxine systeem worden produceerd door stress en niet in de respons aangetoond in het genoom van L. plantarum. op oxidatieve stress repons. Een soortgelijke Met behulp van gen-sequentie vergelijkingen analyse van de andere mogelijke genen van werd uiteindelijk toch een gen geïdentificeerd het thioredoxine systeem liet zien dat de genen dat een rol zou kunnen spelen bij glutathion trxA1, trxA2, trxA3 en trxH betrokken zijn bij de productie in L. plantarum. Dit gen (code naam: oxidatieve stress respons en dus een intergraal open reading frame lp_2336) codeert voor een onderdeel van het thioredoxine systeem in L. glutathione synthase fusie eiwit en verdere ana- plantarum vormen. lyse liet zien dat het vermoedelijk verantwoordelijk is voor glutathione productie in L. plantarum. In de bacterie Escherichia coli speelt het thiore- Samenvattend kan uit dit onderzoek worden ge- doxine systeem een rol bij een groot aantal concludeerd dat zowel het thioredoxine als ook cellulaire en metabole processen. De globale het glutaredoxine systeem een belangrijke en

220 Resúmen, comentarios finales, y perspectivas futuras-

overlappende rol spelen in de bescherming van L. plantarum tegen de schadelijke invloeden van zuurstof en zuurstofradicalen (oxidatievestress).

De reactie van een bacteriële cel op steeds ver- anderde omstandigheden wordt gereguleerd door zogenaamde transcriptie factoren. Deze factoren reguleren elk een set genen die betrokken zijn bij de specifieke respons op verande- rende condities. Een set genen die gereguleerd worden door een transcriptie factor heet een regulon en een aantal van deze regulons vormen samen een regulatoir netwerk. Naast de het verkrijgen van kennis over het thioredoxine en het glutaredoxine systeem, konden de meet- gegevens die beschreven zijn in dit proefschrift ook worden gebruikt om het regulatoire netwerk van L. plantarum ten tijde van oxidatieve stress (gedeeltelijk) te beschrijven. Door middel van uitgebreide analyses zijn een aantal transcriptie factoren gevonden die betrokken zijn bij de reactie op oxidatieve stress. Tenslotte konden ook genen die deel uitmaken van het desbetreffen- de regulon worden voorspeld. Hierdoor is een breed beeld gevormd van de oxidatieve stress respons van L. plantarum. De resultaten van deze studie met betrekking tot het thioredoxine systeem, het glutaredoxine systeem, transcriptionele netwerken en de oxidatieve stress response in L. plantarum geven nieuwe inzichten in de fysiologie van deze bacterie. Deze inzichten kunnen worden gebruikt voor verbetering en optimalisatie van industriële – en voedsel fermentaties.

221 222 Resúmen, comentarios finales, y perspectivas futuras

Resúmen, comentarios finales, y perspectivas futuras

Lactobacillus plantarum es una bacteria que plantarum con alteraciones en estos sistemas. Al tradicionalmente se usa como un cultivo inicia- usar las cepas cultivadas en diferentes condicio- dor para producir alimentos fermentados. La nes de estrés oxidativo (peróxido de hidrógeno, fiabilidad de los cultivos iniciadores es esencial diamida (estrés por tiol), crecimiento aerobio, en cuanto a calidad y propiedades funcionales y crecimiento respiratorio), se efectuó análisis (acidificación, sabor y textura), pero también global de transcripción, y después se validó los en relación con el rendimiento y robusticidad resultados usando diferentes técnicas, entre el- del crecimiento (20). Por ejemplo, la robustez las genómica comparativa, reacción en cultivos probióticos es una característica en cadena de polimerasa cuantitativa (q-PCR), y esencial porque las bacterias deben sobrevivir valoraciones enzimáticas. La genómica compar- al paso por el tubo digestivo, resistir a la flora ativa y el análisis global de transcripción ante el intestinal, persistir en el intestino, y expresar la estrés oxidativo en L. plantarum presentados en función esperada en condiciones de crecimien- esta tesis pueden usarse para optimizar cepas to desfavorables. La calidad de los productos industriales. de fermentaciones industriales en los que se usan bacterias depende de los mecanismos de Estrés oxidativo defensa que han ido evolucionando en las bac- De principio a fin en un proceso de fermentación terias fermentadoras para resistir condiciones industrial, la presencia de oxígeno es inevitable. adversas encontradas durante el procesamien- Desde la época en que el oxígeno empezó a acu- to (oxidación, oxígeno, temperatura, ácido, mularse en la atmósfera de la tierra (hace 3 500 sales). Por ende, para asegurar el uso y ren- millones de años), en los (micro)organismos han dimiento apropiados de bacterias en la indu- evolucionado mecanismos que les permite man- stria, así como para controlar la calidad de los tener un ambiente citosólico reducido (4). La toxic- productos, es necesario entender los diferentes idad del oxígeno se atribuye a especies de oxígeno mecanismos de adaptación presentes en este reactivas (ROS por sus siglas en inglés) como OH. microorganismo. Esta tesis es un estudio con (radical hidroxilo), H2O2 (peróxido de hidrógeno) - enfoque en el entendimiento del estrés oxidativo y O2 (superóxido) (14). Las células vivientes han en el modelo de la bacteria acidoláctica L. plan- evolucionado y creado maneras de afrontar la toxi- tarum. Hay dos sistemas conocidos e involucra- cidad por oxígeno al prevenir la formación de ROS, dos en el estrés oxidativo y en la homeostasis de eliminarlos (degradación enzimática y recolec- la oxidoreducción en bacterias: los sistemas de ción), reparar el daño causado, o sacrificar los blan- la tiorredoxina y de la glutarredoxina. En este cos más vulnerables (20). estudio formamos una colección de cepas de L.

