The melanocortin-1 receptor (MC1R) is a rather unlikely candidate to influence acute pain and opioid analgesia, being best known for its role in skin/hair pigmentation¹ and immunomodulation.² However, quantitative trait locus (QTL) mapping studies in mice led us to the consideration of MC1R involvement in κ-opioid analgesia in female but not male mice³ based on the chromosomal location of the mouse Mc1r gene and the demonstration of MC1R mRNA and protein in the periaqueductal grey,⁴ a midbrain locus of crucial relevance to endogenous analgesic circuitry.⁵ We found that the κ-opioid analgesia phenotype of the female Mc1r-null C57BL/6-Mc1r^e/e (“recessive yellow”, e/e) mutant mouse was similar to that of redheaded women with multiple, protein inactivating MC1R variants.³

The clinical utility of this finding is clearly limited by the uncommon application of κ-opioid acting drugs in pain therapy. On the other hand, drugs acting at the μ-opioid receptor represent the mainstay of clinical treatment of moderate to severe pain. We reasoned that a fuller characterisation of these mutants might reveal other, more clinically relevant phenotypes. To these ends, we tested mutant mice and then humans of various hair colours for their sensitivity to pain and μ-opioid analgesia.

METHODS

Subjects

Naïve, male and female adult (6–10 week old) mice of the C57BL/6J and C57BL/6J-Mc1r^e/e strains were used. Mice were housed with same sex littermates in groups of two to four. Animal experiments were approved by the local animal care and use committee at McGill University.

A total of 44 Dutch and three Scottish volunteers (23 men, 24 women, median age 22, range 18–35 years) participated in this study. All subjects were healthy, and none had participated in pain research previously. All women were on oral contraceptives. Human studies were approved by the local institutional review board at Leiden University Medical Center.

Nociceptive assays

Mice were tested for their basal sensitivity on six different assays of acute and tonic nociception. Detailed protocols have been published previously.⁶ In the tail withdrawal test, mice were lightly restrained and the latency to vigorous removal of the distal half of the tail from 49°C or 47°C water was measured. In the hot plate test, mice were placed on a 53°C or 50°C metal surface within a Plexiglas cylinder, and the latency to first hindpaw lick or shake measured. In the paw withdrawal test, mice were placed atop a glass surface in small Plexiglas enclosures, and a high intensity heat lamp was focused on the right or left mid-plantar hindpaw. The latency to withdraw from the heat stimulus was measured. In the tail clip test, mice were lightly restrained and a binder clip (500 g force) was applied to their tail near the base. The mouse was immediately removed from the holder onto a table top and the latency to lick, bite, or grab the clip was measured. In the abdominal constriction test, mice received an i.p. injection of 0.9% acetic acid, were placed in Plexiglas observation cylinders, and the number of reflexive constrictions of the abdominal musculature counted for 30 min. Data in each assay were obtained by only one experimenter.

Human subjects were tested by same sex experimenters. After arrival of human subjects in the laboratory, an arterial line was placed in the left or right radial artery under local anaesthesia (for blood sampling) and an intravenous line was inserted in the contralateral arm (for drug infusion). Acute pain was induced by an electrical current through two surface electrodes (Red Dot, 3M, London, ON, Canada) placed on the skin over the tibial bone (shin bone) of the left leg. The electrodes were attached to a computer interfaced current stimulator. The noxious stimulus was a 10 Hz tetanic pulse with a duration of 0.1 ms. The intensity of the noxious stimulation was increased from 0 mA in steps of 0.5 mA per 1 s (cut off: 128 mA). The subjects were instructed to press a button on a control box when no further increase in stimulus intensity was acceptable to them and that current was defined as pain tolerance. Upon pressing the button, the stimulus train ended. After a 1 h training session, data acquisition began.

Drugs

Morphine sulfate and morphine-6-glucuronide (M6G) were generously provided by Sabex 2000 (Boucherville, QC, Canada) and CeNeS (Cambridge, UK), respectively. In mice, both drugs were dissolved in saline and administered subcutaneously in an injection volume of 10 ml/kg. The dose given to all human subjects was 0.3 mg/kg (two thirds given as a bolus over a 90 s period, the remainder as a continuous infusion over 58.5 min).

