Seizure Susceptibility, Phenotype, and Resultant Growth Delay in the nclf and mnd Mouse Models of Neuronal Ceroid Lipofuscinoses

Abstract

We examined flurothyl gas-induced seizure latencies and phenotype in 2 mouse models of neuronal ceroid lipofuscinoses: the nclf (Cln6 mutant) variant late-infantile model and the mnd (Cln8 mutant) Northern epilepsy model. Mnd mice on postnatal days 35 to 42 had increased latency to loss of posture compared with wild-type controls. Nclf, mnd, and wild-type mice on postnatal days 21 days to 25 displayed similar latency profiles during repeated seizure induction (kindling) and retesting; seizure phenotypes were different, however. Kindled wild-type mice re-exposed to flurothyl after a 28-day recovery displayed brainstem generalized seizures exclusively. Neuronal ceroid lipofuscinoses mutants demonstrated a lack of brainstem seizures at retesting after 28 days. Repeated induction of generalized seizures delayed weight gain in both nclf and mnd mice compared with wild-type mice. These and our previous results suggest abnormal seizure-related neuronal connectivity and/or plasticity are shared characteristics of the neuronal ceroid lipofuscinoses.

Introduction

Neuronal ceroid lipofuscinoses, also known as Batten disease, are the most prevalent pediatric neurodegenerative diseases, with an estimated incidence of 2 to 4 in 100 000 live births in the US (which translates to 80 to 170 new patients every year).¹ The occurrence of neuronal ceroid lipofuscinoses is the highest in the Scandinavian countries, Finland being the most affected, with an estimated incidence of 1 in 12 500 live births.² Neuronal ceroid lipofuscinoses are characterized histologically by abnormal lysosomal accumulation of autofluorescent storage material and loss of select neuronal populations.³ This family of disorders comprises 8 distinct variants identified by unique patterns of onset and progression.² The different forms are caused by recessive mutations in at least 10 genes (CLN1-10).² Along with vision loss, cognitive decline, psychiatric complications, neurodegeneration, and premature death, seizures are a shared symptom of the neuronal ceroid lipofuscinoses.³

Seizure activity may be a critical driving force behind the progressive neurodegeneration seen with the pediatric-onset neuronal ceroid lipofuscinoses. This hypothesis is supported by studies that indicate that late onset of seizures, or their absence, correlates with milder progression and a better overall prognosis.⁴ In addition, anecdotal information from parents of children with these disorders describes dramatic worsening of condition with repeated seizure activity.

We have previously demonstrated age-dependent alteration of flurothyl-induced seizure latencies and recovery in the Cln3^−/− mouse model of juvenile neuronal ceroid lipofuscinosis,⁵ and wished to determine whether similar seizure-related anomalies existed in 2 other mouse models, the nclf (Cln6 mutant) model of variant late-infantile neuronal ceroid lipofuscinosis⁶ and the mnd (Cln8 mutant) model of Northern epilepsy.⁷

The nclf mouse has progressive retinal degeneration and motor paralysis.⁸ The naturally occurring mnd mutant was originally a potential model of amyotrophic lateral sclerosis⁹ but was later shown to have the mutation in the Cln8 gene, providing a good model for Northern epilepsy/progressive epilepsy with mental retardation.⁷ Mnd mice also display adult-onset progressive motor neuron degeneration, and both the nclf and mnd mice are characterized by accumulation of autofluorescent storage material.^8–9 Studies with the mnd mice have also demonstrated age-dependent changes in behavior that may be highly relevant to the disturbing psychiatric and behavioral symptoms that occur with progression of disease in the pediatric-onset human neuronal ceroid lipofuscinoses.¹⁰

Methods

Animals

Male C57BL/6J wild-type and male nclf and mnd mice on the C57BL/6J background were used for all experiments. Mice were group housed in the University of Rochester vivarium in a temperature-controlled room on a 12-h light/dark cycle. Food and water were provided ad libitum.

Seizure Induction

Animals were transported from the vivarium to the laboratory a minimum of 45 minutes prior to experimental trials to allow for acclimatization. All trials were run between 10 am and 1 pm to minimize circadian influences. Seizures were induced by placing mice individually into a 2.4-liter closed Plexiglas chamber into which a 10% solution of flurothyl (2,2,2-trifluroethyl ether; Sigma-Aldrich, St. Louis, Missouri) was infused as previously described.¹¹ Animals were weighed immediately post-seizure. For kindling studies, generalized seizures were induced once a day for 8 consecutive days (Induction Phase) with subsequent retesting once a week for 4 weeks (Incubation Phase). Data shown are from the fourth retest, 28 days after the end of the kindling.

