Previous Article | Next Article ![]()
Applied and Environmental Microbiology, September 2007, p. 5598-5606, Vol. 73, No. 17
0099-2240/07/$08.00+0 doi:10.1128/AEM.00642-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Department of Agricultural Biotechnology and Center for Agricultural Biomaterials, Seoul National University, 56-1 Sillim-dong, Gwanak-gu, Seoul 151-921, Korea,1 Department of Food and Nutrition, Seoul National University, 56-1 Sillim-dong, Gwanak-gu, Seoul 151-742, Korea,2 Research Institute, Bifido Inc., Ltd., Hongchon 250-804, Korea,3 Department of Hotel Culinary Art, Anyang Technical College, Anyang 430-749, Korea4
Received 20 March 2007/ Accepted 23 June 2007
|
|
|---|
|
|
|---|
The components of the proteolytic systems of lactococci and some lactobacilli have been investigated, and many peptidases from LAB have been purified and characterized. In addition, several peptidase genes have been cloned and sequenced (18, 20).
However, very little is known about the proteolytic enzyme systems of Bifidobacterium spp., and no amino-, di-, and tripeptidase genes have been cloned from bifidobacteria. Bifidobacterium spp. are comparable to LAB in that their general aminopeptidase activity encompasses several dipeptidases and possibly also iminopeptidase and tripeptidases (3, 10, 22). Aminopeptidase and iminopeptidase activities were determined in a cell extract of Bifidobacterium breve by Cheng and Nagasawa (7), and Bifidobacterium infantis, Bifidobacterium longum, and Bifidobacterium adolescentis show amino-, di-, tri-, and carboxypeptidase activities (11). B. longum contains more than 20 predicted peptidases that could provide amino acids from proteinaceous substrates in the gastrointestinal tract and vagina, where carbohydrates are less abundant (27).
In recent years, bifidobacteria have been used in fermented dairy products and nutraceuticals for promoting intestinal health. Although bifidobacteria represent only 3 to 6% of the adult fecal flora, their presence has been associated with beneficial health effects, such as prevention of diarrhea, amelioration of lactose intolerance, and immunomodulation. These correlations have led to the widespread use of bifidobacteria as components of health-promoting foods (27). However, some probiotic bacteria, such as Lactobacillus acidophilus and Bifidobacterium spp., grow slowly in milk because of the absence of proteolytic activity (17).
Proteolytic enzymes, such as di-, tri-, and aminopeptidases, are expected to enhance fermentation because bifidobacteria have a high demand for essential growth factors, such as various amino acids. The importance of free amino acids and peptides to the growth of bifidobacteria has been studied by Cheng and Nagasawa (7).
A dipeptidase is an enzyme that cleaves dipeptides into two amino acids. The dipeptidase activity of B. longum BORI is typically twofold (12) that of various other B. longum strains when using Ala-Ala dipeptide as a substrate. We decided to purify and clone the gene (pepD) coding for this dipeptidase in an Escherichia coli expression vector in order to overexpress this gene and purify the enzyme PepD. This paper describes the purification, characterization, and cloning of a dipeptidase from B. longum BORI and its expression in E. coli.
|
|
|---|
and LMG194 (13) were used for cloning experiments and protein expression, respectively. Strains were purchased from Takara (Shiga, Japan) and Invitrogen (Carlsbad, CA) and cultured at 37°C in Luria-Bertani (LB) broth with vigorous shaking. Ampicillin was used at a concentration of 50 µg·ml–1 for selecting transformed bacteria. The plasmid pGEM-T Easy vector (Promega, Madison, WI) was used for cloning and amplifying the dipeptidase gene (pepD), and the plasmid pBAD-TOPO (Invitrogen) was used for protein expression (13). B. longum BORI was grown in brain heart infusion broth (Difco, Detroit, MI) supplemented with 0.5% glucose and 0.05% (final concentration) L-cysteine·HCl at 37°C. LB broth media and Bacto agar were purchased from Difco. |
View this table: [in a new window] |
TABLE 1. Bacterial strains and plasmids used in this study
|
Enzyme purification.
