Two different people are riding the subway. A third person coughs on these individuals over the course of their trip. One person gets the flu, but the other doesn’t. Somewhere nearby, two more people accidentally eat a piece of meat that wasn’t correctly refrigerated. One develops food poisoning, but the other remains healthy. What factors contribute to these different outcomes?

The key factor is the immune response. Immune cells such as macrophages and granulocytes kill invading microbes. Other immune proteins called cytokines and chemokines aid cellular communication and stimulate the movement of cells towards sites of inflammation. Immune growth factors also form part of the immune response by stimulating the proliferation of specific tissues. If profiles of these immune parameters differ between individuals, then their ability to respond to pathogens will also vary.

In fact, it is well known that the frequencies of different immune cells, proteins, and growth factors vary widely among individuals. However, the factors driving these differences have been debated. Some parts of the immune response may be heritable, or controlled in large part by our human genes. But components of the immune response can also be determined by non-heritable influences or environmental factors. These environmental factors include the microbes we come into contact with on a daily basis, as well as those that comprise our microbiomes. They also include chemicals, toxins, or any other compounds we are exposed to in food, air and water.

This past year, Brodin and team at Stanford University published a seminal study that greatly clarifies how the immune system is governed. The team demonstrate that non-heritable environmental influences play a significantly larger role than inherited factors in shaping the human immune response.

The researchers performed a systems-level analysis of 210 healthy twins between the ages of 8 and 82. They measured 204 immune parameters, including cell population frequencies, cytokine responses, and serum proteins, and found that 77% of these are dominated, and 58% almost completely determined by non-heritable environmental influences.

The team derived their results by measuring the 204 immune parameters in both monozygotic (identical) and dizygotic (fraternal) twins. Identical twins share copies of the same human genes. However, the genomes of fraternal twins diverge significantly so that, on average, they share only about 50 percent of their genes. Brodin and team created a statistical model that measured how closely correlated the levels and/or activity of each immune parameter were in the identical twins. If levels of a given immune parameter were more correlated in the identical twins than in their fraternal counterparts, the parameter was interpreted as “heritable.”

Interestingly, the activity of every immune parameter deemed heritable by the statistical model was able to be altered by the expression of related, non-heritable, environmental factors. This is because no component of the immune response acts in isolation. Multiple pathways connect immune cells and proteins to one another, so that if conditions in the body change, the immune system can best coordinate its response.

After a second analysis, Brodin and team produced the figure to the left. It shows how, throughout the entire immune network under study, nodes associated with heritable immune parameters (yellow) were generally connected to nodes associated with non-heritable, environmental factors (purple). The fact that the purple nodes tend to be larger than the yellow nodes means that hubs, or important connection points in the network, are dominated by the non-heritable, environmental influences. This means that the expression of any heritable immune factor can be directly altered by changes in a person’s environment. In other words, no component of the immune response is isolated from the possible influence of microbial and chemical exposure.

If environmental factors, particularly microbial exposure, drive the non-heritable nature of the immune response, then Brodin and team reasoned that such influences would increase with time. That is because younger twins often share the same environment (home, parents, pets), while older twins have generally lived apart for decades.

To test this possibility, the researchers compared twin-twin correlations for all immune measurements between the oldest and the youngest identical twin pairs in the study (older twins were 72+ years in age, and younger twins were under the age of 20). For several cell population frequencies, they found much reduced correlations with increased age. In the most striking example, the frequency of immune cells called Tregs between the youngest identical twins correlated very strongly at 0.78 but was highly uncorrelated at 0.24 between the oldest identical twin pairs.

The same trend held for serum proteins and many cell signaling responses, causing the team to suggest that “immune divergence between genetically identical twins with age is a common phenomenon.” This suggests that many of the immune parameters originally calculated as “heritable” by the team are highly correlated in identical twins due to shared environmental factors rather than shared human genes.

