Welcome to our Monthly Journal Club! Each month I post a paper or two that I have read and find interesting. I use this as a forum for open discussion about the paper in question. Anyone can participate in the journal club, and provide comments/critiques on the paper by leaving a comment below. I picked this month’s paper because it details an enormous and beautiful study examining how the immune system communicates with the brain to promote anxiety. The paper we are discussing is titled “Stress-induced metabolic disorder in peripheral CD4+ T cells leads to anxiety-like behavior” (click the hyperlink to see the paper) by Jin Jin & colleagues at Zhejiang University in China.
We have all experienced stress. Whether it’s from an upcoming exam, a public performance, or the existential dread experienced day-in and day-out for so many people around the world :) . If stress goes unchecked, it can lead to the development of anxiety, where we start to feel stressed in situations that don’t usually call for this kind of response. How does a stressful experience precipitate anxiety? Why do some people develop anxiety under stress, while others seem resistant?
Many studies have demonstrated that chronic stress negatively influences the immune system. However, it is unclear whether these changes causally contribute to the development of anxiety. Additionally, whether immune-related anxiety is driven by the innate or adaptive immune system remains an unexplored area. Jin Jin and colleagues set out to answer these open questions using a mouse model of chronic stress. To produce stress, the authors subjected mice to brief daily electronic foot shocks (5 times for 3 sec per day for 8 days; ES).
Figure 1: CD4+ T cells play a major role in the development of stress-induced anxiety-like behavior. Mice were given injections of control antibody (IgG), anti-CD4, or anti-CD8 to deplete two major subtypes of T cells. IN a separate group, wild-type (WT) mice were compared to those lacking B and T cells (Rag1-/-). They were given daily electric footshocks (ES) for 8 days to elicit stress responses. A day later, mice were tested for the development of anxiety like behavior in an open field test. Rag-/- mice showed no anxiety like behavior following footshock stress (Panel B), and similarly, mice with their CD4+ T cells ablated also didn’t show anxiety like behavior (panel C). Credit: Fan et al., 2019
In this paradigm, normal mice (wild-type; WT) or those lacking an adaptive immune system (Rag1-/-) were subjected to multiple foot shock sessions over the course of 8 days, and then their behavior was tested for signs of anxiety the following day (see Figure 1). To do this, the authors took advantage of a mouse’s natural tendency to stick to the edges of any area that it is in, avoiding open areas (i.e., thigmotaxis). In this ‘open field test’, WT mice stuck to the edges of the arena much more following electric foot shock-induced stress (i.e., they were more scared of entering the center of the arena). However, Rag1-/- mice showed equivalent anxiety-like behavior regardless of whether they were subjected to prior stress or not. This indicates that the immune system plays a role in the generation of anxiety. To test which cells likely drove this response, the authors depleted two different major types of T cells (CD 4+ helper T cells and CD 8+ cytotoxic T cells) using antibodies against CD4 and CD8. When they did this and then subjected the mice to their stress protocol, mice given control antibodies and those with CD8 cells depleted showed anxiety-like responses in the open field, just like WT mice. However, mice given anti-CD4 antibodies no longer showed any signs of anxiety! This indicates that CD4+ T cells are important in driving anxiety like behavior in response to chronic stress!
Figure 2: RNA-seq of T cells reveals altered gene expression in CD4+ T cells in response to chronic electric footshock stress (Panels A-D). These differentially expressed genes largely contribute to mitochondrial function. On tests of energy production, CD4+ T cells from stressed mice showed severely impaired glycolytic and oxidative phosphorylation capacity (Panel E). (Credit: Fan et al., 2019).
To investigate this finding further, the authors began to look deeper into CD4+ T cells and how they change in response to chronic stress. Using RNA-seq, they were able to identify 128 differentially expressed genes in CD4+ T cells from stress vs. non-stressed mice (Figure 2). Careful examination of what these genes encoded revealed that many of them were essential to mitochondrial function. As mitochondria are essential for energy production, the researchers tested whether CD4+ T cells from stressed mice showed changes in energy utilization and mitochondrial function. Indeed, these cells from stressed mice showed reduced energy production through the glycolysis and oxidative phosphorylation pathways (Figure 2E). When mitochondria were examined, those in CD4+ T cells from stressed mice showed abnormal morphologies and reduced expression of key membrane proteins (Figure 2G). This suggests that stress-induced mitochondrial dysfunction in CD4+ T cells accompanies the development of anxiety-like behavior.
Figure 3: Disrupted mitochondrial function promotes anxiety-like behavior that depends on CD4+ T cell function. Knocking out the mitochondrial membrane protein Miga2 (Miga2-/-) produces severe anxiety-like behavior, which can be rescued by CD4+ T cell depletion. (Credit: Fan et al., 2019).
What could be precipitating these changes in energy metabolism in CD4+ T cells? Prior research has shown that mood disorders are associated with alterations in omega-6 fatty acid and arachidonic acid (AA) concentrations in the brain. AA is a critical modulator of immune processes via its metabolism to leukotriene-B4 (LKB4) and prostaglandin-E2 (PGE2). When the authors infused each of these metabolites into mice, LKB4 produced pronounced anxiety-like behavior irrespective of whether the mice were stressed or not. This effect seemed to depend on CD4+ T cells, and caused significant changes in mitochondrial morphology and function!