223 La capacidad para responder a ROS exige El sistema de tiorredoxina mecanismos cuyo objetivo es minimizar la oxi- El sistema de tiorredoxina es un mecanismo muy dación de tiol y mitigar sus consecuencias. Los conservado en organismos vivos. Este sistema tioles pueden servir como sensores que detectan puede encontrarse dentro de Archaea, así como el estrés oxidativo, y como antioxidantes (3). en Bacteria y Animalia (23). Los miembros de Dos de estos tioles, omnipresentes en la na- este sistema (TRX, TR) también son muy conser- turaleza, son el glutatión (GSH) (17) y la tior- vados; sin embargo, la TR ha evolucionado. La redoxina (TRX) (9). Se ha mostrado que la TRX evolución de la TR desde bacterias hasta mamífe- participa ante el estrés oxidativo en bacterias ros incluye la adquisición de una especificidad (15, 25), así como el GSH (13). Estos tioles se de sustrato más amplia y un sitio catalítico de reciclan dentro de la célula por medio de su re- oxidorreducción adicional (1). El sistema de tior- ductasa respectiva; glutatión reductasa (GR) o redoxina en L. plantarum se caracterizó extensa- tiorredoxina reductasa (TR). Así, el tiol oxidado, mente en el Capítulo 2 y en el Capítulo 3. Al tiol reducido, y reductasa, forman un sistema usar la secuencia de la genoma completamente denominado el sistema de la tiorredoxina o de anotada de L. plantarum WCFS1 (10), prediji- la glutarredoxina. El sistema de tiorredoxina en mos los miembros del sistema de tiorredoxina L. plantarum WCFS1 se estudió en el Capítulo que incluyeron cuatro genes que se predijo que 2 y en el Capítulo 3, donde hemos identificado codificarían para TRX (trxA1, trxA2, trx3, trxH), y a la TR que se codifica a través de trxB1 como dos genes que se predijo que codificarían para una enzima clave en la respuesta de L. planta- TR (trxB1, trxB2). En el Capítulo 2 mostramos rum WCFS1 al estrés oxidativo. Más aún, en el que el gen trxB1 en L. plantarum WCFS1 codi- Capítulo 2 mostramos que la expresión exce- ficó la TR. La expresión excesiva del gen trxB1 siva de trxB1 en L. plantarum WCFS1 produjo en L. plantarum dio por resultado triplicación de una cepa con mayor resistencia hacia estrés por la actividad de TR, mientras que una alteración diamida y peróxido de hidrógeno. Además, en de trxB1 originó un decremento de 2.5 veces de el Capítulo 2 determinamos la transcripción de la actividad de TR (Capítulo 3) en comparación los genes trxA2 y trxB1 en condiciones de estrés con el tipo natural. oxidativo usando técnicas de q-PCR. El efecto protector del GSH en L. plantarum contra el es- Se ha sugerido que el sistema de tiorredoxina trés oxidativo se describe en el Capítulo 5. La tiene importancia en la vida aerobia en Lacto- investigación presentada en esta tesis muestra coccus lactis (8). La adaptación a un ambiente que en L. plantarum los sistemas tanto de tior- aerobio es un proceso conocido para bacterias redoxina como de glutarredoxina tienen impor- acidolácticas (LAB) relacionadas con alimentos. tancia en la respuesta al estrés oxidativo en este Casi todas las LAB son anaerobios facultativos, organismo. pero se ha encontrado que son aerotolerantes, y algunas incluso tienen la capacidad para respi- rar; este es el caso de Lc. lactis en el momen-

