An Ecological and Evolutionary Perspective on Gastrointestinal Health

There have been claims that variations in the composition of the intestinal flora influence individual health going back at least to the early years of the 1900s.  Late in his career, Ilya Mechnikov, co-receipient of the Nobel Prize for Physiology or Medicine in 1908 (along with Paul Ehrlich) and a pioneer in the study of what we now call innate immune mechanisms, promoted a diet based on fermented milk containing live bacterial cultures of lactobacilli (i.e., yogurt) (1).  Mechnikov apparently believed that the bacteria in the yogurt would compete with bacterial species that he thought were, on balance, harmful to human health.  In other words, Mechnikov recognized the potential relevance of gastrointestinal ecology to human health a century ago.

 

A paper just published in Science (2) from the laboratory of Gabriel Núñez provides evidence for a phenomenon that is at least roughly similar to the effect postulated by Mechnikov if only in that competition between gut microbes can substantially influence the trajectory of an intestinal infection.  Kamada et al. studied infections of C57BL/6 mice with Citrobacter rodentium, a natural pathogen of mice.  C. rodentium causes attaching and effacing (AE) lesions on the intestinal epithelium similarly to the important human intestinal pathogens enterohemorrhagic Escherichia coli (EHEC) and enteroptahogenic E. coli (EPEC) providing one rationale for studying the mechanisms involved in infection by this mouse pathogen.

The critical comparison that Kamada et al. study is the time course of infection with C. rodentium in germ-free (GF) and specific pathogen-free (SPF) C57BL/6 mice.  While the number of bacteria detected in the feces increases dramatically over the first week in the GF and SPF hosts, the bacterial loads for the pathogen decrease back to baseline by 21 days post-infection only in the SPF mice.  In the GF mice, the bacterial counts of C. rodentium in the feces remained strongly elevated (i.e., roughly 1×10⁹ colony forming units per gram).  While previous studies have demonstrated that the immune response, both humoral and cell-mediated aspects, is involved in clearing infection by C. rodentium in normal mice, the GF and SPF mice exhibited comparable production of relevant classes of antibody (IgG) and T cells producing interleukin-17 (IL-17) and interferon-γ.  The GF and SPF mice also mounted comparable innate immune responses as assessed by such indices as recruitment of neutrophils and macrophages and colonic expression of various anti-microbial peptides.  Thus, the authors concluded that the presence of commensal bacteria can have a decisive effect on an infection by an enteropathogen.  Our internal ecosystem is arguably a component of the immune system, broadly-construed, with commensals, in at least some cases, serving to select against pathogens.

 

In AE pathogens such as C. rodentium, genes at the locus of enterocyte effacement (LEE), are generally regarded as essential virulence factors for host colonization and are associated with pathogen-mediated tissue damage.  Kamada et al. investigated isogenic C. rodentium strains that do or do not express the LEE genes.  Two different LEE-deficient strains could not successfully infect the SPF mice, whereas the wild-type C. rodentium strain was successful in infecting SPF mice.  However, all of the bacteria, including the LEE-deficient C. rodentium strains could infect GF mice.  So, the need for expression of these virulence–associated genes was contigent on the microbiological environment.

 

An additional finding of particular interest relates to the role of nutritional dependencies in determining which commensals are most effective at competing with C. rodentium in the mouse gut.  When the authors assessed the relative effects of administering E. coli versus either of two Bacteroides species (B. thetaiotaomicron or B. vulgatus) on the numbers of C. rodentium in the intestines of GF mice exposed to that pathogen 21 days previously, they found that co-colonization with E. coli dramatically reduced (200-fold at three days and 500-fold at 14 days) the numbers of C. rodentium while neither of the Bacteroides species had any effect.  The impact of administering E. coli was seen even if the recipient GF mice were already colonized with both C. rodentium and B. thetaiotaomicron or B. vulgatus.

 

The above results may relate to the nutritional preferences of these different bacteria.  Both E. coli and C. rodentium prefer monosaccharides while B. thetaiotaomicron and B. vulgatus can utilize either monosaccharides or polysacahrides as a carbon source.  If Kamada et al. administered both C. rodentium and B. thetaiotaomicron to GF mice and fed the mice a conventional diet containing both monosaccharides or polysaccharides, the presence of the B. thetaiotaomicron bacteria had no effect on the numbers of C. rodentium bacteria.  However, if the dually infected mice were fed on a diet containing only simple sugars, the numbers of C. rodentium declined substantially, implying that if the B. thetaiotaomicron bacteria were forced to compete for the same nutritional source as the C. rodentium, the C. rodentium bacteria were much less able to grow and reproduce.  In the absence of B. thetaiotaomicron bacteria the C. rodentium proliferated as well on the diet consisting only of monosaccharides as on the diet with both simple and complex sugars.

 

References

 

 

 

1. http://www.nobelprize.org/nobel_prizes/medicine/laureates/1908/mechnikov-bio.html

 

2. Kamada N, Kim YG, Sham HP, Vallance BA, Puente JL, Martens EC, Núñez G. Regulated virulence controls the ability of a pathogen to compete with the gut microbiota. Science. 2012 Jun 8;336(6086):1325-9. Epub 2012 May 10. PubMed PMID: 22582016.

 

 

 

3. Sperandio V. Microbiology. Virulence or competition? Science. 2012 Jun 8;336(6086):1238-9. Epub 2012 May 10. PubMed PMID: 22582015.