Maurilia Rojas, Laboratorio de Ciencia y Tecnología de Alimentos, Área Interdisciplinaria de Ciencias Agropecuarias, Universidad Autónoma de Baja California Sur. Km 5.5, Carretera al Sur, 23080 La Paz, Baja California Sur, México. E-mail: firstname.lastname@example.org
Aims: To identify and characterize adhesion-associated proteins in the potential probiotic Lactobacillus fermentum BCS87.
Methods and Results: Protein suspensions obtained from the treatment of Lact. fermentum BCS87 with 1 mol 1−1 LiCl were analysed by Western blotting using HRP-labelled porcine mucus and mucin. Two adhesion-associated proteins with relative molecular weight of 29 and 32 kDa were identified. The N-terminal and internal peptides of the 32 kDa protein (32-Mmubp) were sequenced, and the corresponding gene (32-mmub) was found by inverse polymerase chain reaction. The complete nucleotide sequence of 32-mmub revealed an open reading frame of 903 bp encoding a primary protein of 300 amino acids and a mature protein of 272 residues. A basic local alignment search showed 47–99% identity to solute-binding components of ATP binding cassette transporter proteins in Lactobacillus, Streptococcus and Clostridium. An OpuAC-conserved domain was identified and phylogenetic relationship analysis confirmed that 32-Mmubp belongs to the OpuAC family.
Conclusions: Adhesion of Lact. fermentum BCS87 appeared to be mediated by two surface-associated proteins. 32-Mmubp is a component of ABC transporter system that also functions as an adhesin.
Significance and Impact of the Study: Characterization of 32-Mmubp and 32-mmub will contribute to understanding the host–bacteria interactions of Lact. fermentum with the intestinal tract of pigs.
Lactobacillus fermentum is a normal inhabitant of the gastrointestinal tract of pigs (Tannock et al. 1990). The number of reports of health-promoting effects attributed to Lact.fermentum has been increased in recent years where antagonistic activities against enteropathogens and modulation of immune system are well documented (Vinderola et al. 2004; Lin et al. 2007; Zoumpopoulou et al. 2008). Specific adherence of Lactobacillus to the intestinal tract of rats has been associated with stimulation of the immune system and inhibition of adhesion of pathogens (Herías et al. 1999). Moreover, adhesion is considered as a critical event for colonization not only for lactobacilli but also for pathogenic bacteria (Soto and Hultgren 1999). Mechanisms used by lactobacilli to recognize and adhere to gastrointestinal components were poorly understood until recently. The ability of strains to adhere to mucus is an important characteristic that is evaluated in probiotic Lactobacillus (Ma et al. 2006). Recently, cell surface proteins have been recognized by their ability to bind gastrointestinal mucus and mucins (Rojas et al. 2002; Perea-Vélez et al. 2007; Sun et al. 2007). In Lactobacillus, the mucus-binding protein (Mub) of Lactobacillus reuteri 1063 (Roos and Jonsson 2002), the lectin-like mannose-specific adhesin (Msa) of Lactobacillus plantarum WCFS1 (Pretzer et al. 2005), the mucus adhesion promoting protein (MapA) of Lact. reuteri 104R (Miyoshi et al. 2006) and the Mub of Lactobacillus acidophilus NCFM have been described (Buck et al. 2005). In Lact.fermentum 104R (identified later as Lact. reuteri 104R), a mucus adhesion promoting protein (MAPP) was reported by its ability to bind porcine mucus and mucins (Rojas et al. 2002). However, reports with completely characterized molecules involved in the adhesion of Lact.fermentum strains to intestinal components were not found. Thus, the aim of this study was to identify and characterize surface proteins in the potential probiotic Lact.fermentum strain BCS87 involved in the adhesion to mucus and gastric mucin of pigs. This strain showed the in vitro ability to adhere to mucus and mucin, to grow in mucus, in the presence of bile salts, at a high concentration of sodium chloride and at high temperatures (Macías-Rodríguez et al. 2008). Characterization of adhesive factors in probiotic Lactobacillus strains will contribute to understand the mechanisms used by members of this genus to colonize the gastrointestinal tract of pigs.
