Molecular Diversity of Lactobacillus spp. and Other Lactic Acid Bacteria in the Human Intestine as Determined by Specific Amplification of 16S Ribosomal DNA

Bacterial strains and growth conditions.Reference microorganisms, their sources, and the growth media for each strain used in this study are indicated in Table 1. Escherichia coli transformants were grown overnight with shaking at 250 rpm in Luria-Bertani broth (35). Lactobacillus strains were grown in tubes containing 10 ml of Lactobacillus MRS broth (Difco Laboratories, Detroit, Mich.) and 0.3 g of cysteine hydrochloride liter⁻¹. The remaining strains were grown in either Luria-Bertani or Wilkins West broth that contains (per liter) 33 g of Wilkins-Chalgren anaerobe broth (Oxoid, Basingstoke, United Kingdom), 2 g of glucose, 4 g of arginine HCl, 5 ml of 5% Tween 80, and 0.3 g of cysteine hydrochloride, and the pH was adjusted to 7.0 with HCl. All bacteria were incubated at 37°C. The strict anaerobes, i.e., the Clostridium, Bacteroides, and Peptostreptococcus species, were handled and cultured in an anaerobic glove box with a constant atmosphere of 96% N₂ and 4% H₂ and incubated in an anaerobic jar by using Anaerocult A (Merck Microbiologie, Darmstadt, Germany) to ensure an oxygen-free environment. Prior to inoculation of the remaining species, the tubes containing media were boiled for 20 min and cooled to room temperature to remove oxygen from the medium. After inoculation, which was performed on the bench, tubes were rapidly sealed with a paraffin plug to ensure anoxic conditions.

Recovery, preparation, and storage of fecal and cecal samples.Fecal samples were provided by healthy volunteers of both sexes, and of different ages, in spatula containers and were stored at −20°C until further use. Informed consent was obtained from the parents of children. One healthy full-term baby boy, vaginally delivered and breast-fed, participated in this study. The infant was monitored over time and was breast-fed until day 17, received the first formula food after 2 weeks, and then received progressively more formula. Solid food was given after 3 months. Besides these non-trial-affiliated volunteers, fecal samples were provided by individuals that were involved in a trial wherein subjects were fed with fermented products containing the probiotic Lactobacillus paracasei F19 (15, 24). Upon use samples were thawed on ice-water and mixed, after which they were homogenized in 0.05 M ice-cold potassium phosphate buffer (pH 7.0) at a ratio of 3 g of feces (wet weight) per 50 ml (49). Cecal samples from healthy adults were collected as described previously (8, 22, 23).

DNA isolation.Bacterial DNA from fecal and cecal samples was isolated according to the method of Zoetendal et al. (49). DNA from 5-ml bacterial cultures was isolated by centrifugation of the cells (5 min, 1,500 × g); the cells were then resuspended in 1 ml of Tris-EDTA buffer (pH 8.0) (35). The cell suspension was further treated as described above by bead beating, phenol-chloroform extraction, and ethanol precipitation (49).

Primer design and PCR conditions.All primers used in this study are listed in Table 2. 16S rDNA sequences of phylogenetically related species were retrieved from GenBank (www.ncbi.nlm.nih.gov ) and used to perform multiple alignments by using CLUSTAL W (46). A potential target site starting at position 677 was selected and, based on this site, a 17-mer PCR primer was designed and designated S-G-Lab-0677-a-A-17 (Lab-0677r) according to the OPD nomenclature (1). A second primer, S-G-Lab-0159-a-S-20 (Lab-0159f), was designed based on another conserved site starting at position 158 targeting Lactobacillus and Enterococcus genera (11). The specificity of the primers was screened by submitting the sequence to the Check Probe program of Ribosomal Database Project (RDP; www.cme.msu.edu/RDP ) (21) and the ARB software package (O. Strunck and W. Ludwig [http://www.mikro.biologie.tu-muenchen.de/pub/ARB/documentation/arb.ps ]). The genomic DNA of relevant reference strains and predominant GI-tract species was used as the template in PCR experiments to validate the Lab-0677r and Lab-0159f primer specificities (Table 1).

