Opiate receptor knockout mice define mu receptor roles in endogenous nociceptive responses and morphine-induced analgesia

Abstract

Morphine produces analgesia at opiate receptors expressed in nociceptive circuits. mu, delta, and kappa opiate receptor subtypes are expressed in circuits that can modulate nociception and receive inputs from endogenous opioid neuropeptide ligands. The roles played by each receptor subtype in nociceptive processing in drug-free and morphine-treated states have not been clear, however. We produced homologous, recombinant mu, opiate receptor, heterozygous and homozygous knockout animals that displayed approximately 54% and 0% of wild-type levels of mu receptor expression, respectively. These mice expressed kappa receptors and delta receptors at near wild-type levels. Untreated knockout mice displayed shorter latencies on tail flick and hot plate tests for spinal and supraspinal nociceptive responses than wild-type mice. These findings support a significant role for endogenous opioid-peptide interactions with mu opiate receptors in normal nociceptive processing. Morphine failed to significantly reduce nociceptive responses in hot plate or tail flick tests of homozygous mu receptor knockout mice, and heterozygote mice displayed right and downward shifts in morphine analgesia dose-effect relationships. These results implicate endogenous opioid-peptide actions at mu opiate receptors in several tests of nociceptive responsiveness and support mu receptor mediation of morphine-induced analgesia in tests of spinal and supraspinal analgesia.

Figure 1

Construction of μ opiate receptor knockout mice. (A) Representation of 5′ portions of the murine μ opiate receptor gene, with positions of exon I (closed box), start codon (ATG), BamHI (B), EcoRV (RV), BglII (BII), and EcoRI (RI) restriction endonuclease sites noted. (Bar = 1 kb.) (B) Representation of the pμKO2 targeting vector indicating phosphoglycerate kinase neomycin resistance gene (neo^r) and MC1 thymidine kinase (tk) sequences. The directions of gene transcription are marked by horizontal arrows. Abbreviations and scale as in A. (C) Representation of the predicted mutant allele resulting in the disrupted μ receptor gene. The location of the 5′ and 3′ probes used in the Southern blot analyses are indicated. The putative promoter region and first exon are missing from the mutant allele. Abbreviations and scale as in A and B. (D) Southern blot analysis using a 3′ or 5′ probe to BamHI-digested tail DNA extracted from wild-type (+/+), heterozygote (+/−), and homozygote (−/−) mice. The presence of a single 20-kb fragment indicates a homozygous mutant (−/−) genotype. Wild-type fragments identified by 3′ and 5′ probes are 15 and 5 kb, respectively. (E) Scatchard analyses of saturation radioligand binding data using [³H]DAMGO and membranes prepared from whole brain minus cerebellum specimens from 7-week-old animals. Mean (±SEM; n = 4) values for B_max were 106 ± 13 and 57 ± 11 fmol/mg protein and undetectable for wild-type (+/+), heterozygote (+/−), and homozygote (−/−) μ receptor knockout mice, respectively. Dissociation constant values (K_D) were 0.48 and 0.5 nM and undetectable, respectively. (F) Immunostaining of μ receptor protein in dorsal horn sections through the lumbosacral spinal cord in μ receptor homologous recombinant mice. Darker immunoreactivity is found in superficial laminae of the wild-type (+/+) dorsal horn than in heterozygote (+/−) > homozygote (−/−) mice in two separate experiments. This staining was eliminated by antiserum preabsorption by the peptide to which the antibody was raised but not by preabsorption by an irrelevant peptide (data not shown).

Figure 2

Latencies for nociceptive responses in tail flick (TF) and hot plate (HP) tests in unpretreated mice. Mice of +/+ (n = 14), +/− (n = 24), and −/− (n = 15) μ receptor genotypes underwent tail flick testing in 50 or 53°C water and hot plate testing at 52 or 55°C, as indicated. ∗, P < 0.05 compared with wild-type control values.

Figure 3

Latencies for nociceptive responses in 53°C tail flick and 55°C hot plate tests in pretreated mice. (A) Dose–response relationships for morphine-induced alterations in latencies on 53°C tail flick testing in mice with wild-type (+/+), heterozygote (+/−), and homozygote (−/−) μ opiate receptor genotypes using a cumulative dose–response paradigm as described. Percentage of maximal analgesia was calculated for each mouse as: 100 × {[(latency to tail flick after morphine) − (latency to tail flick at baseline)]/[(15-sec cutoff time) − (baseline latency)]}. ∗, P < 0.05 compared with preinjection control values for the appropriate genotype. Dose–effect relationships were significant for +/+ and +/− mice but not for −/− mice. Among genotype differences, dose–response relationships also were significant for animals of each genotype [P < 0.001, df(2, 120), F = 66 by repeated measures ANOVA]. (B) Dose–response relationships for morphine-induced alterations in latencies on 55°C hot plate testing in mice with wild-type (+/+), heterozygote (+/−), and homozygote (−/−) μ opiate receptor genotypes using a cumulative dose–response paradigm as described. Percentage of maximal analgesia was calculated for each mouse using a 30-sec cutoff time. ∗, P < 0.05 compared with preinjection control values for the appropriate genotype. Dose–effect relationships were significant for +/+ and +/− mice but not for −/− mice. Among genotype differences, between dose–response relationships also were significant for animals of each genotype [P < 0.001, df(2, 124), F = 27 by repeated measures ANOVA]. Max., maximum.