The Evolution & Medicine Review

Last year, my colleagues and I published Nutrient Signaling: Evolutionary Origins of the Immune-Modulating Effects of Dietary Fat. We were prompted to write this paper because of several observations:

1) Chronic inflammatory diseases, such as obesity, diabetes, and atherosclerosis have reached epidemic proportions and account for growing proportion of global mortality and morbidity.

2) A growing body of evidence has shown that modern dietary habits – including overnutrition and exposure to certain nutrients and processed foods – contribute to chronic inflammatory diseases.

3) Nutrients have vastly differing effects on the immune system. Certain nutrients promote pro-inflammatory signaling and dangerous changes in metabolism; other nutrients reduce immune activation, have anti-inflammatory signaling properties, and protect against metabolic diseases. These observations raise the question, why?

Until recently, there has been no satisfactory explanation for these divergent immune and metabolic effects. In particular, the existence of damaging pro-inflammatory signaling pathways triggered by food poses an evolutionary paradox: why should nutrients commonly encountered during human evolution generate a cascade of costly and damaging immune and metabolic disturbances?

In order to answer that question, we looked to the microbiome, the collection of microorganisms that inhabit our bodies and outnumber human cells by an order of magnitude. It has been established that food intake modifies the gut microbiota. It has also been shown previously that gut microbes modify the risk of many chronic inflammatory diseases – such obesity, diabetes, and heart disease. Perhaps nutrient effects on the microbiota hold the key to the question of an immune signaling function of food.

The nutrient signaling model makes the following argument:

1) Some nutrients predictably alter the microbiome in a dangerous way.

2) These negative changes to the microbiome increase the risk of infectious disease, by colonizing the gut with dangerous pathogens, promoting the virulent transformation of commensal bacteria, and by interfering with the barrier function of the gut.

3) These changes often result in increased infectious mortality.  During human evolution, this mortality has driven selection for microbe-triggered mobilization of immune resources directed toward opportunistic and specialized pathogens.

4) Infectious mortality has also selected for nutrient-triggered mobilization of immune resources, allowing certain nutrients to evolve signal properties. These nutrients thus provide an early warning system, alerting the immune system to changes in bacterial populations that are about to occur.

Our model also provides a similar argument for anti-inflammatory nutrients:

1) Some nutrients predictably alter the microbiome in a protective and beneficial way.

2) These beneficial changes to the microbiome decrease the risk of infectious disease, by preventing the colonization of the gut with dangerous pathogens, protecting against the virulent transformation of commensals, and by reinforcing the barrier function of the gut.

3) Microbiota-derived protection from infection has selected for a diminished immune response, thus protecting the host from damaging side effects of inflammation.

4) The same selection has resulted in certain nutrients evolving anti-inflammatory signal properties that anticipate impending beneficial changes to the microbiome.

This model led to two testable hypotheses: First, nutrients that encourage the growth of potential pathogens are expected to cause pro-inflammatory signaling. Second, nutrients that inhibit the growth of potential pathogens are expected to cause anti-inflammatory signaling.

Fats were the nutrients that we focused on, because a high fat, high calorie diet has been shown to be associated with chronic metabolic and cardiovascular diseases in epidemiologic studies. In addition, the consumption of fat been shown to cause low grade inflammation and insulin resistance in both animals and in humans. Fat also has the advantage of being well studied both for its effects on inflammation and for its effects on bacterial growth.

We tested our hypothesis by taking advantage of a unique property of fats, that is, the membrane destabilizing (detergent) effects of certain lipids that can inhibit the growth of various bacteria.

Generally speaking, unsaturated fatty acids (with carbon-carbon double bonds) introduce fluidity when incorporated into cell membranes, potentially destabilizing them and making microbes more susceptible to a coordinated attack by anti-microbial products of the innate immune system.

Membrane destabilization by unsaturated fatty acids occur because double bonds induce a kink in the lipid molecule that interferes with weak intermolecular bonds. This effect causes most unsaturated fatty acids to be liquid at room temperature, while similar saturated fatty acids are solid. Because of these effects, unsaturated fatty acids often inhibit the growth of pathogens and pathobionts (potential opportunistic pathogens).

We compared the effects of unsaturated versus saturated fatty acids on bacterial growth and on inflammation.  We used published reports on the in vitro effects of fatty acids and found a striking linkage between a lipid’s effects on bacteria and on the immune system. We found that saturated fats, relative to their unsaturated counterparts, tend to lack antimicrobial properties while providing a carbon energy that can fuel bacterial growth. These fats were more likely to elicit pro-inflammatory signaling. Unsaturated fats tended to inhibit and kill pathogenic bacteria and promoted anti-inflammatory signaling more than saturated fats. This relationship was highly statistically significant.

We restricted our results to in vitro studies, which allowed us to consider fatty acid effects on bacteria and on the immune system independently. However, this approach was also a weakness because the in vivo relationship of fats and microbiota is highly complex. In the living mammalian host, it has been uncertain whether lipid effects on bacteria modify the microbiome and modulate infection risk.

When we published last year, very little was known about the effects of specific kinds of fats, e.g. omega-3 fatty acids or saturated fat, on the microbiota. A key prediction of the nutrient signaling, however, is that pro-inflammatory fats should also cause an expansion of dangerous bacteria in the gut in vivo. Do different fats really have divergent effects on the microbiota?

More recent studies have shed light on this question.

In the British Journal of Nutrition, Mujico and colleagues showed that a high fat diet (approximately half saturated fat and unsaturated fat) increased the number of Enterobacteriales, a group that contains E. coli, and decreases the number of beneficial barrier bacteria, Bifidobacteria and Lactobacillus. Importantly, they also tested the effects of adding certain unsaturated fat supplements. An oleic acid (olive oil) supplement prevented the increases in E. coli and the declines in Bifidobacteria caused by the high fat diet. An omega-3 polyunsaturated fat supplement also increased the number of beneficial Lactobacillus species and prevented the increase in Enterobacteriales induced by high fat diet. This study suggests that fatty acids generally thought to be healthful – oleic acid found in olive oil, and omega-3 fatty acids found in fish, generally reduce the number of potential pathogens (the E. coli containing group Enterobacteriales) and increase the number of protective species (Lactobacillus and Bifodobacteria) in line with the expectations of the Nutrient Signaling model.

The second question is whether changes in the microbiota actually increase the risk of infection, and thus exert selection because of infectious mortality.

Research supporting this proposition has recently been performed using a mouse model. In a recent report in PLOS One, Haag and colleagues manipulated the gut microbiota of mice by feeding them commensal E. coli, thus raising the intestinal numbers of E. coli by several orders of magnitude. They subsequently inoculated the experimental mice and controls with a pathogen, Campylobacter jejuni.  Control mice with normal gut flora were able to resist colonization and infection with Campylobacter jejuni. The E. coli enhanced mice were unable to resist colonization and suffered infection with Campylobacter.

It remains untested whether a high saturated fat diet also decreases colonization resistance to Campylobacter jejuni. However, with the fast pace of research on the microbiota, the answer to this question will likely arrive soon.