If defective mitochondria and energy production are responsible for the development of anxiety in response to chronic stress, then artificially disrupting mitochondrial function should generate anxiety levels similar to those produced by repeated electric foot shocks. To disrupt mitochondrial function, the researchers knocked out a gene encoding a key mitochondrial membrane protein, Mitogaurdin-2 (Miga2) (Figure 3, Figure 4). When these mice were tested on various behavioral assays, they showed marked signs of anxiety, lending support for the idea that mitochondrial dysfunction drives anxiety-like behavior.
Figure 4: Stress-induced mitochondrial dysfunction in CD4+ T cells promotes anxiety via aberrant increases in xanthine production. Metabolic analyses of serum from WT and Miga2 T-cell KOs (Miga2TKO) revealed marked increases in purine metabolic pathways, and specifically xanthine (~1000x fold increase (10^3)! Xanthine infusions promoted the anxiety like behavior in mice and were drastically elevated in human patients with clinical anxiety. (Credit: Fan et al., 2019).
But how do defective mitochondria influence CD4 T-cells to promote anxiety? The first idea was that in response to stress, T cells aberrantly travel to the brain where they influence neural circuits underlying anxiety. When the authors tested this idea (by blocking CD6 and VLA-4, proteins involved in T-cell migration to brain), they found no evidence that blocking CD4+ T cell migration prevented the development of anxiety. This suggests that instead of T-cells directly traveling to the brain, they release a soluble factor that travels to the brain or causes another cell to influence brain function indirectly.
So, what could these factors be? To investigate this, the authors screened metabolic pathways in normal (WT) and Miga2 knockout mice. They observed marked changes in circulating concentrations of metabolites involved in purine metabolism in the knockout mice (Figure 4). Upon further investigation, they observed large increases in circulating levels of the purine metabolite xanthine (among others) in knockout mice (Figure 4 D,E). Blood samples from humans with anxiety disorders also showed high levels of xanthine (Figure 4 F)! Interestingly, xanthine is a precursor to caffeine and theobromine (a caffeine derivative found in dark chocolate), and xanthine toxicity causes nervousness and tachycardia, which are also observed in patients with anxiety. Similar to the results obtained from infusions of LKB4, infusing mice with xanthine increased anxiety levels, suggesting that disrupted mitochondrial function in CD4+ T cells (in response to LKB4) results in aberrant increases in circulating xanthine which contributes to anxiety.
Figure 5: Miga2 KO mice have increased numbers of oligodendrocytes in the left amygdala, and this is dependent on CD4+ T cells. Knockdown of the putative xanthine receptor AdorA1 (using shRNAs) on oligodendrocytes in the amygdala rescues anxiety-like behavior in the knockout mice. These findings link changes in peripheral immunity and purine metabolism to dysfunctional neural activity and the generation of stress-induced anxiety. (Credit: Fan et al., 2019).
Anxiety, however, is a neurological phenomenon…and so far all we have done is looked at what is happening in the body, not the brain. So, the authors set out to understand how all these changes in the immune system (resulting in elevated xanthine levels) influence neural circuits involved with anxiety (Figure 5). The amygdala is a key brain structure involved in fear and anxiety, so this was a good place to start looking. The researchers observed that Miga2 knockout mice had a larger left amygdala than mice carrying functional copies of Miga2. Based on this finding, they took a deep dive and profiled all cell types within the amygdala of these mice to see how they were altered. Left amygdalae from knockout mice had altered numbers of non-neuronal cells than amygdalae from WT mice. Additionally, infusions of xanthine produced a similar pathology in the amygdala to that of the Miga2 knockouts. Closer examination revealed that knockout mice had many more oligodendrocytes in their amygdala than WT mice, and this could be reversed by depletion of peripheral CD4+ T cells (Figure 5H).
Infusions of xanthine promoted the direct proliferation of oligodendrocytes in the amygdala, suggesting that stress-induced xanthine production promotes aberrant glial cell proliferation in this brain area. Using short-hairpin RNAs (shRNAs), the researchers knocked down the putative receptor for xanthine (AdorA1) specifically on oligodendrocytes within the amygdala (Figure 5I-J). Without this receptor (and therefore without xanthine signaling), KO mice no longer showed anxiety like behavior!
Figure 6: Summary of findings. Stress-induced elevations of LKB4 cause aberrant mitochondrial function in CD4+ T cells. This results in elevations in circulating xanthine levels which signal via A1 receptors on oligodendrocytes in the amygdala. This alters local neural activity to promote anxiety-like behavior! (Credit: Fan et al., 2019).
Together, this giant set of experiments beautifully lays out a complex, multi-system interaction pathway that links chronic stress exposure to the development of pathological anxiety. I am extremely impressed by the sheer number of discoveries in this paper, many of which I did not have time to get to without making this post even longer than it already is! Definitely check out the original paper linked in the opening paragraph if you are interested. An important aspect of this paper is that many of the findings from mouse studies were confirmed in human samples (e.g., circulating xanthine levels in anxiety disorder patients).
Please let me know what you think by leaving a comment below! ‘Till next time, stay curious!