Materials and methods
Bacterial strains and culture conditions
Lactobacillus fermentum strain BCS87 used was isolated from faeces of a healthy 23-day-old piglet from a farm located in an arid coast and showed in vitro probiotic potential (Macías-Rodríguez et al. 2008). A primary culture from a stock stored at −80°C was grown overnight at 37°C in tubes containing 5 ml of MRS broth (Mann Rogosa and Sharp; Difco, MD, USA). Then, the bacterium was inoculated [5% (v/v) of the final volume] into fresh MRS broth and developed for 4 h to perform DNA extractions or inoculated [1% (v/v) of the final volume] into LDM semidefined medium (Conway and Kjelleberg 1989) supplemented with 2% glucose and allowed to grow to reach an optical density (OD600) of 0·9–1·0 for protein extracts preparations.
Extraction of cell surface proteins
Bacterial biomass from 1 l of LDM-grown culture was collected after 5-min centrifugation, at 4°C, 4600 g and washed once in 90 ml phosphate buffered saline (PBS) (145 mmol 1−1 NaCl, 2·87 mmol 1−1 KH2PO4 and 6·95 mmol 1−1 K2HPO4, pH 7·2). The pellet was suspended in 40 ml cold 1 mol l−1 lithium chloride (LiCl) and incubated at 4°C for 1 h. Cell suspension was centrifuged at 15 500 g for 30 min at 4°C, and the supernatant was filtered through a 0·2 μm pore size nitro-cellulose syringe filter. Lithium chloride suspension was dialysed against distilled water in 6000–8000 cut-off tubing (Spectrapor, Los Angeles, CA) and lyophilized (Labconco mod. Freezone 4.5; Kansas City, MO). Protein concentration was determined by Bradford method using a protein assay II kit (Bio-Rad) and a SmartSpect spectrophotometer (Bio-Rad). Protein suspension was stored at −20°C until used.
Mucus and mucin preparation
Mucus and mucin preparation was performed as described (Rojas et al. 2002). Briefly, crude mucus from the small intestine of a 23-day-old healthy piglet was obtained by gentle scraping and suspended in cold HEPES plus Hanks’ balanced salt solution (H–H buffer: 136·87 mmol 1−1 NaCl, 5·37 mmol 1−1 KCl, 1·26 mmol 1−1 CaCl2.2H2O, 0·81 mmol 1−1 MgSO4.7H2O, 0·35 mmol 1−1 Na2HPO4, 2·57 mmol 1−1 KH2PO4, 9·98 mmol 1−1 HEPES, pH 7·4). Mucin type III partially purified from porcine stomach (Sigma) was also used. Both mucus and mucin were conjugated with horseradish peroxidase (HRP) and then stored at −20°C until used.
SDS-PAGE and Western blot protein analysis
Proteins shaved from cells by LiCl treatment and lyophilized as described earlier were diluted in loading buffer at a final concentration of 0·2 μg μl−1. Thirty microlitre of the suspension was run in a 12% denaturing polyacrylamide gel (SDS-PAGE), according to Laemmli (1970), using a Mini-Protean II system (Bio-Rad). Low range prestained or unstained SDS-PAGE molecular weights markers (Bio-Rad) were used. Proteins from nonstained gels were electrophoretically transferred to immobilon PVDF membranes (Millipore) with a semidry transfer system (Bio-Rad). Lactobacillus reuteri 104R previously shown as a strain with strong adhesion properties, through the 29-kDa MAPP protein (Rojas et al. 2002), was used as positive control. Western blot assays were performed as previously described by Rojas et al. (2002) using either HRP-labelled mucus or mucin.
N-terminal and internal peptide sequencing
Proteins of LiCl-treated cells were separated by SDS-PAGE in a 12% polyacrylamide gel (20 × 20 cm) as described earlier. Bands corresponding to proteins adhering to mucus or mucin were excised from the membrane and submitted for Edman degradation. For internal peptides, the band was excised directly from a Coomassie blue stained gel, dried and submitted for sequencing to the PAN Facility, Stanford University Medical Center, USA.
Isolation of chromosomal DNA
Chromosomal DNA was prepared as described by Anderson and McKay (1983). Briefly, 4 ml of a MRS culture prepared as described earlier was harvested by centrifugation at 3300 g for 5 min at 4°C. Pellet was washed in STE buffer [6·7% (w/v) sucrose, 50 mmol 1−1 Tris–HCl, 20 mmol 1−1 EDTA, pH 8·0], resuspended in the same buffer containing 5 mg ml−1 lysozyme additionally and then incubated at 37°C for 1 h. Then, 25 μl of 20% SDS and 4 μl of proteinase K (20 mg ml−1) were added, and the tubes were incubated at 60°C for 1 h. Proteins were removed by phenol–chloroform extraction, and DNA was precipitated from the aqueous phase with 2·5 volumes of cold absolute ethanol. The pellet was washed with 70% ethanol, dried and then suspended in water or TE buffer (10 mmol 1−1 Tris–HCl, 1 mmol 1−1 EDTA, pH 8·0).
