Search

Background Chronic immune-mediated diseases are rapidly expanding and notoriously difficult to cure. Altered relatively stable intestinal microbiota configurations are associated with several of these diseases, and with a possible pre-disease condition (more susceptible to disease development) of the host-microbiota ecosystem. These observations are reminiscent of the behavior of an ecosystem with alternative stable states (different stable configurations that can exist under identical external conditions), and we recently postulated that health, pre-disease and disease represent such alternative states. Here, our aim was to examine if alternative stable states indeed exist in the intestinal ecosystem. Results Rats were exposed to varying concentrations of DSS in order to create a wide range of mildly inflammatory conditions, in a context of diet-induced low microbiota diversity. The consequences for the intestinal microbiota were traced by 16S rRNA gene profiling over time, and inflammation of the distal colon was evaluated at sacrifice, 45 days after the last DSS treatment. The results provide the first formal experimental proof for the existence of alternative stable states in the rat intestinal ecosystem, taking both microbiota and host inflammatory status into consideration. The alternative states are host-microbiota ecosystem states rather than independent and dissociated microbiota and host states, and inflammation can prompt stable state-transition. Based on these results, we propose a conceptual model providing new insights in the interplay between host inflammatory status and microbiota status. These new insights call for innovative therapeutic strategies to cure (pre-)disease. Conclusions We provide proof of concept showing the existence of alternative stable states in the rat intestinal ecosystem. We further propose a model which, if validated in humans, will support innovative diagnosis, therapeutic strategy, and monitoring in the treatment of chronic inflammatory conditions. This model provides a strong rationale for the application of combinatorial therapeutic strategies, targeting host and microbiota rather than only one of the two in chronic immune-mediated diseases. Video Abstract Supplementary information Supplementary information accompanies this paper at 10.1186/s40168-020-00933-7.

Simple Summary Antibiotics perturb the gastrointestinal microbiota by killing bacteria beneficial for animal health and favoring the emergence of potential pathogens. Furthermore, antibiotics favor the emergence of resistant bacteria. Current knowledge on animals’ intestinal microbiota and effects of antibiotics is blurred by the various posology, administration routes, and implemented methodologies for its analysis. We summarized 71 studies analyzing the administration of antibiotics by different routes, conducted on the main food-producing and companion animals, highlighting differences in the methodology applied for the intestinal microbiota and antibiotic resistance analysis. Overall, therapeutic dosage decreased bacterial species diversity and richness in the microbiota and selected antibiotic resistance genes. For non-therapeutic dosage, information on the selection of antibiotic resistance was scarce and the effect on the intestinal microbiota scattered. Understanding the gut microbiota composition and function in animals could open up strategies for its modulation to improve animal health and performance, and to minimize the negative impact of antibiotics. Abstract Antibiotics are major disruptors of the gastrointestinal microbiota, depleting bacterial species beneficial for the host health and favoring the emergence of potential pathogens. Furthermore, the intestine is a reactor of antibiotic resistance emergence, and the presence of antibiotics exacerbates the selection of resistant bacteria that can disseminate in the environment and propagate to further hosts. We reviewed studies analyzing the effect of antibiotics on the intestinal microbiota and antibiotic resistance conducted on animals, focusing on the main food-producing and companion animals. Irrespective of antibiotic classes and animal hosts, therapeutic dosage decreased species diversity and richness favoring the bloom of potential enteropathogens and the selection of antibiotic resistance. These negative effects of antibiotic therapies seem ineluctable but often were mitigated when an antibiotic was administered by parenteral route. Sub-therapeutic dosages caused the augmentation of taxa involved in sugar metabolism, suggesting a link with weight gain. This result should not be interpreted positively, considering that parallel information on antibiotic resistance selection was rarely reported and selection of antibiotic resistance is known to occur also at low antibiotic concentration. However, studies on the effect of antibiotics as growth promoters put the basis for understanding the gut microbiota composition and function in this situation. This knowledge could inspire alternative strategies to antibiotics, such as probiotics, for improving animal performance. This review encompasses the analysis of the main animal hosts and all antibiotic classes, and highlights the future challenges and gaps of knowledge that should be filled. Further studies are necessary for elucidating pharmacodynamics in animals in order to improve therapy duration, antibiotic dosages, and administration routes for mitigating negative effects of antibiotic therapies. Furthermore, this review highlights that studies on aminoglycosides are almost inexistent, and they should be increased, considering that aminoglycosides are the first most commonly used antibiotic family in companion animals. Harmonization of experimental procedures is necessary in this research field. In fact, current studies are based on different experimental set-up varying for antibiotic dosage, regimen, administration, and downstream microbiota analysis. In the future, shotgun metagenomics coupled with long-reads sequencing should become a standard experimental approach enabling to gather comprehensive knowledge on GIM in terms of composition and taxonomic functions, and of ARGs. Decorticating GIM in animals will unveil revolutionary strategies for medication and improvement of animals’ health status, with positive consequences on global health.

