OUP user menu

Correlation of probiotic Lactobacillus salivarius growth phase with its cell wall-associated proteome

Peter Kelly, Patricia B. Maguire, Mary Bennett, Desmond J. Fitzgerald, Richard J. Edwards, Bernd Thiede, Achim Treumann, J. Kevin Collins, Gerald C. O'sullivan, Fergus Shanahan, Colum Dunne
DOI: http://dx.doi.org/10.1016/j.femsle.2005.08.051 153-159 First published online: 1 November 2005


Lactobacillus salivarius subsp. salivarius UCC118 is a probiotic bacterium that was originally isolated from human intestinal tissues and was subsequently shown in a pilot study to alleviate symptoms associated with mild-moderate Crohn's disease. Strain UCC118 can adhere to animal and human intestinal tissue, and to both healthy and inflamed ulcerative colitis mucosa, irrespective of location in the gut. In this study, an enzymatic technique has been combined with proteomic analysis to correlate bacterial growth phase with the presence of factors present in the cell wall of the bacterium. Using PAGE electrophoresis, it was determined that progression from lag to log to stationary growth phases in vitro correlated with increasing prominence of an 84 kD protein associated with in vitro adherence ability. Isolated proteins from the 84 kD band region were further separated by two-dimensional electrophoresis, resolving this band into 20 individual protein spots at differing isoelectric points. The protein moieties were excised, trypsin digested and subjected to tandem mass spectrometry. The observed proteins are analogous to those reported to be associated with the Listeria monocytogenes cell-wall proteome, and include DnaK, Ef-Ts and pyruvate kinase. These data suggest that at least some of the beneficial attributes of probiotic lactobacilli, and in particular this strain, may be due to nonpathogenic mimicry of pathogens and potentially be mediated through a form of attenuated virulence.

  • Growth phase
  • Probiotic
  • Lactobacillus
  • Proteomics
  • Mass spectrometry

1 Introduction

In a number of intestinal disease states altered microflora, impaired gut barrier and/or intestinal inflammation offer a rationale for the effective therapeutic use of probiotic micro-organisms [1]. Such “probiotics”, defined as living micro-organisms which upon ingestion in sufficient numbers, exert health affects beyond inherent basic nutrition [2], have been assessed in human disorders such as but not limited to: allergy [3], urogenital infections [ [, [], colorectal cancer [ [, [] and microbial and antibiotic-associated diarrhoea [ [– [0].

Most commonly, the effects of probiotics require the consumption of considerable numbers of cells which then transiently persist within the physiologically and chemically defined intestinal environments. Factors which influence probiotic efficacy vary from acidic gastric conditions to a neutral or basic pH in the small intestine, presence of biliary and pancreatic secretions, and active healthy or imbalanced mucosal immune responses [1]. Proposed mechanisms through which the ingested probiotic microbes may subsequently benefit their host include production of antimicrobial factors; competition for nutrients; toxin degradation and disruption of eucaryotic toxin receptors; and immunomodulation [1]. While it can be argued that adhesion to epithelial cells and/or mucus appears to mediate colonization of the gastrointestinal tract by lactobacilli, and may be a prerequisite for competitive exclusion of enteropathogenic bacteria and immunomodulation of the host [ [1– [3], it has been postulated that at least some aspects of the probiotic effect may be attributable to nonpathogenic mimicry of pathogens or attenuated virulence [14].

The majority of probiotic products described in the literature are consumed as yoghurt or yoghurt-like dairy products [ [5– [7] and, subsequently, most of the reported technological advances describe methodologies relevant to probiotic viability in milk [18] or innovative products destined for inoculation into milk [19]. However, it is likely that the next generation of probiotic functional/medicinal foods will marry the inherent beneficial traits of currently available and well-studied micro-organisms with modifications designed to enhance immune perception, exclusion of deleterious competitors, and/or delivery of recombinant therapeutic moieties [ [0, [1]. In this study, we have placed an emphasis on the potential for enhancement of probiotic performance (perhaps through immune stimulation) deliverable through cell-wall engineering [22]. In the first step of this process, proteomic analyses were combined with an enzymatic technique to correlate the presence of a previously documented cell wall-associated protein [ [3– [5] with phase of growth of a probiotic Lactobacillus salivarius subsp. salivarius strain. Candidate proteins were identified by tandem mass spectrometry. The results demonstrate that factors such as DnaK and Ef-Ts that have recently been shown as components of the Listeria monocytogenes cell wall [26] are also constituents of the L. salivarius UCC118 cell-wall proteome. These data may provide a rationale for the choosing of specific targets for the enhancements described above. Indeed, this strain which has been shown in a pilot study to alleviate symptoms associated with mild-moderate Crohn's disease [27], and the proteins described in this report, may prove to be appropriate for physiological and genetic enhancement of therapeutic or prophylactic traits. Alternatively, advances in biotechnology, fermentation processes and microencapsulation may facilitate localized therapeutic use of purified forms of the identified proteins.