Oligonucleotide design and synthesis
Oligonucleotides used for PCR amplifications were designed with the Primer Select tool of Lasergene software (ver. 5, Madison, WI) and synthesized at the Instituto de Biotecnología, UNAM (Mexico) and by Sigma-Genosys Company (The Woodlands, TX, USA). Degenerated oligonucleotides were deduced from the sequenced N-terminal and internal peptides. Lowest codon degeneracy of amino acids and the preferential codon usage for Lact.fermentum were considered to decrease the degeneracy of the primers. Other primers were designed from the DNA sequence obtained from PCR fragments. All are listed in Table 1.
Table 1. Oligonucleotides used for PCR amplifications
Amplification and sequencing of gene fragments
A combination of oligonucleotides FAAT87-32 (forward) and RF1087-32 or RE787-32 (reverse) was used to perform an amplification reaction using the chromosomal DNA of Lact.fermentum BCS87 as template. The PCR solution contained a final concentration of 1× Taq polymerase buffer, 3 mmol l−1 MgCl2, 0·2 mmol l−1 of each dNTP, 20 pmol of each primer, 300 ng of chromosomal DNA and 2 U of Taq DNA polymerase in a total volume of 25 μl. Amplification reaction was performed in a thermocycler (GeneAmp 2400; Perkin-Elmer) with the following temperature programme: one cycle at 94°C for 5 min; 30 cycles consisted of a denaturation step at 94°C for 1 min, an annealing step at 54°C for 1 min and an extension step at 72°C for 1 min. A final extension was performed at 72°C for 5 min. PCR products were then analysed in a 1·5% agarose gel. Amplified bands were excised, purified with a Qiagen Gel extraction kit and submitted for sequencing to Macrogen, Korea.
Chromosomal DNA of Lact.fermentum BCS87 (1 μg) was digested using the restriction enzymes BamHI, BglII, EcoRI, HindIII, KpnI, PstI, SacI and SalI, according to manufacturer protocol instructions, run on 1% agarose gel and transferred to nylon membrane by diffusion. A 0·5-kb fragment amplified with primers MEF7 and MER9 was used as a probe. Probe labelling and detection were performed using the ECL kit (Amershan Biosciences) according to the instructions of the manufacturer.
Gene amplification by inverse PCR
In order to find sequences that flank the 0·5-kb fragment obtained by amplification with degenerated oligonucleotides FAAT87-32 and RF1087-32, an inverse PCR was performed as described by Ochman et al. (1988). Chromosomal DNA fragments, obtained from digestions with BamHI, BglII, HindIII, PstI and SacI (1 μg of each digestion), were used for self-ligations (intramolecular ligation). The reactions contained a final concentration of 1× ligase buffer, 5 ng μl−1 of digested DNA and 0·05 U μl−1 of T4 DNA ligase (Invitrogen). Ligation mixtures were incubated at 23°C for 16 h and then each DNA was precipitated with 0·5 mol 1−1 NaCl and three volumes of absolute ethanol at −20°C overnight. Ligated DNA was suspended in 50 μl distilled water and 5 μl of the suspension was used for PCR amplification with oligonucleotides MEF9 and MER11. PCR solution contained a final concentration of 1× Taq polymerase buffer, 1·5 mmol l−1 of MgCl2, 0·2 mmol l−1 of each dNTP, 100 pmol of each oligonucleotide, 5 μl of ligated fragments and 1 U of Taq DNA polymerase in a total volume of 50 μl. Amplification reactions were performed using the following amplification programme: one cycle at 94°C for 5 min; 30 cycles consisted of a denaturation step at 94°C for 1 min, an annealing step at 54°C for 2 min and an extension step at 72°C for 3 min. A final extension was performed at 72°C for 5 min. Amplified products were analysed on 1·5% agarose gels, and bands corresponding to the expected sizes were cut, cleaned and sequenced.
