Genetic Variability of μ-Opioid Receptor in an Obstetric Population

The study protocol was approved by the Institutional Review Boards of Geneva and Columbia University Medical Centers, and subjects gave written informed consent. Subjects were 181 women who gave birth at the Geneva University Hospital (Geneva, Switzerland) or Columbia University Medical Center (New York, New York) between October 2001 and April 2003 and consented to give blood for DNA analysis. Genotyping for the A118G μOR polymorphism was performed as described below. One hundred thirty-seven women from Geneva and 44 women from New York were genotyped. All subjects were healthy parturients with uncomplicated pregnancies and deliveries at term (> 37 weeks’ gestation) with no clinically significant abnormality shown by routine history, physical examination, or laboratory test results. Women with a known history of drug abuse were excluded.

Peripheral blood was collected in 10-ml EDTA tubes from selected subjects. DNA was purified by a nonphenolic method using a Puregene Blood extraction Kit (Gentra, Minneapolis, MN) and tested for molecular weight on gel electrophoresis and purity quality by optical densitometry measure (260/280 nm).

All genotyping was performed at the University of Geneva. For identification of A118G (N40D) in gene OPRM1 , 60 ng DNA from subjects was amplified by polymerase chain reaction under standard conditions using the annealing primers forward (5′ end-labeled with biotin) 5′- CCGGCCGTCAGTACCAT-3′ and reward reverse 5′-GTAGAGGGCCATGATCGTGAT-3′. Polymorphism A118G (GATGGCRACCTGT) was then genotyped on the 247-bp polymerase chain reaction fragment by a minisequencing method §using sequencing (reward reverse) primer 5′-TGGGTCGGACAGGT-3′ as follows.

Polymerase chain reaction products were immobilized with Dynabeads (Dynal, Oslo, Norway) by a 15-min, 65°C incubation in a buffer containing 10 mm Tris-HCl, 2 m NaCl, 1 mm EDTA, and 0.1% Tween 20. Polymerase chain reaction products were then removed from solution using magnetic separation, denatured with NaOH 0.5 m, and washed with 200 mm Tris-Acetate, 50 mm MgAc². The remaining single-stranded DNA was then hybridized with the internal “sequencing” primer by heating the mix to 80°C and slowly cooling it to room temperature. Next, enzyme and substrate mixes were automatically added to each well, and the reactions were processed at 28°C by the sequential addition of single nucleotides with a predetermined order. Luciferase peak heights were proportional to the number of nucleotide incorporations, which have been shown to possess a high degree of correlation (5% error rate) in a number of experimental settings. 13,14 

Proportions of genotypes for OPRM1  are expressed as percentage of total (95% confidence interval). The allele frequencies were calculated using the gene-counting method. The significance of differences in allele and genotype distribution in the two populations, as well as demographic data, was determined using the chi-square or the Fisher exact test, with P < 0.05 being the minimal value accepted as statistically significant.

Self-reported racial distribution and demographic data were similar at both institutions (table 1). Genetic distribution was similar with regard to race and when compared between institutions (table 2). Overall, 121 women were A118 homozygous (67%; 95% confidence interval, 58–73%), 52 were heterozygous (29%; 95% confidence interval, 23–38%), and 8 were G118 homozygous (4%; 95% confidence interval, 2–9%). The overall allelic frequencies using the gene-counting method were 81.2% A118 and 18.8% G118.

Single-nucleotide polymorphisms are naturally occurring variants in the structure of genes that may have a significant influence on the metabolism, clinical effectiveness, and side effect profiles of therapeutically important drugs. Ethnic differences in drug response are in part genetically determined, but genotyping of individuals has significantly greater potential to improve the pharmacotherapy of disease (and pain). As the occurrence and function of genetic variants of drug targets become better understood, this knowledge should lead to the design and use of specific drugs or dosages in patients with differing genotypes.

We report a relatively high frequency of the A118G variant in a diverse obstetric population from two different institutions; our overall G118 allelic frequency was 18.8%. Because race has been shown to affect genetic variability, we compared the genetic distribution between institutions and between the two prevalent racial groups, namely white and Hispanic, and found no difference. Among whites, Bond et al.  11reported that the expected frequencies of homozygous G118 and heterozygous subjects are 2% and 20%, respectively, with an 11.5% G118 allelic frequency, regardless of sex. Overall, among 150 African-American, white, and Hispanic subjects, the G118 allelic frequency was 10.5%. In Asian populations, genotype distributions of several μOR SNPs frequencies have been recently reported, with a G118 frequency in control subjects ranging between 35 and to 47% (Chinese, 35%; Thai, 44%; Malay, 45%; Indian, 47%), 15and the reported frequency among Japanese is 49%. 12Interestingly, a recent study from California studying a group of 109 subjects (61% white) did not find any subjects with the G118 allele. 16 

