Repeated seizures lead to progressive ventilatory dysfunction in SS Kcnj16-/- rats

Abstract

Patients with uncontrolled epilepsy experience repeated seizures putting them at increased risk for sudden unexpected death in epilepsy (SUDEP). Data from human patients have led to the hypothesis that SUDEP results from severe cardiorespiratory suppression after a seizure, which may involve pathological deficiencies in the brainstem serotonin (5-HT) system. Rats with a genomic Kcnj16 mutation (SS^Kcnj16-/- rats) are susceptible to sound-induced generalized tonic-clonic seizures (GTCS) which, when repeated once daily for up to 10 days (10-day seizure protocol), increased mortality, particularly in male rats. Here, we test the hypothesis that repeated seizures across the 10-day protocol will cause a progressive ventilatory dysfunction due to time-dependent 5-HT deficiency. Initial severe seizures led to ictal and postictal apneas and transient decreases in breathing frequency, ventilatory drive, breath-to-breath variability, and brief hypoventilation. These seizure-induced effects on ventilation were exacerbated with increasing seizures and ventilatory chemoreflexes became further impaired after repeated seizures. Tissue analyses of key brainstem regions controlling breathing showed time-dependent 5-HT system suppression and increased immunoreactivity for IBA-1 (microglial marker) without changes in overall cell counts at 3, 7, and 10 days of seizures. Fluoxetine treatment in SS^Kcnj16-/- rats prevented repeated seizure-induced progressive respiratory suppression but failed to prevent seizure-related mortality. We conclude that repeated seizures cause a progressive compromise of ventilatory control in the immediate postictal period largely mediated by serotonin system suppression in brainstem regions of respiratory control. However, other unknown factors contribute to overall survival following repeated seizures in this model.NEW & NOTEWORTHY This study demonstrated that repeated seizures in a novel rat model (SS^Kcnj16-/- rats) caused a progressively greater ventilatory dysfunction in the immediate postictal period associated with brainstem serotonin (5-HT) suppression. Augmenting brain 5-HT with a selective serotonin reuptake inhibitor prevented the progressive ventilatory dysfunction induced by repeated seizures but failed to prevent seizure-related mortality, suggesting that repeated seizures may lead to cardiorespiratory suppression and failure through multiple mechanisms.

Keywords: SUDEP; breathing variability; respiratory depression; seizures; serotonin.

Graphical abstract

Figure 1.

Acute disruption of eupneic (resting) breathing during and after audiogenic seizures in SS^{Kcnj16−/−} rats. Representative raw traces of breathing changes before (baseline), during, and after sound exposure in control (SS^WT; A) and SS^{Kcnj16−/−} (B) rats. Note that while sound stimulation does not cause a seizure and has little effect on breathing in SS^WT rats, the same stimulus led to generalized tonic-clonic seizure (GTCS) in SS^{Kcnj16−/−} rats triggering a stereotypic ventilatory response, including ictal apnea, postictal apnea (at the time of sound termination), and transient postictal respiratory rate suppression (see also insets in B). Seizure severity scores (C) for SS^{Kcnj16−/−} rats (n = 6) that survived the 10-day seizure protocol did not differ across the 10 days of study [P = 0.23; one-way repeated-measures (RM) ANOVA]. Values are represented as means ± SE. n is number of animals.

Figure 2.

Initial severe audiogenic seizures transiently affect aspects of breathing. Data shown are minute ventilation (V̇e; mL/min/100 g; A), breathing frequency (F_B; breaths/min; B), and tidal volume (V_T; mL/breath/100 g; C) in SS^{Kcnj16−/−} rats before and for up to 15 min after a mild (seizure score: 1–2; n = 5) or the first severe (seizure score: 3–4; n = 6) seizure. Data shown are 1 min averages (1–5 min postictal) and 5 min averages (from 5 to 10 and 10 to 15 min postictal). *P < 0.05, **P < 0.01 at each time point vs. baseline (BL), or #P < 0.05 score1–2 vs. score 3–4; two-way repeated-measures (RM) ANOVA with Sidak’s multiple-comparisons test (Time × Seizure Score interaction term shown). Values are represented as means ± SE. n is number of animals.

Figure 3.

Initial severe audiogenic seizures reduce ventilatory drive and breathing variability in the postictal period. Ventilatory drive (V_T/Ti; mL/breath/s; A), the ratio of ventilation to oxygen consumption (V̇e/V̇o₂; mL/min/mL O₂/min; B), and the ratio of ventilation to CO₂ production (V̇e/V̇co₂; mL/min/mL CO₂/min; C) during baseline (BL) conditions and 1, 2, 3, 4, 5, 6–10, and 11–15 min postictal. Example of breath-to-breath analyses (n = 1) of the interbreath interval (IBI) during baseline (D; 414 consecutive breaths) conditions and 5–10 min postictal (E; 414 consecutive breaths). *P < 0.05 vs. BL; #P = 0.0823 vs. BL; one-way repeated-measures (RM) ANOVA. Individual and means ± SE values shown.