All purification procedures were performed at 4°C. To purify the dipeptidase protein from B. longum BORI, cells were harvested by centrifugation at 10,000 x g for 30 min and washed three times with distilled water. The obtained cells were suspended in 50 mM Tris-HCl buffer (pH 7.0) and disrupted by sonication. The cell debris was removed by centrifugation at 10,000 x g for 20 min. Clear supernatant was concentrated by salting out with powdered ammonium sulfate at about 70% saturation. All purification procedures were performed at 4°C. The precipitate was dissolved in 50 mM Tris-HCl buffer (pH 7.0) and dialyzed against the same buffer. The dialysate was applied to a DEAE-cellulose column (2.8 by 60 cm) that had been equilibrated with 50 mM Tris-HCl buffer. The column was washed with the same buffer, and linear gradient elution at a flow rate of 2.0 ml·min–1 was applied with 250 ml of 50 mM Tris-HCl buffer (pH 7.0) and 250 ml of the same buffer containing 1.0 M KCl. In addition, the column was washed with 50 mM Tris-HCl buffer (pH 7.0) containing 1.0 M KCl. The fractions exhibiting dipeptidase activity were pooled and then dialyzed against 50 mM Tris-HCl buffer (pH 7.0). The dialysate was applied to a Q-Sepharose column (2.8 by 60 cm) that had previously been equilibrated with 50 mM Tris-HCl buffer. The column was washed with the same buffer, and linear gradient elution at a flow rate of 2.0 ml·min–1 was applied with 250 ml of 50 mM Tris-HCl buffer (pH 7.0) and 250 ml of the same buffer containing 1.0 M KCl. The fractions containing the dipeptidase activity were pooled, dialyzed against 50 mM Tris-HCl buffer (pH 7.0), and lyophilized.
SDS-PAGE.
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed (Mini-Protein II; Bio-Rad Laboratories, Berkeley, CA) with a 10% polyacrylamide gel. Molecular mass SDS-PAGE protein standards supplied by Elpis Biotech (Taejeon, Korea) were used as reference proteins in the range from 100 to 20 kDa. SDS-PAGE was performed by the procedure of Laemmli (19).
Measurement of dipeptidase activity.
Dipeptidase activity was quantified using the cadmium-ninhydrin assay (12) to follow the hydrolysis of Ala-Gln. The reaction mixture consisted of 50 µl of enzyme preparation, 50 µl of substrate (10 mM in methanol or deionized water), and 400 µl of reaction buffer (50 mM Tris-HCl buffer, pH 8.0). The mixture was incubated for 30 min at 37°C, and the reaction was terminated by the addition of 1 ml of cadmium-ninhydrin reagent. The contents were thoroughly mixed, heated at 84°C for 5 min, and cooled immediately on ice, and then the absorbance at 507 nm was measured. One unit of enzyme activity was defined as that required to increase the absorbance by one unit per minute under the assay conditions.
Measurement of protein concentration.
The protein was measured using a Bradford protein assay kit with bovine serum albumin as the standard. The standard bovine serum albumin solution was purchased from Bio-Rad Laboratories, and the absorbance at 595 nm was measured.
Effects of pH and temperature on dipeptidase activity.
The effects of pH values in the range 2 to 10 on enzyme activity were examined in buffers at 50 mM, comprising glycine-HCl (pH 2 to 3), maleic acid-Tris-NaOH (pH 4 to 6), Tris-HCl (pH 7 to 9), and glycine-NaOH (pH 10). To assess the effect of pH on enzyme stability, the enzyme was dissolved in buffers at 50 mM at pH values ranging from 2 to 10 and incubated for 30 min at 37°C. The residual activity was subsequently measured at 37°C with Ala-Gln as the substrate. The optimum temperature for dipeptidase activity was determined within the range from 10°C to 80°C in 50 mM Tris-HCl buffer (pH 8.0). The thermal stability of the enzyme was assessed by incubating the enzyme solutions for 10 min at temperatures ranging from 10°C to 80°C. The residual activity was subsequently measured at 37°C with Ala-Gln as the substrate.
N-terminal and internal amino acid sequence analysis.