2 Materials and methods

2.1 Culturing of Lactobacillus salivarius subsp. salivarius UCC118

L. salivarius subsp. salivarius UCC118 was propagated in MRS broth (de Mann Rogosa & Sharpe; Oxoid, Hampshire, England) incubated anaerobically in GasPak™ jars (BBL) with CO2 generating kits (Anaerocult A™;Merck) at 37 °C for 2, 10 or 18 h. Post-incubation, bacterial cell numbers were corrected to 108 cells/ml irrespective of lag, log or stationary growth phase sample preparations. Prior to any of the following procedures, bacterial cells were washed three times in quarter strength Ringer's solution (Oxoid).

2.2 One-dimensional electrophoresis of cell-wall associated proteins

Following incubation of bacterial cells in 2 ml TEL solution (100 mM Tris–HCl pH 8, 5 mM EDTA and 1.0% lysozyme) for 3 h at 37 °C, supernatant was collected by centrifugation at 3000g for 10 min. The obtained cell extracts were electrophoresed on denaturing polyacrylamide gels (10% SDS) as previously described [28]. Total protein concentrations were corrected to equivalence prior to electrophoresis. Protein spots were visualized by Comassie blue stain. All images were captured using a digital camera (Finepix model 2600, Fuji, Japan).

2.3 Two-dimensional electrophoresis (2-DE) of cell-wall associated proteins

2-DE was performed as previously described by Gorg et al. [29], with the following minor modifications [ [0, [1]. Following preparation, 100 μg of cell wall proteins from the lag, log and stationary phase of UCC118, samples were resuspended in rehydration buffer (7 M Urea; 2 M Thiourea; 2% 3-[(3-Cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS) (Sigma); 0.5% carrier ampholytes; 0.2% dithiothreitol (DTT) and loaded onto 18 cm, pI 3–10 immobiline dry strips (Amersham Pharmacia Biotech, UK) and focused for 60,000 V h. Following equilibration, the strips were sealed on top of 10% polyacrylamide gels (20 cm × 20 cm). Protein spots were visualized using silver staining [30]. The 84 kD protein band was then excised from one-dimensional PAGE gels, the proteins eluted overnight by diffusion and the eluted proteins separated as above by 2-DE using “zoom-in” gels (Amersham Pharmacia Biotech UK) (pH 3–10, 4–7, 4.5–5.5 and 5–6) in the first dimension. Protein spots were visualized by SYPRO® Ruby fluorescent stain (Molecular Probes™) as previously described [30] and were used to excise spots for tandem mass spectrometry.

2.4 Tandem MS analysis

In-gel digestion with trypsin was performed as previously described [ [0, [1]. For liquid chromatography tandem MS analysis (LCMSMS), tryptic digests were injected onto a PePSep (polypeptide separation) column (75 μm ID, 150 mm) (LC Packings, Leeds, UK) using an autosampler (briefly, 3–5 μl injected onto a PepSeq trap column (300 μm ID, 5 mm) at 30 μl/min). Peptides were separated using a CapLC nano-HPLC system (Waters Micromass, Manchester, UK) with a gradient from 93% A (0.08% HCOOH), 7% B (0.08% HCOOH, 90% MeCN) to 55% A, 45% B over 37 min (total cycle time: 50 min). The column was connected to a QTOF Micro mass spectrometer (Waters Micromass, Manchester, UK) with a nanoflow probe and the flow through the column set to 200 nl/min. Spectra were acquired in survey scan mode (scan duration 0.9 s, interscan delay 0.1 s, m/z 350–1500, automatic selection of double and triple charged ions for MSMS acquisition; parameters for MSMS spectra: scan duration 2.5 s, interscan delay o.1 s, m/z 70–1500). Spectra were deconvoluted and deisotoped using the MaxEnt3 algorithm (Waters Micromass, Manchester, UK) and sequences were determined using computer-assisted manual sequencing using the program PepSeq (Waters Micromass, Manchester, UK). The obtained sequences were searched against an in-house genomic database of strain UCC118. 460 genomic “contigs” of UCC118 were translated in all three reading frames and each open reading frame (ORF) of at least 20 amino acids used to perform BLAST [32] searches against the SwissProt-TrEMBL database [33]. ORFs were crudely annotated with text descriptions from the top 100 BLAST hits with an e-value <10−6. Peptides were then searched against these ORFs using an in-house tool, PRESTO (Peptide Regular Expression Search Tool, http://bioinformatics.rcsi.ie/redwards/presto), a search tool for searching peptide sequences or motifs using an algorithm based on Regular Expressions. It also has a MS–MS mode for searching peptides sequenced from tandem mass-spectrometry data. It can handle a variety of regular-expression type peptide sequences and search them against databases allowing mismatches if desired.