Primers were synthesized commercially by MWG Biotech AG, Ebersberg, Germany. Bacterial primer S-D-Bact-0011-a-S-17 (Bact-0011f), formerly named 27f (18), was paired with the designed reverse primer Lab-0677r, while primer Lab-0159f was paired with the universal reverse primer S-*-Univ-0515-b-A-24 with a GC clamp (Uni-0515-GCr). The GC clamp attached at the 5′ end of one of the primers creates products suitable for separation by DGGE and/or TGGE. PCR was performed employing the Taq polymerase kit from Life Technologies (Gaithersburg, Md.). PCR mixtures of 50 μl contained 20 mM Tris-HCl (pH 8.4), 50 mM KCl, 3 mM MgCl₂, 50 mM concentrations of each deoxynucleoside triphosphate, 1.25 U of Taq polymerase, 10 pmol of each primer, and ca. 250 ng of genomic DNA isolated from pure cultures. Samples were amplified in a PE Applied Biosytems GenAmp PCR system 9700 (Foster City, Calif.) by using the following program: predenaturation at 94°C for 5 min; 35 cycles of denaturation 94°C for 30 s, variable annealing temperature for 20 s, and extension at 68°C for 40 s; and a final extension at 68°C for 7 min.

PCR to investigate the general or Lactobacillus-specific GI-tract bacterial community by DGGE was performed with the following 16S rDNA primer combinations: (i) S-D-Bact-0124-a-S-27 with a GC clamp (Bact-0124-GCf) and Univ-515r or (ii) Lab-0159f and Uni-0515-GCr. Nested PCR was performed with these primers on previously generated products from amplification with Bact-0011f and Lab-0677r. The cycling program consisted of 94°C for 5 min; 35 cycles of 94°C for 30 s, 56°C for 20 s, and 68°C for 40 min; and finally 7 min at 68°C. PCR products that were used as templates in nested PCRs were purified with the Qiaquick PCR purification kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions.

DGGE analysis of PCR amplicons.PCR products generated with primers Bact-0124f-GC and Univ-0515r or primers Lab-0159f and Univ-0515r-GC were separated by DGGE according to the specifications of Muyzer et al. (29) by using the Dcode system (Bio-Rad Laboratories, Hercules, Calif.) with the following modifications. Polyacrylamide gels (dimensions, 200 by 200 by 1 mm) consisted of 8% (vol/vol) polyacrylamide (37.5:1 acrylamide-bisacrylamide) and 0.5× TAE (35). Denaturing acrylamide of 100% was defined as 7 M urea and 40% formamide. The gels were poured from the top by using a gradient maker and a pump (Econopump; Bio-Rad Laboratories, Hercules, Calif.) set at a speed of 4.5 ml/min, and gradients of 30 to 60% were used for the separation of the generated amplicons. Before polymerization of the denaturing gel (28-ml gradient volume), a 7.5-ml stacking gel without denaturing chemicals was added, and the appropriate comb was subsequently inserted. Electrophoresis was performed for 16 h at 85 V in a 0.5× TAE buffer at a constant temperature of 60°C. Gels were stained with AgNO₃ according to the method of Sanguinetti et al. (36).

Cloning of the PCR-amplified products.PCR amplicons were purified with the Qiaquick PCR purification kit according to the manufacturer’s instructions. PCR products were cloned into E. coli JM109 by using the Promega pGEM-T vector system (Promega, Madison, Wis.). PCR was performed on cell lysates of ampicillin-resistant transformants by using pGEM-T specific primers T7 and Sp6 to confirm the size of the inserts. To establish the diversity within the group of 25 selected clones of each origin, amplicons of the correct size were subjected to restriction fragment length polymorphism analysis by using the restriction enzyme MspI. Plasmids containing a unique insert of the appropriate size or corresponding to a band in the community fingerprint were purified by the QIAprep spin miniprep kit (Qiagen, Hilden, Germany) and were subjected to DNA sequence analysis.

Sequence analysis.One microgram of purified pGEM plasmid was used for sequence analysis of the cloned 16S rDNA fragments. Sequencing reactions were performed with the Sequenase (T7) sequencing kit (Amersham Life Sciences, Slough, United Kingdom) according to the manufacturer’s specifications by using the IRD-800 5′-prime end-labeled primers T7 and Sp6. Sequences were automatically analyzed on a LI-COR DNA sequencer 4000L (Lincoln, Nebr.) and corrected manually.

Phylogenetic placement.Phylogenetic analysis was performed by using the ARB software package. Sequences were aligned and a rooted neighbor-joining tree (E. coli positions 31 to 648) was constructed by using Bacillus subtilis as an outgroup species.