N-terminal amino acid sequencing of a protein blotted on a polyvinylidene difluoride membrane was analyzed at the Korea Basic Science Institute (Seoul, Korea) using a protein sequencer (PerkinElmer 491; Applied Biosystems, Foster City, CA). Edman degradation was performed using the standard program supplied by Applied Biosystems. The N-terminus-blocked proteins were further analyzed by internal amino acid sequencing. The blocked spots were cut out with a blade, subjected to in-gel trypsin digestion, and analyzed by reverse-phase high-performance liquid chromatography (SMART HPLC; Pharmacia Diagnostics, Uppsala, Sweden). Twenty-microliter aliquots of peptide samples were separated on a Sephasil C18 column (Amersham Biosciences) (2.1 x 100 mm) equilibrated with acidified water containing 0.1% (wt/vol) trifluoroacetic acid. The column was washed with acidified water for 5 min, and the peptide sample was eluted with a linear gradient of acetonitrile (0 to 60% [vol/vol]) in acidified water over a 60-min period at a flow rate of 200 ml·min–1. Fractions of 150 ml were collected.
Amplification of the pepD gene from B. longum BORI genomic DNA.
PCR was used to amplify pepD from genomic DNA of B. longum BORI using GeneReleaser (Bioventure, Murfreesboro, TN). Two primers, pepD-F (forward, 5'-CACACCATGGCCTGCACCACGATTCTG-3') and pepD-R (reverse, 5'-TGGGACTAGTCTAAAAGTCGGACATGTGGAAGC-3'), were designed based upon the B. longum NCC2705 genomic sequence (GenBank accession number NC_004307.2, gene BL0479), where the underlining indicates a translation start or stop codon. NcoI and SpeI restriction sites (shown in boldface type in the above sequences) were incorporated into pepD-F and pepD-R, respectively, in order to facilitate the cloning of pepD to the pBAD-TOPO derivative vector for subsequent screening. The pBAD-TOPO derivative was constructed by inserting the PCR product containing the SpeI site into the pBAD-TOPO vector (Invitrogen). Also, another primer, pepD-His-R (reverse, 5'-ATACACTAGTAAAGTCGGACATGTGGAAGC-3'; the SpeI restriction site shown in boldface type), was designed to amplify pepD for the expression of a C-terminal polyhistidine-tagged fusion protein. The pepD-His-R primer has no stop codon in order to continue the nucleotide sequence coding for the polyhistidine tag of the pBAD-TOPO derivative. The PCR product obtained from the pepD-F and pepD-R primer set was named pepD, and that obtained from the pepD-F and pepD-His-R primer set was named pepD-His, which is the C-terminal polyhistidine-tagged fusion protein. The additional C-terminal amino acid sequence deduced from the PepD-His-encoding gene was TSGPKGELEGKPIPNPLLGLDSTRTGHHHHHH, which is from amino acids 554 to 585 in the sequence of PepD-His (the underlining indicates the V5 epitope and polyhistidine tag region). The PCR was performed with 25 ng of DNA in a final volume of 50 µl containing deoxyribonucleoside triphosphate (each at 0.25 to 0.5 mM), oligonucleotides (50 pM), and 1.0 to 3.0 U of LA-Taq polymerase (Takara). Amplification was performed on a thermal cycler (PROGENE; Techne, Cambridge, United Kingdom) with 30 cycles of denaturation at 95°C for 30 s (5 min in the first cycle), annealing at 60°C for 60 s, and elongation at 72°C for 90 s (10 min in the last cycle).
General cloning techniques and sequence analysis.
The PCR products encoding PepD and PepD-His were digested with NcoI and SpeI and then ligated with the NcoI-SpeI-digested pBAD expression vector (Invitrogen) to generate pBAD-pepD and pBAD-pepD-His, respectively. The ligation mixture was used in E. coli DH5
heat shock transformation. The transformed cells were spread on LB agar containing 50 µg·ml–1 ampicillin. After incubation at 37°C overnight, the transformants were selected, and their recombinant plasmids were purified (plasmid miniprep kit; QIAGEN, Hilden, Germany). The sequence of the cloned pepD gene was then determined (BigDye Terminator and ABI377 system; Applied Biosystems).
Expression of two recombinant dipeptidases of PepD and PepD-His in E. coli.
