Extracellular Proteins from Lactobacillus plantarum BMCM12Prevent Adhesion of Enteropathogens to Mucin

Borja Sanchez • Marıa C. Urdaci

Received: 1 February 2012 / Accepted: 14 March 2012 / Published online: 30 March 2012

� Springer Science+Business Media, LLC 2012

Abstract The aim of this study was to study the inter-

ference of the extracellular proteins produced by Lacto-

bacillus plantarum BMCM12 with the adhesion of some

well-known gut pathogens. The extracellular proteins

secreted by L. plantarum BMCM12 in MRS broth were

precipitated, resolved by SDS-PAGE, and identified by

tandem mass spectrometry. Discordances between the

observed and the theoretical molecular masses of several

proteins suggested the presence of protein glycosylation,

corroborated with specific glycoprotein staining after pro-

tein de-glycosylation using trifluoromethanesulfonic acid.

Experiments of exclusion, competition, or prevention of

the pathogen adhesion to mucin were performed using

BMCM12 extracellular proteins, using Escherichia coli

LMG2092 and Salmonella enterica subsp. enterica

LMG15860. Extracellular proteins from BMCM12 reduced

significantly the adhesion of the pathogens when they were

added prior to adhesion assays. These proteins play thus

important roles in preventing pathogen adhesion to the

mucin layer.

Introduction

Lactobacillus plantarum is one of the species with the

largest genome in comparison to other lactic acid bacteria

(LAB) [1]. This provides a high versatility in both carbon

source utilization and attachment to different substrates, as

well as a great capacity of adaptation to changing envi-

ronmental conditions [2]. Some L. plantarum strains, such

as L. plantarum 299v, have been shown to confer benefits

on human health and are, nowadays, commercialized as

probiotics for human nutrition [3]. The genome of

L. plantarum has been shown to contain several genes

coding for extracellular proteins, which might have rele-

vant roles in the interaction of the bacterium with its sur-

roundings and with the human host. In this context, certain

extracellular proteins might be responsible for some pro-

biotic traits, including host immunomodulation [4]. Thus,

identification and characterization of extracellular proteins

are crucial steps for understanding the physiology and

mechanisms of action of probiotic bacteria.

Glycosylation is a common post-translational modifi-

cation of proteins [5], which serves a wide range of func-

tions, among which protection against proteolysis and

modulation of their biological activity in Eukarya and

Archaea divisions [6, 7]. Experimental evidence dealing

with the potential functions of bacterial glycoproteins is

still limited [8]. The best known bacterial protein glyco-

sylation pathway is the one from Campylobacter jejuni, in

which Asparagines within D/E-Z-N-X-S/T motifs (being Z

and X not Prolines) are glycosylated through the action

of the products encoded in the protein glycosylation locus

[9]. Recently, a general O-glycosylation system has been

described in the genus Bacteroides, a Gram-negative,

human commensal bacteria [10]. The system is basically

characterized by the incorporation of fucose residues into

B. Sanchez (&)

Instituto de Productos Lacteos de Asturias, Consejo Superior de

Investigaciones Cientıficas (IPLA-CSIC), Ctra. Infiesto s/n,

33300 Villaviciosa, Asturias, Spain

e-mail: borja@ipla.csic.es

B. Sanchez � M. C. Urdaci

UMR 5248 CBMN CNRS-Universite Bordeaux 1-ENITAB,

Laboratoire de Microbiologie et Biochimie Appliquee, 1 cours

du General de Gaulle, 33175 Gradignan Cedex, France

123

Curr Microbiol (2012) 64:592–596

DOI 10.1007/s00284-012-0115-6

Bacteroides glycoproteins, which are crucial for these

bacteria in order to efficiently colonize the mammal gut

[10]. The vast majority of the bacterial glycoproteins

described so far appeared to be secreted or associated on

the cellular surface, suggesting potential roles in the

interaction with their surroundings, or with the host in the

case of pathogens [11].