Edition of the sequences obtained by PCR was performed with the dnastar’s Lasergene sequence analysis software (ver.5). Sequence similarity was found using Blast (http://www.ncbi.nlm.nih.gov/BLAST/). Translated amino acid sequences, calculation of relative molecular weight and the isoelectric point were obtained by public tools at Expasy (http://www.expasy.ch/tools/). Other online sequence analysis services such as ClustalW (http://clustalw.genome.ad.jp/), Signal P 3.0 (http://www.cbs.dtu.dk/services/SignalP/) and tmhmm (http://www.cbs.dtu.dk/services/TMHMM/) were also used. To find conserved domains in the 32-kDa protein (referred later as 32-Mmubp; 32-kDa mucus- and mucin-binding protein), a search with specialized Blast tools and Pfam was performed (http://blast.ncbi.nlm.nih.gov/Blast.cgi; http://pfam.sanger.ac.uk/). Finally, phylogenetic relationships were calculated by maximum parsimony (MP) using the program paup 4.0b (Swofford 2003) with a heuristic search employing general search options. Initially, the full multiple alignment of the seed group of the OpuAc family (PF04069) was downloaded from the Pfam website (Pfam ver. 22.0, July 2008). In order to analyse relationships between the sequences, this alignment was used to construct a neighbour-joining (NJ) tree using the program paup 4.0d56 (D.L. Swofford unpublished data). Sequences from the same taxa were removed from the multiple alignments. If multiple copies of the same sequence were present (e.g. 100% identical sequences from different strains), only one copy was retained. By doing so, the general topology of the phylogenetic tree was not affected by reducing the number of sequences. Confidence limits for phylogenetic trees were estimated from bootstrap analyses (1000 replications for neighbour-joining searches).
Sequence of the 32-mmub gene has been deposited at the GenBank database under the accession number FJ026800.
Extraction of cell surface proteins and Western blot analysis
To understand the relevance of proteinaceous surface components in the adhesion of the potential probiotic Lact.fermentum strain BCS87 to mucus and mucin, a protein suspension was obtained by treatment with 1 mol 1−1 LiCl. After the treatment, a protein concentration of 2·9 μg ml−l was quantified by Bradford method. A Western blot assay using HRP-labelled mucus and mucin showed that two of 24 protein bands observed in the Coomassie Blue stained gel were able to adhere to both mucus and mucin (Fig. 1a). These proteins showed a relative molecular weight (MW) of approx. 29 and 32 kDa. This report was focused on the band of 32 kDa MW.
N-terminal and internal peptide sequencing
In order to identify the 32-kDa protein (32-Mmubp) from Lact.fermentum BCS87, sequences corresponding to N-terminal and two internal peptides were determined. N-terminal sequence was – NSKTPIIVGSKSFTE (15 amino acids), whereas sequences corresponding to the two internal peptides were respectively DDKHLWPAYNLVPLVR (peptide F10; 16 amino acids) and KYGLYTISDLQK (peptide E7; 12 amino acids).
Amplification and sequencing of the 32-Mmubp encoding gene (32-mmub)
The N-terminal and internal peptides of 32-Mmubp were used to design degenerated oligonucleotides in order to amplify an internal fragment of the 32-mmub gene. A combination of FAAT87-32 and RF1087-32 oligonucleotides resulted in an amplified band of approx. 550 bp. This fragment was sequenced and revealed a coding sequence (CDS) encompassing the N-terminal sequence and ended by part of the internal peptide E7 of 32-Mmubp. Primers MEF7 and MER9 deduced from the sequence of the initial 0·55-kb PCR fragment were used to generate a DNA probe. The restriction map of the DNA region encompassing 32-mmub was determined by Southern blot on digested chromosomal DNA of Lact.fermentum BCS87. Approximate sizes of hybridized restricted fragments were: BamHI 2·5 kbp, BglII 2 kbp, EcoRI 8 kbp, HindIII 2·3 kbp, KpnI 8 kbp, PstI 2·2 kbp, SacI 2 kbp and SalI 1·2 and 0·6 kbp. Because the expected size of the gene was around 1 kb, we retained BamHI, BglII, HindIII, PstI and SacI digestions assuming these fragments could contain the whole gene and might have a size compatible with PCR amplification. Those digestions were used for self-ligation and inverse PCR amplifications. Four amplified products of approx. 1·6 kbp (BglII), 1·9 kbp (HindIII), 1·8 kbp (PstI) and 0·9 kbp (SalI) were obtained and subsequently sequenced. Finally, from the four products, a fragment of 2803 bp that contains the complete CDS of the 32-mmub gene was deduced.
