Cell wall proteome of Clostridium thermocellum and detection of glycoproteins

Microbiological Research 167 (2012) 364–371

Contents lists available at SciVerse ScienceDirect

Microbiological Research journal homepage: www.elsevier.de/micres

Cell wall proteome of Clostridium thermocellum and detection of glycoproteins Tingting Yu, Xinping Xu, Yanfeng Peng, Yuanming Luo ∗∗ , Keqian Yang ∗ State Key Laboratory of Microbial Resources, Institute of Microbiology, Chinese Academy of Sciences, Beijing 100101, China

a r t i c l e

i n f o

Article history: Received 20 December 2011 Received in revised form 21 February 2012 Accepted 23 February 2012 Keywords: Clostridium thermocellum Two-dimensional gel electrophoresis Cell wall proteome Glycoprotein

a b s t r a c t Clostridium thermocellum, a thermophilic anaerobe, has the unusual capacity to convert cellulosic biomass into ethanol and hydrogen. In this work, the cell wall proteome of C. thermocellum was investigated. The proteins in the cell wall fraction of C. thermocellum prepared by the boiling SDS method were released by mutanolysin digestion and resolved on two-dimensional (2D) gel. One hundred and thirtytwo proteins were identified by mass spectrometry, among which the extracellular solute-binding protein (CbpB/cthe 1020), enolase, glyceraldehyde-3-phosphate dehydrogenase and translation elongation factor EF-Tu were detected as highly abundant proteins. Besides the known surface localized proteins, including FtsZ, MinD, GroEL, DnaK, many enzymes involved in bioenergetics, such as alcohol dehydrogenases and hydrogenases were also detected. By glycan stain and MS analysis of glycopeptides, we identified CbpB as a glycoprotein, which is the second glycoprotein from C. thermocellum characterized. The fact that CbpB was highly abundant in the cell wall region and glycosylated, reflects its importance in substrate assimilation. Our results indicate cell wall proteins constitute a significant portion of cellular proteins and may play important physiological roles (i.e. bioenergetics) in this bacterium. The insights described are relevant for the development of C. thermocellum as a biofuel producer. © 2012 Elsevier GmbH. All rights reserved.

1. Introduction The bacterial cell wall provides the cell with a layer of physical protection while allowing growth and division. It is mainly composed of peptidoglycan polymers, which form a bag-like structure surrounding the cell. In Gram-positive bacteria, the cell wall is made up of peptidoglycans and covalently linked anionic polymers such as teichoic acids. A subset of cellular proteins are attached to the cell wall through interactions with peptidoglycan or other accessory molecules. These proteins are often surface exposed, and could be involved in host adhesion, invasion and toxicity in some pathogenic bacteria. Recently, surface and cell wall proteomes have received increased attention, mostly because there is an urgent need to understand host–bacteria interactions and to identify bacterial surface exposed antigens. These attempts identified extracellular proteins that are either loosely attached to the cell envelope through noncovalent interactions, or those that are tightly associated with the cell wall. Characterization of these proteins requires the isolation of different cell envelope fractions and then solubilization of proteins from the isolated fractions. To release the cell wall-associated proteins, hydrolysis of cell wall structures

∗ Corresponding author. Tel.: +86 10 64807459; fax: +86 10 64807800. ∗∗ Corresponding author. Tel.: +86 10 64807432; fax: +86 10 64807429. E-mail addresses: [email protected] (Y. Luo), [email protected] (K. Yang). 0944-5013/$ – see front matter © 2012 Elsevier GmbH. All rights reserved. doi:10.1016/j.micres.2012.02.006

with various wall lytic enzymes, such as lysozyme (Kelly et al. 2005; Izquierdo et al. 2009), lysostaphin (Gatlin et al. 2006) or mutanolysin (Nandakumar et al. 2005) is the desired method. The released proteins could be further detected by both gel based and gel-free methods before the final protein identification step using MS technology (Calvo et al. 2005; Tjalsma et al. 2008). The gel based method, however, is still more straight forward in comparative investigation and posttranslational modification studies (Zhou et al. 2010). Clostridium thermocellum is a Gram-positive, anaerobic, thermophilic bacterium with the unusual capability to convert cellulosic biomass into soluble sugars and to ferment sugar substrates into industrially important products such as ethanol and hydrogen. For ethanol production, the thermophilic characteristic of C. thermocellum brings several advantages over its mesophilic counterpart, such as Clostridium cellulovorans. First, high temperature fermentation will reduce ethanol recovery cost, as well as the cost of cooling normally associated with meso-temperature fermentations. Second, high temperature fermentation is less prone to contamination. Third, the biomass-degrading enzymes from this bacterium are more stabile at high temperatures than enzymes from mesophilic organisms, thus enhance the performance of these enzymes. The surface region of C. thermocellum had been investigated mostly to characterize the composition and functional assembly of the cellulosomes. The cytoplasmic membrane of C. thermocellum was shown by electron micrograph to be surrounded by a thin peptidoglycan layer, a regular S-layer and an amorphous

