Address correspondence and reprint requests to Dr. T. N. Seyfried at Department of Biology, Chestnut Hill, MA 02167, U.S.A.email@example.com
Summary: Purpose: The EL mouse is an animal model for multifactorial idiopathic epilepsy. Although EL mice have been studied extensively for >45 years, the etiology of male sudden death and its relation to seizures have not been defined. Here we investigated the cause of EL male sudden death and its relation to epilepsy.
Methods: For histopathologic analysis, the terminally ill EL mice (n = 15) were killed, and the tissues were fixed. Blood chemical composition was compared between the terminally ill EL (n = 9) and the healthy age-matched EL (n = 17) and DDY (n = 11) males. To determine the effect of the ketogenic diet (KD) on sudden male death, young male EL mice (P30) were randomly separated into two groups that were fed ad libitum with either Agway lab chow (control n = 38) or with the KD (treated, n = 39) for 5 months. The genetic predisposition to sudden death was analyzed in the backcross generation (n = 106) of a cross between EL and the nonepileptic ABP strains.
Results: Sudden death coincided with the onset of seizures (70–80 days) and affected 94% of male EL mice by age 300 days. Urethral plugs were observed histologically in 13 of 15 longitudinally sectioned penises. Concentrations of blood urea nitrogen, creatinine, phosphorus, and calcium in the terminally ill mice were significantly elevated when compared with those of healthy animals. None of the mice treated with the KD experienced sudden death, whereas 15 (39%) of the untreated control mice died by age 5 months. The sudden death in male EL mice was inherited as an autosomal recessive sex-limited lethal trait.
Conclusions: The cause of sudden death in male EL mice arises from abnormal ejaculation, which produces a urethral plug with consequent urinary retention and acute severe uremia. The coincident onset of seizures and sudden death in EL males suggests that a sexual dysfunction is associated with epilepsy in this model.
The incidence of sexual dysfunction (SD) is higher in patients with epilepsy than in patients with other chronic neurologic diseases, and among those with epilepsy, SD is most frequent in patients with partial seizures and complex limbic seizures (1–4). About 30–60% of men with epilepsy have various abnormalities of sexual behavior including diminished sexual libido and/or potency, abnormal morning penile tumescence, and erectile dysfunction (1,2,3,4,5). Recent studies have shown that the sexual dysfunction in those with epilepsy results from physiologic disturbances associated with recurrent seizures rather than from psychological illness (5). Seizure activity in limbic structures may alter the neuronal regulation of sexual response and may affect the hypothalamic–pituitary axis, which regulate gonadal function (4). Investigations of sexual dysfunction in epilepsy have been difficult because of the confounding effects of antiepileptic drugs and the subjective nature of patient behavioral analysis (1,2,4). The absence of an animal model for sexual dysfunction associated with epilepsy has restricted research in this area.
The epileptic EL mouse has been studied extensively as a natural model for human idiopathic epilepsy and complex partial seizures (6,7). Seizures in EL mice originate in or near the parietal lobe and then spread to the hippocampus and to other brain regions (8,9). The seizures are accompanied by EEG abnormalities, vocalization, incontinence, loss of postural equilibrium, excessive salivation, hippocampal gliosis, and head, limb, and chewing automatisms (6). Phenytoin (PHT) and phenobarbital (PB) can effectively reduce seizure susceptibility (10–13), and the high-fat, low-carbohydrate ketogenic diet (KD) can inhibit epileptogenesis in EL mice (14). As with inherited human idiopathic epilepsies, gene–environmental interactions play a significant role in the determination of seizure frequency and onset in EL mice (7).
In addition to seizures, male EL mice experience a high frequency of sudden death (15,16). We present evidence that the sudden death in male EL mice is not directly caused by seizures and arises from abnormal ejaculation. The sudden death is inherited as an autosomal recessive sex-limited lethal trait and can be prevented by a high-fat KD, which inhibits epileptogenesis in EL mice. A preliminary report of these findings has appeared (17).