Additional file 8 : Fig. 8. Two host states. Panel a, Host inflammatory status (distal colon histology score) distribution over all experimental groups (0% up to 3% DSS). Panel b, Distal colon histology score distributions for groups treated with 1, 2, or 3% DSS, and untreated control group. Colors represent treatment groups as indicated. Bimodal distributions are observed when data from all groups are combined (panel a, Additional Table 6), and within the 2% and 1% DSS groups (panel b).

Additional file 5 : Fig.5. Principal coordinates analysis. 3rd and 4th axes of the principal coordinates analysis of microbiota data shown in Fig. 2a.

Additional file 3 : Fig. 3. Effect of DSS treatments on the host. Panel a, Development of rat bodyweight from the start of the first DSS treatment (T0) onward, relative to weight at T0. Curves represent means of treatment groups. Colors represent treatment groups as indicated. Red horizontal lines at the bottom of the figure indicate DSS treatment periods. Statistically significant differences are observed between 0% and 3% DSS groups at T2, T3, T16, T17, T18 and T19 (p < 0.05; Kuskal-Wallis test with Dunn’s post test, adjusted for multiple testing). Differences at other time-points or between other treatment groups and the 0% DSS group were not significant. Panel b, Colon length at T75 (45 days after last DSS treatment), in different treatment groups. Each mark represents one rat. Horizontal lines indicate median values. Differences between treatment groups are not statistically significant (Mann Whitney test). Panel c, Distal colon histology scores (total inflammation scores from Additional Table 5) at T75 in different treatment groups. Each mark represents one rat. Horizontal lines indicate median values. Statistically significant differences are observed between 0% and 2% DSS groups (p=0.0068) and between 0% and 3% DSS groups (p=0.0001) (Mann Whitney test).

Additional file 16 : Table 6. Bimodal distribution of host inflammatory status. Statistical support for bimodal distribution of host inflammatory status (implying two host states).

Additional file 6 : Fig. 6. Enterotyping. Microbiota data from T-7 to T68 were aggregated at genus-level, filtered for “unknown” and “uncultured” attributions, and analyzed using the clustering approach described in [6] (“enterotyping”). Panel a, clustering score (Calinski-Harabasz index [36]) as a function of the number of clusters shows a maximum at 3 clusters. Panel b, clustering with 3 clusters. Panel c, correspondence between clusters in panel b and PCoA1-based microbiota states from Fig. 2 (1, basal state; 2, alternative state). Numbers indicate number of samples (Table 1) in each category. Microbiota state 2 roughly corresponds to cluster A. Panel d, Akkermansia, Phascolarctobacterium and Bacteroides distributions in clusters A, B and C. Combined data from T-7 to T68; each dot represents one intestinal microbiota sample. Abundance is expressed as number of sequence reads on a total of 38,000. Only genera for which the median abundances in the two microbiota states differ at least 1.2-fold with q < 0.05 (Wilcoxon test with FDR adjustment) are presented. The table presents q-values for pairwise comparisons of relative abundances between clusters, for each of the three species (Kruskal-Wallis test with posthoc Dunn’s test and Holm correction for multiple comparisons).