Alterations of the encoded receptor may have functional consequences for receptor activity such as binding affinity and potency. Bond et al.  11demonstrated a threefold increase in binding affinity and potency of β endorphins at the G118 variant of the μOR in vitro . The A118G variant predicts an Asn to Asp change at amino acid 40 of the receptor. Because the Asn residue is a putative site for N-glycosylation at the N-terminal region of the μOR, this variant may interfere with subsequent receptor glycosylation and alter agonist binding or activity. One possible mechanism of the pharmacologic differences between the genetic variants might be that the binding affinity of ligands to the variant receptor may be dependent on the size of the opioid ligand. Bond et al.  11found that for most small opioid ligands, there was no change in binding affinity with A118G polymorphism, but there was a significant increase in binding affinity for the larger endogenous opioid peptides, such as β endorphin. Furthermore, β endorphin seemed to be three times more potent at the altered μOR. These findings, however, have not been consistently demonstrated. Befort et al.  17found that although an extremely rare SNP in the third transmembrane domain with a proline substitution for a serine (S268P) severely impaired receptor signaling, the A118G variant did not modify the functional properties of the receptor.

In vivo , the influence of the A118G polymorphism on the clinical response to morphine and morphine-6-glucuronide (M6G), an active metabolite of morphine, has been evaluated. 18Pupillary constriction, a measure of the central effects of opioids, was reduced in subjects with A118G variant after the administration of M6G but not morphine. These same authors sought to investigate the risk for M6G accumulation and opioid overdose and toxicity in two patients with severe kidney failure genotyped for SNPs of the μOR. 19One patient tolerated morphine treatment despite high plasma M6G and was found to be homozygous for the G118 allele. The other patient, homozygous for the wild-type allele A118, was found to have severely impaired vigilance during the first day of morphine treatment. The authors postulated that the G118 allele may be one of several genes or factors conferring protection against M6G-related opioid side effects and perhaps toxicity. Recently, two studies have reported that individuals with a G118 allele (five heterozygotes, one homozygote) had an increased cortisol response to the administration of naloxone compared with A118 homozygotes, suggesting that there may indeed be clinical consequences to this polymorphism. 20,21 

There is evidence for a key role for the opioid system in the pathophysiology of substance abuse 22,23; therefore, the μOR is an ideal candidate gene for susceptibility or protection from addictive behaviors. Nonetheless, studies on the association of the A118G variant with alcohol dependence or drug abuse have not been consistent. Bond et al.  11reported that the A118 allele was present in a significantly higher proportion of opioid-dependent Hispanic individuals than non–opioid-dependent individuals. An association between the A118 allele and alcohol dependence among white alcoholics was also reported, 24but this finding was not confirmed by others. 12,25–27Another study in white subjects abusing alcohol, cigarettes, cocaine, marijuana, or other illegal drugs found that there was a significant association between substance dependence and the A118 allele. 28Interestingly, the response to naltrexone, a μOR antagonist used for treatment of alcohol and opioid dependence, has been shown to be significantly better among white alcoholics carrying the G118 allele, with significantly lower rates of relapse and a longer time to return to heavy drinking, than alcoholics homozygous for the A118 allele. 29 

The most commonly used approach for the study of “pain genes” is gene expression assays relying on previously identified genes or complementary DNA sequences of interest. Despite great advances in transgenic technology and genetic mapping of selected candidate genes, we are far from having elucidated the genetics of pain. Studies of the μOR gene may provide insight into the mechanisms involved in the interindividual differences in physiologic responses to endogenous opioids, pain perception, and pain thresholds. Specific SNPs of the μOR may alter the response to opioids and also influence side effects, toxicity, and tolerance profiles. Information on μOR genotype may enable us to predict the response to μOR manipulation and allow opioid analgesic regimens to be tailored to individuals’ genetic makeups. This should result in better pain management with fewer or milder side effects for labor and delivery, postoperative care, and chronic pain states. In the future, gene therapy of the μOR could be used for patients with chronic pain who are refractory to opioids. In addition, identifying SNPs relevant to opioid analgesia is likely to contribute to a better understanding of clinical conditions beyond analgesia, such as immune modulation, opioid reward mechanisms, and drug abuse.

To date, a handful of mutations/polymorphisms affecting pain in humans have been investigated. In our study, we found that a genetic variant of the μOR associated with an increased response to β endorphin is present in more than 30% of our obstetric population. Further studies investigating the clinical implications of this finding, such as the impact of this genetic variant on labor pain and neuraxial opioid requirements for labor analgesia, will be of interest.