Figure 4.

Repeated severe seizures exacerbate ventilatory suppression. Minute ventilation (V̇e; mL/min/100 g; A), breathing frequency (F_B; breaths/min; B), and tidal volume (V_T; mL/breath/100 g; C), ventilatory drive (V_T/Ti; mL/breath/s; D), and the convection requirement (V̇e/V̇o₂; mL/min/mL O₂/min; E) in SS^{Kcnj16−/−} rats before and for up to 15 min after severe (score: 3–4; n = 6) seizures across the 10-day protocol. *P < 0.05 or **P < 0.01 vs. baseline (BL), #P < 0.05 compared with equivalent time point on days 1–3. Two-way repeated-measures (RM) ANOVA with Dunnet’s multiple-comparisons test (Time × Seizure Score interaction term shown). Values are represented as means ± SE. n is number of animals.

Figure 5.

Representative confocal images of brainstem tissue sections from SS^{Kcnj16−/−} rats that did not have induced seizures (naïve) or 10 seizures immunofluorescence-stained serotonin (5-HT) system markers (A and B) or IBA-1 (C). 5-HT (red) and tryptophan hydroxylase (TPH, cyan) immunoreactivity (ir) was colocalized to expected regions within the rostral raphe magnus (RMg; Bregma −11.0 mm) and pallidus (RPa) in naïve tissues but both 5-HT-ir and TPH-ir appeared reduced after 10 seizures (A; ×4). 5-HT-ir (red; ×20) within the pre-Bötzinger complex (preBötC) and nucleus ambiguus (NA; Bregma −12.6 mm) was also visibly reduced (B), whereas IBA-1-ir (red; ×20) appeared to increase in these same regions. Scale bars are 500 µm in A and 100 µm in C. Images are brightness and contrast enhanced the same in naïve and 10 seizure conditions to illustrate the effect.

Figure 6.

Daily fluoxetine treatment in SS^{Kcnj16−/−} rats prevents the exacerbation of ventilatory suppression with repeated seizures. Minute ventilation (V̇e; mL/min/100 g; A), breathing frequency (F_B; breaths/min; B), and tidal volume (V_T; mL/breath/100 g; C) in SS^{Kcnj16−/−} rats (n = 7) during baseline (BL) conditions and for up to 15 min after severe (score: 3–4) seizures across the 10-day protocol. *P < 0.05 or **P < 0.01 vs. baseline (BL). Two-way repeated-measures (RM) ANOVA with Tukey’s multiple-comparisons test (Time × Day interaction term shown). There was no effect of Day (P = 0.4156) across the 10-day protocol. Values are represented as means ± SE. n is number of animals.

Figure 7.

Effects of daily fluoxetine treatment in SS^{Kcnj16−/−} rats (red) on maximal ventilatory responses, seizure severity, apnea duration, and mortality compared with no treatment (blue) across the 10-day seizure protocol. Audiogenic seizure-induced maximal changes in minute ventilation (Max. ΔV̇e; mL/min/100 g; A) and breathing frequency (Max. ΔF_B; breaths/min; B) increased in untreated SS^{Kcnj16−/−} rats (n = 6) across the 10-day seizure protocol but were unaffected in fluoxetine-treated SS^{Kcnj16−/−} rats (n = 7). Average seizure scores (C) were comparable between untreated (n = 6) and fluoxetine-treated (n = 7) SS^{Kcnj16−/−} rats although were slightly lower (P = 0.0086) with fluoxetine on days 4–7. Ictal (primary) apnea duration (s; D) and total apnea duration (ictal + secondary; s; E) were consistently shorter in fluoxetine-treated SS^{Kcnj16−/−} rats (n = 7; P < 0.05) compared with untreated SS^{Kcnj16−/−} rats (n = 6) at all time points. Fluoxetine treatment did not affect survival probability compared with untreated SS^{Kcnj16−/−} rats (P = 0.207; F). Two-way repeated-measures (RM) ANOVA with multiple-comparisons test (Time × Day interaction term shown). Values are represented as means ± SE. *P < 0.05, **P < 0.01 vs. days 1–3 within treatment group; #P < 0.05 vs. no treatment at equivalent time point. Survival rates were assessed using a Log-rank (Mantel–Cox) test. n is number of animals.