224 Resúmen, comentarios finales, y perspectivas futuras

to de la adición de hemina y la presencia de cada por el gen trxA2 L. plantarum WCFS1 tiene oxígeno (2, 16). El crecimiento en condiciones la mayor similitud con la TRX bien caracterizada aerobias de una cepa de L. plantarum con al- de Escherichia coli (4) y Bacillus subtilis (6). teración de trxB1 dio por resultado inhibición de 19% en comparación con el tipo natural (Capí- La gama de reacciones afectadas por la TRX en tulo 3). El crecimiento aerobio y respiratorio en las bacterias comprende proteínas citoplásmi- L. plantarum se describe en el Capítulo 4, donde cas reductoras y degradación de peróxido de presentamos la función esencial de un sistema hidrógeno al proporcionar equivalentes reduc- de tiorredoxina funcional en la adaptación de tores a la peroxirredoxina, así como al ser una L. plantarum hacia condiciones de crecimiento subunidad esencial de la DNA polimerasa del aerobias y respiratorias. bacteriófago T7, y donador de hidrógeno para la ribonucleótido reductasa (25). En estudios Se efectuó un análisis funcional del gen trxB2 proteómicos de E. coli reveló que TRX se rela- usando experimentos con q-PCR, herramientas cionó con un total de 80 proteínas que implican bioinformáticas (Capítulo 2) y análisis de datos la participación de TRX en más de 26 procesos del transcriptoma de una cepa de L. plantarum celulares (11). Los análisis de transcriptoma de con alteración de trxB2 (Capítulo 4). Estos ex- L. plantarum WCFS1 presentados en los Capítu- perimentos mostraron que el gen trxB2 no con- los 2, 3 y 4 sugieren que el sistema de tiorredox- tuvo un sitio catalítico -CxxC- (rasgo común de ina participa en varios procesos celulares en una TR); experimentos con q-PCR mostraron que esta bacteria. En el Capítulo 3, por ejemplo, di- el gen trxB2 se indujo bajo estrés por calor y que chos análisis, sugirieron que el sistema de tior- una alteración del gen trxB2 en L. plantarum dio redoxina en L. plantarum WCFS1 participa en por resultado una expresión considerablemente una amplia gama de procesos, entre ellos el más alta de genes que se predijo que codifican metabolismo de azúcar y purina. Además, en el proteínas especificamente inducidas ante el es- Capítulo 2 y en el Capítulo 4 las alteraciones del trés por calor en comparación con el tipo natu- sistema de tiorredoxina en L. plantarum WCFS1 ral. Por ende, el resultado sugiere una función dieron por resultado expresión relativa más alta del gen trxB2 en la respuesta al estrés por calor importante de genes que se predijo que partici- más que en la respuesta al estrés oxidativo en L. pan en mecanismos de reparación de DNA y re- plantarum WCFS1. El análisis del transcriptoma spuesta ante circunstancias de estrés, así como de las diferentes cepas presentadas en el Capí- genes relacionados con la actividad de vías bio- tulo 2 y en el Capítulo 4, elucidó una posible sintéticas para purinas, y para aminoácidos que participación de los genes trxA1, trxA2, trxA3 y contienen azufre. trxH en la respuesta al peróxido de hidrógeno. Estudios de homología de las proteínas codifica- Respuesta al estrés oxidativo en L. das por estos cuatro genes al nivel de aminoá- cidos sugieren que la proteína producida codifi-

225 plantarum WCFS1 y análisis de promotor para entender la respu- Se encuentra bien establecido que en las LAB, esta al estrés oxidativo y definir mecanismos así como en otras bacterias, han evolucionado nuevos en L. plantarum WCFS1. En el Capí- mecanismos de defensa para sobrevivir a cam- tulo 2 identificamos dos motivos (motifs)- con bios ambientales y estreses industriales (nutrien- servados, usando análisis de promotor de los tes, cofactores, temperatura, ROS, etc.) (19). Se genes afectados de manera importante debido sabe que los mecanismos comprendidos en la a estrés por peróxido de hidrógeno, así como respuesta al estrés en bacterias están regulados a expresión excesiva de trxB1. El primer motivo en los ámbitos tanto de transcripción como de (motif) promotor encontrado en muchos de los metabolito (14). En bacterias, la transcripción genes afectados por peróxido de hidrógeno cor- empieza cuando un complejo de RNA polim- respondió al bien caracterizado regulón LexA- erasa (RNAP) y un factor sigma (s) se unen a DinR en B. subtilis que se sabe que induce pro- sitios específicos en el DNA denominados pro- teínas que participan en la reparación del DNA motores. Hasta donde sabemos, la regulación y la supervivencia celular (24). Además, se ende la transcripción bacteriana puede estar me- contró un motivo (motif) conservado en el grupo diada por factores sigma (que afectan la capa- de genes afectados por una expresión excesiva cidad de unión de la RNAP) y por factores de de trxB1. Hasta donde sabemos, este es un transcripción (TF) (que afectan la actividad de la regulón nuevo en L. plantarum que podría ser RNAP). Aun cuando en teoría cada gen puede específico para TRX. Además, en el Capítulo 4 tener un regulador, los organismos han creado usamos instrumentos bioinformáticos, así como redes para operar de manera eficiente y minimi- análisis de secuencia de promotor para deter- zar la carga genética en la célula (20). El grupo minar la participación funcional de lp_0889 y de genes o regulón que está influido por un fac- lp_1360. Estos últimos genes lp_0889, lp_1360 tor sigma o por uno de los TF se llama una red. se caracterizaron más como homólogos del bien Los genes dentro de una red por lo general conocido regulador oxidativo OhrR de B. subtilis comparten un motivo promotor muy conserva- (3) y se predijo que su red de genes relacionada do. Identificar regulaciones y redes reguladoras incluye genes: gapB, pox3, kat, mrsA3. es esencial para controlar, predecir o someter a Los análisis de transcriptoma presentados en procesos de ingeniería la conducta de LAB (20). esta tesis también han añadido información Wels et al (22) efectuaron el trabajo pionero ha- para entender los procesos de adaptación cia descubrir las redes reguladoras presentes en después de estrés oxidativo en L. plantarum L. plantarum usando técnicas in-silico (bioinfor- WCFS1 en el ámbito de transcriptoma. En el mática) y análisis de promotor. Capítulo 4 informamos que la adaptación al oxígeno en L. plantarum WCFS1 se correla- En el estudio descrito en esta tesis se ha hecho ciona con un grupo de nueve genes, incluso uso de genómica comparativa, predicciones de los que codifican para piruvato oxidasas pox5 ortología, así como de recursos bioinformáticos y pox3, y los reguladores de transcripción pu-

