29 May 2012. Researchers have identified the cortical neurons that mediate antidepressant responses, according to a report appearing online May 25 in Cell. A team of scientists led by Nathaniel Heintz of New York’s Rockefeller University report that layer 5 corticostriatal pyramidal neurons expressing the adaptor protein p11 exhibit a robust response to treatment with the antidepressant fluoxetine, including an upregulation of serotonin receptors, that is p11 dependent.
Major depressive disorder (MDD) is known to involve a dysregulation of a number of brain regions involved in mood, including the prefrontal cortex, basal ganglia, and amygdala (Drevets, 2000). Serotonin is a key player in MDD, and selective serotonin reuptake inhibitors (SSRIs) are the most widely used antidepressants. However, the mechanisms underlying the pathology of MDD and the targets of successful treatments are thought to involve distinct brain areas. Thus, characterization of not only the pathophysiology of the illness, but also the cell types and molecular mechanisms that mediate the antidepressant response, are crucial to the development of novel treatment strategies.
One protein recently implicated in the antidepressant response is p11. Encoded by the S100a10 gene, p11 controls serotonin signaling through interactions with serotonin receptors that target them to the cell surface. Lower levels of p11 have been found in the cortex of patients with MDD as well as in a mouse model of the illness, and antidepressants increase the levels of p11 in mice (see SRF related news story). In the present study, Heintz and his group investigated whether cortical cells that express p11 are important in the action of antidepressants.
First author Eric Schmidt and colleagues used the so-called “bacTRAP” approach to selectively assess S100a10 cells. Recently developed by this group, this method allows for expression profiling of individual cell types (Doyle et al., 2008) without needing to isolate single cells beforehand (Heiman et al., 2008). Instead, bacterial artificial chromosome (BAC) transgenic mice, in which an EGFP-tagged ribosomal transgene is expressed in a specific cell population, are used for subsequent affinity purification and quantification of the mRNAs expressed within that cell type.
Using this approach, the researchers found that the majority of transgene expression occurred in pyramidal cells in superficial layer 5 of the cortex, and using retrograde tracing, they found that these S100a10 cells project to the dorsal striatum. All cells that expressed the transgene also contained p11. The S100a10-expressing pyramidal cells seem to be a previously uncharacterized cell population, as immunoprecipitates from these cells did not contain mRNAs such as Ntsr1, Etv1, and Glt25d2, which are found in other types of pyramidal cells.
Affecting antidepressant response
Using the SSRI fluoxetine (FLX), researchers then probed the role of S100a10 cells in the antidepressant response. Expression of cortical p11 was increased following two weeks of FLX treatment, and 62 different genes were differentially regulated by the drug after a slightly longer treatment. In contrast, the expression of only four genes was modulated by FLX in another population of pyramidal neurons, Glt25d2 cells, suggesting that the S100a10 cells are particularly sensitive to antidepressant treatment. Since p11 is known to affect serotonin receptor (5-Htr) function (see SRF related news story), the researchers next examined if 5-Htrs in S100a10 cells were changed following FLX treatment. Indeed, 5-Htr4 expression was increased in S100a10, but not Glt25d2, cells. The researchers hypothesize that this elevation in 5-Htr4 may make corticostriatal neurons more sensitive to the increased serotonin levels resulting from the drug, producing a beneficial increase in signaling from the cortex to the striatum.
To assess whether p11 played a role in the 5-Htr changes, the researchers then mated the S100a10 bacTRAP mice with p11 knockout animals, creating mice with p11 deleted from S100a10 cells (S100a10 p11 knockouts). Six different subtypes of 5-Htrs were downregulated in these animals, suggesting that a knockout of p11 produces a loss of serotonergic tone. Moreover, the knockout animals also failed to exhibit the dramatic response to FLX treatment observed in the S100a10 bacTRAP mice—most of the genes that were altered by FLX treatment in the S100a10 neurons were no longer changed in the S100a10 p11 knockouts.
Schmidt and colleagues then performed behavioral testing in cortex-specific p11 knockout mice to test the therapeutic effects of FLX treatment. In the novelty suppressed feeding paradigm, animals are placed in an open field with food in the center. Those treated with antidepressants take less time to approach the food in the novel environment than do untreated animals (Dulawa and Hen, 2005). However, in the current study, unlike wild-type animals, the cortex-specific p11 knockout mice did not exhibit a shorter latency to feed after treatment with FLX, indicating a diminished behavioral response to the antidepressant. Importantly, there were no alterations in baseline anxiety in the cortex-specific p11 knockouts, suggesting that the results reflect failure of these animals to respond to antidepressant treatment. Similar results were also observed using the tail suspension test, another task that is commonly used to assay antidepressant responses.
Taken together, the results of the current study demonstrate that S100a10-expressing corticostriatal projection neurons are critical for the molecular and behavioral responses of SSRIs. They also demonstrate the effectiveness of the bacTRAP translational profiling approach in detecting cell types that respond to drugs. The authors are hopeful that this approach will ultimately help patients, calling their study “an exciting step forward in the march toward development of more effective treatments for depression.”—Allison A. Curley.
Schmidt EF, Warner-Schmidt JL, Otopalik BG, Pickett SB, Greengard P, Heintz N. Identification of the Cortical Neurons that Mediate Antidepressant Responses. Cell . 2012 May 25 ; 149(5):1152-63. Abstract