T. Yu et al. / Microbiological Research 167 (2012) 364–371

outer layer formed by the protuberate cellulosome (Lemaire et al. 1998). Cellulosome is a multienzyme complex responsible for the efficient degradation of cellulose substrate into soluble sugars. Among the 81 predicted cellulosomal component proteins of C. thermocellum, 67 have been detected by proteomic approaches (Zverlov et al. 2005; Gold and Martin 2007; Raman et al. 2009). Four S-layer homology (SLH) repeats containing proteins, OlpB, Orf2p, SdbA, and OlpA have been characterized. They were shown to interact with the cell wall and to mediate the noncovalent attachment of the cellulosome or free cellulases to the cell surface (Salamitou et al. 1994; Lemaire et al. 1995, 1998; Leibovitz et al. 1997). The major S-layer protein (SlpA) is involved in the formation of Slayer in C. thermocellum (Lemaire et al. 1998). Membrane proteome of C. thermocellum was also investigated to identify differentially expressed proteins between the wild type and ethanol-adapted strains (Williams et al. 2007). However, the roles of cell wall proteins in the special physiology of C. thermocellum are still poorly understood. Therefore, we focused on the proteins associated with the cell wall, so to expand the proteomic coverage of the cell envelope proteins of C. thermocellum. For this purpose, a SDS-boiling method which dissolves cell membrane and cellular proteins was adopted, so the insoluble peptidoglycan with associated proteins could be isolated. To further investigate the cell wall proteins, glycan staining technique was used and putative glycoprotein signals were observed. Preliminary characterization of CbpB showed it is a glycoprotein. 2. Materials and methods 2.1. Bacterial strain and growth conditions C. thermocellum DSM 1237 (ATCC 27405) was grown anaerobically at 60 ◦ C in DSM 122 broth as described elsewhere (http:// www.dsmz.de/microorganisms/medium/pdf/DSMZ Medium122. pdf) with the following modifications: the concentration of MgCl2 ·6H2 O was reduced from 2.6 to 1.3 g/l, K2 HPO4 ·3H2 O was reduced from 5.5 to 3.6 g/l to minimize precipitation of inorganic salts after mixing and l-glutathione was replaced by cysteine hydrochloride at the final concentration of 0.50 g/l. 5.0 g/l of cellobiose was included as the carbon source. The medium was adjusted to pH 7.2–7.4 with NaOH and maintained under a 100% oxygen-free nitrogen atmosphere. 2.2. Cell wall protein preparation Cell wall fraction was prepared using a previously described method (Walter et al. 1999). Briefly, a 24 h culture (OD600 1.6) was harvested by centrifugation (7000 × g, 15 min, 4 ◦ C), washed 2 times with PBS and re-suspended in PBS. The cell suspension was mixed with equal volume of 8% SDS, boiled for 10 min and incubated in the hot water without boiling for another 5 min. Boiling of the cells in 4% SDS dissolved membrane, released cytoplasmic and membrane proteins into solution, also solubilized the well known surface proteins, such as S-layer proteins. After centrifugation, the precipitate was boiled again in SDS and then stirred overnight at room temperature. After boiling for a third time, the insoluble cell wall fraction was collected by centrifugation (15 000 × g, 30 min, 22 ◦ C) and washed 3 times with TE buffer (25 mM Tris–HCl, pH 7.5, 20 mM EDTA) containing 10% iso-propanol to remove SDS, once with TE containing 1 M NaCl to remove released proteins that attached to the cell wall fraction through ionic interactions and finally three more times with only TE until no proteins could be detected in the supernatant by SDS-PAGE. All washes were performed by centrifugation at 15 000 × g for 30 min, with the first three washes at 22 ◦ C and the latter at 4 ◦ C. To release cell wall associated proteins, the

365

purified peptidoglycan was treated with mutanolysin (Sigma) in TE (containing 1 mM phenylmethanesulfonyl fluoride (PMSF) and 10 ␮g/ml leupeptin) at a final concentration of 20 ␮g/ml for 4 h at 37 ◦ C. Insoluble material from the digestion were removed by centrifugation at 20 000 × g for 15 min at 4 ◦ C. The supernatant was collected as cell wall associated proteins and stored at −20 ◦ C. For 2-DE, proteins were precipitated with 15% trichloroacetic acid (TCA) at 4 ◦ C for 30 min. The sample was centrifuged (15 000 × g, 15 min, 4 ◦ C), and the pellet was washed twice with ice-cold acetone, air dried and solubilized in lysis buffer (7 M urea, 2 M thiourea, 4% CHAPS, 1 mM PMSF, 10 ␮g/ml leupeptin) for 2 h at 4 ◦ C. The supernatant was collected and stored at −80 ◦ C for further use. Protein concentration was determined using the Bradford method (Bradford 1976). 2.3. Two dimensional electrophoresis Just before the isoelectric focusing (IEF) step, a cell wall protein sample (400 ␮g) was adjusted to 350 ␮l using the rehydration solution (8 M urea, 4% CHAPS, 0.002% bromophenol blue, 20 mM DTT and 0.5% carrier ampholytes pH 3–10). After loading the sample to the IPG-strips (pH 3–10 NL, 18 cm, GE Healthcare), the IEF was preformed on an Ettan IPGphor3 system (GE Healthcare) using the following voltage profiles: 30 V for 12 h, 200 V for 1 h, 500 V for 1 h, 1000 V for 1 h, gradient increase to 8000 V for 1 h and a final phase of 8000 V to a total of 80 000 V h. The IPG strips were then equilibrated for 15 min in equilibration buffer (0.375 M Tris–HCl, pH 8.8, 6 M urea, 2% SDS, 20% glycerol) containing 16 mM DTT and another 30 min in equilibration buffer with 2.5% iodoacetamide, before finally placed on the top of a 12% SDS–PAGE gel for the second dimension separation. The gel was run for 15 min at constant current of 10 mA and then 6 h at 20 mA on a PROTEAN IIXL system (Bio-Rad), with the temperature kept at 16 ◦ C using water cooling. The gel was stained overnight using PhastGel BlueR (Amersham Biosciences) or by glycan stain as described below in accordance with the manufacturer’s instructions. 2.4. Protein identification Gel images were digitalized by ImageScanner III (GE Healthcare) at dpi 300 and analyzed using Image Master 6.0 2-D platinum software (GE Healthcare). Three cell wall protein samples prepared from three independent cultures were analyzed by 2-DE. Spots from the 2D images were automatically detected and manually checked to eliminate any artifacts. Spots present in at least two replicate gels were considered for subsequent analysis. A total of 384 spots were selected, excised from the gel and submitted to ingel-digestion according to established protocols (Shevchenko et al. 2006). Briefly, the gel spots were washed three times with deionized water, destained with 0.1 M NH4 HCO3 /50% acetonitrile (ACN) and then vacuum dried. The gels were then rehydrated with modified sequencing grade trypsin (12.5 ng/␮l, Sigma) at 4 ◦ C for 45 min. Then, with the addition of 10 ␮l of 20 mM NH4 HCO3 , the enzymatic digestion was performed for 16 h at 37 ◦ C. The peptide solutions were recovered, dried in a speed vacuum and stored at −20 ◦ C for MS analysis. For MALDI-TOF MS and MS/MS analysis, the dried peptides were dissolved in 2 ␮l of the matrix solution (5 mg/ml ␣-cyano4-hydroxycinnamic acid [CHCA] in 50% ACN/0.1% trifluoroacetic acid [TFA]), and 0.7 ␮l of the mixture was deposited onto the target plate. MS and MS/MS spectra were acquired using the ABI 4700 Proteomics Analyzer MALDI-TOF-TOF mass spectrometer (Applied Biosystems) in the reflector positive mode. The laser was operated at a 200 Hz repetition rate with wavelength of 355 nm and the accelerated voltage was operated at 2 kV. The MS spectrum was calibrated using a peptide standard kit (Applied Biosystems)