226 Resúmen, comentarios finales, y perspectivas futuras

tativos codificados por Lp_0889 y Lp_1360. En posición funcional del sistema de tiorredoxina y el Capítulo 3 determinamos que un sistema de de glutarredoxina en L. plantarum en respuesta tiorredoxina por completo funcional es esencial al estrés oxidativo. para un crecimiento sostenido en condiciones aerobias y respiratorias. En el Capítulo 5 la respuesta hacia estrés por peróxido de hidrógeno COMENTARIOS FINALES en una cepa con lp_2336 alterado en compara- Los análisis impulsados por datos descritos en ción con el tipo natural difirió en la expresión esta tesis dan más información sobre las redes de los genes que se predijo que codifican para reguladoras de transcripción presentes en L. el sistema de tiorredoxina y genes relacionados plantarum. Los datos se han apoyado con litera- con estrés (trxA1, trxB2, spx, clp, mrsA3). El gen tura científica sobre la respuesta al estrés oxida- lp_2336 es un homólogo de una proteína de tivo de B. subtilis (1-3, 5), L. plantarum WCFS1 difusión glutatión sintasa, GSHF. Esta proteína (7, 9) y otras bacterias (25). En consecuencia, la de fusión se encarga de la producción de GSH participación del sistema de tiorredoxina y de in vitro en Listeria monocytogenes (5). Por ende, glutarredoxina en estas redes se ha elucidado estos resultados sugieren que hay una super- y presentado en un modelo (Fig. 1). La diferen-

FIGURA 1. Redes de transcripción activadas en L. plantarum WCFS1 como resultado de estrés oxidativo. 2 Estrés de oxidorreducción

Oxígeno

Estrés oxidativo

Peróxido de Calor hidrógeno Desequilibrio de tiol

Lp_1360 Lp_3247 Lp_0889 CstR hrcA

Limitación SOS de azufre FurR

spx

pox5 gpx tpx recA lexA uvrC dinB metEcysK fur trxB1 Lp_3128 kat ahpC mrsA3 lp_0064 pox3 gapB npr2 mntH2 gor trxA clpCclpP clpE dnaK groEL lp_3324 227 cia principal entre la respuesta hacia el estrés por resultado una respuesta al estrés oxidativo al nivel del transcriptoma bien estudiada en B. “simulada” dentro de las células y, en conse- subtilis revisada en la introducción de esta te- cuencia, una cepa con mejor resistencia a la sis, y el modelo aquí presentado (Fig. 1) es la diamida y el peróxido de hidrógeno (Capítulo ausencia de sB en L. plantarum. Se sabe que 2). Esta conclusión implica que podemos des- el factor sB controla más de 150 genes en B. encadenar resistencia más alta al estrés oxi- subtilis (1). La respuesta al estrés oxidativo en dativo en una cepa industrial al influir sobre la esta bacteria consta de una serie de regulones cantidad de TR presente dentro de las células. y mecanismos de respuesta al estrés entrelaza- Esta estrategia podría aplicarse en procesos in- dos. Se han omitido muchos genes de este mod- dustriales parcialmente anaerobios, como en la elo a fin de evitar que la representación visual preparación de quesos frescos y embutidos. La se convierta en una telaraña desorientadora. La concentración de TR intracelular podría dirigirse necesidad de esta omisión subraya la necesidad al usar técnicas de ingeniería metabólica (esta apremiante de nuevos instrumentos bioinfor- tesis), pero en teoría también al influir sobre las máticos que puedan apoyar la visualización de cantidades de TRX al ajustar la concentración una vasta cantidad de datos generados medi- de oxígeno en la fermentación. Otro resultado ante el análisis de transcriptoma global usado interesante y aplicable obtenido en esta inves- en este trabajo. Estos instrumentos deben incluir tigación es que el GSH protege a L. plantarum anotación, visualización, agrupación y bibliote- contra el peróxido de hidrógeno (Capítulo 5). cas para evaluar la función biológica de genes. De este modo, el ambiente de crecimiento op- timizado que protegería a L. plantarum contra estrés oxidativo se obtendrá en un medio con PERSPECTIVAS FUTURAS alto contenido de GSH. Para obtener un medio El control de cultivos iniciadores y de cultivos rico en GSH, la composición de los medios se probióticos en condiciones industriales es es- puede cambiar al complementar GSH de mane- encial para proporcionar un producto de sabor ra directa, y someter a las cepas a procesos de agradable, atractivo, sano, y seguro. El estrés ingeniería para producir GSH (esta tesis), o al oxidativo es una de las condiciones duras que usar complementos con alto contenido de GSH estos microbios fermentativos han conseguido (plantas, vegetales o GSH encapsulado). La soportar durante su uso en procesos de fermen- robustez de cepas industriales también es im- tación industriales. La genómica comparativa y portante en el área de alimentos probióticos. La el análisis de transcriptoma global de la respu- funcionalidad de ciertas bacterias probióticas se esta al estrés oxidativo de L. plantarum presen- fundamenta en su supervivencia al pasar por el tados en esta tesis pueden usarse para optimi- tubo digestivo, y la capacidad para persistir y a zar cepas industriales. veces colonizar la flora intestinal. Además, puede ser importante para la actividad probiótica Los resultados presentados en este estudio que los microbios liberen en estas condiciones muestran que la producción excesiva de TR da severas en el tubo digestivo los factores que