MATERIALS AND METHODS
The inbred EL/Suz (EL) mice and the nonepileptic DDY mice were originally obtained from J. Suzuki (Tokyo Institute of Psychiatry) and from Clea, Japan, respectively. Both strains are now available from The Jackson Laboratory (Bar Harbor, ME, U.S.A.). The inbred nonepileptic ABP/LeJ (ABP) strain was purchased from the Jackson Laboratory. Mice were housed in plastic cages with Sani-chip bedding that was changed once per week. Male mice used for the studies were not bred and were housed together with one to three other males. No evidence of fighting between the male mice was observed throughout the study. The mice were kept on a 12-h light/dark cycle with food and water provided ad libitum. All mice used were maintained in the Boston College Animal Care Facility, and the procedures for their use were in strict accordance with the NIH Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care Committee. EL and ABP mice were crossed to produce reciprocal F1 hybrids. The 106 ABPELF1 × EL backcross mice were produced by reciprocal crosses of ABPELF1 and EL mice. Mice were carefully examined twice a day, usually at 8 a.m. and 8 p.m., and mice that had become suddenly terminally ill or had died in a 12-h period were identified. We identified the terminally ill mice by their nonambulatory state. All terminally ill and dead mice had distended bladders, as detected by abdominal palpation and necropsy, respectively, and did not display any signs of illness during the previous scheduled observation. All terminally ill mice usually died within 1–2 h. Before death, these terminally ill mice were anesthetized, and their blood was collected. The mice were then killed with CO2, and their tissues were collected for histologic analysis. Blood-sample collection and histologic analysis were not performed on dead mice identified during surveillance because of the poor quality of the tissues. Considering that all mice that became suddenly terminally ill died shortly after their identification and had distended bladders similar to those seen in the dead mice, we can extend our finding in terminally ill mice to mice that were found dead.
The mice were tested for seizures by using a gentle handling procedure, as previously described, and were characterized as seizure susceptible if they experienced a generalized seizure (7). This involved loss of postural equilibrium accompanied by excessive salivation and head, limb, and chewing automatisms. Mice that expressed vocalization or twitching that did not progress to a generalized seizure, were not considered susceptible.
The KD was obtained from Purina Laboratories (Richmond, IN), as we recently described (14), and contained the following ingredients: vitamin-free casein, 15%; Solca Floc (nonnutritive fiber), 3%; RP Vitamin Mix with 3.2 % fiber carrier, 3.3%; RP Mineral Mix no. 10 with 2.1% fiber carrier, 8.25%; DL-methionine, 0.12%; choline chloride, 0.33%; and lard with 0.02% ethoxyquin as antioxidant, 70%. The KD contained no carbohydrates and had 4.75:1 ketogenic (fats/proteins + carbohydrates) ratio. The diet caloric value was 6.85 Kcal/g; fat and protein comprised 90% and 10% of the calories, respectively. The Agway lab chow (control diet) was obtained from Agway Inc., Syracuse, NY. According to the manufacturer's specification, this diet delivers 4.4 Kcal/g gross energy; fat, carbohydrate, protein, and fiber composed 5.5%, 52.0%, 22.5%, and 4.5% of the diet, respectively. The effect of the KD on EL sudden death was evaluated in littermates fed ad libitum either the KD (n = 39) or the Agway control diet (n = 38) for 5 months starting at age 30 days. The χ2 test with a 2 × 2 contingency table was used to evaluate the effect of the KD on sudden death.
Blood urea nitrogen and electrolytes
An automatic analyzer (Hitachi 747) was used to determine the levels of urea nitrogen and electrolytes in blood according to standard methods (18). Blood samples were collected from suddenly terminally ill mice before death and from age-matched healthy EL and DDY male mice. All mice were anesthetized with isoflurane (Halocarbon Laboratories, River Edge, NJ, U.S.A.), and the blood was collected in heparinized tubes by puncture of the retroorbital sinus.
For gross pathologic analysis, suddenly terminally ill male EL mice were killed, and necropsies were performed. For histopathologic analysis, tissue was fixed by immersion into 10% buffered formalin. The urogenital system was removed, processed, and then embedded in paraffin. Paraffin-embedded tissues were cut to 5-μm sections and stained with hematoxylin and eosin. Fifteen terminally ill and six healthy age-matched control mice were used for histologic analysis.
Genetic marker analysis
Mouse genomic DNA was isolated from mouse tails, as we previously described (19). Simple sequence-length polymorphisms (SSLPs) were used to identify genetic loci associated with sudden death. SSLPs were characterized by the polymerase chain reaction (PCR) with oligonucleotide primers purchased from Research Genetics, Carlsbad, CA. PCR products were electrophoresed on 6% polyacrylamide gel. The gels were exposed to x-ray film overnight at –80°C. Two individuals manually typed the SSLP markers. The χ2 analysis was used to determine the probability that the distribution of genotype within the “dead” subgroup of mice differed significantly from the expected distribution, assuming independent segregation of alleles.
Development of sudden death and seizure susceptibility in male EL mice
EL male mice appeared healthy and did not experience seizures during testing at age 30 days. The first male EL mouse died suddenly at age 71 days. The incidence of sudden death gradually increased with age, and by 300 days, 74 (94%) of the 79 male mice were dead (Fig. 1A). All these mice became suddenly terminally ill and died without signs of chronic disease or seizure. The sudden death was limited to males, as no female mice became suddenly terminally ill and died during the course of the study.