The plasmids pBAD-pepD and pBAD-pepD-His (carrying pepD and pepD-His) proliferated in E. coli DH5
and were extracted, purified, and transformed again in E. coli LMG194, with the transformants screened on an agar plate containing 50 µg·ml–1 ampicillin. The transformant cells were cultured in LB broth containing 50 µg·ml–1 ampicillin, shaken overnight at 37°C, transferred into flasks containing 50 ml of fresh LB broth, and grown at 37°C with vigorous shaking. The pepD and pepD-His genes were expressed by induction with 0.02% L-arabinose, which was added to the culture broth of E. coli LMG194 harboring each plasmid at the beginning of the exponential phase (optical density at 600 nm of 0.5). Six hours after induction, in the case of E. coli LMG194 harboring pBAD-pepD-His, 50 ml of the culture solution was centrifuged at 10,000 x g for 10 min at 4°C, and the pelleted cells were resuspended in 5 ml of lysis buffer (50 mM Tris-HCl, 300 mM NaCl, and 10 mM imidazole, pH 8.0). In the case of E. coli LMG194 harboring pBAD-pepD, cell growth was remarkably retarded and hence was stopped 18 h after induction. The suspension was sonicated at least four times by a sonic dismembrator (Fisher Scientific, Pittsburgh, PA) using 30-s pulses that were applied at 1-min intervals. The lysate was clarified at 10,000 x g for 10 min at 4°C, and the soluble fraction was analyzed by SDS-PAGE.
Purification of PepD-His.
The recombinant PepD-His (rPepD-His) was purified using an immobilized nickel-nitrilotriacetic acid (Ni-NTA) affinity column. The soluble fraction containing the recombinant protein was loaded onto the Ni-NTA column that had previously been equilibrated in the same lysis buffer, and the column was washed twice with five times the bed volume of 50 mM Tris-HCl, 300 mM NaCl, and 20 mM imidazole (pH 8.0). The recombinant protein was then eluted using elution buffer (50 mM Tris-HCl, 300 mM NaCl, and 250 mM imidazole; pH 8.0), and the eluted protein was analyzed by SDS-PAGE.
MALDI-TOF MS.
Matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry (MS) was performed (Voyager-DE STR biospectrometry workstation; Applied Biosystems) using ions formed with a N2 pulsed laser beam (337 nm) accelerated to 25 kV.
|
|
|---|
|
View this table: [in a new window] |
TABLE 2. Purification of the dipeptidase from B. longum BORI
|
![]() View larger version (48K): [in a new window] |
FIG. 1. Purification of B. longum BORI dipeptidase and expression of recombinant PepD and PepD-His in E. coli. (A) SDS-PAGE of the enzyme obtained from the final Q-Sepharose chromatography step. (B) SDS-PAGE of the induced rPepD in E. coli LMG194 harboring pBAD-pepD. Lanes: M, molecular mass standards; 1, disrupted cell extract containing rPepD. (C) SDS-PAGE of the induced rPepD-His in E. coli LMG194 harboring pBAD-pepD-His. Lanes: M, molecular mass standards; 1, purified rPepD-His from Ni-NTA affinity chromatography; 2, disrupted cell extract containing rPepD-His.
|
![]() View larger version (13K): [in a new window] |
FIG. 2. Effects of environmental conditions on the activity of the dipeptidase purified from B. longum BORI determined on Ala-Gln. (A) pH dependence of enzyme activity as measured at 37°C. (B) Residual activity of the enzyme after 30 min of preincubation at various pH values as measured at 37°C for 30 min. (C) Effects of temperature on the activity of the enzyme at pH 8. (D) Residual activity of the enzyme after 10 min of preincubation at various temperatures as measured at 37°C for 30 min.
|
|
View this table: [in a new window] |
TABLE 3. Substrate specificity of the dipeptidase purified from B. longum BORI
|
![]() View larger version (63K): [in a new window] |
FIG. 3. Nucleotide sequences of 1.6-kb fragments and deduced amino acid sequence of pepD from B. longum BORI. The N-terminal conserved sequence (amino acids 3 to 27) of the U34 family are double underlined, and the catalytic cysteine and the Arg and Asp essential residues are boxed. The determined internal amino acid sequences of the purified enzyme and rPepD-His were identical at the boldface and underlined letters (amino acids 231 to 239 and 395 to 405 of the sequence deduced from pepD, respectively). The nucleotide sequence is available in the NCBI and GenBank databases under accession number NC_004307.2.