366

T. Yu et al. / Microbiological Research 167 (2012) 364–371

containing des-Argl-bradykinin (m/z 904.4675), Angiotensin I (m/z 1296.6851), Glul-fibrinopeptide B (m/z 1570.6774), adrenocorticotropic hormone (ACTH) (1–17, m/z 2093.0861), ACTH (18–39, m/z 2465.1989) and ACTH (7–38, m/z 3657.9294) and MS/MS calibration was performed by the MS/MS fragment peaks of Glulfibrinopeptide B. The MS/MS spectra were acquired by the data dependent acquisition method with the 10 strongest precursors selected from one MS scan. Neither baseline subtraction nor smoothing was applied to record spectra. MS and MS/MS data were analyzed and peak lists were generated using GPS Explorer 3.5 (Applied Biosystems). MS peaks were selected within a mass range between 850 and 3500 Da with the minimum S/N of 20; peak density was filtered with no more than 50 peaks per 200 Da. MS/MS peaks were selected with mass range of 60–20 Da below the precursor ion and with the minimum S/N of 10. The MASCOT 2.0 search engine (Matrix Science, London, U.K.) was employed to search the MS and MS/MS data against the C. thermocellum protein sequence database (13 011 sequences; 4 242 250 residues) downloaded from National Center for Biotechnology Information (NCBI) on December 1, 2008. Searching parameters were set as follows: trypsin digestion with one missed cleavage; a charge state of +1; monoisotopic; carbamidomethylation of cysteine as fixed modification; oxidation of methionine as variable modification; mass tolerance of precursor ion and fragment ion at 0.2 Da and 0.5 Da respectively. For all proteins identified, MASCOT score greater than 54 was accepted as significant (p value < 0.05). 2.5. Bioinformatics analysis Sequences of identified proteins were converted to FASTA files and organized into groups of ten sequences. These groups were then submitted to the SignalP 3.0 Server (http://www.cbs.dtu.dk/services/SignalP/) for signal peptide prediction. Variables were set as Gram-positive bacteria and using both the Neural Networks and Hidden Markov models as methods. For lipoprotein and transmembrane domain prediction, the above-mentioned groups were submitted to the LipoP 1.0 Server (http://www.cbs.dtu.dk/services/LipoP/) and the TMHMM Server 2.0 (http://www.cbs.dtu.dk/services/TMHMM/) separately. Predicted results were manually checked to exclude the interference of signal peptides that were already detected by SignalP. Nonclassical protein secretion was analyzed using SecretomeP (http://www.cbs.dtu.dk/services/SecretomeP/). Proteins with an NN-score exceeding the normal threshold of 0.5 were accepted, excluding those predicted to contain a signal peptide. Protein functions were assigned according to Clusters of Orthologous Groups of Proteins (COG) (http://www.ncbi.nlm.nih.gov/COG). PSORTb version 3.0 (http://www.psort.org/psortb/) was employed to predict protein cellular localization (Yu et al. 2010). 2.6. Glycoprotein detection and identification Proteins in SDS–PAGE and 2-DE gels were stained with Pro-Q Emerald 488 glycoprotein gel stain kit (P21875, Molecular Probes) according to the manufacturer’s instructions. Briefly, gels were fixed, washed and oxidized with periodic acid. After thoroughly washing to remove residual periodic acid, gels were stained in the dark for 2 h for SDS–PAGE gels and 2.5 h for 2 D gels. Images were acquired utilizing Typhoon 9400 instrument (GE Healthcare) with a 488 nm laser excitation and 520 nm bandpass emission filter. After fluorescent signal capture, the gels were stained with Coomassie Blue as mentioned above to detect all proteins. Putative glycoproteins were detected by grayscale adjustment of images according to the CandyCane molecular weight standards (Molecular Probes) co-electrophoresed.

Putative glycoproteins were further analyzed using the in-gel glycopeptides enrichment method described by Larsen et al. with minor modifications (Larsen et al. 2005). Briefly, putative glycoprotein spots were cut from Coomassie Blue stained 2-DE gels and in-gel digested by trypsin as described for protein identification. The resulting peptides solutions were then incubated with 1 ␮g Pronase E (sigma) at 37 ◦ C for 72 h. Glycopeptides in the samples were purified and concentrated consecutively using HyperSep C8 and Hypercarb solid phase (SPE) columns (Thermo Scientific, Part Number 60109-210 and 60109-212 respectively). Both columns were wetted with ACN and conditioned with 0.05% TFA. The digestion mixtures were first applied to the C8 column, washed with 0.05% TFA and eluted with 25% ACN in 0.05% TFA. The elution fraction was vacuum dried, redissolved in 0.05% TFA and pooled with the unbound flow-through and wash solution. In the way, the nonglycopeptides were greatly removed by firmly binding to the C8 column. The pooled mixture was then loaded onto the preconditioned porous column and washed with 0.05% TFA to remove salt. Glycopeptides were eluted with 10% ACN in 0.05% TFA and vacuum dried. For MALDI-TOF MS and MS/MS analysis, a matrix preparation method previously optimized for oligosaccharides detection was adopted. 5-Chloro-2-mercaptobenzothiazole (CMBT) and 2,5-dihydroxybenzoic acid (DHB) were prepared as described (Pfenninger et al. 1999). 0.5 ␮l of CMBT was applied to the target as the first layer and dried; glycopeptide samples resuspended in H2 O was mixed with equal volume of DHB and a total of 0.7 ␮l was spotted on top of the microcrystalline CMBT. All MALDI MS and MS/MS data were acquired with Applied Biosystems 4700 proteomics analyzer mass spectrometer in the positive reflectron ion mode. The laser intensity was optimized to averaging 3700 shots to give the best signal-to-noise ratio and resolution.

3. Results 3.1. Sample preparation and protein identification by MALDI-TOF MS and MS/MS In this study, we adopted a cell wall-associated protein preparation method based on an earlier report on the isolation of AbpS, a covalent anchored cell wall protein from Streptomyces reticuli (Walter et al. 1999). To minimize the contamination by cellulosomal proteins, cells of C. thermocellum DSM1237 (ATCC 27405) were harvested in the late-exponential phase when the cellulosomes start to detach from the cell (Coughlan et al. 1985; Mayer et al. 1987). The large amount of S-layer proteins which constitute as much as 10–15% of total cellular proteins (Boot and Pouwels 1996) were not detected in our results after the boiling SDS treatment. The purified cell wall fraction was resistant to lysozyme. Therefore, another endo-N-acetyl muramidase, mutanolysin, which specifically digests the ␤-1,4-glycosidic bond between the Nacetylmuramic acid (NAM) and N-acetylglucosamine (NAG) in the repeating disaccharide unit of peptidoglycan as lysozyme was tried and was able to efficiently release proteins from the cell wall fraction of C. thermocellum. Mutanolysin is active against peptidoglycan modified by O-acetylation of NAM at C-6 (Vollmer 2008). Whether the same modification exists in the peptidoglycan of C. thermocellum is still not known. The solubilized proteins were further separated by 2-DE, 384 spots were excised from the 2-D gel and a total of 132 proteins were identified by MALDI-TOF PMF and MS/MS. The 2D gel and lists of all identified proteins are provided as Additional files 1 and 2. Various bioinformatics softwares were employed to predict the secretion and the location of the proteins identified; the results

