In this study, we performed an extensive search for functional SNPs within the OPRM1
gene locus and made several fundamental observations. First, we identified several novel and potentially functional SNPs. Particularly, a novel SNP rs563649, which is situated within the newly identified exon 13, was the strongest independent contributor to measured pain-sensitivity responses (Table ). In contrast to SNP rs563649, the SNPs that have been previously associated with OPRM1-dependent phenotypes displayed rather modest effects. Specifically, the non-synonymous SNP variant A118G
), which has been shown to be associated with human pressure pain thresholds (26
), showed much smaller differences. In agreement with previous findings, in our cohorts homozygotes for the G
allele displayed the highest mean values for mechanical pain thresholds, whereas homozygotes for the A
allele displayed the lowest mean values for mechanical pain thresholds; however, this difference did not reach significance (data not shown). Importantly, the previously published association achieved statistical significance only among males (26
), whereas our cohort tested for experimental pressure pain consisted of females only.
Further, we have identified a potentially functional haplotype associated with high pain sensitivity (Table ), which is tagged by the minor allele T
of rs563649. The frequency of this haplotype corresponded to the frequency of the minor T
allele and the SNP rs563649 was the strongest individual contributor to the association, suggesting that the minor T
allele of this newly identified SNP is a strong candidate for the functional site of the haplotype. This haplotype spans over the OPRM1
gene locus and consists of all the alleles that have previously been shown to be associated with lower pain thresholds (rs1799971, A
), negative mood (rs495491, G
) and increased substance dependency, including opioid dependence (rs609148, G
allele and rs495491, G
), all of which are consistent with a low level of opioid receptor activity. On the other hand, whereas the T allele of rs563649 in our cohort is associated with a decreased analgesic response to opioids, again consistent with low levels of receptor activity (21
), the A
allele of rs1799971, which is situated on the haplotype tagged by the T allele of rs563649, is associated with an increased analgesic response to opioids (16
Together, our data give rise to the possibility that some of the previously reported positive associations can be explained, at least in part, by the high LD between the reported markers and the SNP rs563649 identified in this study (57
). Furthermore, it is likely that several functional SNPs co-exist within the haplotype, where alternative alleles play specific roles in distinct but related pain phenotypes, are carefully balanced and thus transmitted jointly. This haplotype was observed in 6% of the subjects, thus representing a substantial proportion of the human population and is of clinical significance.
It should be noted that our association analysis has been conducted on two moderately sized cohorts. The primary results were obtained in the larger UNC cohort that consisted of a homogeneous sub-sample of Caucasian females. The second UF cohort is smaller and consists of both genders. Although the association tests were highly significant in the first cohort only, and it is important to replicate our results in larger cohorts, the direction of effects was the same for both cohorts with respect to pain sensitivity, and associations with morphine response phenotypes were in the expected direction, suggesting that the T allele codes for a low-efficiency receptor variant. Furthermore, the validity of our findings is supported by molecular biology experiments presented here, which demonstrate the existence of human exon 13 and a functional effect of the identified SNP rs563649.
To understand how the newly identified SNP may produce functional effects, we constructed a reporter vector that includes the entire predicted IRES element. Although the exact molecular mechanism by which rs563649 regulates OPRM1 function and pain signaling requires further studies, out data suggest that the presence of minor T
allele should lead to higher expression levels of corresponding MOR-1K
isoforms. Furthermore, the localization of a strong functional SNP within the human analog of mouse exon 13 provides evidence for the biological significance of MOR-1K
isoforms. We showed that MOR-1K
isoforms with variable exon 13–exon 2 junctions are expressed in a tissue-specific manner and likely contribute to tissue-specific post-transcriptional regulation. Because the T
allele of rs563649 is associated with higher translation efficiency and higher pain sensitivity, we suggest that the truncated MOR-1K form contributes to hyperalgesic-like rather than analgesic states and plausibly represents one of the molecular mechanisms underlying the excitatory effect of opiods, contributing to tolerance, drug dependence and opioid-induced hyperalgesia (60
). Furthermore, similarities in the tissue-specific expression patterns between MOR-1
, with lower abundance of MOR-1K
relative to MOR-1
isoforms, support the possibility of a regulatory function for MOR1-K
. To date, no systematic studies involving the genetics, molecular and cell biology of this form of OPRM1
receptor have been reported; and its biological function clearly warrants further research.
We also have shown that almost all of the known mouse exons have corresponding orthologs within the human OPRM1
gene locus and thus the genetic structure of human OPRM1
is much more complex than is currently appreciated. According to recent studies, at least 50% or more of human genome is expressed in alternatively spliced isoforms (63
). There is a number of both computational and molecular biological difficulties that make computational predictions and cloning of the genetic variants of OPRM1
highly challenging. The novel approaches used in this study overcame these difficulties. The discovery of a much more complicated receptor structure makes possible new interpretations of existing results and permits the generation of novel testable hypotheses in diverse disciplines that have interest in studying the OPRM1 function.
Genomic organization of the expanded human OPRM1
locus is highly similar to the organization of the mouse OPRM1
locus. However, alternative splicing events within this locus display some substantial differences between human and mouse, and thus findings from rodent studies should be considered with caution when applied to the function of human opioid receptor. Specifically, our RT–PCR and 5′RACE data suggest that exon 13 containing OPRM1
mRNA variants in human and mouse are highly divergent (Fig. A–C; Supplementary Material, Fig. S5
). New tissue-specific alternative splice isoforms increase functional diversity of the gene and create additional options for regulation, as demonstrated by the strong effect of SNP rs563649 on pain sensitivity (64
). Importantly, location of this functional SNP in a region of variable inter-species conservation shows that biological function does not always correlate with sequence conservation.
Our results also suggest that newly discovered alternative exons, rather than constitutive exons, may represent targets for genetic variability which modifies OPRM1 receptor function. Since modifications of the function of the main receptor form may have dramatic consequences on fitness and may not reach significant frequency in the general population, genetic variations in alternative exons that are expressed at low levels or only under specific conditions may lead to more subtle phenotypic differences that underlie the observed variation in OPRM1-dependent phenotypes. Furthermore, our data and methodological approaches are of relevance to genome-wide association studies when SNPs associated with clinical phenotypes are located in genomic areas that are devoid of known structural genetic elements (65
In summary, the identification of a potentially new functional SNP with a substantial minor allele frequency within alternative exons of OPRM1 is of considerable importance to the field of pain genetics. Although the association of SNPs rs563649 with sensitivity to noxious stimuli and morphine responses requires further confirmation in larger cohorts, our results provide strong evidence that this SNP has a far greater effect on receptor function than other currently known OPRM1 polymorphisms. Collectively, our data open conceptually new approaches to examining the structure and function of the OPRM1 gene.