3 Results

3.1 One-dimensional electrophoresis of cell-wall associated proteins

The L. salivarius UCC118 protein preparations, which included cell wall-associated proteins and “secretome” of the bacteria, were isolated at the three distinct growth phases, lag phase proteins following 2 h growth, log phase proteins after 10 h and stationary phase proteins after 18 h growth. These proteins were separated by one-dimensional protein gel electrophoresis. The protein concentrations in each of the samples were identical, as measured by Biorad protein assay (30 mg/ml). At the lag phase of growth very few protein bands were expressed in comparison to other phases of growth (Fig. 1). The 84 kD protein band of interest was absent during the lag phase, present in the log phase, but the stationary phase of growth had highest levels of 84 kD protein expression. This corresponds with relative adherence abilities during each of the phases of growth (data not shown).

Figure 1

Illustrates the growth curve of Lactobacillus salivarius subsp. salivarius UCC118 incubated anaerobically in MRS broth. The corresponding stained single dimension SDS–PAGE of Lactobacillus salivarius UCC118 protein preparation at varying phases of growth are shown. The arrow indicates the position of the 84 kD band.

3.2 2-DE of cell-wall associated proteins

The L. salivarius UCC118 cell wall protein preparation from the lag, log and stationary phases were further separated using 2-DE (pI 3–10) and silver stained. Representative 2D gels from each phase are depicted in Fig. 2. Varying levels of protein expression were noted across each of the growth phases and again confirmed that the 84 kD adhesin area was not expressed in lag phase but was present during the exponential and stationary phases (Fig. 2). Indeed, this area had an identical morphology at the log and stationary phases and three discreet subpopulations of protein spots were visible.

Figure 2

Representative silver stained 2-D gels (n= 5) of the Lactobacillus salivarius UCC118 cell wall protein preparation at the three phases of growth. Gel A: the lag phase, gel B: log phase, and gel C: stationary phase. The arrow indicates the position of the 84 kD adhesin band.

The protein band was further separated using overlapping pH ranges in the first dimension (Fig. 3). The first and second subpopulations were each resolved into four protein spots and the third subpopulation into 12 protein spots, indicating that this adhesion band comprises a total of 20 individual protein/peptide spots. All protein spots in the band focused at a pH below 7.

Figure 3

‘Zoom-in' 2D gels stained with SYPRO Ruby™ of the isolated 84 kD band over the following pH ranges: pH 3–10, 4–7, 4.5–5.5 and 5–6. Position A: the first subpopulation of proteins consisting of four protein spots in the IP range 4–5, position B: the second subpopulation consisting of four protein spots in the IP range 4–5, and position C: the third subpopulation consisting of 12 protein spots in the IP range 4–7.

Acrylamide segments correlating to each of the 20 distinct proteins were excised for sequencing. These were pooled. Only three proteins were identified using this approach (Table 1): the 70 kD heat shock protein DnaK (HSP70), the transcription elongation factor EF-Ts and pyruvate kinase. DnaK is a chaperone molecule which aids proper protein folding in bacteria and also assist in protein translocation across membranes [34]. Elongation factors assist the addition of every amino acid to the growing polypeptide chain in the elongation stage of protein synthesis. EF-Ts like EF-Tu is involved in the aminoacyl-tRNA binding phase of the elongation cycle [35]. Pyruvate kinase is a glycolytic enzyme which is necessary to enable bacteria to metabolise and utilise carbohydrates.

View this table:
Table 1

Lactobacillus salivarius UCC118 proteinsd identified by tandem mass spectrometry

m/zChargeMWtaΔ (ppm)bSequencecRemarksProtein
597.2721193.57233ALVAVEGDMEKMethionine oxidised to sulphoneEF-Ts
785.3321569.71037EENFAEEVMSQIKMethionine oxidised to sulphoxideEF-Ts
606.2521211.54645ISMDDSIEGTKMethionine oxidised to sulphoxidePyruvate kinase
  • aCalculated molecular weight of the pseudomolecular ion [M + H]+.