T. Yu et al. / Microbiological Research 167 (2012) 364–371

367

Table 1 Putative extracellular proteins among the identified cell wall proteins. Gene locus Cthe Cthe Cthe Cthe Cthe Cthe Cthe Cthe Cthe Cthe Cthe Cthe Cthe Cthe Cthe Cthe Cthe Cthe Cthe a b c d e

0029 0093 0681 1020 1862 1964 2268 3202 1555 0345 1433 3120 0722 0755 2183 2705 1211 2174 2324

Function descriptions

Predicted localizationa

Hypothetical protein Septum site-determining protein MinD IMP dehydrogenase/GMP reductase Extracellular solute-binding protein ABC transporter related protein FAD-dependent pyridine nucleotide-disulphide oxidoreductase Sodium-transporting two-sector ATPase CRISPR-associated protein, Csh2 family NLPA lipoprotein Malate dehydrogenase (MAD) Short-chain dehydrogenase/reductase SDR Pyruvate flavodoxin/ferredoxin oxidoreductase-like protein tRNA (5-methylaminomethyl-2-thiouridylate)-methyltransferase Aspartate aminotransferase UDP-glucose pyrophosphorylase Transketolase subunit B Tryptophan synthase subunit beta Transcription termination factor Rho Resolvase-like protein

CM CM CM CM CM CM CM CM CM U U U C C C Cd Ce Ce Ce

Signal peptide

IIb

Ic Ic Ic Ic Ic Ic

Protein localization predicted by PSORTb version 3.0 (CM, cytoplasmic membrane; U, unknown or multiple localization; C, cytoplasmic). Putative lipoproteins predicted by LipoP 1.0 server with signal peptide cleavage site for signal peptidase II. II, signal peptidase II cleavage site. Signal peptide containing proteins predicted by SignalP 3.0 server. I, signal peptidase I cleavage site. Proteins with one trans-membrane domain predicted by TMHMM Server 2.0. Non-classical secreted protein predicted by SecretomeP.

were summarized in Table 1. 7 proteins (cthe 0345, cthe 0722, cthe 0755, cthe 1020, cthe 1433, cthe 1555 and cthe 2183) were predicted by SignalP to contain signal peptide, indicating their secretion through the classical Sec pathway. Among these seven proteins, the extracellular solute-binding protein (CbpB/cthe 1020) was also predicted by LipoP to be a lipoprotein and it is the only protein predicted to contain signal peptidase II cleavage site. Transketolase subunit B (cthe 2705) was identified by TMHMM to have one transmembrane domain (TMH). For the vast majority of proteins identified, no classical secretion signals or surface localization signatures could be detected; only three proteins, tryptophan synthase subunit beta (cthe 1211), transcription termination factor Rho (cthe-2174) and resolvase-like protein (cthe 2324) were identified as non-classical secreted proteins by SecretomeP. PSORTb 3.0 predicted 8 membrane located proteins (cthe 0029, cthe 0093, cthe 1020, cthe 1555, cthe 1862, cthe 1964, cthe 2268 and cthe 3202) and two proteins of multiple locations (cthe 0345 and cthe 1433). Overall, only 19 proteins could be assigned an extracellular location, while the majority are cytoplasmic proteins that are not expected to appear in the cell wall location.

3.3. Detection and identification of a new glycoprotein From the 2-DE image of C. thermocellum cell wall proteins, it is obvious that some proteins migrate to multi-spots, which is a common feature of post-translationally modified proteins. When these cell wall proteins were separated on SDS-PAGE and stained with a glycan stain, a 50 kDa band showed strong fluorescent signal (Fig. 1a), comparable to the signals from glycoproteins of the CandyCane molecular markers. To unambiguously identify the protein, the protein sample was separated on 2-DE gel and the intensive positive spots were identified by MS to be CbpB. The authenticity of the fluorescent signals is reflected by the low fluorescence signals from other highly expressed proteins, which showed similar intensity as CbpB in Coomassie Blue stain (Fig. 1c and d). To further verify the glycosylation of the protein, a gel-based glycopeptide enrichment method was employed. Initially, trypsin was used to generate peptides; however, the trypsin-generated peptides showed a spectrum dominated by non-glycopeptides,

3.2. Functional classification of identified cell wall proteins Based on COG classification, the 132 identified proteins were grouped into 15 functional categories (Table 2). The highest portion (20.5%) was proteins involved in energy production and conversion. Eight proteins are designated to be cell division (FtsZ/cthe 0445 and MinD/cthe 0093), cell wall biogenesis (cthe 0424, cthe 0978, cthe 2183 and cthe 2629) or cell motility (cthe 2242 and cthe 2819) related. Among proteins with metabolic functions, a whole collection of enzymes from the glycolytic pathway except the hexokinase, phosphoglycerate mutase and pyruvate kinase were present. Notably, four ribosomal proteins (cthe 0714, cthe 1006, cthe 2721 and cthe 2903), four translation elongation factors, EF-Ts (cthe 1005), EF-G (cthe 1794 and cthe 2729) and EF-Tu (cthe 2730), and three chaperones, GroEL (cthe 2892), DnaK (cthe 1322) and trigger factor (TF, cthe 2739) were detected.

Fig. 1. Gel images of C. thermocellum cell wall glycoproteins. Proteins were separated by both SDS-PAGE (a and b) and 2-DE (c and d) and stained with Coomassie (b and c) or Pro-Q Emerald 488 glycoprotein stain (a and d). The ∼50 kDa band which reacted with the glycan stain was marked by an arrow. Positive spots from 2-DE gel were excised and analyzed by MALDI-TOF MS and MS/MS, and found to be the extracellular solute binding protein (Cthe 1020). M, the CandyCane glycoprotein markers, which contains four glycoproteins: avidin (18 kDa), ␣1 -acid glycoprotein (42 kDa), glucose oxidase (82 kDa), and ␣2 -macroglobulin (180 kDa) and the non-glycosylated proteins: L carbonic anhydrase (29 kDa), bovine serum albumin (66 kDa) and phosphorylase b (97 kDa). S, cell wall proteins of C. thermocellum.