Generalized seizures were classified from grade 1 to 7 using a previously published scoring system.¹² Grades 1 through 2 represent forebrain seizures. Grade 3 through 7 seizures can be elicited in the absence of forebrain connections and are generally termed “brainstem seizures.” Generalized brainstem seizures are preceded by forebrain seizures and are designated as forebrain-to-brainstem seizures.¹²

Data Analysis

Means of latencies to loss of posture as well as means of body weight were compared using the Student unpaired t test.

Results

Age-Dependent Increase in the Latency to Flurothyl-induced Seizure in mnd Mice

Similarly to previous observations for the Cln3^−/− mouse model of juvenile neuronal ceroid lipofuscinoses,⁵ mnd male mice ages 35 days to 42 days demonstrated a statistically significant, 26% increase in latency to loss of posture in a single trial flurothyl exposure (P < .05). No latency differences were observed between mnd and wild-type mice ages 21days to 25 days or 6 months to 9 months (Figure 1). Seizure latencies for nclf mice were statistically identical to those of wild-type mice at all 3 ages (Figure 1).

Figure 1.

Age-dependent increase in the latency to flurothyl-induced seizure in mnd mice. Male wild-type, mnd, and nclf mice at 3 different ages (21 days to 25 days, 35 days to 42 days, and 6 months to 9 months) were used in the experiments. Seizure was induced by inhalation of flurothyl gas, and the latency until loss of posture was measured. Columns and bars represent mean ± standard error of the mean (n = 4–7). At the age of 35 days to 42 days, the mean latency to flurothyl-induced loss of posture was 532.1 ± 47.8 s for mnd mice and 420.9 ± 17.6 s for 35-day to 42-day wild-type mice (P < .05 by unpaired t test). No significant differences were observed between mnd and wild-type mice at earlier (21 days to 25 days) or later (6 months to 9 months) developmental time points. No statistically significant differences existed between nclf and age-matched wild-type mice.

Immature Wild-type, nclf and mnd Mice Show Similar Seizure Latencies but an Altered Seizure Phenotype During Kindling and Retesting

Wild-type, nclf, and mnd mice at 21 days to 25 days of age underwent kindling: seizure induction by flurothyl gas once a day for 8 consecutive days. Seizure induction latencies during the 8 days of kindling in wild-type, nclf, and mnd mice were very similar (Figure 2A). A trend toward increased latency in the mnd mice was observed but did not reach statistical significance on any but the eighth day (P < .05 by unpaired t test). No differences in seizure induction latencies were observed between groups during a 28-day post-kindling retest (once a week for 4 weeks; data for the 28th day retest are shown in Figure 2A). An expected shift from forebrain to forebrain-to-brainstem seizures was observed in the kindled wild-type mice upon retesting 28 days later (Figure 2B). Mutant strains, however, failed to show this phenotypic change (Figure 2B). None (0%) of the nclf or mnd mice, but all (100%) of wild-type mice, demonstrated a forebrain-to-brainstem seizure (Figure 2B). This phenotypic switch following flurothyl-kindling has been proposed to result from synaptic reorganization.^11–12 An absence of this phenotypic switch in the nclf and mnd mice indicates there may be shared disruption and reorganization of seizure-related neuroanatomical pathways in the brainstem among the neuronal ceroid lipofuscinoses.

Figure 2.

Immature wild-type, nclf, and mnd mice show similar seizure latencies but an altered seizure phenotype during kindling and retesting. Wild-type, nclf, and mnd mice at 21 days to 25 days of age underwent kindling (seizure induction by flurothyl once a day for 8 consecutive days) and subsequent retesting (seizure induction by flurothyl once a week for 4 weeks). Columns and bars represent mean ± standard error of the mean (n = 3–5). (A) Seizure induction latencies during the 8 days of kindling (day 1 through day 8) and in the 4th (28th day) post kindling retest (RT-28 days). Statistical analysis was performed using unpaired t test: P < .05 as compared with wild-type; NS: not significant. (B) Seizure phenotypes during the 8 days of kindling and in the 4th (28th day) post kindling retest. Eight-day kindling in wild-type mice produced the expected shift from cortical to brainstem seizures when the animals were retested 28 days later. All (100%) wild-type mice demonstrated a brainstem seizure, whereas none (0%) of the nclf or mnd mice converted to the expected phenotype.