Brodin and team then calculated how acquisition of the pathogen cytomegalovirus (CMV) can shape the immune response. Once acquired, CMV persists in its host for life. The researchers studied a group of identical twins discordant for CMV infection (one twin was infected with CMV, and the other was not). They compared their twin-twin correlations for all measurements of immune parameters to those of a separate group of identical twins who both tested negative for the virus. The identical twins discordant for CMV infection showed greatly reduced correlations for many immune cell frequencies, cell signaling responses, and cytokine concentrations. In general, the influence of CMV was so broad that it affected 119 of all 204 measurements dispersed throughout the immune network. This means that acquisition of just one pathogen can dramatically modulate a person’s overall immune profile.

Finally, the researchers immunized a separate study group of twins with seasonal flu vaccine. They studied the manner in which these subjects created proteins called antibodies in response to the vaccine. The researchers were surprised to find no detectable contributions from heritable factors on any of the vaccine responses. When the analysis was repeated to exclude subjects with high antibody titers from previous flu vaccines, the team obtained the same result. They subsequently contended that “infectious disease exposure can shape subsequent immunity.” In other words, the manner in which an individual responds to the flu vaccine appears to be largely shaped by a series of factors related to infectious history – the number of times they have been infected by the flu virus, the strain of the flu virus acquired when infected, and the manner in which previous flu vaccines have generated an immune response.

Commentary

I read the full text of the paper in one sitting, and re-read it several times. For years, I’ve put forth a hypothesis called successive infection. The hypothesis contends that inflammatory disease symptoms result from the sum of a person’s infectious history and other environmental exposures. During the successive infectious process, each pathogen a person acquires alters the immune response, often in a manner that makes it easier for other pathogens to also persist in the same host.

The Brodin study now allows us to quantifiably estimate the manner in which just a single pathogen can alter immune parameters as part of the successive infectious process. The ability of CMV to impact 58% of immune parameters measured is higher than even I would have guessed. Several of my papers discuss how CMV can disable activity of the VDR nuclear receptor. It’s possible that some of the immune parameters altered by chronic CMV infection are modified via this pathway. Brodin and team’s flu vaccine data further supports the successive infection model, demonstrating yet another instance in which a person’s immune response is dictated by exposure to infectious agents and not human genes.

A second cornerstone of the successive infection hypothesis is that the immune responses of people diverge as they age and are exposed to different microbes over time. In some cases this may occur in a manner that may promote disease. Brodin’s data supports this assumption, showing that even immune parameters that appear heritable in younger twins are actually impacted by the environment over time.

I’ve also contended for years that inflammatory diseases are driven by dysregulation of the human microbiome and metagenome rather than by faulty human genes acting alone. In Brodin’s study, even the heritable immune parameters under study were directly connected to those modulated by environmental variables – variables such as microbiome composition.

These findings have major implications for the direction of future medical research. Each year, agencies that fund medical research, such as the National Institutes of Health (NIH), must carefully allocate their limited resources to research teams that can best study human disease. Some of these teams study how components of our human genes might contribute to chronic illness. But, as described here, limited therapeutic options have emerged from these genome-wide association studies.

Brodin and team’s data provide yet one more reason why this may be the case. It goes without saying that the immune response plays a central role in the development of any human chronic disease. If the immune response is driven largely by environmental factors and not human genes, then studies of the human genome alone cannot provide a complete picture of disease mechanisms. This strongly suggests that the research community should de-prioritize genome-wide association studies, and focus instead on better understanding the human microbiome, and how composition of the microbiome is impacted by other environmental variables such as diet, medication, and toxic exposure.

A little more on the findings

Seventy-two of the immune parameters studied by the team were different kinds of immune blood cells. Among these, a few had very strong influences from heritable factors, including naive, CD27+ cells, and central memory CD4+ T cells. However, for the most part, non-heritable influences were clearly dominant. In fact, for 61% of all immune cell populations, the influence of heritable factors was undetectable. This was true of both cells associated with the adaptive immune response such as T and B cells, and of cells associated with the innate immune response such as granulocytes, monocytes and natural killer cells.

The team also estimated the influences of heritable and non-heritable factors associated with 24 cytokines, 10 chemokines, 6 growth factors, and 3 other serum proteins identified in both sets of twins. Some cytokines were particularly heritable, such as IL-12p40. But for many other measurements, such as IL-10 and a group of chemokines, the heritable influence was low.

An interactive map of the study’s results is available.