In this study, we have focused on the interaction of the

extracellular proteins secreted by strain L. plantarum

BMCM12, a bacterium isolated from the traditional

African bread ‘‘chikwangue’’. These proteins prevented the

adhesion of two well-known enteropathogens, Escherichia

coli LMG2092 and Salmonella enterica subsp. enterica

LMG15860, to mucin.

Materials and Methods

Culture Conditions

Isolated colonies were obtained on MRS agar (Becton–

Dickinson France SAS, Le Pont-De-Claix, France) and

were used to inoculate 10 ml of MRS broth (Becton–

Dickinson), which were incubated overnight aerobically,

and without shacking, at 37 �C. These cultures were used

to inoculate (1 % v/v) 50 ml of fresh MRS, where strain

BMCM12 was grown until the early stationary phase of

growth (monitored following the OD600 of cultures).

Escherichia coli LMG2092 and Salmonella enterica subsp.

enterica LMG15860 were grown ON from stocks stored

at -80 8C in brain–heart infusion broth (BHI) (Becton–

Dickinson) at 37 8C in an anaerobic cabinet (Bactron

Anaerobic/Environmental Chamber, Sheldon Manufactur-

ing Inc., Cornelius, OR) in an atmosphere of 5 % CO2–5 %

H2–90 % N2. These cultures were used for inoculate fresh

media (1 % v/v), and the pathogens were collected at sta-

tionary phase of growth.

Extracellular Protein Extraction

For the precipitation of secreted proteins, aliquots of 5 ml

of cultures at stationary phase were harvested by centri-

fugation (10 min, 3500 g, 4 8C), the supernatant being

filtered (0.45 lm). Proteins were precipitated following

two different procedures. First, two volumes of cold etha-

nol (Merck KGaA, Darmstadt, Germany) were added to

supernatants, and proteins were left to precipitate at 4 �C

overnight. Second, a trichloroacetic acid (TCA) (Sigma-

Aldrich, Saint-Quentin Fallavier, France)-based precipita-

tion was performed [12]. Ethanol- and TCA-precipitated

proteins were recovered by centrifugation (10 min, 9300 g,

4 8C), and pellets were washed twice with chilled acetone

(Merck). Pellets were allowed to dry at room temperature

and proteins were re-solubilized in an ultrasonic bath

for 10 min (Deltasonic, Meaux, France) in 40 ll of

19 Laemmli buffer [13].

Protein De-glycosylation

Protein de-glycosylation was performed in screw-cap vials

following a trifluoromethanesulfonic acid (TFMS)-based

method (Sigma-Aldrich), a protocol that completely

removes both O- and N-linked glycans from glycoproteins

[14]. Amounts of 1.5 mg of secreted proteins, extracted

and precipitated as described above, were placed in ice, and

150 ll of chilled TFMS:toluene (Merck) (2:1) were added.

Samples were mixed gently until complete protein solubi-

lization, and were subsequently incubated for 30 min at

-20 �C. Four microliter of chilled bromophenol blue in

ethanol (2 mg/ml) was added and samples were placed in

an ethanol bath, previously pre-cooled to -20 �C. Two-

hundred microliter of pyridine solution (Merck) (60 %

anhydrous pyridine in 1:1 methanol:water), also pre-cooled

to -20 �C, were carefully added until complete sample

neutralization. Finally, 400 ll of a solution of ammonium

bicarbonate 0.5 % (w/v) (Sigma-Aldrich) were added, and

proteins precipitated following a methanol/chloroform

protocol [15].

Protein Manipulations

Amounts of 40 lg of protein were resolved by SDS-PAGE

in 12.5 % (w/v) polyacrylamide gels. Cytoplasmic extracts,

obtained by sonication (Vibracell 75021 Ultrasonic Pro-

cessor, Fisher Scientific Bioblock, Illkirch, France) for 3–7

cycles of 3 min (amplitude 12, duty 33 %), were used as

controls. The presence of glycoproteins was shown with a

specific staining able to reveal the presence of both N- and

O-glycosylated proteins (GelCode Glycoprotein Staining

Kit, Thermo Scientific, Rockford, IL), following the man-

ufacturer’s instructions. Proteins were totally stained using

Coomassie staining (Pierce). For protein identification,

selected bands were excised from gels and digested with

trypsin using standard protocols, the resulting peptide

mixture being analysed by tandem mass spectrometry (MS/

MS), as already described .