Evidence for Seizure-Related Growth Delay in nclf and mnd Mice

Daily seizure activity for 8 consecutive days (kindling), and subsequent, once-a-week seizure induction (retesting) for 4 weeks resulted in a statistically significant decrease in total body weight in nclf (P < .01) and mnd (P < .001) mice when compared with naïve same-strain controls (Figure 3). Repeated seizure activity (kindling and retesting for 4 weeks) did not change the weight of wild-type mice (Figure 3). Young adult nclf and mnd naïve mice were consistently bigger than wild-type naïve mice; therefore, we made weight comparisons to same strain naïve animals only.

Figure 3.

Repeated flurothyl-induced seizure activity inhibits the growth of nclf and mnd mice. Wild-type, nclf, and mnd mice at 21 days to 24 days of age underwent kindling (seizure induction by flurothyl once a day for 8 consecutive days) and subsequent retesting (seizure induction by flurothyl once a week for 4 weeks). After the last retesting (at 57 days to 60 days of age), the mice were weighed. Age-matched naïve (no seizure induction), same-strain control mice also were weighed. Columns and bars represent mean ± standard error of the mean (n = 3–5). Repeated seizure activity significantly decreased the total body weight of nclf and mnd mice when compared with naïve same-strain controls (P < .01 for nclf and P < .001 for mnd by unpaired t test). Wild-type animals showed normal growth despite repeated seizure activity (NS: not significant by unpaired t test).

Discussion

We have documented an age-dependent increased latency to flurothyl-induced generalized seizures in 2 out of the 3 neuronal ceroid lipofuscinosis mouse models we have examined (Figure 1 and ⁵). In both Cln3^−/− and mnd mice, this alteration of seizure induction threshold occurs only during a specific developmental window, at 35 days to 42 days postnatally, and is not present at earlier or later developmental time points. The findings suggests that abnormal synapse formation, pruning, and/or strengthening of synapses may occur. However, the evidence for this remains indirect and putative aberrant postnatal wiring might also result from earlier, even from embryologic, insults.

A high mortality rate in Cln3^−/− mice following a forebrain-to-brainstem seizure⁵ originally led us to hypothesize that absence of the neuronal ceroid lipofuscinosis-linked protein perturbs brainstem anatomy and/or postnatal nervous system maturation, and plasticity. This hypothesis is supported by our current findings that both nclf and mnd neuronal ceroid lipofuscinosis-mutant mice appear resistant to flurothyl kindling-induced remodeling (Figure 2B), which is a plasticity-dependent phenomenon. Behavioral studies in mnd mice have shown age-dependent changes in habituation, fear conditioning, and increases in both initiation and length of intruder-induced aggressive behavior.¹⁰ These results are consistent with the idea that neuronal ceroid lipofuscinosis-related mutations may result in aberrant brainstem and limbic connectivity and/or altered plasticity. Glutamate receptors play an essential role in neuronal plasticity,^13–15 and therefore, abnormal glutamate receptor function can be an important contributor to disease development and progression. Aberrant glutamate receptor function have been demonstrated in the Cln3^−/− mouse model of juvenile neuronal ceroid lipofuscinosis.^16–19 The mRNA expression of type 1 metabotropic glutamate receptor (mGluR1) was found to be downregulated in the brain of 10-week-old Cln3^{−/ −} mice.¹⁶ Our previous studies also demonstrated that an abnormally increased AMPA-type glutamate receptor activity largely contributes to the motor coordination deficit in Cln3^{−/ −} mice as early as 1 month of age.^18–19

Lastly, both mnd and nclf mice had highly significant decreases in total body weight after kindling compared with age-, sex-, and strain-matched mutants in which seizures had not been induced (Figure 3). It is not surprising that repetitive seizures in newly weaned (21- to 25-day-old) animals might delay or otherwise interfere with normal development. It is interesting, however, that this effect was limited to nclf and mnd mice; whereas wild-type kindled and naïve animals did not lose weight.

Taken together, our results support a working hypothesis that mutations in some (or all) of the neuronal ceroid lipofuscinosis-related proteins alter the biochemical and neuroanatomic pathways that mediate seizure initiation, propagation, and recovery.