Oral Microbiota Shift after 12-Week Supplementation with Lactobacillus reuteri DSM 17938 and PTA 5289; A Randomized Control Trial

Abstract

Background: Lactobacillus spp. potentially contribute to health by modulating bacterial biofilm formation, but their effects on the overall oral microbiota remain unclear.

Methods and findings: Oral microbiota was characterized via 454-pyrosequencing of the 16S rDNA hypervariable region V3-V4 after 12 weeks of daily Lactobacillus reuteri DSM 17938 and PTA 5289 consumption. Forty-four adults were assigned to a test group (n = 22) that received lactobacilli lozenges (108 CFU of each strain/lozenge) or a control group that received placebo (n = 22). Presence of L. reuteri was confirmed by cultivation and species specific PCR. Tooth biofilm samples from 16 adults before, during, and after exposure were analyzed by pyrosequencing. A total of 1,310,292 sequences were quality filtered. After removing single reads, 257 species or phylotypes were identified at 98.5% identity in the Human Oral Microbiome Database. Firmicutes, Bacteroidetes, Fusobacteria, Proteobacteria, and Actinobacteria were the most abundant phyla. Streptococcus was the most common genus and the S. oralis/S. mitis/S. mitis bv2/S. infantis group comprised the dominant species. The number of observed species was unaffected by L. reuteri exposure. However, subjects who had consumed L. reuteri were clustered in a principal coordinates analysis relative to scattering at baseline, and multivariate modeling of pyrosequencing microbiota, and culture and PCR detected L. reuteri separated baseline from 12-week samples in test subjects. L. reuteri intake correlated with increased S. oralis/S. mitis/S. mitis bv2/S. infantis group and Campylobacter concisus, Granulicatella adiacens, Bergeyella sp. HOT322, Neisseria subflava, and SR1 [G-1] sp. HOT874 detection and reduced S. mutans, S. anginosus, N. mucosa, Fusobacterium periodicum, F. nucleatum ss vincentii, and Prevotella maculosa detection. This effect had disappeared 1 month after exposure was terminated.

Conclusions: L. reuteri consumption did not affect species richness but induced a shift in the oral microbiota composition. The biological relevance of this remains to be elucidated.

Trial registration: ClinicalTrials.gov NCT02311218.

Conflict of interest statement

Competing Interests: Test and placebo lozenges were obtained from the manufacturer BioGaia AB (Lund, Sweden). The supplier had no role in the study design or implementation, data analysis, decision to publish, or manuscript preparation. The authors declare no conflicts of interest.

Figures

Fig 1. Flow chart diagram.

Number of study participants and samples analyzed in the test and placebo groups at each sampling occasion.

Fig 2. Rarefaction curves of baseline operational taxonomic unit prevalence rates according to the number of reads.

Comparisons include all 15 subjects at baseline. Red dots indicate subjects in the placebo group; blue dots indicate subjects in the test group.

Fig 3. Relative abundances of 9 identified bacterial phyla in 45 plaque samples.

Sequences were matched to Human Oral Microbiome Database using Quantitative Insights into Microbial Ecology (QIIME) for phylum-level taxonomic identification. Data are presented as stacked bars and phyla are ordered by decreasing abundance and stratified by study group and sample occasion (i.e., baseline, 12-week exposure, and 1-month follow-up).

Fig 4. Rarefaction curves of operational taxonomic unit prevalence rates according to the number of reads in the test and placebo groups.

Data are presented as means with standard errors. Comparisons include 8 subjects in the placebo-treated group by sampling occasion (red symbols) and 7 subjects in the test group by sampling occasion (blue symbols). Differences within or between the groups were not statistically significant.

Fig 5. PCoA clustering analysis of baseline and 12-week samples in the placebo group.

Red dots indicate baseline samples and blue dots 12 week samples.

Fig 6. PCoA clustering analysis of baseline and 12-week samples in the test group.

Green dots indicate baseline samples and yellow dots 12 week samples.

Fig 7. Partial least-squares analysis (PLS) of microbiota associated with 12 weeks consumption of L. reuteri.

(A) Scatter-loading plot illustrating clustering of baseline versus 12 week treatment samples from the test group; The scores t1 and t2 are the new PCA created variables summarizing the x-variables. The red dots indicate placebo samples and blue dots test samples. The oval circle illustrates the tolerance ellipse based on Hotelling´s of T2, any observation located outside of the elipse would be an outlier. (B) Loading plot illustrating taxa associated with baseline versus 12 week treatment in the test group. The PLS model employed test and placebo group allocation as y and pyrosequencing taxa, L. reuteri and S. mutans by cultivation and PCR as the x-block. Red triangles indicate taxa associated with the baseline microbiota in the test group and blue triangles those associated with the microbiota after 12 week treatment in the same group. Black symbols indicate variables that were non-influential in the projection.

Fig 8. PCoA clustering analysis of 12-week and 1 month follow-up samples in the test group.

Yellow dots indicate baseline samples and green dots 1 month follow-up samples.