228 Resúmen, comentarios finales, y perspectivas futuras

desencadenan el efecto probiótico. Por ende, ación, peróxido de hidrógeno, disulfuro, limit- la robustez de los procesos de ingeniería en las ación de azufre, y estrés por calor, la jerarquía cepas probióticas podría ser la clave para el de la respuesta al estrés — si existe alguna — se éxito en el logro de la funcionalidad deseada. entiende menos. La superposición entre respu- Varios indicios obtenidos en este estudio para estas de estrés se ha establecido bien en bacte- mejorar la robustez de L. plantarum contra rias. Por ejemplo, se sabe que en B. subtilis, el condiciones de estrés oxidativo también podrían estrés oxidativo afecta a proteínas que también usarse para mejorar la robustez en otras bac- son inducidas bajo estrés por calor (15). En esta terias acidolácticas (probióticas). En lugar de tesis mostramos que el sistema de tiorredoxina buscar robustez mejorada, también podríamos es crucial en la respuesta al estrés oxidativo usar el conocimiento para obtener cepas que (Capítulo 3) y en la adaptación a condiciones muestren robustez reducida. Esta característica de crecimiento aerobias y respiratorias (Capí- es importante para cultivos adjuntos en diversas tulo 4); aun así, el sistema de tiorredoxina tam- fermentaciones donde estos microbios determi- bién se induce bajo estrés por calor, en especial nan la calidad del producto terminal en lo que el gen trxB2 (Capítulo 2). Si bien es fácil enlazar se refiere a sabor y textura. el estrés oxidativo y la respuesta al estrés por calor, hay otras redes de estrés más difíciles de La amplia aplicabilidad de la investigación sobre visualizar, como la relación entre la producción el sistema de tiorredoxina efectuada en esta tesis excesiva de TR y proteínas normalmente induci- se fundamenta en el hecho de que este sistema das bajo limitación de azufre. se encuentra muy conservado en organismos de diversos grupos taxonómicos. Por ende, las lec- En esta tesis se ha presentado trabajo pionero ciones obtenidas a partir de la investigación de dirigido a la comprensión de los mecanismos bacterias pueden llevar a posibles aplicaciones de respuesta presentes en el ámbito de tran- en eucariotas, incluso seres humanos. En la scriptoma en L. plantarum como resultado de investigación del cáncer, se sabe que las células estrés oxidativo. El propósito de este trabajo es cancerosas tienen un alto contenido de TRX, y servir como un instrumento, y como un bloque que las concentraciones de TRX reducida altas de construcción, para estudios adicionales. Las pueden causar enfermedad propensa a cáncer conclusiones a las cuales se llegó en relación (25). Por ende, el conocimiento del sistema de a la respuesta al estrés en L. plantarum se han tiorredoxina puede aumentar nuestro cono- basado en diseño experimental bien pensado, cimiento con miras a descubrir los mecanismos análisis estadístico objetivo de los datos y, en que están dentro de células cancerosas. muchos casos, análisis funcional adicional. Por ejemplo, el índice de crecimiento reducido El trabajo descrito en esta tesis indica que la observado con la aireación en una cepa de L. respuesta al estrés oxidativo es una red en- plantarum con alteración de trxB1 se investigó trelazada de regulones únicos y complejos. Si usando microarreglos (Capítulo 3). Este análisis bien hemos observado una relación entre aire- sugirió que el fenotipo observado se enlazó con

229 el metabolismo del azúcar y la purina. En consecuencia, cuando cultivamos la misma cepa en otro azúcar (celobiosa) como única fuente de carbono, logramos influir sobre el fenotipo y mejorar el índice de crecimiento del mutante en condiciones de cultivo aerobias. Sin embargo, fue imposible efectuar un análisis funcional de cada indicio o validar cada gen afectado e importante obtenido en nuestros resultados. Por lo tanto, sin duda en este método impulsado con muchos datos, junto con interpretación de datos, se usaron puntos de corte estadísticos, y por tanto la presencia de resultados positivos y falsos es inevitable. Además, las redes reguladoras presentes en L. plantarum y operativas bajo estrés oxidativo antes descritas, sólo incluy- en regulación en el ámbito de transcripción. Se sabe que la respuesta al estrés en, por ejemplo, Lc. lactis, está regulada en los ámbitos de factores s (transcripción), de proteína y en sistemas reguladores por dos componentes, y de metabolito (detectores del flujo de metabolito) (14). Por ende, el siguiente paso es usar un método de biología de sistemas y complementar el trabajo en esta tesis con datos sobre proteoma, y metaboloma sobre la respuesta al estrés oxidativo en L. plantarum.