|
The pepD and pepD-His genes were subcloned into E. coli DH5
by using the pBAD-TOPO derivative vector as described in Materials and Methods. The two new plasmids containing the pepD and pepD-His genes, designated pBAD-pepD and pBAD-pepD-His, respectively, were sequenced to ensure that the PCR had not introduced any errors.
pepD was successfully expressed in E. coli LMG194 (Fig. 1B). The rPepD and rPepD-His proteins were expressed by induction with 0.02% L-arabinose. Further, rPepD-His was purified using a Ni-chelating resin column as described in Materials and Methods (Fig. 1C). Both rPepD and rPepD-His exhibited no dipeptidase activity, even though they were overexpressed in E. coli. SDS-PAGE revealed that rPepD and rPepD-His expressed in E. coli both had a molecular mass of about 80 kDa, which was not as expected from the theoretical molecular mass deduced from the pepD sequence. MALDI-TOF MS was used to determine whether this unexpected result was simply due to an unusual conformation or to the formation of a complex with another protein(s) when the protein was expressed in E. coli.
MALDI-TOF MS of the native dipeptidase purified from B. longum BORI produced a mass spectrum with a clear peak at 49,496 Da, as shown in Fig. 4A. This result was consistent with the molecular mass determined from SDS-PAGE (Fig. 1A). MALDI-TOF MS of purified dipeptidase (PepD-His) including the V5 epitope and polyhistidine residues also resulted in a characteristic mass spectrum, as shown in Fig. 4B, where the single fragment at a mass-to-charge ratio (m/z) of 62,932 corresponds to PepD-His. Actually, 63,191 is the theoretical m/z value deduced from the pepD-His gene sequence, which additionally encodes the V5 epitope and polyhistidine tag region. This discrepancy between the molecular masses is probably attributable to the MALDI-TOF MS procedures; hence, we consider that the amino acid sequence analysis and MALDI-TOF MS data were consistent. However, this result was not consistent with the molecular mass (about 80 kDa) determined from SDS-PAGE (Fig. 1C), which indicates that the PepD protein has an unusual conformation. We were also unable to determine the N-terminal sequence of the purified rPepD-His as with the purified native enzyme. The internal amino acid sequence of rPepD-His was determined to be VTSENFYWANR, which is identical to the sequence between amino acids 395 to 405 deduced from pepD (Fig. 3). Another dipeptidase gene, pepDB, had no region corresponding to the internal amino acid sequence determined from the purified rPepD-His (data not shown).
![]() View larger version (18K): [in a new window] |
FIG. 4. MALDI-TOF MS spectra of native dipeptidase purified from B. longum BORI (A) and His-tagged dipeptidase expressed in E. coli (B).
|
Sequence analysis of PepD from B. longum BORI.
The amino acid sequences deduced from the open reading frame were homologous to dipeptidases from various microorganisms, such as B. adolescentis (83% identity, gi 119025186), Oenococcus oeni PSU-1 (54% identity, gi 116491743), B. longum NCC2705 (52% identity, gi 23465037; Table 4, row S), Lactobacillus casei ATCC 334 (50% identity, gi 116493607), Lactococcus lactis subsp. cremoris SK11 (45% identity, gi 116512355), and Lactobacillus helveticus CNRZ32 (44% identity, gi 18202547). The sequence analysis also revealed that the G+C content of the pepD gene was 63.7%.
|
View this table: [in a new window] |
TABLE 4. Multiple-sequence alignment of the first ß-hairpin motif in the PVA/U34 family and other Ntn hydrolases
|
On the other hand, the analysis using the PSI-BLAST program indicated that only amino acids 1 to 428 of the total 533 amino acids in the PepD sequence were conserved in the U34 peptidase family (data not shown). The theoretical molecular mass of these 428 amino acids was 48.3 kDa, which was similar to that of the native peptidase purified from B. longum BORI (Fig. 1A). Additionally, MALDI-TOF MS revealed that native pepD purified from B. longum BORI had a molecular mass of 49 kDa. These data suggest that the native dipeptidase is subject to a posttranslational modification involving protein splicing, unlike recombinant dipeptidases (rPepD and rPepD-His) expressed in E. coli.