368

T. Yu et al. / Microbiological Research 167 (2012) 364–371

Table 2 Functional classification of the 132 identified C. thermocellum ATCC 27405 cell wall proteins.a Code

Function descriptions

J K L D M N O C G E F H I R S

Translation, ribosomal structure and biogenesis Transcription Replication, recombination and repair Cell cycle control, cell division, chromosome partitioning Cell wall/membrane/envelope biogenesis Cell motility Posttranslational modification, protein turnover, chaperones Energy production and conversion Carbohydrate transport and metabolism Amino acid transport and metabolism Nucleotide transport and metabolism Coenzyme transport and metabolism Lipid transport and metabolism General function prediction only Function unknown

–

Not in COGs Total

a

Number

%

13 3 1 2 4 2 7 27 12 20 8 7 2 3 1

9.8 2.3 0.8 1.5 3.0 1.5 5.3 20.5 9.1 15.2 6.1 5.3 1.5 2.3 0.8

20 132

15.2

Assignments and their functional descriptions are based on NCBI COG classifications (http://www.ncbi.nlm.nih.gov/COG).

4. Discussion

Fig. 2. MALDI MS analysis of glycopeptides from extracellular solute-binding protein (Cthe 1020). (a) MS spectrum of the sodium (m/z 1368.6) and potassium (m/z 1384.6) adducts of the glycopeptide produced by non-specific digestion of Cthe 1020. (b) MS/MS spectrum of the precursor ion at m/z 1368.6. The peak at m/z 1206.5 corresponds to the loss of a terminal hexose residue; the peaks at m/z 1151.5, 827.4 and 665.3 correspond to successive losses of hexose residues.

which significantly suppressed the signals of glycopeptides. Therefore, a combination of trypsin and pronase digestion was used to minimize the size of glycopeptides and the interference of nonglycopeptides. The glycopeptides were enriched and purified as described in the method. MS analysis of glycopeptides from three independent experiments all gave dominant peaks at m/z 1368.6 and 1384.6 (Fig. 2a), both of which were subjected to MS/MS analysis. In fragmentation spectrum of m/z 1368.6, signals differing by the mass of hexose (Hex, 162 Da) residues were detected (Fig. 2b). Peak of m/z 1206.5 could have lost the terminal Hex from the precursor ion. In addition, the ion series of 1151.5, 827.4 and 665.3, with decreases of 2 Hex residues or 1 Hex residue respectively, clearly indicate that m/z 1368.6 is a glycopeptide containing at least three Hex residues. For precursor ion m/z 1384.6, no distinguishable fragment signals were detected possibly due to weak signal levels. As this ion pair differed by 16 Da, it is possible that the m/z 1384.6 represents the potassium form of ion m/z 1368.6, a common observation in ionization of glycans or glycopeptides without basic amino acid. These preliminary results verify CbpB as a glycoprotein. We are still working on the protein glycosylation site and the structure of the glycan.

Previously, the ‘boiling in 2% SDS’ method was used to fractionate cell surface proteins of C. thermocellum, which released the known cell surface proteins, such as the cellulosomal scaffolding protein CipA and the SLH containing proteins OlpA and OlpB, in the soluble fraction (Lemaire et al. 1995). After such treatment, these proteins were no longer detectable in the remaining insoluble peptidoglycan fraction even after further SDS treatment or mechanical disruption of cells. Our results also showed the complete elimination of these known surface proteins. However, they also reported that certain cells were resistant to 2% SDS treatment, and following sonification of the insoluble fraction could disrupt these SDS-resistant cells and release cytoplasmic proteins similar to those released by the first SDS treatment (Lemaire et al. 1995). But, in contrast to their method, we used the mild mutanolysin treatment to release proteins from the insoluble peptidoglycan fraction, which is not expected to disrupt the SDS-resistant cells. And the released proteins showed a profile that is different from the SDS released proteins, thus supporting the hypothesis that the mutanolysin released proteins were associated with the cell wall. Although there is a small possibility that certain proteins were retained by the cell wall polymers due to physical hindrance, we believe that the selective retention of these 132 proteins is most likely due to their special or strong interaction with the cell wall polymers. Cell wall-associated proteins in Gram-positive bacteria can be categorized into three major groups based on linkage mechanisms: (i) covalently cell wall-bound proteins that are attached to the peptidoglycan by sortase mediated mechanisms; (ii) noncovalently cell wall-bound proteins with specific cell wall-binding domains (CWBDs); and (iii) the so-called moonlighting proteins, referring to cytoplasmic proteins that are exported and localized to the cell wall region by unknown mechanisms (Dreisbach et al. 2011). Interestingly, comparative genomic analysis revealed that the surface-associated proteins of C. thermocellum ATCC 27405 distinguished greatly from those of other Clostridium species (Bruggemann and Gottschalk 2008). For example, the well-known sortase genes are absent in this bacterium; accordingly, no LPxTG motif containing proteins (substrates of sortase) could be identified in C. thermocellum. Besides, proteins with CWBD2 (PF04122) or choline-binding domain (PF01473), which are prevalent among other clostridia, are also absent in C. thermocellum. On the contrary, C. thermocellum contains 26 S-layer proteins with predicted