Additional file 7 : Fig. 7. Correspondence between enterotyping-based clusters and PCoA1-based microbiota states. Juxtaposition of Table 1 (from main text, left) showing PCoA1-based microbiota states as a function of time for each rat, and the corresponding table showing enterotyping-based clusters from Additional Fig. 6b (right). Alternative microbiota state 2 and cluster A are highlighted. Rats (lines of the table) are grouped according to the host-ecosystem state at T68/75 (last column of Table 1, cf Fig. 5), then by resemblance of microbiota-by-time profile from T27 onward (cf Table 1).

Additional file 2 : Fig. 2. Effect of diet shift on microbiota composition. Panel a, Time course of relative abundance (number of sequence reads on a total of 38,000) of selected bacterial genera, before and after diet shift at T-31 (cf Additional Table 2). Colored lines each represent one rat. Dashed lines connect median values. Panel b, Abundance distributions of selected Lactobacillus species before (T-34) and after (T-20) diet shift (n=10). Abundance is expressed as number of sequence reads on a total of 38,000. Each dot represents one rat. P-values, Wilcoxon test.

Host–microbial co-metabolism products are being increasingly recognised to play important roles in physiological processes. However, studies undertaking a comprehensive approach to consider host–microbial metabolic relationships remain scarce. Metabolomic analysis yielding detailed information regarding metabolites found in a given biological compartment holds promise for such an approach. This work aimed to explore the associations between host plasma metabolomic signatures and gut microbiota composition in healthy adults of the Milieu Intérieur study. For 846 subjects, gut microbiota composition was profiled through sequencing of the 16S rRNA gene in stools. Metabolomic signatures were generated through proton NMR analysis of plasma. The associations between metabolomic variables and α- and β-diversity indexes and relative taxa abundances were tested using multi-adjusted partial Spearman correlations, permutational ANOVA and multivariate associations with linear models, respectively. A multiple testing correction was applied (Benjamini–Hochberg, 10 % false discovery rate). Microbial richness was negatively associated with lipid-related signals and positively associated with amino acids, choline, creatinine, glucose and citrate (−0·133 ≤ Spearman’s ρ ≤ 0·126). Specific associations between metabolomic signals and abundances of taxa were detected (twenty-five at the genus level and nineteen at the species level): notably, numerous associations were observed for creatinine (positively associated with eleven species and negatively associated with Faecalibacterium prausnitzii). This large-scale population-based study highlights metabolites associated with gut microbial features and provides new insights into the understanding of complex host–gut microbiota metabolic relationships. In particular, our results support the implication of a ‘gut–kidney axis’. More studies providing a detailed exploration of these complex interactions and their implications for host health are needed.

Additional file 9 : Fig. 9. State transitions in the host-microbiota ecosystem. Host-microbiota ecosystem model from Fig. 5. Possible trajectories of the host-microbiota ecosystem from a “healthy” state to a “(pre-) disease” state, comprising a microbiota state-transition and a host inflammatory state-transition, are indicated by the three filled arrows. These transitions may be sequential (upper and lower arrows) or simultaneous (diagonal arrow).

Additional file 4 : Fig. 4. Effect of DSS treatments on microbiota diversity. Microbiota alpha diversity (Shannon index) at time-points indicated at the bottom of the figure. Each dot represents one rat. Colors represent treatment groups as indicated. Red arrows indicate DSS treatments (T0 to T2, T14 to T16, and T28 to T30, respectively). At T3, alpha diversity was significantly lower in the 0.5%, 1% and 2% DSS groups than in the 0% DSS control group (p < 0.01; Kuskal-Wallis test with Dunn’s post test, adjusted for multiple testing). No differences between treatment groups were observed at T-1, nor at T68. In the 1% and 2% DSS groups, alpha diversity was lower at T3 than at T-1, while the opposite was true in the 0% DSS control group (p < 0.05; Wilcoxon test). None of the treatment groups showed significant differences in alpha diversity between T-1 and T68 (Wilcoxon test).