230 Resúmen, comentarios finales, y perspectivas futuras

REFERENCES

1. Arner, E. S., and A. Holmgren. 2000. Physiological functions of thioredoxin and thioredoxin reductase. Eur J Biochem 267:6102-9. 2. Duwat, P., S. Sourice, B. Cesselin, G. Lamberet, K. Vido, P. Gaudu, Y. Le Loir, F. Violet, P. Lou biere, and A. Gruss. 2001. Respiration capacity of the fermenting bacterium Lactococcus lactis and its positive effects on growth and survival. J Bacteriol 183:4509-16. 3. Fahey, R. C. 2001. Novel thiols of prokaryotes. Annu Rev Microbiol 55:333-56. 4. Fedoroff, N. 2006. Redox regulatory mechanisms in cellular stress responses. Ann Bot (Lond) 98:289-300. 5. Gopal, S., I. Borovok, A. Ofer, M. Yanku, G. Cohen, W. Goebel, J. Kreft, and Y. Aharonowitz. 2005. A multidomain fusion protein in Listeria monocytogenes catalyzes the two primary activities for glutathione biosynthesis. J Bacteriol 187:3839-47. 6. Hecker, M., J. Pane-Farre, and U. Volker. 2007. SigB-dependent general stress response in Bacillus subtilis and related gram-positive bacteria. Annu Rev Microbiol 61:215-36. 7. Hecker, M., and U. Volker. 2001. General stress response of Bacillus subtilis and other bacteria. Adv Microb Physiol 44:35-91. 8. Helmann, J. D., M. F. Wu, A. Gaballa, P. A. Kobel, M. M. Morshedi, P. Fawcett, and C. Paddon. 2003. The global transcriptional response of Bacillus subtilis to peroxide stress is coordinated by three transcription factors. J Bacteriol 185:243-53. 9. Holmgren, A. 1985. Thioredoxin. Annu Rev Biochem 54:237-71. 10. Kleerebezem, M., J. Boekhorst, R. van Kranenburg, D. Molenaar, O. P. Kuipers, R. Leer, R. Tarchini, S. A. Peters, H. M. Sandbrink, M. W. Fiers, W. Stiekema, R. M. Lankhorst, P. A. Bron, S. M. Hoffer, M. N. Groot, R. Kerkhoven, M. de Vries, B. Ursing, W. M. de Vos, and R. J. Siezen. 2003. Complete genome sequence of Lactobacillus plantarum WCFS1. Proc Natl Acad Sci U S A 100:1990-5. 11. Kumar, J. K., S. Tabor, and C. C. Richardson. 2004. Proteomic analysis of thioredoxin-targeted proteins in Escherichia coli. Proc Natl Acad Sci U S A 101:3759-64. 12. Leichert, L. I., C. Scharf, and M. Hecker. 2003. Global characterization of disulfide stress inBacillus subtilis. J Bacteriol 185:1967-75. 13. Li, Y., J. Hugenholtz, T. Abee, and D. Molenaar. 2003. Glutathione protects Lactococcus lactis against oxidative stress. Appl Environ Microbiol 69:5739-45. 14. Miyoshi, A., T. Rochat, J. J. Gratadoux, Y. Le Loir, S. C. Oliveira, P. Langella, and V. Azevedo. 2003. Oxidative stress in Lactococcus lactis. Genet Mol Res 2:348-59. 15. Scharf, C., S. Riethdorf, H. Ernst, S. Engelmann, U. Volker, and M. Hecker. 1998. Thioredoxin is an essential protein induced by multiple stresses in Bacillus subtilis. J Bacteriol 180:1869-77. 16. Sijpesteijn, A. K. 1970. Induction of cytochrome formation and stimulation of oxidative dissimilation by hemin in Streptococcus lactis and Leuconostoc mesenteroides. Antonie Van Leeuwenhoek 36:335-48. 17. Smirnova, G. V., and O. N. Oktyabrsky. 2005. Glutathione in bacteria. Biochemistry (Mosc) 70:1199-211. 18. Stevens, M. J. A. 2008. Transcriptome Response of Lactobacillus plantarum to Global Regulator Deficiency, Stress and other Environmental Conditions. PhD thesis. Wageningen University. 19. Stortz, G., and R. Hengge-Aronis. 2000. Bacterial Stress Responses, Washington DC. 20. van de Guchte, M., P. Serror, C. Chervaux, T. Smokvina, S. D. Ehrlich, and E. Maguin. 2002. Stress responses in lactic acid bacteria. Antonie Van Leeuwenhoek 82:187-216. 21. Vido, K., H. Diemer, A. Van Dorsselaer, E. Leize, V. Juillard, A. Gruss, and P. Gaudu. 2005. Roles of thioredoxin reductase during the aerobic life of Lactococcus lactis. J Bacteriol 187:601-10. 22. Wels, M. 2008. Unraveling the regulatory network of Lactobacillus plantarum WCFS1. PhD thesis. Wagenin gen University. 23. Williams, C. H., Jr. 2000. Thioredoxin-thioredoxin reductase--a system that has come of age. Eur J Biochem 267:6101.