T. Yu et al. / Microbiological Research 167 (2012) 364–371

S-layer homology domain (PF00395), which are rarely found in other clostridial. We used PSORTb 3.0 to predict putative cell wall proteins of C. thermocellum, and obtained 47 putative cell wall proteins. Among these 47 proteins, 10 are S-layer like domain containing proteins, 7 are cellulosomal anchoring proteins and 11 are cell wall hydrolases. Most of these proteins are known to locate and function at the cell surface. However, these proteins were not detected in this study, probably because their interactions with cell wall components are not strong enough to withstand the SDS treatment used in this study to prepare the cell wall samples. Besides, the training algorithm of PSORTb 3.0 is based on conventional characteristics of secreted or cell wall anchoring proteins, it is reasonable the program could not predict the proteins detected in this study, which lack classical signal peptides or cell wall anchoring domain. Overall, the difference in surface-associated proteins between C. thermocellum and other clostridials may reflect the adaptation of C. thermocellum to its special ecological niche and also explain the lack of predicted cell wall-associated proteins using universal bioinformatics tools in our work. One interesting trend in cell wall proteome is the detection of a large number of typical cytoplasmic proteins (moonlighting proteins) by various sample preparation methods in Gram-positive and -negative bacteria and in eukaryotes (Hernaez et al. 2010; Insenser et al. 2010; Tunio et al. 2010; Wolfe et al. 2010). Glycolytic enzymes, chaperones and translation factors, in particular, were frequently found. Until now, how these proteins are exported and localized to the cell wall region remains a mystery as they lack both known secretion signals and CWBDs. There have been suggestions that these proteins are released by cell lysis (Gatlin et al. 2006) or a mechanism involving prophages (Young and Blasi 1995). But recent research in Staphylococcus aureus demonstrated that the major autolysin (Atl) is responsible for the secretion of 22 moonlighting proteins. Comparative proteome research with intracellular protein fractions showed that the excretion was discriminative as the most abundant cytoplasmic proteins were not found in the secretome (Pasztor et al. 2010). Selective excretion of these proteins may involve a post-translational modification dependent mechanism, as demonstrated with enolase, only a small fraction auto-modified by its substrate, 2-phosphoglycerate, was exported from the cytoplasm (Boel et al. 2004). Although the prevalence of this mechanism is not known, the detection of certain cytoplasmic proteins in the cell wall region is being supported by more and more evidences (Tunio et al. 2010). C. thermocellum can grow efficiently only on ␤-1,3 and ␤1,4 glycans and prefers long cellodextrins. Strobel et al. (1995) have suggested that cellodextrin import in C. thermocellum was mediated by ABC transporters, which are multimeric complexes consisting of two hydrophobic integral membrane domains and two hydrophilic cytoplasmically located ATPases. Most prokaryote ABC importers utilize an extracellular solute-binding protein (SBP) to bind the substrate with high affinity and present it to the transmembrane channel. In Gram-negative bacteria, such proteins localize in the periplasm while in Gram-positive bacteria, SBPs are usually lipoproteins either tethered to the cell surface via N-terminal cysteines attached to the cytoplasmic membrane or fused to the transporter component (Davidson et al. 2008). In C. thermocellum, five sugar ABC transporter systems have been characterized by sequence homology of putative SBP to known sugar-binding proteins (Nataf et al. 2009), among which CbpB was detected in this work. Lipoproteins are generally classified as membrane-associated proteins, but they were also found in the growth medium of Bacillus subtilis due to secondary proteolytic cleavage (Antelmann et al. 2001) or cell wall localized due to interaction with other exported proteins (Dreisbach et al. 2011). SBPs have been grouped into eight clusters based on their sequence

369

similarities, and proteins of each group were found to have similar solute-binding specificities (Tam and Saier 1993). CbpB falls into the first cluster which has specificity for malto-oligosaccharides, multiple oligosaccharides, ions and ␣-glycerol phosphate. It is worth noting that this protein is one of the most abundant observed in the 2D gels. Besides, among the five putative sugar ABC transporter system predicted, the membrane-spanning domain protein MsdB2 (cthe 1018) was the only one identified from purified membrane fractions (Williams et al. 2007). Although the expression of the other membrane domain protein MsdB1 (cthe 1019) has not been detected yet, the cthe 1018-1020 system might serve as the major transporters in cellodextrin assimilation. Notably, this gene cluster lacks its own ATPases. In some Gram-positive organisms, the ATP-binding subunit of an ABC system is not part of a specific transporter complex, instead it is shared by multiple transporters (Davidson et al. 2008). In this work, the highly abundant CbpB was found to be a glycoprotein. Detection of glycoprotein in Clostridium is not surprising because bacterial protein glycosylation had first been reported on the S-layer protein of two hyperthermophilic Clostridium species (Sleytr and Thorne 1976). In C. thermocellum, the SL component of cellulosome (the 250 kDa major component of cellulosome which is now called the scaffolding protein) was reported to be a glycoprotein containing two glycan chains of tetra- and di-saccharides, with the major tetrasaccharide O-linked to threonine residues of SL via galactopyranose (Gerwig et al. 1989, 1993). In addition, 13 cellulosomal proteins showed positive signals in glycoprotein staining, indicating they are likely glycosylated (Kohring et al. 1990). Glycosylation of solute-binding proteins has been repeatedly reported in prokaryotes. In Campylobacter jejuni, the foodborne pathogens with a well-characterized N-glycosylation pathway, both the HisJ protein (cj0734c, a component of the histidine uptake system) and the CjaA protein (cj0982c, a solute-binding protein of cysteine transporter) were shown to be glycosylated (Nita-Lazar et al. 2005; Wyszynska et al. 2008). In addition, two solute-binding proteins (Ng0372 and Ng1494) from Neisseria gonorrhoeae were reported to be glycoproteins (Vik et al. 2009). In archaea, for example, in Pyrococcus furiosus, Thermococcus litoralis and Sulfolobus solfataricus, whose sugar uptake systems have been extensively studied, glycosylation of solute-binding proteins seems to be a common feature (Horlacher et al. 1998; Elferink et al. 2001; Koning et al. 2002). Although glycosylation may not be essential for solute binding, as proteins heterologously expressed in E. coli remained active, it maybe important to protein stability and protein–envelope interaction (Albers et al. 2004). However, the mechanisms and the functional significance of the modifications remain unclear and needs to be further investigated. In this study, 20.5% of identified proteins were energy production related, including enzymes involved in pyruvate catabolism, hydrogen, ethanol and acetate synthesis. Interestingly, the bifunctional acetaldehyde/alcohol dehydrogenase (adhE, cthe 0423) was identified in this work. In C. thermocellum, multiple genes encoding alcohol dehydrogenases were present, but AdhE was proposed to be the key enzyme for ethanol production (Rydzak et al. 2009). Importantly, Brown et al. (2011) showed that mutation of this enzyme led to improved ethanol tolerance, mainly due to shift of cofactor specificity from NADH to NADPH, thus affecting electron flow. Detection of such enzymes at the cell surface region is well documented. For example, Entamoeba histolytica encodes a 97 kDa multifunctional alcohol/acetaldehyde dehydrogenase that was reported to bind fibronectin, collagen type II and laminin (Yang et al. 1994). And a 37 kDa ADH from Candida albicans was found on the cell surface and in the culture supernatant, acting as the cell surface integrin (receptors for extracellular matrix proteins) (Klotz et al. 2001). Besides, the pyruvate dehydrogenase E1 ␤ subunit of Mycoplasma