To determine the developmental relation between sudden death and seizure onset, we tested naive male EL mice for seizures at age 30, 67, 120, 150, and 180 days. Each mouse was tested only once. Seizure susceptibility in EL mice increased with age, and by 67 days, 42% of EL males experienced seizures (Fig. 1B). These findings indicate that the onset of seizure susceptibility coincided with the onset of sudden death.
Blood urea nitrogen (BUN) and blood electrolytes
Serum levels of BUN, creatinine, and phosphorus were significantly higher in male EL mice that became suddenly terminally ill than in DDY males or in EL males that appeared healthy (Table 1). The BUN and electrolyte levels were similar in age-matched healthy DDY and EL males. The values in these healthy mice also are similar to those reported previously in healthy C57BL/6 mice (20). All of the identified terminally ill and dead male EL mice had markedly distended bladders. These findings indicate that male EL mice died of acute severe uremia.
Asterisks indicate that the plasma component was significantly higher in the obstructed mice than that in the healthy DDY and EL control mice: p < 0.0001 (two-tailed Student's t test).
Mice were between ages 76 and 401 days.
Healthy mice were active, responsive, and ambulatory.
EL mice with urinary retention were nonambulatory.
n, the number of independent mice analyzed.
20.72 ± 0.89
27.29 ± 1.76
246.9 ± 9.2*
0.14 ± 0.02
0.15 ± 0.02
1.72 ± 0.19*
8.08 ± 0.36
7.78 ± 0.81
25.76 ± 1.36*
9.85 ± 0.08
9.45 ± 0.18
10.08 ± 0.36
Gross morphology and histology of the urogenital system
The urogenital system of terminally ill male EL mice (n = 15) exhibited marked distention of the bladder and mild hydronephrosis. Cystic distention was relieved by severing the distal pelvic urethra, which remained dilated after voiding. Hemorrhage was often observed in the urinary bladder and in seminal vesicles. Marked congestion was present in the erectile tissues (i.e., corpus cavernous urethrae) enclosing the urethral diverticula. Urethral plugs were observed histologically in 13 of 15 longitudinally sectioned penises. We believe the absence of urethral plug in two of the examined terminally ill mice was due to poor tissue quality and technical difficulties during tissue sectioning. Plugs filled the lumen of the penile urethras and were observed to extend from the pelvic flexure to the external orifice and into the urethral diverticula (Fig. 2A and B). The plugs were composed of an acellular, eosinophilic coagulum and contained entrapped spermatozoa. Plugs also were associated with bacterial infection and neutrophil infiltration (Fig. 3). As a result of marked bladder distention, kidneys showed signs of mild hydronephrosis associated with atrophy of the renal papilla (Fig. 4).
To confirm the positive association between urethral plugs and sudden death, we analyzed the urogenital system of six healthy male EL mice at risk of developing the obstruction. No urethral plugs or signs of obstructive uropathy were found on histopathologic analysis of these mice (data not shown). These findings indicate that sudden death in male EL mice arises from abnormal ejaculation, which produces a urethral plug with consequent urethral obstruction and acute severe uremia.
Genetics of sudden male death
Male EL, ABP, ABPELF1, and ABPELF1 × EL backcross mice were monitored for 300 days beginning at age 30 days to evaluate the role of genetic factors in the sudden death. Male ABP and ABPELF1 mice appeared healthy throughout the study and did not died suddenly. Conversely, male ABPELF1 × EL (AEE) backcross mice began to die suddenly at age 63 days and, by age 300 days, 22 (43%) of these 51 mice were dead (Fig. 5). As seen with male EL mice, the male AEE backcross mice that became suddenly terminally ill and died had markedly distended bladders and hemorrhages in seminal vesicles. None of the female AEE backcross mice died suddenly or became terminally ill. The number of dead males found in the AEE backcross did not differ significantly from the number expected (25.5), based on an autosomal recessive mode of inheritance (χ2, 1.23; p > 0.2, df = 1). Taken together, these findings indicate that the sudden death is inherited as an autosomal recessive sex-limited lethal trait.
Because seizure onset and sudden death were developmentally coincident, we tested the association between sudden death and previously mapped chromosomal loci that influence epilepsy in EL mice (21,22). None of the identified EL seizure-susceptibility loci (El1, El2, and El3) were associated with sudden death in male EL mice (Table 2).
Table 2. Tests of the associations of sudden death in the ABPELF1 × EL backcross population with seizure-susceptibility loci
The genotypes of ABPELF1 × EL backcross male mice were determined at seizure-susceptibility loci.
b The χ2 (1 df) was used to determine that the distribution of the genotypes in “dead” subgroup of mice did not differ significantly from the expected distribution, assuming independent assortment of alleles.
