A variant in the OPRM1 gene
Significant DNA sequence variation in humans also exists in the OPRM1 gene, and the frequencies of the variations differ between populations.12 A number of groups have investigated a particularly common DNA variant in the OPRM1 gene that is a functional exonic variation in the extracellular region of the protein. The variant is an alanine (A) to guanine (G) nucleotide substitution at position 118 from the start of translation, resulting in a change in the amino acid sequence at that site of an asparagine to an aspartic acid (N40D), which alters the chemical properties of the peptide. This SNP is commonly known by its database identification number— rs1799971.

The asparagine variant of the protein leads to a 3-fold increase in β-endorphin binding to the receptor.13 In vitro studies suggest that the aspartic acid variant results in the production of 1.5 to 2.5 times less transcription of the gene's messenger RNA, as well as a 10-fold decrease in protein levels.14

Clinical studies of the SNP reveals a number of interesting associations with dysfunctions in the hypothalamic-pituitary-adrenal (HPA) axis. In a study of 39 healthy men, baseline levels of cortisol were not affected by the rs1799971 DNA variant; however, cortisol response to blockade of the opioid receptor with naloxone was dependent on the genotype at the A118G position. Higher serum cortisol concentrations, and faster cortisol responses, were detected in individuals with the G (aspartic acid) allele,which suggests that there is a connection between the G allele and altered HPA axis responsivity.15 Similar studies have replicated this finding and some found that participants with at least 1 copy of the G allele also exhibited higher cortisol levels at baseline, further suggesting a genetic marker for disorders of HPA axis dysfunction.16-18 Proopiomelanocortin is a peptide prehormone, which is cleaved to form adrenocorticotropic hormone, and β-endorphin, which has a high affinity for µ-opioid receptors. This affinity is variable, depending on the sequence variation present in the OPRM1 gene.13

No association was detected for the rs1799971 SNP and symptoms of anxiety and depression in a cross section of 867 community-living adults, a longitudinal study of 660 children, or 30 healthy subjects.17,19

Variations in antidepressant response, tolerance, and remission
We have examined the association of genetic variability in the OPRM1 gene and response, remission, and tolerance to the antidepressant citalopram, using a subset of participants from the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study. This is the largest study to date that analyzes markers across the entirety of the OPRM1 gene, and the first study investigating OPRM1 association with antidepressant treatment response, remission, and tolerance. It includes analysis of white Hispanics, whites, and non-Hispanic blacks separately, and controls for alcohol and drug dependence.

We proposed that individuals without the capacity to respond to citalopram may have a defect in the endogenous opioid system, preventing full alleviation of symptoms in response to antidepressant medication. We focused on 53 genetic variants in the OPRM1 gene for this investigation. The preliminary data revealed 8 SNPs to be associated with antidepressant response (Table). Association was defined as a P value from an additive model of less than .05, and an odds ratio with 95% confidence intervals that did not exceed 1.

Response was defined as having a 50% reduction in the Quick Inventory of Depressive Symptomatology (QIDS) score from baseline on the last visit.20 Of the 8 SNPs, 5 were found in the white Hispanic sample and 3 were in the white sample. No findings for the non-Hispanic black sample were detected for the response phenotype; however, 1 SNP was associated with the remission phenotype in the non-Hispanic black sample.

This SNP, and 6 detected in the white Hispanic sample, compose the SNPs seen in the remission phenotype. Remission was defined as having a QIDS score of 5 during follow-up. No SNPs in the white sample were associated with the remission phenotype. Four SNPs were found in both the remission and response phenotypes for the white Hispanic sample. Only 2 SNPs were found to be associated with the tolerance phenotype in the white sample, when tolerance was defined based on study exit data of the STAR*D trial. If participants continued their treatment with citalopram, they were considered tolerant; however, if at the end of the study they discontinued citalopram or left the study because of adverse effects, they were considered intolerant.

The most significantly associated SNP detected was correlated with the response phenotype in whites (rs540825). This SNP is located in an alternatively spliced exon of the OPRM1 gene and is found only in the MOR1X isoform. Preliminary data suggest that when expressed in cultured cells, this isoform differs significantly from the typical MOR1 isoform in its regulation by membrane trafficking (M. Tanowitz, M. von Zastrow, personal oral communication, 2008). The existence of this isoform is not well studied in humans, but we found it to be expressed in the human brain. Further signaling and trafficking experiments along with fine-mapping around the area of the associated SNP will aid in determining the role of this SNP in antidepressant response. It is possible that the interaction of the MORs and antidepressant response is a result of the activation of downstream signaling of the µ-opioid system. An alternative isoform of the MOR may be expressed at lower levels than the typical MOR1 isoform. Thus, if an alternative isoform of the MOR is the pre-dominant receptor expressed in the brain and results in degradation rather than recycling of the receptor, the downstream signaling will be negatively affected, potentially influencing a behavioral response. Furthermore, it is plausible that the associated SNP in the alternative exon may decrease the amount of MOR available, perhaps leading to an attenuation of the downstream signaling response that then adversely affects antidepressant responsiveness.

Summary
Substantial evidence in rodents exists to support the role of the µ-opioid system in antidepressant response and the pathophysiology of depressive-like symptoms. In humans, the placebo effect is clearly mediated by the µ-opioid system, which potentially explains a portion of the response seen with many pharmaceuticals. Furthermore, the present data also suggest a role in antidepressant response and tolerance, as well as remission from major depression in humans. The interindividual differences in variations in the OPRM1 gene, and the associations described above, have implications in personalized medicine. One could imagine selecting one antidepressant over another based on genotypic constitution, and predictability of response rate, tolerance to the particular medication. Together this body of literature points to a new direction for research into novel antidepressant treatments and the pathology of MDD.

 
Pages: 1  2