Neurogeneticist Lea K. Davis, Ph.D., has devoted her career to finding the biological roots of psychiatric conditions and is now turning her focus to a link between depression and inflammation.
“We have found a strong association between depression and consistently elevated white blood cell counts,” said Davis, an associate professor of genetic medicine at Vanderbilt University Medical Center. “Now, we will try to learn how this association originates and the nature of a feedback loop we suspect takes place to ingrain the cycle.”
Davis was principal investigator on a recent study published in JAMA Psychiatry detailing a Vanderbilt-led collaboration between four medical centers within the PsycheMERGE network, a national consortium utilizing EMRs and the NIH sponsored eMERGE network for mental health research. In the study, Davis and colleagues report how associations found at Vanderbilt between depression polygenic scores and white blood cell count proved replicable across the diverse populations of Icahn School of Medicine at Mount Sinai, Mass General Brigham and the Million Veteran Program.
One of the most interesting findings from the study was that the highest risk scores for depression were correlated with only about a 5 percent increase in white blood cell count, which is still within the normal reference range.
“This is not an immune response at the level of an autoimmune disease, but rather a sustained low-grade activation,” Davis said.
Big Data Connects Comorbidities
More broadly, Davis is using Vanderbilt’s DNA biobank, BioVU, in conjunction with EMR data to elucidate the trajectories taken by mental health and wellness initiatives as they impact other body systems. For example, her research team is studying relationships between comorbid conditions – such as anxiety, post-traumatic stress disorder and chronic pain – with coronary artery disease.
“We have found a strong association between depression and consistently elevated white blood cell counts.”
“The idea is to connect the genetic risk for these conditions and others, like major depression or schizophrenia or bipolar disorder, with the clinical biomarkers that we’ve measured on patients over their lifetime,” Davis said.
Major depression was the first condition they focused on after building the infrastructure for these analyses.
The White Blood Cell–Depression Connection
To start, the research team looked at all patients with a sample deposited in BioVU who had DNA variations associated with major depression and calculated their polygenic risk scores. They then related these risk scores to 315 different clinical laboratory measurements, such as absolute monocyte and neutrophil counts, as well as cholesterol and triglyceride levels.
“While depression is multifactorial, we wanted to test the degree to which genetic liability lines up with various clinical biomarkers,” Davis said.
“We think there is a kind of feedback loop where people who are at higher genetic risk also have this more sustained activation of the immune system, which then predisposes them to have a higher risk for a clinical manifestation of depression.”
The relationship with high white blood cell count was the most significant, followed by elevated cholesterol. However, when the team corrected for the actual clinical diagnosis of depression, the cholesterol association attenuated but there was no difference in the association with the immunological markers.
“This suggested that the markers of cholesterol were really a consequence of the clinical manifestation of depression. But what we were seeing with the immunological markers was linked to the genetic contribution to depression,” Davis explained.
A Bidirectional Relationship
Next, Davis and the team conducted a mediation analysis to probe whether higher white blood cell counts mediate the genetic risk for depression or vice versa.
“What was really interesting was that we found evidence for both,” Davis said. “We think there is a kind of feedback loop where people who are at higher genetic risk also have this more sustained activation of the immune system, which then predisposes them to have a higher risk for a clinical manifestation of depression. Once depression sets in, it may also trigger a neuroimmune response, perpetuating the cycle.”
Replication and Next Steps
Davis then turned to the PsycheMERGE network to validate her team’s findings and assess reproducibility.
“We were able to robustly replicate these associations with each of these partners, which is really remarkable because each biobank taps into very different underlying populations.”
Despite having found an association that is uncorrupted by other comorbidities, even the presence of the condition itself, and that is also bidirectional and replicable across diverse populations, questions remain, primarily about how these two conditions interact mechanistically.
“From what we’ve learned from animal model research, I think it’s a long cascade of signaling molecules, starting with a biological response to a threat, which is inflammation,” Davis said.
Some clinical solutions may already be in place, since many common antidepressants, perhaps serendipitously, have an anti-inflammatory effect.
“There hasn’t yet been a large study across many classes of drugs looking at white blood cell count in particular, but I can tell you that study is on its way,” she said.