Novel neurosteroid pregnanolone pyroglutamate suppresses neurotoxicity syndrome induced by tetramethylenedisulfotetramine but is ineffective in a rodent model of infantile spasms

Abstract

Background: Neurosteroids are investigated as effective antidotes for the poisoning induced by tetramethylenedisulfotetramine (TMDT) as well as treatments for epileptic spasms during infancy. Both these conditions are quite resistant to pharmacotherapy; thus, a search for new treatments is warranted.

Methods: In this study, we determined the efficacy of two novel neurosteroids, pregnanolone glutamate (PAG) and pregnanolone pyroglutamate (PPG), and tested these drugs in doses of 1-10 mg/kg (ip) against the TMDT syndrome and in our rodent model of infantile spasms.

Results: Only PPG in doses 5 and 10 mg/kg suppressed the severity of the TMDT syndrome and TMDT-induced lethality, while the 1 mg/kg dose was without an effect. Interestingly, the 1 mg/kg dose of PPG in combination with 1 mg/kg of diazepam was also effective against TMDT poisoning. Neither PAG nor PPG were effective against experimental spasms in the N-methyl-D-aspartate (NMDA)-triggered model of infantile spasms.

Conclusions: While evidence suggests that PAG can act through multiple actions which include allosteric inhibition of NMDA-induced and glycine receptor-evoked currents as well as augmentation of ɣ-aminobutyric acid subtype A (GABAA) receptor-induced currents, the agent appears to neither have the appropriate mechanistic signature for activity in the infantile spasm model, nor the adequate potency, relative to PPG, for ameliorating the TMDT syndrome. The full mechanisms of action of PPG, which may become a potent TMDT antidote either alone or in combination with diazepam are yet unknown and thus require further investigation.

Keywords: Antidote; Neurosteroids; Neurotoxicity; Severe seizures; Tetramethylenedisulfotetramine.

Figure 1.. Effects of novel neurosteroids in the TMDT syndrome

A. Timeline of the experiment. TMDT was injected subcutaneously (sc, 0.3 mg/kg of body weight) at time 0. Over time, body twitches, then clonic and tonic-clonic seizures developed, sometimes continuing into status epilepticus with the lethal ending (followed up to 24 hours). Treatments (novel neurosteroids or vehicle) were injected ip right after the occurrence of the first clonic seizure. B. Latency to the onset of the first clonic seizure (before any treatment was injected). The distribution of latencies with no significant differences among pre-destined groups confirms good randomization. CDX vehicle N=12. Each treatment subgroup (dose/drug) N=10. C. Latency to the onset of the second clonic seizure, appearing now after the treatment. In some animals, the second clonic seizure did not reappear. The latency to onset was not significantly different among the groups as well. D. Number of clonic seizures per individual subject. There was no difference among the groups. E. Latency to the onset of tonic-clonic seizure from the TMDT injection. There was no difference among the groups, though in some groups tonic-clonic seizures were suppressed as reflected in (F). F. Graphical representation of survival analysis reflecting cumulative protection from the occurrence of tonic-clonic seizures. There was a significant difference among the groups. G. Number of tonic-clonic seizures in each subject. Group treated with PPG 10 mg/kg i.p. had significantly suppressed number of tonic-clonic seizures compared to the CDX vehicle. H. Latency to death from the TMDT injection. There was no difference among the groups, though in some groups not all subjects died. This is further reflected in (I). I. Graphical representation of survival analysis reflecting cumulative protection. There was a significant difference among the groups. J. TMDT syndrome score marking the most severe event that occurred after the TMDT injection. Both 5 and 10 mg/kg doses of PPG significantly suppressed the score compared to the CDX controls. Statistical analysis was performed using one-way ANOVA (B, C, E, H; mean±SEM shown), survival analysis with Mantel-Cox log rank test (F, I, individual values shown) or Kruskal-Wallis test (D, G, J; median and individual values shown) followed by Dunn’s post hoc test. Asterisks (*) mark p<0.05. ANOVA – analysis of variance CDX – cyclodextrin ip – intraperitoneal TMDT – tetramethylenedisulfotetramine PAG – pregnanolone glutamate PPG – pregnanolone pyroglutamate sc – subcutaneous

Figure 2.. Effects of PPG combination with diazepam (DZP) in the TMDT syndrome

A. Timeline of the experiment. TMDT was injected subcutaneously (sc, 0.5 mg/kg of body weight) at time 0. Over time, body twitches, then clonic and tonic-clonic seizures developed, sometimes continuing into status epilepticus with the lethal ending (followed up to 24 hours). Treatments (novel neurosteroid PPG 1 mg/kg, diazepam-DZP 1 mg/kg, or combination of both PPG+DZP or vehicle) were injected ip right after the occurrence of the first clonic seizure. CDX vehicle N=10; PPG N=7; DZP N=8; PPG+DZP N=7. B. Latency to the onset of the first clonic seizure (before any treatment was injected). The distribution of latencies with no significant differences among pre-destined groups confirms good randomization. C. Latency to the onset of the second clonic seizure, appearing now after the treatment. In some animals, the second clonic seizure did not reappear. The latency to onset was not significantly different among the groups as well. D. Number of clonic seizures per individual subject. Bars represent medians. There was no difference among the groups. E. Latency to the onset of tonic-clonic seizure from the TMDT injection. There was no difference among the groups, though in the PPG+DZP group tonic-clonic seizures were suppressed as reflected in (F). F. Graphical representation of survival analysis reflecting cumulative protection from the occurrence of tonic-clonic seizures. There was a significant difference among the groups. G. Number of tonic-clonic seizures in each subject. Bars represent medians. The number of tonic-clonic seizures was significantly suppressed in the group treated with PPG+DZP compared to the CDX vehicle. H. Latency to death from the TMDT injection. There was no difference among the groups (ANOVA), though in some groups not all subjects died. This is further reflected in (I) I. Graphical representation of survival analysis reflecting cumulative protection. There was a significant difference among the groups. J. TMDT syndrome score marks the most severe event that occurred after the TMDT injection. Bars represent medians. Only the combination of 1 mg/kg PPG with 1 mg/kg of DZP of PPG significantly suppressed the score compared to the CDX controls. Statistical analysis was performed using one-way ANOVA (B, C, E, H; mean±SEM shown), survival analysis with Mantel-Cox log rank test (F, I; individual values shown) or Kruskal-Wallis test (D, G, J; median and individual values shown) followed by Dunn’s post hoc test. Asterisks (*) mark p<0.05. ANOVA – analysis of variance CDX – cyclodextrin DZP – diazepam ip – intraperitoneal TMDT – tetramethylenedisulfotetramine PPG – pregnanolone pyroglutamate sc – subcutaneous

