First demonstration that brain CYP2D-mediated opiate metabolic activation alters analgesia in vivo

Abstract

The response to centrally acting drugs is highly variable between individuals and does not always correlate with plasma drug levels. Drug-metabolizing CYP enzymes in the brain may contribute to this variability by affecting local drug and metabolite concentrations. CYP2D metabolizes codeine to the active morphine metabolite. We investigated the effect of inhibiting brain, and not liver, CYP2D activity on codeine-induced analgesia. Rats received intracerebroventricular injections of CYP2D inhibitors (20 μg propranolol or 40 μg propafenone) or vehicle controls. Compared to vehicle-pretreated rats, inhibitor-pretreated rats had: (a) lower analgesia in the tail-flick test (p<0.05) and lower areas under the analgesia-time curve (p<0.02) within the first hour after 30 mg/kg subcutaneous codeine, (b) lower morphine concentrations and morphine to codeine ratios in the brain (p<0.02 and p<0.05, respectively), but not in plasma (p>0.6 and p>0.7, respectively), tested at 30 min after 30 mg/kg subcutaneous codeine, and (c) lower morphine formation from codeine ex vivo by brain membranes (p<0.04), but not by liver microsomes (p>0.9). Analgesia trended toward a correlation with brain morphine concentrations (p=0.07) and correlated with brain morphine to codeine ratios (p<0.005), but not with plasma morphine concentrations (p>0.8) or plasma morphine to codeine ratios (p>0.8). Our findings suggest that brain CYP2D affects brain morphine levels after peripheral codeine administration, and may thereby alter codeine's therapeutic efficacy, side-effect profile and abuse liability. Brain CYPs are highly variable due to genetics, environmental factors and age, and may therefore contribute to interindividual variation in the response to centrally acting drugs.

Figure 1

Inhibiting brain CYP2D reduced the first 60 min of codeine-induced analgesia. Compared to vehicle pretreatment, propranolol pretreatment resulted in (A) significantly lower percentage of maximal analgesic effect (%MPE) at 15, 20, 30 and 40 min after codeine injection (n=4-16/time point), and significantly lower area under the analgesia-time curve (AUC) for (B) 0-30 min and (C) 0-60 min, but not for (D) 60-120 min or 0-120 min after codeine injection (n=16). Compared to vehicle pretreatment, propafenone pretreatment resulted in (E) significantly lower %MPE at 20, 30 and 40 min after codeine injection (n=7-12/time point), and significantly lower AUC for (F) 0-30 min and (G) 0-60 min, but not for (H) 60-120 min or 0-120 min after codeine injection (n=12). Error bars indicate SEM. *p<0.05, **p<0.01 using a paired t-test. Veh=vehicle, PL=propranolol, PF=propafenone.

Figure 2

Inhibiting brain CYP2D did not affect morphine-induced analgesia. Compared to vehicle pretreatment, propranolol pretreatment did not result in (A) significantly different percentage of maximal analgesic effect (%MPE) after morphine injection (p>0.06 at all time points, n=12), or significantly different area under the analgesia-time curve (AUC) for (B) 0-30 min, (C) 0-60 min, (D) 60-120 min or 0-120 min after morphine injection (p>0.4 for all time periods, n=12). Compared to vehicle pretreatment, propafenone pretreatment did not result in (E) significantly different %MPE after morphine injection (p>0.09 at all time points, n=6), or significantly different AUC for (F) 0-30 min, (G) 0-60 min, (H) 60-120 min or 0-120 min after morphine injection (p>0.2 for all time periods, n=6). Error bars indicate SEM. Paired t-tests were used. Veh=vehicle, PL=propranolol, PF=propafenone.

Figure 3

Inhibitor-pretreated rats had lower morphine levels in the brain but not in plasma at 30 min after codeine injection. Compared to vehicle-pretreated rats, propranolol-pretreated rats had significantly lower morphine concentrations in (A) the brain but not in (B) plasma at 30 min after codeine injection (n=11/group), and significantly lower morphine to codeine ratios in (C) the brain but not in (D) plasma at 30 min after codeine injection (n=11/group). Compared to vehicle-pretreated rats, propafenone-pretreated rats had significantly lower morphine concentrations in (E) the brain but not in (F) plasma at 30 min after codeine injection (n=8/group), and significantly lower morphine to codeine ratios in (G) the brain but not in (H) plasma at 30 min after codeine injection (n=8/group). Error bars indicate SEM. *p<0.05, **p<0.01 using an unpaired t-test. Veh=vehicle, PL=propranolol, PF=propafenone.

Figure 4

Analgesia correlated with brain, and not plasma, morphine levels at 30 min after codeine injection. Percentage of maximal analgesic effect (%MPE) trended toward correlating with morphine concentration in (A) the brain (p=0.07, n=20) and not in (B) plasma (p>0.8, n=21). Percentage of maximal analgesic effect correlated significantly with morphine to codeine ratios in (C) the brain (p<0.004, n=20) and not in (D) plasma (p>0.8, n=21). Pearson correlation coefficients were used.

Figure 5

Propranolol-pretreated rats did not have lower morphine levels in the brain at 60 or 90 min after codeine injection. At both 60 min (n=7/group) and 90 min (n=8/group) after codeine injection, there was no significant difference in (A) brain morphine concentrations (p>0.1, p>0.9, respectively) or (B) brain morphine to codeine ratios (p>0.4, p>0.2, respectively) between propranolol-pretreated rats and vehicle-pretreated rats. Error bars indicate SEM. Unpaired t-tests were used. Veh=vehicle, PL=propranolol.

Figure 6

Inhibiting brain CYP2D in vivo lowered ex vivo codeine metabolism to morphine by brain membranes but not by liver microsomes. (A) In brain membranes incubated with 500 μM codeine, velocity of morphine formation was significantly lower in propranolol-pretreated rats than in vehicle-pretreated rats (n=8/group). (B) In liver microsomes incubated with 500 μM codeine, there was no significant difference in the velocity of morphine formation between propranolol-pretreated rats and vehicle-pretreated rats (n=8/group). Error bars indicate SEM. *p<0.05 using an unpaired t-test. Veh=vehicle, PL=propranolol, V=velocity of morphine formation.