Nucleotide sequence accession numbers.Thirty-nine 0.7-kb and nine 0.4-kb sequences of the 16S rDNA determined in this study were deposited with the GenBank database under accession numbers AF335874 to AF335918 and AF368382 to AF368390 , respectively.

Development and evaluation of a Lactobacillus group-specific primer.Based on the alignment of the complete 16S rRNA sequences of various members of the three recognized Lactobacillus subgroups (4) and those of several other related lactic acid bacteria, a potential PCR primer binding site was identified starting at E. coli position 677 of the 16S rDNA (Table 3). Primer S-G-Lab-0677-a-A-17, complementary to this part of the 16S rRNA, was designed taking into consideration (i) the nucleotide mismatches with related species, (ii) the complementarity with the target at the 3′ end, possibly ending with a C; (iii) a G+C content of >50%, and (iv) a size of ca. 20 nucleotides. Sequence comparison by using the Check Probe, ARB, and BLAST programs confirmed that the targeted region was conserved only among all of the 16S rRNA sequences of Lactobacillus, Leuconostoc, Pediococcus, Weissella, and Aerococcus species present in the Ribosomal Database Project and GenBank, except for L. maltaromicus, L. vitulinus, L. catenaformis, and L. helveticus. The first three of the latter lactobacilli are more related to Clostridium and Carnobacterium spp. than to the genus Lactobacillus (38). Their exclusion may have only limited consequences since these particular Lactobacillus spp., as well as members of the Pediococcus, Weissella, and Aerococcus species, are generally not indigenous to the human GI tract. Moreover, specific PCR products were obtained with DNA from two strains of L. helveticus, suggesting that the mismatches found in the database with the primer are most likely the result of sequencing errors (see below). The terminal 3′ end of the primer was also complementary to Eubacterium biforme and Eubacterium cylindroides 16S rDNA sequences. Most importantly, the primer had one mismatch at the 3′ end with 16S rDNA sequences of several other related species, including some bacteria commonly found in large numbers in the GI tract, such as Clostridium, Eubacterium, Bacteroides, and Bifidobacterium spp., that should prevent annealing (Table 3).

The primer was experimentally tested by performing PCR on genomic DNA isolated from a range of Lactobacillus species, related lactic acid and other bacteria, as well as cloned 16S rDNA sequences that have been demonstrated to be present in large numbers in the GI tract (49). The optimal annealing temperature was empirically determined by raising it in steps of 2°C from 56 to 68°C, by using genomic DNA of 27 bacterial species and plasmid DNA from nine clones (Table 1). The optimum was found to be 66°C and, after a maximum of 35 cycles, products were obtained with DNA from all tested Lactobacillus, Leuconostoc, and Weissella spp., whereas the 16S rDNA of all other species was not amplified (Table 1).

Biodiversity of the Lactobacillus genus evaluated by Lab-0677r.The diversity of members of the Lactobacillus genus was investigated by using primer S-G-Lab-0677-a-A-17 in combination with primer Bact-0011f on bacterial DNA isolated from fecal and other intestinal samples that generated 0.7-kb PCR products. Sequence analysis of unique clones of fecal origin from five healthy adults and an infant, as well as samples taken from the cecal chyme during intestinal intubation in four healthy volunteers, resulted in sequences with significant identity (97 to 99%) to the 16S rDNA of known Lactobacillus spp. (Table 4). Among all 39 sequences retrieved, L. ruminis-like sequences were the most frequently encountered in both fecal and cecal samples. Leuconostoc-like sequences that showed similarity to Leuconostoc mesenteroides and Leuconostoc argentinum were also found in these samples. Eubacterium biforme-like sequences were found in the feces of three adults (Table 4). Phylogenetic analysis clearly showed that sequences were retrieved from all three subgroups of the genus Lactobacillus (Fig. 1). Most representatives were positioned in the Lactobacillus casei-Pediococcus subgroup (I), with the majority being most similar to L. ruminis. Sequences retrieved from the infant feces grouped with L. salivarius, L. rhamnosus, and L. paracasei. Some sequences grouped within the L. delbrueckii subgroup (II), while the Leuconostoc paramesenteroides subgroup (III) was represented by sequences that were similar to Leuconostoc mesenteroides and Leuconostoc argentinum. Overall, the sequence results confirmed the validity of the developed primer.