Adhesion Assays

Adhesion to Type II mucin (Mucin, type II, Sigma-

Aldrich), was tested following the procedure described

before by our research group (16), using a starting inocu-

lum of 108 CFUs, as determined by plate count. Twenty

micrograms of extracellular proteins solubilized in PBS

were added 60 min prior, during or after 60 min of path-

ogen incubation with the mucin monolayer. ON bacterial

B. Sanchez, M. C. Urdaci: Adhesion Inhibition by Lactobacillus Extracellular Proteins 593

123

cultures, in the early stationary phase of growth, were used

in all cases. Assays were performed in triplicate and the

data were expressed as the percentage of adhesion, calcu-

lated with the following formula: (CFU recovered/CFU

added) 9 100.

Statistical Analysis

Throughout the manuscript, data were subjected to one-

way analysis of variance with the SPSS 18.0 software

(SPSS Inc., Chicago, IL).

Results

A representative SDS-PAGE gel showing the extracellular

proteins produced by L. plantarum BMCM12 in MRS is

shown in Fig. 1. All proteins were identified by tandem

mass spectrometry (MS/MS) against the genome of L.

plantarum WCFS1 with the exception of PL25 (Table 1),

which is likely a specific protein of the BMCM12 strain.

The strong protein band observed in the molecular mass

zone of 66 kDa (marked with an asterisk in Fig. 1) pro-

duced poor tryptic profiles.

As can be seen in Table 1, several secreted proteins

(notably PL26, PL27, and PL28) presented aberrant

migration in SDS-PAGE, as deduced by comparing

observed/theoretical molecular masses. Thus, we hypoth-

esized that some proteins might be glycosylated. Our first

approach was to treat the crude extracellular protein

extracts with TFMS, which completely removes N- and

O-linked glycans and that is one of the first choices for

studying protein glycosylation in bacteria. As can be seen

in Fig. 1 (lane 7), glycosylation is not the cause of the

molecular mass shift observed for PL26, PL27, and PL28.

Surprisingly, two bands presented a molecular mass shift in

the high molecular mass zone of the gel after TFMS

de-glycosylation, which were named PL21 and PL22,

respectively. Since several techniques need to be combined

to provide evidence for protein glycosylation, PL21 and

PL22 glycosylation was further confirmed by specific

staining (Fig. 1).

Three different adhesion tests were performed in order

to determine the effect of the extracellular proteins on

the adhesion of E. coli LMG2092 and S. enterica subsp.

enterica LMG15860 to mucin (Fig. 2). In the first one, the

extracellular proteins were added before testing pathogen

adhesion, reflecting the preventive effects of this fraction to

inhibit pathogen adhesion. In the second and third setups

extracellular proteins were added during or after the

pathogens, reflecting the ability of these proteins to com-

pete or remove previously attached pathogens, respec-

tively. Significant reductions in pathogen adhesion were

only observed in the first case, where proteins were pre-

incubated with the mucin monolayers.

Discussion

Secretion of glycoproteins by probiotic bacteria is a subject

of current interest because of their potential implications

in gut physiology. These proteins are among the first

molecules that interact with the host cells, and might be

involved in processes such as surface recognition, host

immunomodulation, and molecular cross-talking. Among

L. plantarum extracellular proteins we have detected two

Fig. 1 Representative polyacrylamide gel showing the extracellular

proteins produced by L. plantarum BMCM12 strain in MRS broth.

