Address correspondence and reprint requests to Dr. C.A. Frye at Department of Psychology, The University at Albany–SUNY, 1400 Washington Avenue, Albany, NY 12222, U.S.A. E-mail: email@example.com
Summary: Purpose: To investigate the importance of progesterone (P4) metabolism by the 5α-reductase type I enzyme in mitigating P4 antiseizure effects.
Methods: Ovariectomized, female homozygous and heterozygous 5α-reductase type I knockout mice (n = 23) and their wild-type siblings (n = 31) were administered P4 (1.0 mg), and their pentylenetetrazol (PTZ)-induced ictal behaviors were compared with those of vehicle-administered mice (n = 49).
Results: Mice deficient in the 5α-reductase type I enzyme administered P4, or vehicle-administered control mice, had significantly shorter latencies and increased incidence of PTZ-induced hindlimb extension and death than did wild-type mice administered P4.
Conclusions: These data suggest that P4's metabolism by the 5α-reductase type I enzyme may mitigate some of P4's antiseizure effects in the PTZ-induced seizure model.
Clinical evidence suggests that progesterone (P4) has antiseizure effects. Seizure incidence is often increased during the perimenstrum (i.e., as is the case with catamenial epilepsy), when P4 levels are relatively low; typically less seizure activity occurs midcycle, when P4 concentrations are higher. Progesterone therapy is successfully used to manage seizure disorders among some women with epilepsy (1).
There is evidence to suggest that a metabolite of P4 may mediate some of P4's antiseizure effects. Both P4 and its 5α-reduced metabolite, 3α-hydroxy-5α-pregnan-20-one (3α,5α-THP) have anticonvulsant effects. In rodents, when P4 or 3α,5α-THP is administered systemically, kainic acid-, pentamethylenetetrazol (metrazol)-, or perforant pathway stimulation–induced seizures are attenuated (2,3). There is a stronger negative correlation between circulating or central 3α,5α-THP concentrations and seizure threshold than for P4 and seizures. For example, central 3α,5α-THP concentrations correlate better with seizures than do P4 concentrations in naturally cycling or hormone-administered rats (4,5). Progesterone or 3α,5α-THP administration to rats produces similar thresholds for chemical or electroconvulsant-induced seizures, yet P4 administration increases central concentrations of P4 and 3α,5α-THP, but in 3α,5α-THP-administered rats, only 3α,5α-THP and not P4 concentrations are increased (2). Decreased seizure susceptibility in women with catamenial epilepsy is better correlated with increases in 3α,5α-THP rather than increases in P4 itself (6). The latency for P4 anticonvulsant effects is longer for P4 than for 3α,5α-THP. More than 50 years ago, Selye (3) reported that P4 protected rodents from pentylenetetrazol (PTZ)-induced seizures (3). Progesterone's anticonvulsant effects had a longer latency than did the anticonvulsant effects of 3α,5α-THP, which implies that the P4 antiseizure activity may require sufficient time to allow its metabolism to 3α,5α-THP.
The importance of P4 metabolism in ictal behavior was examined in 5α-reductase knockout mice. Two isoforms of the 5α-reductase enzyme have been cloned and their distribution characterized (7,8). The type I isoform is constitutively expressed in the rodent central nervous system (CNS) at all stages of brain development. The type II isoform is transiently expressed in late fetal/early postnatal life (7). 5α-Reductase knockout mice are a valuable research tool in which the expression of the type I 5α-reductase gene is perturbed, which results in deficiencies in its protein product, the 5α-reductase enzyme (9), and thereby 3α,5α-THP. This experiment tested the hypothesis that wild-type mice administered P4 would show decreased ictal activity compared with 5α-reductase knockout mice administered P4, whose ictal activity was expected to be comparable to that of vehicle-administered control mice (if the P4 antiseizure effects require metabolism to 3α,5α-THP).
All methods were preapproved by the Institutional Animal Care and Use Committee. B6, 129S-srd1-al tmi mahe female mice were derived from 10 breeder pairs (homozygous –/– father and heterozygous +/– mother) obtained from Jackson Laboratories (Bar Harbor, ME, U.S.A.). The offspring cannot be differentiated by their phenotype; thus polymerase chain reaction was used to determine the genotype of resulting offspring (10). Subjects were either homozygous or –/– for the 5α-reductase type I enzyme (n = 15), heterozygous or –/+ for the 5α-reductase type I enzyme (n = 19), or were wild-type or +/+ for the 5α-reductase type I enzyme (n = 69). Mice were group housed until surgery, after which time, mice were individually housed in a room on a 12:12-h reversed light/dark cycle with continuous access to Purina Rat Chow and water in their cages.
Mice were ovariectomized when anesthetized with sodium pentobarbital (70 mg/kg or to effect). A minimum of 2 weeks later, mice were injected s.c. with P4 (1,000 μg in 0.1 ml) or vehicle (sesame oil) 1 h before behavioral testing. Within each genotypic group, mice were randomly assigned to receive P4 or vehicle. Of the 15 mice that were homozygous or –/– for the 5α-reductase type I enzyme, 12 were administered P4, and three received vehicle. Of the 19 mice that were heterozygous or –/+ for the 5α-reductase type I enzyme, 11 were administered P4, and eight received vehicle. Of the 69 mice that were wild-type or +/+ for the 5α-reductase type I enzyme, 31 were administered P4, and 38 received vehicle.