The complete nucleotide sequence of 32-mmub gene revealed an ORF (open reading frame) of 903 bp encoding a predicted primary protein of 300 amino acids (Fig. 2). The N-terminal and internal peptides (E7 and F10) matched with the corresponding deduced amino acid sequence. A signal peptide of 28 amino acids and a cleavage site between residues 28 and 29 were detected with the SignalP 3.0 prediction software. Moreover, the prediction of transmembrane helices showed that the first 1–7 amino acids are predicted to be inside of the cell, whereas residues 7–29 could be in the membrane, and finally the region encompassing amino acids 30–300 could be outside. These results are in accordance with N-terminal and internal peptides sequenced for the 32-Mmubp, which are predicted to be outside. The mature protein consists of 272 residues with a molecular mass of 29 974 Da, which is similar to the apparent molecular mass observed by SDS-PAGE (Fig. 1). Moreover, an isoelectric point of 9·78 and a positive net charge of 21·22 at pH 7·0 were predicted. A similarity search using the deduced amino acid sequence of the 32-Mmubp was performed with the algorithm blastp. Results showed between 47% and 99% identity to solute-binding components of ABC transporters proteins (ATP-binding cassette proteins) of Clostridium beijerincki (YP 001310619), Clostridium acetobutylicum (NP 348102), Streptococcus thermophilus (YP 141667, YP 001310619), Lact. reuteri (YP 001841171, AAY86791, YP 001270791, EAS89056) and 99% identity only to Lact.fermentum (BAG27284). Alignment performed with those proteins showed conserved peptides, even if the percentage of identity is not high (Fig. 3). As the genome of Lact.fermentum IFO 3956 was recently published (Morita et al. 2008), a search of homology (Blast) between the complete 2803-bp fragment obtained by inverse PCR and this genome was performed. Results showed 98% identity with an ABC transporter system of Lact.fermentum IFO 3956. This observation suggests that 32-Mmubp in Lact.fermentum BCS87 is part of the ABC transporter system that is organized in the same manner (data no shown). A search of putative conserved domains performed using Blast and Pfam showed that 32-Mmubp belongs to the PBPb superfamily. 32-Mmubp showed similarity to substrate-binding domains of ABC type glycine/betaine transport systems of the OpuAc family (PF04069). Family members of OpuAC are often integral membrane proteins or predicted to be attached to the membrane by a lipid anchor. This family contains 928 sequences widely distributed in bacteria (340 species) and archaea (11 species). A phylogenetic relationship analysis, using available amino acid sequences of members of the OpuAC domain, showed a conserved relationship of the 32-Mmubp sequence with all sequence members of this family. A tree constructed with 37 sequence members of OpuAC, and the 32-Mmubp sequence is shown in Fig. 4.
Together with mucins and antimicrobial peptides, intestinal microbiota constitutes the first defensive barrier against the invasion of pathogenic bacteria (Liévin-Le and Servin 2006). However, mechanisms by which intestinal beneficial microbiota colonizes the gastrointestinal tract are poorly understood until now. It has been accepted that surface molecules that act as mucus-binding proteins could be vital during colonization (Boekhorst et al. 2006). In Lact.fermentum strains, the in vitro adhesion appears to be correlated with their probiotic ability (Lin et al. 2007; Li et al. 2008). In this report, two adhesion-associated proteins involved in adhesion of potential probiotic Lact.fermentum strain BCS87 were identified. These proteins were extracted with LiCl, suggesting that they are noncovalently associated with the cellular surface of Lact.fermentum BCS87 (Lortal et al. 1992). Both proteins showed the ability to bind intestinal mucus and gastric mucin, suggesting that components of mucins could be the receptors for these proteins. These receptors have been previously reported by Fang et al. (2000) and Vimal et al. (2000), who observed that enterotoxigenic Escherichia coli K88ac+ and Salmonella typhimurium specifically bind mucus proteins in the small intestine of pigs and rats respectively. In addition to secretory mucins (encoded by genes MUC2, MUC5AC, MUC5B and MUC6), membrane-associated mucins (encoded by genes MUC1, MUC3A, MUC3B, MUC4, MUC12, MUC13, MUC16 and MUC17) provide a steric barrier that can limit direct access of pathogens to intestinal cells (Zoghbi et al. 2006). In Lactobacillus strains, it was demonstrated that the ability to adhere to mucin-secreting HT29 cells is related with an increase in epithelial cell MUC3 expression (Mack et al. 2003). In this study, it was not possible to evaluate the adherence of Lact.fermentum BCS87 specifically to membrane-associated mucins. However, the ability of Lact.fermentum BCS87 to adhere to partially purified mucin suggests that this could be a candidate to evaluate their contribution on