The same samples were firstly submitted to specific glycoprotein

staining (left) and then to classic Coomassie staining (right). Lane 1horseradish peroxidase (positive control); lane 2 soybean trypsin

inhibitor (negative control); lane 3 BMCM12 cytoplasmic extract;

lane 4 proteins precipitated from MRS; lane 5 proteins secreted by the

BMCM12 strain (EtOH-precipitated); lane 6 proteins secreted by the

BMCM12 strain (TCA-precipitated); lane 7 TCA-precipitated pro-

teins after de-glycosylation with TFMS. MM molecular markers

(LMW-SDS Marker Kit; GE Healthcare, Bordeaux, France)

594 B. Sanchez, M. C. Urdaci: Adhesion Inhibition by Lactobacillus Extracellular Proteins

123

proteins that appear to be glycosylated. For one of the two

proteins, the major autolysin Acm2, the nature of the gly-

cosylation has been recently described, consisting mainly

in N-acetyl-glucosamine [16].

Bands PL23 and PL24 were identified as muramidase

and GAPDH, respectively. Muramidases are surface pro-

teins responsible for peptidoglycan hydrolysis during

bacterial growth, and are also involved in other biological

functions such as cell wall turnover and cell separation and

division [4]. On the contrary, GAPDH is a cytoplasmic

protein that is frequently found on the surface of many

LABs [4]. More concisely, L. plantarum GAPDH has been

shown to bind human colonic mucin and blood antigens

[17–19]. PL29 was showed to carry a chitin-binding

domain by bioinformatic analysis. Homologous proteins

harboring this domain are found in other bacteria, were

they have been proposed as important colonization factors

[20, 21]. This protein could thus perform important

roles on L. plantarum adhesion to the gastrointestinal

epithelium.

PL21 and PL22 showed a clear shift in their observed

molecular masses after protein de-glycosylation. As said

above, PL21 and PL22 glycosylation was confirmed by

specific staining. The theoretical sequence of both proteins

presented a signal peptide and domains typical of surface

proteins, such as LysM domains, NLPC60 domains,

transmembrane helices or, in the case of PL22, a C-ter-

minal anchor LPXTG. Glycosylation might be produced by

one or several of the glycosyl-transferases contained in the

Table 1 Extracellular proteins produced by Lactobacillus plantarum BMCM12

Banda Protein Accessionb MMc pId MS/

MSeMWEf SignalCg Surfaceh PSORTbi Glyc.j

PL21 Extracellular protein gi|28378774 48.3 9.11 2 72 ANA-AS LysM (2),

NLPC60

Extracellular ?

PL22 Extracellular protein gi|28379317 35.0 9.55 1 114 AKA-DT LysM, LPXTG Unknown ?

PL23 Muramidase gi|28271996 82.1 8.99 2 62 ASA-NQ SH3_5 (4) Extracellular –

PL24 Glyceraldehyde

3-phosphate

dehydrogenase

gi|28377642 36.4 5.30 7 173 – – Cytoplasmic –

PL25 Not identified – – – – – – – – –

PL26 Extracellular protein gi|28270057 22.1 8.78 3 380 AQA-TA LysM Unknown –

PL27 Extracellular protein gi|28272281 21.3 8.86 5 334 ANA-DS LysM Extracellular –

PL28 Hypothetical protein gi|28379289 21.5 9.70 1 114 VKA-TN – Unknown –

PL29 Extracellular protein gi|28378386 22.2 8.89 2 133 VSA-HG – Extracellular –

S Serpin B1 (from Sus scrofa) gi|417185 42.5 5.99 2 132 – – – –

All bands were identified against the genome of Lactobacillus plantarum WCFS1a Labels refer to bands in Fig. 1b Protein accession numberc Theoretical molecular massd Theoretical isoelectric pointe Fragmented MS/MS peptides allowing the identification of the protein [23]f MOWSE score resulting from the ion MS/MS search against the non-redundant NCBI protein database. All scores are statistically significant

(p \ 0.05)g Signal peptidase cleavage sites were predicted using SignalP [24]h Presence of domains characteristic from surface proteins. Number of domain repetition is indicated between parenthesesi Final subcellular localization was predicted using the PSORTb package. Possible localisations are cytoplasmic, cytoplasmic membrane, cell

wall or extracellular. When the probability for being surface or extracellular is the same, ‘‘unknown’’ is returned [25]j Potential protein glycosylation is indicated with the symbol ?