• Open in new tab
• Download powerpoint

FIG. 1.

Phylogenetic tree based upon the neighbor-joining method of partial 16S rDNA sequences (E. coli positions 31 to 648) from clones derived by PCR with the primer Bact-0011f and the specific primer Lab-0677r. Suffixes: F, feces; C, cecal chyme; B, baby. Reference sequences are included which were found to be the closest relatives of the retrieved clones. Three subgroups are indicated: I, L. casei-Pediococcus group; II, L. delbrueckii group; III, Leuconostoc paramesenteroides group. B. subtilis is used as an outgroup. The scale bar represents (calculated) distance. The origin of the clones is presented in Table 4.

DGGE analysis of Lactobacillus amplicons.To facilitate rapid monitoring of Lactobacillus-specific sequences, use was made of sequence-specific separation of PCR amplicons by DGGE that requires having the GC clamp attached to one of the primers. However, a low product yield was observed when either of the primers, Bact-0011f or Lab-0677r, was equipped with a GC clamp, and therefore we opted for the use of a nested-PCR approach. This involved a first round of PCR with primers Bact-0011f and Lab-0677r amplifying the Lactobacillus-like community, followed by a second PCR with established GC-containing universal primers. To prevent amplification of residual genomic DNA or low-yield aspecific amplicons formed in the first PCR, only samples yielding sufficient product (15 ng/μl) were used for a second round of PCR with primers Bact-0124-GCf and Uni-0515r.

Succession of Lactobacillus spp. after birth.The intestinal microbial community of infants is developing and therefore highly unstable (20; C. Favier, unpublished data). A study of the development of the Lactobacillus-like community in babies was performed by PCR with primers Bact-0011f and Lab-0677r on fecal DNA from baby F from delivery (day 1) to 5 months later (day 147) at regular intervals. No PCR products were obtained up to day 55, indicating that the template was absent or that the amount of template was too low to be detected. The PCR amplicons obtained from days 55 to 147 were used for the nested PCR with primers Bact-0124-GCf and Uni-0515r, and the resulting 0.4-kb fragments were separated in a 30 to 60% DGGE gel (Fig. 2). Two amplicons with different intensity were present from day 55 throughout the experiment, and the sequences of these 0.4-kb fragments were identical to clones earlier retrieved from the infant, i.e., B103 (L. rhamnosus) and B121 (L. casei) (Table 4). Successional change was observed with amplicon 3, which was identical to clone B123 (L. salivarius), that appeared on day 93 and faded between days 129 and 147.

• Open in new tab
• Download powerpoint

FIG. 2.

Development of the Lactobacillus-like community over time in an infant. DGGE analysis of amplicons generated by nested PCR from fecal samples taken on day 55 up to day 147. Fragments that are indicated by arrows and numbers were identified by the Lab-clone library of the infant as described in Table 4. The origin of the fragments and the corresponding clone are indicated as follows: 1, L. rhamnosus (B103); 2, L. casei (B121); and 3, L. salivarius (B123).

Effect of consumption of lactobacilli on the Lactobacillus sp. diversity in children.The fate of orally administered L. paracasei F19 in a clinical trial with children was investigated with primer Lab-0677r by using nested PCR and DGGE. Fecal samples from children between 10 and 18 months of age that were given 10¹⁰ CFU of L. paracasei F19 twice a day for 3 weeks or a placebo for the same time period were analyzed in a blind fashion. Although L. paracasei F19 was not detected in the dominant bacterial community of these young children (47), it could be detected as a double band when the Lab-0677-specific PCR-DGGE approach was used (Fig. 3). This analysis also revealed that in some children the specific community is also unstable. This instability of the Lactobacillus-like community was particularly evident in two individuals and appeared independent of the L. paracasei F19 administration. In individual I the community was changing throughout the experiment, but despite this change strain F19 could be detected during the administration period. In individual III, the double band of strain F19 or a Lactobacillus species with the same melting behavior was present even before the trial started and was enhanced during the experiment. However, after the cessation of placebo-administration (Fig. 3, individual III) it disappeared. In a subject where the Lactobacillus-like community appeared to be stable (Fig. 3, individual II), the administration of L. paracasei F19 had a temporary impact on the composition but it returned to its original composition after cessation of the strain administration.

• Open in new tab
• Download powerpoint

FIG. 3.