231 24. Winterling, K. W., A. S. Levine, R. E. Yasbin, and R. Woodgate. 1997. Characterization of DinR, the Bacillus subtilis SOS repressor. J Bacteriol 179:1698-703. 25. Zeller, T., and G. Klug. 2006. Thioredoxins in bacteria: functions in oxidative stress response and regula tion of thioredoxin genes. Naturwissenschaften 93:259-66.

232 Training and Supervision Plan (VLAG)

Training and Supervision Plan (VLAG)

Courses: • BIT1- bioinformatics course (2004, Vlag). • 4th Int. Advanced course on Chemistry and biochemistry of antioxidants • 1st FEBS Advanced Lecture Course: Systems Biology • Safe Handling with radioactive materials and sources • Functional genomics (UU) • Philosophy and Ethics of Food Science and Technology

Conferences: • WCFS Food Summit (2003) • Current Themes in Ecology april 2 • ALW Genetica meetings (2003-2006) • Lactic Acid bacteria 8 (2005, poster presentation) • ASM General meeting, poster presentation 21-26 may 2006 • Kluyver Centre for Genomics of industrial Fermentation (2004-2007) • NVMM oral presentation 10-12 april 2006 • NBC-11 16-17 maart 2006 (poster presentation)

Training: • Oxford 23 – 30 march 2006, St Edmunds Hall, Oxford, England General courses, language use, presentation courses, statistics, e.g. • Scientific Writing 8/05- 3/07 2003 (WUR) 24 contact hours • Course Debaat and Dialoog (WCFS) 7april

233 List of publications

Thioredoxin reductase is a key factor in the oxidative stress response of Lactobacillus plantarum WCFS1

L. Mariela Serrano, Douwe Molenaar, Michiel Wels, Bas Teusink, Peter A. Bron, Eddy J. Smid Microbial Cell Factories August 2007

Global transcriptional analysis reveals the specific role of thioredoxin reductase in oxidative stress response in Lactobacillus plantarum WCFS1

L. Mariela Serrano, Douwe Molenaar,Bas Teusink, Willem M. de Vos, Eddy J. Smid Manuscript in preparation

The thioredoxin system plays an important role in adaptation of Lactobacillus plantarum WCFS1 to aerobic and respiratory growth.

L. Mariela Serrano, Christof Francke, Douwe Molenaar, Bas Teusink, Willem M. de Vos, Eddy J. Smid Manuscript in preparation

Glutathione protects Lactobacillus plantarum WCFS1 against hydrogen peroxide stress

L. Mariela Serrano, Douwe Molenaar, Bas Teusink, Willem M. de Vos, Eddy J. Smid Manuscript in preparation

234 List of publications

About the author

ters via Solventothermal Techniques” was evaluated with distinction. On September 2000 Mariela moved overseas to Delft, The Netherlands. At the TUDelft, she com- pleted the MSc in Chemical Engineering on August 2002. Her MSc thesis project was supervised by Prof. Jack Pronk and Pascal Daran-Lapujade and entitled “Tran- script and Enzyme Activity Analysis of the Central Metabolism of Saccaromyces cerevisiae grown in Chemostat Cultures.” This project was essential for getting her involved in Biotechnology and fascinat- ing world of micro-organisms. On Febru- ary 2003, Mariela moved to Zutphen, The L. Mariela Serrano was born on the 8th Netherlands and began with her Ph.D. re- of January 1978 in Quito, Ecuador. She search at NIzo Foor Research on oxidative graduated in May 1995 from the American stress response in Lactic Acid Bacteria high school in Quito. Before starting with which resulted in this thesis. Her Ph.D. re- her bachelor’s degree she went for a one- search was done in collaboration with the year study abroad program with the pro- University of Wageningen and the Top In- gram American Field Service (AFS). This stitute of Food and Nutrition ,TI Food and activity brought her to Dieren, The Neth- Nutrition (formerly known as WCFS). erlands where she lived with the Oudenrijn Currently Mariela is working as a scientist at family from august 1995 until July 1996. CSK Food Enrichment where she is putting In September 1996 Mariela moved to Lake into practice all acquired knowledge and Forest in Chicago, United States. There techniques. she pursued a Bachelor’s in Arts at Lake Forest College from 1996-2000, her major was Chemistry. Her thesis entitled “Syn- thesis of Rhenium Thiophosphate Clus-

235 236 Acknowledgements

Acknowledgements

And of course at this point I am able to thank all the people without whom this book had never become a reality. I just hope I don’t forget anyone, but knowing I will here it comes...