0

1

2

3

4

5

6

7

BMCM12 BMCM12+ E. coli E. coli+ S. enterica S. enterica+

% a

dhes

ion

******

***

Fig. 2 Adhesion of L. plantarum BMCM12, E. coli LMG2092, and

Salmonella enterica subsp. enterica LMG15860 to mucin monolayers

preincubated (?) or not with 20 lg of BMCM12 extracellular

proteins (***p \ 0.01)

B. Sanchez, M. C. Urdaci: Adhesion Inhibition by Lactobacillus Extracellular Proteins 595

123

L. plantarum genome, but this fact deserves further

research.

Extracellular protein fractions were used for testing the

inhibition of the adhesion of the model enteropathogens

E. coli LMG2092 and Salmonella enterica subsp. enterica

LMG15860 to mucin, for which adhesion to epithelium

crucial for their virulence. Proteins were added to mucin

prior, during and after the addition of the pathogen, a

clearly effect in the adhesion being only observed in the

first case. Therefore, these proteins can only interfere, in

our experimental setting, when attached first to mucin,

being unable to compete and to displace previously

adhered pathogen cells. In this regard, we had already

observed that many of the proteins secreted by the strain

BMCM12 could bind mucin [22].

Conclusion

To sum up, extracellular proteins produced by the strain

BMCM12 inhibit the adhesion of two enteropathogens,

representing a molecular mechanism by which probiotics

may decrease enteropathogen invasion. Further research

will elucidate the precise interaction between these extra-

cellular proteins (among which some glycoproteins), and

the human mucin.

Acknowledgments Borja Sanchez was the recipient of a Juan de la

Cierva post-doctoral research contract from the Spanish MICINN.

This study has been supported with grant (RM2010-00012-00-00)

from the Spanish MICINN.

References

1. Chevallier B, Hubert JC, Kammerer B (1994) Determination of

chromosome size and number of rrn loci in Lactobacillus plan-tarum by pulsed-field gel electrophoresis. FEMS Microbiol Lett

120:51–56

2. Kleerebezem M, Boekhorst J, van KR, Molenaar D, Kuipers OP,

Leer R, Tarchini R, Peters SA, Sandbrink HM, Fiers MW et al

(2003) Complete genome sequence of Lactobacillus plantarum.

WCFS1 PNAS USA 100:1990–1995

3. Molin G (2001) Probiotics in foods not containing milk or milk

constituents, with special reference to Lactobacillus plantarum299v. J Clin Nutr 73:380S–385S

4. Sanchez B, Bressollier P, Urdaci MC (2008) Exported proteins in

probiotic bacteria: adhesion to intestinal surfaces, host immuno-

modulation and molecular cross-talking with the host. FEMS

Immunol Med Microbiol 54:1–17

5. Mirgorodskaya E, Krogh TN, Roepstorff P (2000) Characteriza-

tion of protein glycosylation by MALDI-TOFMS. Meth Mol Biol

146:273–292

6. Opdenakker G, Rudd PM, Ponting CP, Dwek RA (1993) Con-

cepts and principles of glycobiology. FASEB J 7:1330–1337

7. Yurist-Doutsch S, Chaban B, VanDyke DJ, Jarrell KF, Eichler J

(2008) Sweet to the extreme: protein glycosylation in Archaea.