Effect of orally administered L. paracasei F19 on the Lactobacillus-like community over time. DGGE analysis of amplicons generated in a nested PCR with primers Bact-0124-GCf and Uni-0515r. Children were administered L. paracasei F19 (I and II) or placebo (III and IV) for 4 weeks. Fecal samples were taken before the experiment (lanes 0), after 2 weeks of administration (lanes 2), and 2 weeks after administration had ceased (lanes 6). The PCR amplicon for L. paracasei F19 was concomitantly separated. The fragments in the profiles corresponding to F19 are indicated by the box.

Lactobacillus-like community development in time and space in adults.While Lactobacillus and Leuconostoc sequences were readily obtained from adult fecal samples, analysis of the amplicons obtained by nested PCR revealed an overrepresentation of Eubacterium biforme sequences and only small amounts derived from the Lactobacillus group (data not shown). Therefore, another more specific primer Lab-0159f, which targets the Lactobacillus and Enterococcus genera, was used in a nested-PCR strategy with Uni-0515-GCr. Although the Lab-0159f primer resulted in aspecific PCR products when used as a primary PCR primer on fecal samples (data not shown), it was very selective when used in the nested-PCR approach to reamplify amplicons obtained with Bact-0011f and Lab-0677r and circumvented reamplification of Eubacterium biforme-like sequences (Table 1). Comparison of the dominant bacterial community with the Lactobacillus-like community revealed a decreased diversity for the Lactobacillus-like sequences where relatively few bands were present (Fig. 4). Identification of the bands resulted in the retrieval of Lactobacillus- and Leucono stoc-like sequences with identities to known species ranging from 98 to 99%.

• Open in new tab
• Download powerpoint

FIG. 4.

DGGE analysis of the dominant (d) and Lactobacillus-like community (s) diversity in fecal and cecal samples from six adults (A to D, H, I), after PCR with primers Bact-0124-GCf and Uni-0515r and nested PCR with primers Lab-0159f and Uni-0515-GCr, respectively. The dominant fragments in the Lactobacillus-like patterns indicated by arrows and numbers were identified by clones from the existing clone library of Lactobacillus-like sequences, as described in Table 4. The origin of the fragments and the corresponding clones are indicated as follows: 1, L. sakei (F45); 2, L. ruminis (F68); 3, L. ruminis (F70); 4, L. delbrueckii (F93); 5, L. fermentum (F81); 6, L. ruminis (F1); 7, L. gasseri (S1); and 8, Leuconostoc mesenteroides (S27).

The fecal bacterial community of several healthy adults was monitored over a period of 20 months (Fig. 5). The DGGE patterns of the dominant bacterial community of each individual were specific and stable (data not shown), which is in agreement with previous results (50). The DGGE fingerprints of the dominant Lactobacillus-like community obtained with the nested PCR approach with the specific Lab-0159f primer revealed a variable stability of the Lactobacillus-like community among the different individuals (Fig. 5). The diversity of species increased in both individuals K and L, but individual M appeared to harbor a stable community of lactobacilli. PCR-DGGE is a semiquantitative method in which variation in band-intensities reflects the relative number of specific amplicons. For example, in individual K (Fig. 5) at 6 months the amount of sequences close to L. gasseri (fragment 1) decreased relative to that of the newly emerging sequences resembling L. paracasei (fragment 3). Identification of the dominant 0.4-kb fragments in the three adults resulted in sequences most similar (98 to 99%) to L. casei, L. paracasei, L. gasseri, L. sakei, and L. ruminis. Notably, in adults L and M sequences originating from L. ruminis-like sequences were dominantly present throughout the observed time frame.

• Open in new tab
• Download powerpoint

FIG. 5.

Monitoring of the Lactobacillus-like community of adults in time. DGGE analysis of amplicons generated by nested PCR with primers Lab-0159f and Uni-0515-GCr, originating from three individuals (K, L, and M) from whom fecal samples were obtained at 0, 6, and 20 months. The dominant fragments in Lactobacillus-like patterns indicated by arrows and numbers were sequenced and compared to known sequences in GenBank, as described in Table 4. The origin of the fragments and the corresponding clones (see Table 4) are indicated as follows: 1, L. gasseri (F706); 2, L. casei (F703); 3, L. paracasei (F723); 4, L. ruminis (F748); 5, L. ruminis (F754); 6, L. sakei (F749); 7, L. casei (F747); 8, L. ruminis (F761); and 9, L. ruminis (F763).