One hour after P4 or vehicle administration, mice were placed in a plastic arena (23.5 × 20.5 × 19.5 cm) to habituate; this was followed by PTZ administration (85 mg/kg, i.p.). Immediately after PTZ injection, ictal behaviors were recorded for 10 min, and whether death occurred was monitored for 30 min. Ictal activity included the latency to, and incidence of, tonic–clonic seizures, hindlimb extension, and death.
There were 31 wild-type mice that were administered P4 (progesterone wild-type mice). There were no statistically significant differences between the behavior of the homozygous (n = 12) and heterozygous (n = 11) mice that were administered P4; therefore, these groups were combined to constitute the 5α-reductase knockout mice administered P4 (n = 23; progesterone knockout mice). There were also no statistically significant differences in the behavior of the mice that were administered vehicle; thus the 38 wild-type, the eight heterozygotes, and the three homozygotes were combined to constitute the control group of vehicle-administered mice (n = 49; vehicle control). The percentage of mice that displayed tonic–clonic seizure, hindlimb extension, or death was compared across groups using the nonparametric statistics (the sign test). The latencies to tonic–clonic seizure, hindlimb extension, or death were compared across groups using one-way analysis of variance and post hoc tests to determine group differences. Alpha level for statistical significance was p ≤ 0.05.
The percentage of wild-type mice administered P4 that exhibited hindlimb extension (sign test, p < 0.05; see Fig. 1, middle left) or that died of PTZ-induced seizures (sign test, p < 0.05; see Fig. 1, bottom left) was significantly lower compared with the percentage of mice deficient in 5α-reductase that were administered P4 or the percentage of vehicle-administered control mice.
Progesterone administration to wild-type mice significantly increased the latencies to tonic–clonic seizures [F(2, 99) = 2.935, p = 0.05; see Fig. 1, top left], hindlimb extension [F(2, 99) = 3.298, p = 0.04; see Fig. 1, middle left], and death [F(2, 99) = 10.267, p = 0.001; see Fig. 1, bottom left], compared with mice deficient in 5α-reductase that were administered P4 or vehicle-administered control mice.
These findings supported our hypothesis that P4 administered to wild-type mice would significantly reduce ictal activity compared with that of 5α-reductase knockout mice administered P4, whose behavior was comparable to vehicle-administered control mice. In support, P4 given to mice deficient in the type I 5α-reductase enzyme decreased the latency and increased the incidence of PTZ-induced hindlimb extension and death compared with wild-type mice administered P. Wild-type mice administered P4 had longer latencies to hindlimb extension and death than did vehicle-administered control mice, which were not different from the 5α-reductase knockout mice administered P4. These data suggest that, in mice, P4's metabolism by the 5α-reductase type I enzyme may mitigate some of P4's antiseizure effects in the PTZ-induced seizure model.
The present findings are consistent with previous data suggesting that P4's metabolism by 5α-reductase is important for P4's antiseizure effects. Reducing P4 metabolism in rats by administering 5α-reductase inhibitors, such as 4MA or finasteride, reduces 3α,5α-THP concentrations and attenuates antiseizure effects of P4 in kainic acid and perforant-pathway seizure models (4,11). Clinical findings provide additional support that P4's metabolism by 5α-reductase is important for its antiictal effects. There is a case study of a woman with catamenial epilepsy whose seizures were well controlled with P4 therapy, but when treated for pattern baldness with the 5α-reductase inhibitor, finasteride, experienced precipitous increases in seizures that abated when finasteride treatment was discontinued (A.G. Herzog, unpublished observations).
The antiseizure effects of acute administration of P4 in the present experiment may provide insight regarding P4's mechanism of action. 3α,5α-THP has its actions through γ-aminobutyric acid (GABA)A–benzodiazepine receptor complexes (GBRs), which have been implicated in epilepsy; whereas P4 typically has actions at intracellular progestin receptors (PRs), which are not usually associated with epilepsy. Progesterone has a high affinity for intracellular PRs, but 3α,5α-THP is devoid of activity for PRs, which suggests that the effects of P4 and 3α,5α-THP on seizure threshold are not the sole consequence of actions at traditional intracellular PRs (12). Many antiepileptic drugs (AEDs) used therapeutically have their effects by increasing the activity of the primary inhibitory neurochemical in the brain, GABA (13). Progestins can act at GBRs; however, P4 itself binds only weakly to GBRs (14). When P4 is metabolized by the 5α-reductase enzyme to 3α,5α-THP, it is 500 times more potent than P4, and 50 times more potent than diazepam at modulating GBRs (15).
In summary, mice deficient in 5α-reductase activity did not show the reduction in ictal activity that wild-type mice did when administered P4. The ictal behavior of knockout mice administered P4 was not different from that of vehicle-administered control mice. These data suggest that, in mice, P4's metabolism by the 5α-reductase type I enzyme may mitigate some of P4's antiseizure effects in the PTZ-induced seizure model.
Acknowledgment: This research was supported by The National Science Foundation (95-14463; 98-96263), The Epilepsy Foundation of America, and The Donaghue Foundation for Biomedical Research.