the stimulation of expression of MUC3 genes. To determine whether 32-Mmubp of Lact.fermentum BCS87 is expressed by other Lactobacillus strains, a PCR using MEF7 and MER9 oligonucleotides was performed. Chromosomal DNA of 26 strains belonging to Lact.fermentum, Lactobacillus johnsonii, Lactobacillus delbrueckii subsp. bulgaricus, Lactobacillus crispatus, Lactobacillus vaginalis and Lact. reuteri species was used as template to amplify an internal product of 32-mmub gene. A 0·55-kbp amplified product was observed in Lact.fermentum strains but not in other species, suggesting that 32-mmub gene is conserved in this species (data no shown). Gene that codes for 32-Mmubp (32-mmub gene) is reported; a signal peptide in the deduced amino acid sequence between amino acids 1 and 28, a cleavage site between residues 28 and 29 (Fig. 2) and a high isoelectric point were predicted for this protein. These characteristics suggest that 32-Mmubp is released to the medium, but it could be anchored to cell wall by electrostatic interactions with acidic groups as suggested by Turner et al. (1997) for a basic protein of Lact. fermentum BR11. 32-Mmubp from Lact. fermentum BCS87 is a member of an ABC transporter system. Similarly in Lact. reuteri NCIB 11951, a collagen-binding protein was identified and characterized and showed similarity to solute-binding components of bacterial ABC transporters (Roos et al. 1996). These findings suggest that proteins involved in transport of metabolites may have more than one function or a different function from what was predicted by bioinformatic tools as previously observed by Wall et al. (2003) and Båth et al. (2005). A bioinformatic approach based on homology and sequence domain search was used to support the function assignment of the 32-Mmubp as a binding protein. Derived from this analysis, a phylogenetic tree was constructed using sequences of the seed group of the OpuAC family (Fig. 4). This family is part of a high-affinity multicomponent binding proteins-dependent transport system involved in bacterial osmoregulation. In the phylogenetic tree, the 32-Mmubp sequence showed that which are conserved between prokaryotic protein sequences of substrate-binding regions on ABC type glycine/betaine transport systems. Some members of the corresponding taxa having similar ecological niches to those occupied by lactobacilli (gastrointestinal and respiratory tracts), i.e. Helicobacter pylori and Mycobacterium tuberculosis, do not group together suggesting that adhesion mechanisms is not a phylogenetic-associated trait. The full species tree provided by Pfam for this family of proteins (data not shown) includes 351 species with members representing the whole bacteria and archaea domains. A tree based on sequences of rDNA represents the closest approach to species phylogeny. Therefore, a strong incongruence between the species and protein trees as observed for members of the OpuAC protein family and corresponding taxa might be an evidence of a high degree of lateral transfer. It has been demonstrated that horizontal transfer by genetic transformation of bacteria occurs in the environment between the species inhabit similar ecological niches (Lorenz and Wackernagel 1994). Horizontal transfer of genes between bacteria and archaea was previously reported by Noll et al. (2008), who showed that maltose ABC transporter operons were carried over by horizontal transfer between the Thermococcales (members of archaea domain) and Thermotogales (members of bacteria domain). In fact, the genome sequence of the bacterial hyperthermophile Thermotoga maritima revealed evidence of extensive horizontal gene transfer with archaea (Nelson et al. 2001). Many of the genes that have been shared with archaea encode for ABC transporters (Nanavati et al. 2006). However, an explanation why micro-organisms living in quite different niches could share these type of transport systems is still unknown.
In this study, an additional function as a mucus adhesion promoter is assigned to 32-Mmubp of Lact.fermentum BCS87. However, as two adhesive proteins were identified by Western blot on the surface of Lact.fermentum BCS87, it is necessary to determine their individual contribution to mucus and mucin adhesion on this strain. Actually, the sequence of N-terminal and two internal peptides of the 29-kDa adhesive protein did not present identity with the previously reported MAPP protein of Lact.fermentum 104R (data not shown). Therefore, the study of the gene encoding this protein is ongoing. Finally, we can conclude that description of 32-Mmubp and 32-mmub involved in the adhesion to mucosal components is important to understand the host–bacteria interactions of the potential probiotic Lact.fermentum BCS87 in the intestinal tract of pigs.
This study was supported by CONACYT project no. 29410-B and Ecos-Nord collaborative project between Mexico and France No. M05-A01. We thank CONACYT for financial assistance (Graduate fellowship no.144406).