|
|
|---|
In this study, a dipeptidase from the cell extract of B. longum BORI was purified about 109-fold with an activity yield of 26.8%. The dipeptidase was found to be a monomer with a molecular mass of about 49 kDa. Almost no activity was detected below pH 5.0, but the enzyme showed high activity in alkaline conditions. The activity was about 50% of the maximum at 70°C but was very low at 80°C. Also, the dipeptidase retained 58.3% of its activity after 10 min of preincubation at 80°C. The enzyme was heat stable, unlike a dipeptidase from Lactobacillus curvatus DPC2024 (21). The sequence of the first 20 amino acids of the dipeptidase from L. curvatus DPC2024 was the same as that of a dipeptidase from Lactobacillus sakei subsp. sakei 23K, which is a member of the M20 family of peptidases (Table 4, row R). On the other hand, N-terminal amino acid sequencing of both the purified native dipeptidase from B. longum BORI and the purified rPepD-His in E. coli was not successful, presumably because of a block, such as N-terminal posttranslational modification. We suggest that the initiator methionine of PepD was hydrolyzed, and N acetylation of the terminal alanine residue to yield acetyl-Ala-Cys-Thr-Thr-protein occurred similarly to that in cellular polypeptide Op18 (14).
Application of the PSI-BLAST program revealed that the PepD sequence was highly homologous to the sequences of the U34 family of peptidases. Also, the N-terminal conserved domain of U34 family members is highly homologous to that of penicillin V acylase (PVA) (Table 4) (23). Proteins of the PVA/U34 family are members of a large superfamily of amidohydrolases called the N-terminal nucleophile (Ntn) hydrolases, which exhibit variable sequences and great diversity in substrate specificity. They include enzymes such as choloylglycine hydrolases, acid ceramidases, and isopenicillin N acyltransferases and a subgroup of eukaryotic proteins with unclear function (16). The most representative sequence for all PVA/U34 family proteins resides in the motif containing the catalytic cysteine residue and corresponds to the N-terminal ß-hairpin motif in the structure of Bacillus sphaericus PVA (Table 4, row J) (23, 29). Other common features of this motif include the hydrophobic pattern and positions occupied mainly by small residues near the catalytic cysteine. The ß-hairpin motif is longer in close homologs of U34 family dipeptidases than in close homologs of PVAs (Table 4). Two other residues (Arg 22 and Asp 25) are also highly conserved in most PVA/U34 proteins. The cysteine side chain serves as the nucleophile, and the free
NH2 serves as the proton donor and acceptor in the catalytic process. PVA/U34 proteins have been designated as cysteine-type Ntn hydrolases (23). These previous studies suggest that the PepD protein with the same catalytic cysteine also has an N-terminal ß-hairpin structure in B. longum BORI and thus potentially adopts the Ntn hydrolase fold (Table 4, row A).
On the other hand, crystallization of PVA from B. sphaericus showed that the mature form of PVA was created by autoproteolytic processing originating from a nucleophile such as the catalytic cysteine residue (6). In addition, various Ntn hydrolases catalyze internal peptide bond rearrangements through an N
O or N
S acyl shift by a Ntn amino acid created by autoproteolytic processing (24, 25). These previous studies support that the native dipeptidase resulted from protein splicing such as autoproteolytic processing in B. longum BORI, since otherwise the molecular mass of rPepD-His expressed in E. coli would have been equal to the theoretical mass deduced from the pepD sequence. Hence, other factors might be involved in protein splicing in B. longum BORI, unlike the case in E. coli.
Taken together, these results suggest that the dipeptidase from B. longum BORI is a cysteine-type Ntn hydrolase and has a ß-hairpin motif similar to that of PVA, which results in it being processed by protein splicing. Currently, the C-type U34 dipeptidases such as this PepD from B. longum BORI are assigned to the C69 family of peptidases.
PepD from B. longum BORI is the first of the U34 dipeptidases to be purified and characterized and to be potentially processed by protein splicing. Future studies should attempt to elucidate whether other factors are related to protein splicing and whether this enzyme can hydrolyze other substrates such as PVA.
We also thank the Korea Basic Science Institute and the Seoul National University National Center of Interuniversity Research Facilities for performing internal amino acid sequencing and MALDI-TOF MS analyses, respectively.
Published ahead of print on 29 June 2007. ![]()
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»