370

T. Yu et al. / Microbiological Research 167 (2012) 364–371

pneumoniae was also found at a surface location and was involved in the binding to fibronectin (Dallo et al. 2002). 5. Conclusions Investigation of cell wall proteome in C. thermocellum revealed many unexpected but interesting proteins at this location. Enzymes involved in energy metabolism and bioenergetics, constitute a significant portion of proteins detected, reflecting the special energy production capabilities of this bacterium. Another interesting aspect of the cell wall proteome is the presence of large numbers of “cytoplasmic” proteins, which had been frequently identified in surface or cell envelope proteomes, many proved to serve moonlighting functions in cell invasion or adhesion. Our study confirms their localization in the cell wall region and suggests their moonlighting functions maybe a general phenomenon. An integrated view of the sub-proteome and their organization in the cell wall region will facilitate further study on the biomass hydrolysis and ethanol production functions in this bacterium. Acknowledgments This work was supported by Ministry of Science and Technology of China, Grant 2007CB707800. We wish to thank Qin Sun for her assistance in running 2D gel electrophoresis and Qian Wang for performing the MS analysis. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.micres.2012.02.006. References Albers SV, Koning SM, Konings WN, Driessen AJ. Insights into ABC transport in archaea. J Bioenerg Biomembr 2004;36(1):5–15. Antelmann H, Tjalsma H, Voigt B, Ohlmeier S, Bron S, van Dijl JM, et al. A proteomic view on genome-based signal peptide predictions. Genome Res 2001;11(9):1484–502. Boel G, Pichereau V, Mijakovic I, Maze A, Poncet S, Gillet S, et al. Is 2-phosphoglycerate-dependent automodification of bacterial enolases implicated in their export? J Mol Biol 2004;337(2):485–96. Boot HJ, Pouwels PH. Expression, secretion and antigenic variation of bacterial S-layer proteins. Mol Microbiol 1996;21(6):1117–23. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding. Anal Biochem 1976;72:248–54. Brown SD, Guss AM, Karpinets TV, Parks JM, Smolin N, Yang S, et al. Mutant alcohol dehydrogenase leads to improved ethanol tolerance in Clostridium thermocellum. Proc Natl Acad Sci USA 2011;108(33):13752–7. Bruggemann H, Gottschalk G. Comparative genomics of clostridia: link between the ecological niche and cell surface properties. Ann NY Acad Sci 2008;1125:73–81. Calvo E, Pucciarelli MG, Bierne H, Cossart P, Albar JP, Garcia-Del Portillo F. Analysis of the Listeria cell wall proteome by two-dimensional nanoliquid chromatography coupled to mass spectrometry. Proteomics 2005;5(2):433–43. Coughlan MP, Hon-Nami K, Hon-Nami H, Ljungdahl LG, Paulin JJ, Rigsby WE. The cellulolytic enzyme complex of Clostridium thermocellum is very large. Biochem Biophys Res Commun 1985;130(2):904–9. Dallo SF, Kannan TR, Blaylock MW, Baseman JB. Elongation factor Tu and E1 beta subunit of pyruvate dehydrogenase complex act as fibronectin binding proteins in Mycoplasma pneumoniae. Mol Microbiol 2002;46(4):1041–51. Davidson AL, Dassa E, Orelle C, Chen J. Structure, function, and evolution of bacterial ATP-binding cassette systems. Microbiol Mol Biol Rev 2008;72(2):317–64. Dreisbach A, van Dijl JM, Buist G. The cell surface proteome of Staphylococcus aureus. Proteomics 2011;11(15):3154–68. Elferink MG, Albers SV, Konings WN, Driessen AJ. Sugar transport in Sulfolobus solfataricus is mediated by two families of binding protein-dependent ABC transporters. Mol Microbiol 2001;39(6):1494–503. Gatlin CL, Pieper R, Huang ST, Mongodin E, Gebregeorgis E, Parmar PP, et al. Proteomic profiling of cell envelope-associated proteins from Staphylococcus aureus. Proteomics 2006;6(5):1530–49. Gerwig GJ, Kamerling JP, Vliegenthart JF, Morag E, Lamed R, Bayer EA. The nature of the carbohydrate-peptide linkage region in glycoproteins from the cellulosomes of Clostridium thermocellum and Bacteroides cellulosolvens. J Biol Chem 1993;268(36):26956–60.