Figure 3.. Effects of novel neurosteroids in the rat model of infantile spasms

A. Timeline of the PAG/PPG experiment. On P12, all pups (prenatally exposed to betamethasone as a condition of the model) were injected with NMDA (7.5 mg/kg ip). After the experimental spasms subsided, rats were randomized into treatment groups (vehicle=CDX, PAG, or PPG in doses 1,5, or 10 mg/kg) and injected ip with respective treatments. The same approach (minus randomization) was used on P13. On P14, no spasms were triggered and only the treatment was injected. On P15, there was no treatment and only the spasms were triggered with NMDA. N of animals per individual group after the quality control has been applied: CDX=12, PAG1=9, PAG5=9, PAG10=9, PPG1=9, PPG5=9, PPG10=8. B. Timeline of the positive control with ACTH treatment experiment. On P12, all pups (prenatally exposed to betamethasone as a condition of the model) were injected with NMDA (7.5 mg/kg ip). After the experimental spasms subsided, rats were randomized into either the treatment group with ACTH or the control group (normal saline). ACTH (0.1 mg/kg) was administered sc at 14:00 and 20:00 on P12, at 08:00, 14:00 and 20:00 on P13 and P14. Spasms were repeatedly triggered on P13 and P15. Vehicle group N=13, ACTH group N=11. C. Latency to the onset of the first spasm measured from the administration of NMDA in individual designated groups on P12 – before treatment. The graph illustrates the efficacy of randomization. There was no significant difference among the groups, i.e., randomization was good. D. Number of spasms determined after administration of NMDA in individual designated groups on P12. At this moment, no treatment was administered yet, so the graph illustrates efficacy of randomization. There was no significant effect, i.e., randomization was good. E. Latency to onset of the first spasm measured from the administration of NMDA once the treatment was completed in P15 rats. There was no significant effect of any treatment. F. Number of spasms determined after administration of NMDA once the treatment was completed in P15 rats. There was no significant effect of any treatment. G. Index composed of the number of spasms on P15 (post-treatment) over the number of spasms on P12 (pre-randomization) for each animal. There was no significant effect of any treatment. H. Normalized body weight gain from P12 to P15 (during chronic treatments). Index represents the difference in body weight at P15 minus P12 over starting weight at P12 for each animal. None of the treatments affected body weight gain. I. Latency to the onset of the first spasm (left) and the number of spasms (right) in individual designated groups on P12 before any treatment has occurred. The graph illustrates the efficacy of randomization. There was no significant difference between the groups, i.e., randomization was good. J. Latency to the onset of the first spasm (left) of the first spasm in groups on P15, after treatment with ACTH was completed. ACTH significantly delayed the onset of spasms compared to saline controls as well as suppressed the number of spasms, which was our primary outcome. K. Ratio between the number of spasms on P15 (after treatment was completed) over P12 values (before treatment baseline) for individual animals (left). There was a significant decrease in the ratio indicating a strong effect of ACTH. The figure is complemented by the relative weight gain ratio (P15-P12/P12; right). There were no changes in weight gain irrespective of treatment. Statistical analysis was performed using one-way ANOVA (C, D, E, F, G, H; mean±SEM shown) or Student’s t-test (I, J, K; mean±SEM shown). Asterisks (*) mark p<0.05. ACTH – adrenocorticotropic hormone, corticotropin ANOVA – analysis of variance CDX – cyclodextrin ip – intraperitoneal NMDA – N-methyl-D-aspartic acid PAG – pregnanolone glutamate PPG – pregnanolone pyroglutamate sc – subcutaneous

Figure 4.. Effects of pregnanolone pyroglutamate (PPG) in the acute treatment in the rat model of infantile spasms

A. Timeline of the experiment. On P15, the pups (prenatally exposed to betamethasone as a condition of the model) were injected with a single dose of PPG (10 mg/kg ip) that was maximally effective in the TMDT syndrome model. After 60 mins, spasms were triggered by NMDA (15 mg/kg ip). N for CDX=9; N for PPG10=8. B. Latency to the onset of the first spasm measured from the administration of NMDA. There was a significant effect of PPG 10 mg/kg dose in terms of increased latency to onset. C. Number of spasms determined after administration of NMDA. There was no effect of PPG on this major indicator of treatment efficacy. Statistical analysis was performed using Student’s t-test (B, C; mean±SEM shown). Asterisks (*) mark p<0.05. CDX – cyclodextrin ip – intraperitoneal NMDA – N-methyl-D-aspartic acid PPG – pregnanolone pyroglutamate