Changes in intestinal microflora of Caenorhabditis elegans following Bacillus nematocida B16 infection

Abstract

The effect of pathogenic bacteria on a host and its symbiotic microbiota is vital and widespread in the biotic world. The soil-dwelling opportunistic bacterium Bacillus nematocida B16 uses a "Trojan horse" mechanism to kill Caenorhabditis elegans. The alterations in the intestinal microflora that occur after B16 infection remain unknown. Here, we analyzed the intestinal bacteria presented in normal and infected worms. The gut microbial community experienced a complex change after B16 inoculation, as determined through marked differences in species diversity, structure, distribution and composition between uninfected and infected worms. Regardless of the worm's origin (i.e., from soil or rotten fruits), the diversity of the intestinal microbiome decreased after infection. Firmicutes increased sharply, whereas Proteobacteria, Actinobacteria, Cyanobacteria and Acidobacteria decreased to different degrees. Fusobacteria was only present 12 h post-infection. After 24 h of infection, 1228 and 1109 bacterial species were identified in the uninfected and infected groups, respectively. The shared species reached 21.97%. The infected group had a greater number of Bacillus species but a smaller number of Pediococcus, Halomonas, Escherichia and Shewanella species (P < 0.01). Therefore, this study provides the first evaluation of the alterations caused by pathogenic bacteria on symbiotic microbiota using C. elegans as the model species.

Figure 1. Structural analysis of C. elegans intestines and detailed pictures of endophytic bacteria obtained via TEM.

Photographs of C. elegans originating from soil. (A) Light microscopy showed that the worms were alive and that their intestines were normal after feeding on E. coli for 48 h; (B) Light microscopy showed that the worms were dead and their intestines were disorganized after feeding on B. nematocida strain B16g-1 for 4 h prior to transfer to E. coli plates for 24 h; (C) Cross-section of uninfected worms (4000-fold amplification); (D) Cross-section of 24-h-infected worms (4000-fold amplification). Symbiotic microbiota are indicated with white arrows in (C,D).

Figure 2. Venn diagram comparing the uninfected and 24-h-infected groups.

(A) Data for worms from soil; (B) Data for worms from rotten fruits.

Figure 3. Statistical results of the comparison of the significant microflora of soil worms between the uninfected and 24-h-infected groups.

(A) At the phylum level, Actinobacteria, Cyanobacteria and Planctomycetes differed significantly between the two samples. (B) At the genus level, Bacillus, Pediococcus, Halomonas, Shewanella and Escherichia differed significantly between the two groups. The vertical axis represents the sequence reads. (C) Percentage of phylum and genus levels. ^aP < 0.05.

Figure 4

(A) Pie charts of the bacterial distribution at the phylum level in normal and infected worms from soil. a. Uninfected controls; b. 12 h after infection and shifting from B16 to OP50; c. 24 h after infection and shifting from B16 to OP50. K and P represent the kingdom and phylum, respectively. (B) Changes in the major bacterial populations of normal and infected worms from soil at the phylum level. (C) Column diagram of the bacterial distribution at the phylum level in normal and infected worms from rotten fruit.

Figure 5. Comparison of species abundance between the normal and 24-h-infected worms originating from soil at the genus level.

Figure 6. Changes in microbial diversity in the guts of normal and 24-h-infected worms from soil.

Six samples were analyzed, and the clustering of the microbial communities was performed using the PCoA of the weighted UniFrac matrix. The percentage of variation indicated by the principal coordinates is shown on the axes. Color coding: blue, samples from the 24-h-infected group; red, samples from the uninfected control. PCoA: Principal co-ordinate analysis.