Mol Microbiol 68:1079–1084

8. Hitchen PG, Dell A (2006) Bacterial glycoproteomics. Microbi-

ology 152:1575–1580

9. Alaimo C, Catrein I, Morf L, Marolda CL, Callewaert N, Valvano

MA, Feldman MF, Aebi M (2006) Two distinct but inter-

changeable mechanisms for flipping of lipid-linked oligosaccha-

rides. EMBO J 25:967–976

10. Fletcher CM, Coyne MJ, Villa OF, Chatzidaki-Livanis M,

Comstock LE (2009) A general O-glycosylation system impor-

tant to the physiology of a major human intestinal symbiont Cell

137:321–331

11. Benz I, Schmidt MA (2002) Never say never again: protein

glycosylation in pathogenic bacteria. Mol Microbiol 45:267–276

12. Sanchez B, Chaignepain S, Schmitter JM, Urdaci MC (2009) A

method for the identification of proteins secreted by lactic acid

bacteria grown in complex media. FEMS Microbiol Lett 295:

226–229

13. Laemmli UK (1970) Cleavage of structural proteins during the

assembly of the head of bacteriophage T4. Nature 227:680–685

14. Edge AS (2003) Deglycosylation of glycoproteins with tri-

fluoromethanesulphonic acid: elucidation of molecular structure

and function. Biochem J 376:339–350

15. Wessel D, Flugge UI (1984) A method for the quantitative

recovery of protein in dilute solution in the presence of detergents

and lipids. Anal Biochem 138:141–143

16. Fredriksen L, Mathiesen G, Moen A, Bron PA, Kleerebezem M,

Eijsink VGH, Egge-Jacobsen W (2011) The major autolysin

Acm2 from Lactobacillus plantarum undergoes cytoplasmic

O-glycosylation J Bacteriol 193(22): 6323–6330

17. Kinoshita H, Uchida H, Kawai Y, Kawasaki T, Wakahara N,

Matsuo H, Watanabe M, Kitazawa H, Ohnuma S, Miura K, Horii

A, Saito T (2008) Cell surface Lactobacillus plantarum LA 318

glyceraldehyde-3-phosphate dehydrogenase (GAPDH) adheres to

human colonic mucin. J Appl Microbiol 104:1667–1674

18. Vollmer W, Joris B, Charlier P, Foster S (2008) Bacterial pep-

tidoglycan (murein) hydrolases. FEMS Microbiol Rev 32:

259–286

19. Kinoshita H, Wakahara N, Watanabe M, Kawasaki T, Matsuo H,

Kawai Y, Kitazawa H, Ohnuma S, Miura K, Horii A, Saito T

(2008) Cell surface glyceraldehyde-3-phosphate dehydrogenase

(GAPDH) of Lactobacillus plantarum LA 318 recognizes human

A and B blood group antigens. Res Microbiol 159:685–691

20. Kirn TJ, Jude BA, Taylor RK (2005) A colonization factor links

Vibrio cholerae environmental survival and human infection.

Nature 438:863–866

21. Kawada M, Chen CC, Arihiro A, Nagatani K, Watanabe T,

Mizoguchi E (2008) Chitinase 3-like-1 enhances bacterial adhe-

sion to colonic epithelial cells through the interaction with bac-

terial chitin-binding protein. Lab Invest 88:883–895

22. Sanchez B, Schmitter JM, Urdaci MC (2009) Identification of

novel proteins secreted by Lactobacillus plantarum that bind to

mucin and fibronectin. J Mol Microbiol Biotechnol 17:158–162

23. Pappin DJ (2003) Peptide mass fingerprinting using MALDI-TOF

mass spectrometry. Meth Mol Biol 211:211–219

24. Gardy JL, Laird MR, Chen F, Rey S, Walsh CJ, Ester M,

Brinkman FS (2005) PSORTb v. 2.0: expanded prediction of

bacterial protein subcellular localization and insights gained from

comparative proteome analysis. Bioinformatics 21:617–623

25. Emanuelsson O, Brunak S, von Heijne G, Nielsen H (2007)

Locating proteins in the cell using TargetP, SignalP and related

tools. Nat Protoc 2:953–971

596 B. Sanchez, M. C. Urdaci: Adhesion Inhibition by Lactobacillus Extracellular Proteins

123