Analgesia

Naïve mice were tested at each of four doses (5, 7.5, 10, or 15 mg/kg) for morphine, and each of three doses (5, 10, or 15 mg/kg) for M6G, to construct dose-response curves. Sample sizes in these experiments were n = 7–24/genotype/sex/dose. Mice were tested for baseline sensitivity on the 49°C tail withdrawal test, immediately injected with morphine or M6G, and then retested at 15, 30, and 60 min post-injection for morphine, and additionally at 120, 180, 240, and 300 min post-injection for the longer acting M6G.

Prior to drug infusion in humans, three baseline tests were performed. After the drug infusion was initiated at time t = 0, the pain tests were performed at the following times: t = 5, 10, 20, 30, 40, 50, 60, 65, 70, 80, 90, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190, 200, 220, 240, 260, 280, 300, 320, 340, and 360 min.

Pharmacokinetic analysis

Mice of both genotypes and both sexes were injected with morphine (10 mg/kg) or M6G (10 mg/kg), and left undisturbed in their home cages until sacrificed by decapitation at 15, 60, or 180 min post-injection. Trunk blood was collected on ice and immediately centrifuged for 10 min at 3500/min to isolate plasma, which was then stored at −20°C until analysis. Plasma morphine and M6G concentrations were determined with liquid chromatography tandem mass spectrometry. The lower and upper limits of quantification were set at 1.25 and 5000 ng/ml, respectively. Each drug/genotype/sex/time condition was run in triplicate.

At fixed times (t = 2, 5, 10, 20, 30, 40, 50, 60, 62, 65, 70, 80, 90, 105, 120, 150, 180, 240, 300, and 360 min after start of drug infusion), 5 ml of arterial blood in human subjects was drawn for determination of plasma concentrations of M6G. At time points where blood sampling coincided with pain assessment, the pain testing preceded the sampling. Finally, in each subject, an extra sample of blood was obtained for genotyping. Plasma M6G concentrations were determined as described above, with the coefficient of variation varying from 4 to 8% over the calibration range of 2–10 000 ng/ml.

Genotyping

The MC1R polymorphism detection strategy involved sequencing the 5′ two thirds of the open reading frame, where all of the known variants occur except for D294H, which was specifically screened for using a Taq1 loss of digestion strategy. See Mogil et al³ for details.

Data analysis

Nociceptive assay data in mice were analysed by two way ANOVA (genotype, sex). Where significant interactions were obtained, genotype or sex differences were evaluated by two tailed Student’s t test. Half maximal antinociceptive doses (AD₅₀s) and associated 95% confidence intervals were calculated from % analgesia (area under the curve) data using the method of Tallarida and Murray.⁷

For each human subject, the area under the time-effect curves (AUEC) was calculated using the trapezoidal rule and standardised by the length of the study (360 min). The data were separated by genotype with one group consisting of subjects with at least two variant alleles of the MC1R gene previously associated with MC1R loss of function (and, thus, red hair) (ins29, R151C, R160W, D294H) and the other group consisting of subjects with the consensus sequence, one variant allele only, or two variant alleles at amino acids not associated with MC1R loss of function (V60L, V92M, and R163Q)⁸ (see table 2 in Appendix 1 for details).

A two way ANOVA (genotype, sex) was performed on baseline pain, AUEC, and plasma M6G concentration data. Where significant interactions were obtained, genotype or sex differences were evaluated by two tailed Student’s t test.

RESULTS

As shown in fig 1, e/e mutants displayed reduced nociceptive sensitivity. On most assays, a significant effect of genotype was observed in both sexes; the exception was the tail withdrawal test, in which only male e/e mice displayed significantly higher latencies. The reduced sensitivity of e/e mutants did not appear to be opioid mediated, since the differences persisted after pretreatment with naloxone (10 mg/kg, s.c.) (data not shown).