Gerwig GJ, de Waard P, Kamerling JP, Vliegenthart JF, Morgenstern E, Lamed R, et al. Novel O-linked carbohydrate chains in the cellulase complex (cellulosome) of Clostridium thermocellum. 3-O-Methyl-N-acetylglucosamine as a constituent of a glycoprotein. J Biol Chem 1989;264(2):1027–35. Gold ND, Martin VJ. Global view of the Clostridium thermocellum cellulosome revealed by quantitative proteomic analysis. J Bacteriol 2007;189(19): 6787–95. Hernaez ML, Ximenez-Embun P, Martinez-Gomariz M, Gutierrez-Blazquez MD, Nombela C, Gil C. Identification of Candida albicans exposed surface proteins in vivo by a rapid proteomic approach. J Proteomics 2010;73(7):1404–9. Horlacher R, Xavier KB, Santos H, DiRuggiero J, Kossmann M, Boos W. Archaeal binding protein-dependent ABC transporter: molecular and biochemical analysis of the trehalose/maltose transport system of the hyperthermophilic archaeon Thermococcus litoralis. J Bacteriol 1998;180(3):680–9. Insenser MR, Hernaez ML, Nombela C, Molina M, Molero G, Gil C. Gel and gel-free proteomics to identify Saccharomyces cerevisiae cell surface proteins. J Proteomics 2010;73(6):1183–95. Izquierdo E, Horvatovich P, Marchioni E, Aoude-Werner D, Sanz Y, Ennahar S. 2-DE and MS analysis of key proteins in the adhesion of Lactobacillus plantarum, a first step toward early selection of probiotics based on bacterial biomarkers. Electrophoresis 2009;30(6):949–56. Kelly P, Maguire PB, Bennett M, Fitzgerald DJ, Edwards RJ, Thiede B, et al. Correlation of probiotic Lactobacillus salivarius growth phase with its cell wall-associated proteome. FEMS Microbiol Lett 2005;252(1):153–9. Klotz SA, Pendrak ML, Hein RC. Antibodies to alpha5beta1 and alpha(v)beta3 integrins react with Candida albicans alcohol dehydrogenase. Microbiology 2001;147(Pt 11):3159–64. Kohring S, Wiegel J, Mayer F. Subunit composition and glycosidic activities of the cellulase complex from Clostridium thermocellum JW20. Appl Environ Microbiol 1990;56(12):3798–804. Koning SM, Konings WN, Driessen AJ. Biochemical evidence for the presence of two alpha-glucoside ABC-transport systems in the hyperthermophilic archaeon Pyrococcus furiosus. Archaea 2002;1(1):19–25. Larsen MR, Hojrup P, Roepstorff P. Characterization of gel-separated glycoproteins using two-step proteolytic digestion combined with sequential microcolumns and mass spectrometry. Mol Cell Proteomics 2005;4(2):107–19. Leibovitz E, Ohayon H, Gounon P, Beguin P. Characterization and subcellular localization of the Clostridium thermocellum scaffoldin dockerin binding protein SdbA. J Bacteriol 1997;179(8):2519–23. Lemaire M, Miras I, Gounon P, Beguin P. Identification of a region responsible for binding to the cell wall within the S-layer protein of Clostridium thermocellum. Microbiology 1998;144:211–7. Lemaire M, Ohayon H, Gounon P, Fujino T, Beguin P. OlpB, a new outer layer protein of Clostridium thermocellum, and binding of its S-layer-like domains to components of the cell envelope. J Bacteriol 1995;177(9):2451–9. Mayer F, Coughlan MP, Mori Y, Ljungdahl LG. Macromolecular organization of the cellulolytic enzyme complex of Clostridium thermocellum as revealed by electron microscopy. Appl Environ Microbiol 1987;53(12):2785–92. Nandakumar R, Nandakumar MP, Marten MR, Ross JM. Proteome analysis of membrane and cell wall associated proteins from Staphylococcus aureus. J Proteome Res 2005;4(2):250–7. Nataf Y, Yaron S, Stahl F, Lamed R, Bayer EA, Scheper TH, et al. Cellodextrin and laminaribiose ABC transporters in Clostridium thermocellum. J Bacteriol 2009;191(1):203–9. Nita-Lazar M, Wacker M, Schegg B, Amber S, Aebi M. The N-X-S/T consensus sequence is required but not sufficient for bacterial N-linked protein glycosylation. Glycobiology 2005;15(4):361–7. Pasztor L, Ziebandt AK, Nega M, Schlag M, Haase S, Franz-Wachtel M, et al. Staphylococcal major autolysin (Atl) is involved in excretion of cytoplasmic proteins. J Biol Chem 2010;285(47):36794–803. Pfenninger A, Karas M, Finke B, Stahl B, Sawatzki G. Matrix optimization for matrix-assisted laser desorption/ionization mass spectrometry of oligosaccharides from human milk. J Mass Spectrom 1999;34(2):98–104. Raman B, Pan C, Hurst GB, Rodriguez Jr M, McKeown CK, Lankford PK, et al. Impact of pretreated switchgrass and biomass carbohydrates on Clostridium thermocellum ATCC 27405 cellulosome composition: a quantitative proteomic analysis. Plos One 2009;4(4):5271. Rydzak T, Levin DB, Cicek N, Sparling R. Growth phase-dependant enzyme profile of pyruvate catabolism and end-product formation in Clostridium thermocellum ATCC 27405. J Biotechnol 2009;140(3–4):169–75. Salamitou S, Lemaire M, Fujino T, Ohayon H, Gounon P, Beguin P, et al. Subcellular localization of Clostridium thermocellum ORF3p, a protein carrying a receptor for the docking sequence borne by the catalytic components of the cellulosome. J Bacteriol 1994;176(10):2828–34. Shevchenko A, Tomas H, Havlis J, Olsen JV, Mann M. In-gel digestion for mass spectrometric characterization of proteins and proteomes. Nat Protoc 2006;1(6):2856–60. Sleytr UB, Thorne KJ. Chemical characterization of the regularly arranged surface layers of Clostridium thermosaccharolyticum and Clostridium thermohydrosulfuricum. J Bacteriol 1976;126(1):377–83. Strobel HJ, Caldwell FC, Dawson KA. Carbohydrate transport by the anaerobic thermophile Clostridium thermocellum LQRI. Appl Environ Microbiol 1995;61(11):4012–5. Tam R, Saier Jr MH. Structural, functional, and evolutionary relationships among extracellular solute-binding receptors of bacteria. Microbiol Rev 1993;57(2):320–46.

T. Yu et al. / Microbiological Research 167 (2012) 364–371 Tjalsma H, Lambooy L, Hermans PW, Swinkels DW. Shedding & shaving: disclosure of proteomic expressions on a bacterial face. Proteomics 2008;8(7):1415–28. Tunio SA, Oldfield NJ, Berry A, Ala’Aldeen DA, Wooldridge KG, Turner DP. The moonlighting protein fructose-1, 6-bisphosphate aldolase of Neisseria meningitidis: surface localization and role in host cell adhesion. Mol Microbiol 2010;76(3):605–15. Vik A, Aas FE, Anonsen JH, Bilsborough S, Schneider A, Egge-Jacobsen W, et al. Broad spectrum O-linked protein glycosylation in the human pathogen Neisseria gonorrhoeae. Proc Natl Acad Sci USA 2009;106(11):4447–52. Vollmer W. Structural variation in the glycan strands of bacterial peptidoglycan. FEMS Microbiol Rev 2008;32(2):287–306. Walter S, Rohde M, Machner M, Schrempf H. Electron microscopy studies of cell-wall-anchored cellulose (Avicel)-binding protein (AbpS) from Streptomyces reticuli. Appl Environ Microbiol 1999;65(3):886–92. Williams TI, Combs JC, Lynn BC, Strobel HJ. Proteomic profile changes in membranes of ethanol-tolerant Clostridium thermocellum. Appl Microbiol Biotechnol 2007;74(2):422–32. Wolfe LM, Mahaffey SB, Kruh NA, Dobos KM. Proteomic definition of the cell wall of Mycobacterium tuberculosis. J Proteome Res 2010;9(11):5816–26.

371

Wyszynska A, Zycka J, Godlewska R, Jagusztyn-Krynicka EK. The Campylobacter jejuni/coli cjaA (cj0982c) gene encodes an N-glycosylated lipoprotein localized in the inner membrane. Curr Microbiol 2008;57(3):181–8. Yang W, Li E, Kairong T, Stanley Jr SL. Entamoeba histolytica has an alcohol dehydrogenase homologous to the multifunctional adhE gene product of Escherichia coli. Mol Biochem Parasitol 1994;64(2):253–60. Young R, Blasi U. Holins: form and function in bacteriophage lysis. FEMS Microbiol Rev 1995;17(1–2):191–205. Yu NY, Wagner JR, Laird MR, Melli G, Rey S, Lo R, et al. PSORTb 3.0: improved protein subcellular localization prediction with refined localization subcategories and predictive capabilities for all prokaryotes. Bioinformatics 2010;26(13):1608–15. Zhou W, Capello M, Fredolini C, Piemonti L, Liotta LA, Novelli F, et al. Mass spectrometry analysis of the post-translational modifications of alpha-enolase from pancreatic ductal adenocarcinoma cells. J Proteome Res 2010;9(6): 2929–36. Zverlov VV, Kellermann J, Schwarz WH. Functional subgenomics of Clostridium thermocellum cellulosomal genes: identification of the major catalytic components in the extracellular complex and detection of three new enzymes. Proteomics 2005;5(14):3646–53.