When I started this Ph.D. I had just moved to Zutphen “Amsterdam of the East” leaving behind the international city of Delft, where everything was done primarily in English. Suddenly I found myself travelling by train for more than two hours every day, surrounded by Dutch speaking colleagues and digging into the new world of Microbiology.

The first months I spent in the office of Jeroen, reading about lactic acid bacteria, my new pet organism, and brainstorming what I was going to do the next four years. In no time, I had a place at Marc’s Lab (lucky me); a daily supervisor, Masja, and a lot of meetings to attend to. After a couple of months there was a big change in project C008, Eddy Smid became the project leader and thus my thesis supervisor. Knowing that a thesis supervisor is an essential factor for the success of your Ph.D., this change made me nervous. Nevertheless, after getting to know Eddy, I realized that there was nothing to fear. Eddy, thanks for being a comprehensible and sensitive supervisor; you are a great researcher with the determination of a rock, and always found an alternative, a new perspective, a new chance. You provided me with optimism even in the darkest moments (and there were a few!) What I learned from you is that even though discussion is essential, performing it is even better. Of course throughout the research what could I have done without the hawk-supervision of Willem M. de Vos. Willem, without our brainstorm sessions I would certainly had been blinded from the big picture. Even though following your trait of thought was impossible; you inspired me to be better. You surprised me always with your knowledge, memory, and pace of thought. I truly believe that you are a supernatural human being.

I am a true believer that a researcher without a good working environment would not be able to perform. Therefore I would like to thank Nizo Food Research, the department of Microbiology at the University of Wageningen, and TI Food and Nutrition (formerly WCFS).

And to be more specific. To the known Nizo-Kelder specially: Ingrid, Anne, Marke, Patrick, Iris, Tanja, Saskia, Bert, Marjo, and Roger for their help, assistance both technical and spiritual. You guys were always open for a question, for listening and foremost for lending a hand at all times. I was lucky to meet such a bunch. It was great working with you! Special thanks to Iris, Roger, Jaqueline, Anne, and their respective families. You shared knowledge with me but also became close friends. Thanks for the famous nights of bonanza, camping at Oerol, beer tasting, and Kerst-klaas. Each of you is extremely special, and I will never forget all the adventures we have gone through together.

237 There are more people at Nizo whom I will like to thank for their support in my research Annerreiau, the microtiterplate expert, Barry B. for his help with different spray drying experiments, Roelie and Guido for all the HPLC measurements they carried out for me. Specially thanks to Douwe, thank your for your patience and work in data processing!

At the department of Microbiology, Francis Cottar, thanks for everything. I bombarded you with emails these last years.

At TI Food and Nutrition, all personnel were always ready to lend a hand. Michiel S., Mali, and Greer thanks. Also Greer I am very grateful that you offered to proofread this thesis, I am sure you needed a double whiskey for that!

I would also like to thank other Ph.D.’s students from the “Denk-tank” from whom I have learned quite a lot. Jolanda the quiet researcher; the white-trousers AIO: Arno, who always had an anecdote or a joke to share; the fifth Beatle, Marc, who introduced me to Bob Dylan and ABBA and for the few “Rescue my Soul” -moments when things were not going as planned. Other Ph.D. students: Gabriele, “roomie,” Marieke, and Eva, I cherish the friendships we have made through the different trips to Japan and California as well as other VLAG activities.

Support within my project and at the lab was essential for me; on the other hand, support outside the working arena was as important if not more. Adriana, Tiadrina, who has given me a flow of strength through these years, I am most grateful for that and ofcourse for the spanish translation. María Inés and Joanne I am so lucky to have friends like you. You have proven to me that distance is not an ob- stacle but only a bunch of numbers. Thanks for being there and for being my friends. To my family as big as that is. My dutch family including my Oma, Janny, Anton, Paulien, Yasmin, and Diederik for caring for me and taking me into their family. Specially thanks to Paulien, that between her work, family, and reconstruction at home helped me out in making this book so beautiful!!!!

When the Oudenrijn’s welcomed me in 1995 they did not realized that I would never leave J. I am so glad that my destiny brought me to you. Thanks for all your support.

To my parents, María and Julio who gave me the wings to explore and a set of strong legs to stand on my own. Without you papis, I would not understand the meaning of unconditional love and I surely would not have found the strength and courage to continue last year. To my siblings María Rosa, Car-

238 Acknowledgements

los Alberto, and Jose Julio who are always there for me and from whom I have learned more than I give them credit for.

To my four beautiful nieces Madison, Danica, Luciana and Sterre as well as to my nephews Yannik and Tygo. One smile on your faces was the best medicine at all times.

Four years – oops! I mean 4.5 - of work, research, struggles, discoveries, fits, gratifications, have been a great and tricky roller coaster.

Thank you all for riding it with me,

L. Mariela Serrano Alias Bella

239 Cover: “La Plaza, Ensamblaje en Madera” Oswaldo Viteri

Design and Lay-out: Paulien van den Oudenrijn

Printed by: Ponsen & Looijen B.V. , Wageningen

Printing sponsors:

240