  • bDifference between experimental and theoretical molecular weight of the pseudomolecular ion in parts per million.

  • cAmino acids yielding ions are represented in bold.

  • dAcrylamide segments correlating to each of the 20 distinct proteins were excised for sequencing. Only these sequences were identified.

4 Discussion

Given the previously demonstrated potential of L. salivarius UCC118 in alleviating intestinal disorders [ [3, [5], elucidating the proteinaceous cell wall components mediating its ability (i.e., to adhere to intestinal cells and/or to promote host immune perception) will open a number of avenues of more sophisticated experimentation.

The three proteins isolated from the 84 kD band; pyruvate kinase, DnaK and EF-Ts, are normally recognized as cytosolic proteins. However the identification of such proteins on the cell wall or secretome of this probiotic bacterium is not without precedent. Recently DnaK, EF-Tu and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were isolated form the cell wall of Listeria monocytogenes [26]. The mechanism by which such proteins leave the bacterial cell (other than due to chemical or physical disruption) has yet to be determined but a possible function of these proteins when associated with the cell wall is suggested by findings of in vitro studies which demonstrate that DnaK, EF-Tu, enolase and GAPDH are capable of plasminogen binding [26]. It has been proposed that, in vivo, surface-displayed GAPDH and enolase of Streptococcus pneumoniae recruit plasminogen from the host to the surface of the pathogen and facilitate subsequent activation of the protease by tissue-type plasminogen activator tPA [36].

Plasminogen is a 92 kD glycoprotein which is present in human plasma, in its active form it plays a pivotal role in fibrinolysis but it also has broad substrate specificity against extracellular matrix components like fibronectin, laminin and induces activation of matrix-degrading proteases, such as collagenases which in the case of S. pneumoniae may allow the bacterium to pass through the extracellular matrix of its host [ [7, [8].

Previously, we have shown that the proteins responsible for adhesion of L. salivarius ssp. salivarius to intestinal epithelium are present within the 84 kD band of its cell wall [23]. Using the examples of Listeria monocytogenes and S. pneumoniae, it is not illogical to suggest that the proteins isolated for the 84 kD band, which are identical or closely related to similar cell wall proteins of these two pathogenic bacteria, are responsible for bacterial adherence to gut epithelium by recruiting plasminogen. This hypothesis relies on a theory of attenuated virulence, whereby the beneficial probiotic bacterium may in a nonpathogenic fashion mimic the mode of adhesion of a bacterial pathogen to the host cell in order to mediate its beneficial properties.

Of the main criteria for selecting probiotic strains, their ability to adhere to intestinal epithelia is often described as paramount [ [, [4, [9]. Indeed, adhesion to epithelial cells and/or mucus appears to mediate colonization of the gastrointestinal tract by lactobacilli [40] and may be a prerequisite for competitive exclusion of enteropathogenic bacteria [ [1, [2] and immunomodulation of the host [43].

The fact that three proteins, analogous to those found in the cell wall proteome of Listeria monocytogenes, were isolated form the 84 kD band of L. salivarius by a completely different extraction method may be taken to support these findings [26]. In these in vitro studies it is unlikely that the observed proteins are cytosolic contaminants, as the 84 kD band was absent in the lag phase and not as strongly expressed in the log phase of growth as in the stationary phase when the cell wall proteome was enzymatically stripped from the bacterium. Even in the lag and log phases of growth, cytosolic proteins would be plentiful and should be present if they were contaminants.

The data presented here may aid in resolving discussion regarding the absolute necessity for viable or even whole probiotic organisms rather than the delivery of defined benefits/traits or factors on a more contingent basis. Future work may define the therapeutic and technological potential of the 84 kD protein components with respect to relevant genetic coding elements, possible tissue tropisms, effective manufacture of purified proteins and eventual delivery of stable (maybe microencapsulated) proteins for localized therapeutic use.


The authors thank Maurice O'Donoghue for technical assistance. This study was supported in part by grant aid under the Food Sub-Programme of the Operational Programme for Industrial Development administered by the Irish Department of Agriculture and Food, part-financed by the European Regional Development Fund; the Programme for Research in Third Level Institutions (PRTLI) administered by the Irish Higher Education Authority; the Irish Health Research Board; Science Foundation Ireland; and the European Commission (PROBDEMO: FAIR-CT96-1028; PROGID: QLK1-2000-00563).


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
  34. [34].
  35. [35].
  36. [36].
  37. [37].
  38. [38].
  39. [39].
  40. [40].
  41. [41].
  42. [42].
  43. [43].
View Abstract