Influence of the KD on sudden male death
We recently showed that the high-fat KD could retard the incidence and onset of seizures in male EL mice (14). Based on these observations, we asked if the KD might also influence sudden death in EL males. With exception of oily-appearing fur, the KD-fed males appeared active and healthy throughout the study. After 5 months of treatment, the incidence of sudden death was significantly lower in the male EL mice on the KD diet (none of 39) than in the male mice on the control diet (39%, 15 of 38; p < 0.001) (Fig. 6). These findings show that the KD prevents sudden death in EL males.
In this study we showed that sudden death in male epileptic EL mice results from acute uremia due to obstructive uropathy. In contrast to previous suggestions, sudden death in these mice was not directly caused by clinical seizures (16). Male mice normally emit a viscous sperm-containing fluid during copulation. The amino acid composition of this fluid is similar to that of the copulatory vaginal plug (23,24). The incomplete emission of this copulatory fluid in male EL mice produces a solidified obstruction (urethral plug) in the distal end of the urogenital tract. As a result, all male EL mice that became suddenly terminally ill and died had distended bladders. The urethral plugs were present in the majority of terminally ill EL males, but were absent in healthy EL males. The marked elevation of BUN and blood electrolytes observed in all terminally ill mice also is consistent with physical blockage of the urethra, which prevented voiding and caused acute severe uremia in the affected EL males. These findings concur with those in a previous pathologic study of the EL mouse (15). The presence of bacterial colonies suggests that the plug was formed before death and is not the result of terminal ejaculation. These data suggest that male EL mice experience sexual dysfunction in addition to seizures.
The EL mouse is an animal model for limbic complex partial seizures, and the onset of seizures and obstructive uropathy with consequent death are developmentally coincident in these mice. Interestingly, the highest incidence of sexual dysfunction in humans with epilepsy occurs in association with limbic partial seizures (1,4). Our findings suggest that the idiopathic epilepsy and limbic seizures in EL mice are associated with male sexual dysfunction and that the EL mouse may be a useful animal model for studying mechanisms and potential treatments for sexual dysfunction in humans with epilepsy.
Obstructive uropathy and sudden death were reported previously in long-term studies of male mice. In contrast to our findings in EL males, the obstructive uropathy reported in these studies generally occurred in old mice and was found in association with large urine volumes, male fighting, self-mutilation, or ether anesthesia (25–29). Abnormalities in ejaculation or urogenital reflexes were proposed as the cause of obstructive urethral plugs in the KK-Ay and STR/1N mouse strains, respectively (30,31). However, EL is the only known strain in which abnormal ejaculation and obstructive uropathy occur at sexual maturity and are associated with epileptic seizures.
We found that the sudden death in male EL mice is inherited as an autosomal recessive sex-limited lethal trait. It is common to find inherited traits that are either autosomal recessive, sex-limited, or lethal, but it is rare to find an inherited trait that is autosomal recessive, sex-limited, and lethal (32). Additionally, the inheritance of obstructive uropathy and sudden death in EL mice was not associated with the seizure-susceptibility genes (El1, El2, and El3) previously found in these mice (21,22). These findings suggest that different genetic mechanisms determine seizure susceptibility and obstructive uropathy in male EL mice or that a new seizure locus, not previously identified, is associated with both seizures and obstructive uropathy. We do not, however, exclude the possibility that recurrent seizure activity influences urogenital reflexes in male EL mice. Thus the expression of a conditional mutation affecting the urogenital reflexes would be seen only at the onset of seizure susceptibility. Further studies are needed to examine the relation between seizures, obstructive uropathy, and urogenital reflexes in male EL mice.
Diet is known to influence obstructive uropathy in mice. For example, a high-fat diet reduced the incidence of obstructive uropathy in STR/N strain, whereas a high-caloric diet worsened the disease in KK-Ay mice (27,31). Ninomiya et al. (31) suggested that diet could effect seminal vesicles and the metabolism of the coagulating gland, which could influence the chemical composition of the copulatory plug. The mechanism by which the KD eliminated sudden death and obstructive uropathy in male EL mice is unknown. Further studies are needed to determine if the KD prevents obstructive uropathy through an effect on the nervous system, the urogenital system, or on the chemical composition of the copulatory fluid.
In summary, we found that the sudden death in male EL mice arises from abnormal ejaculation, which produces a urethral plug with consequent urinary retention and acute severe uremia. The coincident onset of seizures and obstructive uropathy in EL males suggests that a sexual dysfunction is associated with epilepsy in this model. The KD possibly eliminates sudden death by preventing abnormal ejaculation and urethral obstruction. The EL mouse thus may be a useful model for studying the mechanisms by which epilepsy affects reproductive function and for evaluating potential treatments for sexual dysfunction in those with epilepsy.
Acknowledgment: This research was supported by the Boston College Research Fund and NIH grant (HD 39722).