When I looked at the genetics of placebo effects last June I was struck by the apparent overlap between the genes associated with depression and placebo responses. Does this mean there are biological mechanisms common to antidepressant drug and placebo responses, I wondered, which might partly explain the notorious difficulty demonstrating that antidepressant drugs work better than placebo in clinical trials?
Here I will discuss the genetics of antidepressant drug responses. This is a “bottom up” look at depression biology, which complements other methods like neuroimaging, phenomenological observation and diagnosis, and psychotherapy research. Here I will focus on pharmacodynamics—how the drugs work in the brain—and not address the genetics of drug metabolism by the liver, which also contributes importantly to treatment response. Most of the research has addressed depression, but some of the findings have been extended to treatment of anxiety disorders with antidepressants.
First, some background information about how antidepressant drugs work, from Goodman and Gilman’s pharmacology textbook. Reuptake by the presynaptic neuron is the main mechanism by which action of the monoamines serotonin, dopamine, and norepinephrine at postsynaptic receptors is ended. Each has a transporter protein on the presynaptic terminal. The most commonly used antidepressants, the selective serotonin reuptake inhibitors (SSRI’s), selective serotonin and norepinephrine reuptake inhibitors (SNRI’s), and tricyclic antidepressants, block the transporter proteins, leaving more neurotransmitter molecules in the synapse and thereby enhancing postsynaptic receptor effects. The older monoamine oxidase inhibitor (MAOI) drugs inhibit the breakdown of these neurotransmitters and increase their storage in presynaptic secretory granules. Other drugs enhance monoamine effects by other, sometimes complex, means: trazodone acts directly on postsynaptic receptors.
Longer-term use of antidepressants evokes adaptive and regulatory mechanisms both presynaptically and postsynaptically. These include reduced production of transporter proteins, increased receptor density, increased receptor-G protein coupling and cyclic AMP signaling, induction of growth factors like brain-derived neurotrophic factor (BDNF), and facilitation of neurogenesis in the hippocampus.
Many genes have been associated with antidepressant response. Sarah Helton and Falk Lohoff of the National Institute on Alcohol Abuse and Alcoholism have summarized those involved in the serotonin system. Variants in genes for enzymes involved in the biosynthesis and metabolism of monoamines include those for tryptophan hydroxylase, which catalyzes one of the steps in serotonin biosynthesis; monoamine oxidase A (MAO-A), which is important in the breakdown of serotonin, dopamine, and norepinephrine; and catechol O-methyl transferase (COMT), which breaks down dopamine and norepinephrine. Several variants in each of these genes have been implicated, but results have not been consistent.
The gene for the serotonin transporter protein, which is thought to be the site of action of the SSRI’s, has received a great deal of attention. A variant in the promoter region of the gene, which controls transcription of the DNA sequences encoding the transporter protein into messenger RNA, has been associated with both vulnerability to depression and other psychiatric problems and poor response to antidepressant medication. The short version of this allele leads to less production of transporter protein and less serotonin reuptake and, interestingly, is associated with increased depression and poor medication response. This sounds paradoxical, since less serotonin reuptake should lead to more serotonin at postsynaptic receptors and less depression. I have not been able to find an explanation for this. Variants at several other sites in the gene have also been associated with antidepressant response.
Sixteen different serotonin receptors have been identified. Of these, 5-HT1a and 5-HT2a are inhibitory autoreceptors which reduce the release of serotonin from the axonal terminal. Variants in the genes for these proteins and in those encoding serotonin receptors have been associated with antidepressant medication response, but conflicting findings seem to be the rule.
In another review, Yvet Kroeze, Huiqing Zhou, and Judith Homberg of Radboud University in the Netherlands go into more detail. They note that SSRI’s transiently raise serotonin concentration, while chronic treatment increases baseline levels. An increase in serotonin, however, does not directly elevate mood, and the mechanisms by which the drugs relieve depression and anxiety disorders are not fully clear. Also, besides the variant in the promotor region of the serotonin transporter gene, other epigenetic mechanisms such as DNA methylation and histone acetylation/methylation may be important in serotonin signaling.
Interestingly, these authors cite several studies in which a single nucleotide polymorphism in the promoter region of the serotonin transporter gene (different from the short allele mentioned above) was associated with high placebo responses rates in trials of citalopram for generalized anxiety. This has apparently been interpreted as lack of citalopram efficacy, which may be an error, since a high placebo response rate does not necessarily mean the drug ineffective. (Peter Kramer discusses this issue at length in his recent book; see my review in Psychiatric Times.
The dopamine and serotonin systems interact, and variants in the COMT and dopamine transporter genes have been associated with SSRI response, though, again, findings have not always agreed. Similarly, variants in genes involving glycine, another neurotransmitter with both excitatory and inhibitory effects, have been associated with antidepressant responses, as have genes in the glutamergic system. One of the more widely studied genetic variants associated with antidepressant response is in the BDNF gene. And genetic variants in the hypothalamus-pituitary-adrenal (HPA) axis have also been associated with antidepressant response.
The inflammatory system is also involved with depression. Kroeze and colleagues explain that inflammation is triggered by cytokines and eicosanoids, which recruit immune cells and promote healing. Cytokines include interleukins, which communicate with white blood cells; chemokines, which regulate cell movement; and interferons, which are anti-viral. Eicosanoids include prostaglandins, which regulate fever and blood vessel dilation; and leukotrienes, which attract leukocytes. Proinflammatory cytokines have many effects in the brain, including reducing serotonin synthesis, increasing neurotoxic metabolites, and increasing and reducing synthesis of both serotonin and dopaminee. They affect the HPA axis and can inhibit neurogenesis in the hippocampus. Interestingly, the last can be restored by nonsteroidal anti-inflammatory drugs (NSAID’s). Also, several genetic variants involving interleukins have been associated with antidepressant drug responses.
In another area, a study has associated several variants in the CLOCK gene that affects circadian rhythms with response to the SSRI fluvoxamine.
Micro RNA’s (miRNA’s) are short RNA molecules which bind to complementary regions on messenger RNA (mRNA) strands, resulting in gene repression or slicing. In animal research, fluoxetine increased miRNA levels and reduced the expression of the serotonin transporter.
Clearly, this is a very complex field. The effects of antidepressant medication are affected by variants in a host of genes encoding proteins involved in monoamine synthesis and metabolism, monoamine reuptake, monoamine receptors, nerve growth factors, the HPA axis, inflammation, and circadian rhythms. These effects may be at the level of DNA coding for proteins, gene promoter regions, other gene introns and exons (DNA sequences in a gene which do not directly code for the amino acid sequences in proteins), and other epigenetic effects, including miRNA’s, DNA methylation, and histone modifications.
This complexity is staggering. From a drug development point of view, there are a huge number of possibilities for intervening in the biological systems of depression and anxiety. Unfortunately the prospects for a clear understanding the mechanisms of psychiatric symptoms and disorders and how such things as placebo effects might interact with drug effects are less sanguine.