Next, we constructed analgesic dose-response curves for e/e and C57BL/6 mice of both sexes given systemic morphine (5–20 mg/kg) and its bioactive metabolite, M6G (5–15 mg/kg). Although there has been some confusion in the literature, it is now quite clear that M6G acts largely at the μ-opioid receptor,⁹ albeit likely at one coded for by an alternatively spliced transcript of the Oprm gene.¹⁰ As shown in fig 2A and B for M6G (and table 1 in Appendix 1), e/e mutants of both sexes displayed significantly greater inhibition of thermal nociception from these drugs than did their C57BL/6 counterparts at all doses tested. The M6G data were particularly robust, with differences in both the peak analgesic effect and the duration of analgesia.

Given the strength of the M6G finding in mice, we attempted a translational genetic association study in humans with this clinically useful analgesic. A total of 22 subjects (11 men, 11 women) had two or more variant alleles of the MC1R gene in amino acids known to abolish MC1R functionality (R151C, R160W, and D294H); all of these subjects had red hair and fair (type I or II) skin. The remaining 25 subjects (12 men, 13 women) had zero or only one variant alleles at these loss of function loci, and thus possessed functional MC1Rs (see table 2 in Appendix 1), even though seven of these subjects also had red hair (presumably due to genes other than MC1R). Baseline pain tolerance differed significantly between genotypes, with greater currents tolerated by MC1R variant subjects, 20.9 (SEM 1.7) mA, compared to control subjects, 15.8 (1.2) mA (p = 0.018; fig 1F). There was no significant main effect of sex or genotype×sex interaction.

Analgesic responses after M6G were greater in MC1R variant subjects compared to control subjects. The areas under the time-effect (pain tolerance relative to baseline) curves were 1.18 (0.04) mA for control subjects versus 1.49 (0.09) mA for MC1R variant subjects (p = 0.003; fig 2D). These values indicate that M6G produced an average increase in tolerable current of 18% (or 4 mA) and 49% (or 10 mA) above baseline in control and MC1R variant subjects, respectively (fig 2E). There was no significant main effect of sex or genotype×sex interaction.

The genotypic differences observed were related to the pharmacodynamics of M6G acting at μ-opioid receptors, since plasma M6G concentrations did not differ between genotypes in either mice (fig 2C) or humans (fig 2F) at multiple time points post-M6G injection.

DISCUSSION

Adequate pain control in acute pain patients is of vital importance. Poor pain relief may result in complications and prolonged hospitalisation, and the patient may ultimately develop chronic pain or complex pain syndromes.¹¹ Pain control is often difficult due to large between-patient variability in pain responses and analgesic efficacy.¹² In this study, we demonstrate a significant link between MC1R gene sequence, pain tolerance, and efficacy of the μ-opioid, M6G. In striking agreement with the mouse mutant studies, we observed greater M6G induced analgesic responses in those red-haired humans who are also loss of function “mutants” at the MC1R gene.

The observation that in humans and mice, the M6G/MC1R interaction is sex independent contrasts with our earlier finding that MC1R status affected pentazocine (κ-opioid) analgesia in females only.³ Whether these are general differences or are restricted to the specific μ- and κ-opioids already tested requires further investigation. The decreased pain sensitivity in MC1R non-functioning humans and e/e mice compared to controls suggests that endogenous activation of the MC1R may counteract an endogenous pain inhibition system; that is, produce anti-analgesia. Our failure to observe this difference in basal pain sensitivity previously³ is likely due to a variable genotypic effect size across different pain modalities (thermal and ischemic pain³v electrical current pain).

Our choice to study the role of MC1R in M6G analgesia in our human population was obvious based on the results of the mouse studies. We believe that these data nicely illustrate the power of direct mouse to human translation in genetic studies of pain. Demonstration of the possible role of MC1R in human clinical pain awaits large scale genetic association studies. We suggest that such association studies are more profitably attempted after the successful demonstration of a pharmacogenetic effect against experimental laboratory pain, which can be more carefully controlled.

APPENDIX 1