This article is mentioned in:

Introduction

Mossy fiber sprouting (MFS) has been observed in experimental models of temporal lobe epilepsy (TLE) (1) and in the epileptic human hippocampus (2), and the involvement of MFS in TLE has long been demonstrated. Previous studies have supported the hypothesis that MFS alters synaptic connectivity and circuit organization, thus forming recurrent excitatory connections, which contribute to an enhanced susceptibility to seizures (3). However, other studies have demonstrated that MFS is a consequence of TLE, an event that is not correlated with seizures (4). Therefore, the mechanisms underlying MFS remain unclear and the sequence of occurrence of MFS and TLE is controversial.

Tau protein is an important microtubule-associated protein that is localized in neuronal axons and facilitates microtubule dynamics, axonal transport, neurite outgrowth and axonal elongation (5). These functions of tau are modulated by site-specific phosphorylation, mainly occurring at Ser202/Thr205 (6). Furthermore, various epilepsy models have confirmed the significantly increased phosphorylation of tau protein following seizure, and the resulting induced MFS is important in TLE (7,8). Although different kinases may modify tau, it has been suggested that glycogen synthase kinase-3β (GSK-3β) is important in regulating tau phosphorylation, including at the Ser202/Thr205 residue, under physiological and pathological conditions (9).

GSK-3β is a constitutively active multi-functional serine/threonine kinase that is involved in regulating diverse physiological pathways, including neuronal structure, apoptosis, axon growth and guidance, cytoskeletal stability and synaptic plasticity (10). Furthermore, previous studies have demonstrated that GSK-3β affects axoplasmic transport and axonal growth through the phosphorylation of tau protein (11). Therefore, GSK-3β may also function in epileptogenesis, particularly in MFS by a mechanism involving tau phosphorylation. In the present study, the progression of MFS was observed during kindling in association with the expression levels of GSK-3β mRNA and protein, and with GSK-3β activity, in order to investigate the correlation between TLE, MFS and GSK-3β and the involvement of GSK-3β in epileptogenesis.

Materials and methods

Animal groupings

Male Sprague-Dawley (SD) rats (n=180; age, 6–8 weeks; weight, 180–120 g) obtained from the Animal Experimental Centre (Central South University, Chengsha, China) were divided into the control and PTZ groups by the random number method. The study was conducted in accordance with the guidelines of the Central South University and the Guide for the Care and Use of Laboratory Animals (12), and was approved by the Institutional Animal Care and Use Committee of Central South University. All efforts were made to minimize animal suffering and to reduce the number of animals used in each group. The groups were further divided into five subgroups, each containing 18 rats, which were sacrificed at different time points (3 days and 1, 2, 4 and 6 weeks) following the initial PTZ injection (Sigma-Aldrich, St. Louis, MO, USA). Rats of the five subgroups were sacrificed and perfused for Timm staining and scoring, immunohistochemistry and in situ hybridization of GSK-3β and GSK-3β kinase activity.

Animal models and behavior monitoring

Rats in the PTZ group received one dose of 30 mg/kg PTZ intraperitoneally (i.p.) daily until they were kindled or sacrificed. The control rats were injected with an equal dose of saline. Rats were observed every 2 h daily for the occurrence of PTZ-induced seizures prior to kindling. Convulsive behavior was classified into five stages as proposed by Racine (13). Rats were considered to be kindled when a seizure attack (score, ≥3) occurred following each PTZ injection for five consecutive days. Seizure activity in kindled rats without PTZ induction was perceived as spontaneous recurrent seizures (SRSs). Kindled rats were monitored for seizure occurrence using a video camera positioned above the cages until they were sacrificed.

Timm staining

At the aforementioned time points, rats were sacrificed (n=60). The rats were anesthetized with 10% chloral hydrate and intracardially perfused with 150 ml saline, followed by 200 ml 0.1 mol/l phosphate buffer (pH 7.2–7.6) containing 0.4% sodium sulfide and 250 ml 4% paraformaldehyde at 4°C. The brains were carefully removed and fixed in 4% paraformaldehyde for 24 h. Subsequent to this, the brains were transferred into 0.1 mol/l phosphate buffer containing 30% sucrose at 4°C until they sank for 72 h, following which they were embedded in OCT and stored at −70°C for further analysis. The embedded brains were cut into 25-μm coronal sections using a vibratome (CM1950; Leica, Wetzlar, Germany) and the microscope slides with brain slices were placed into a Timm developer (MaxQ7000; Thermo Scientific, Waltham, MA, USA) for 60–80 min at 26°C in the dark and then removed for a 5-min wash in water to terminate the reaction. The slides were dehydrated, cleaned and mounted with gum consisting of 60 ml 50% Arabic gum, 10 ml 2.55% citric acid, 10 ml 2.35% sodium citrate, 30 ml 5.67% hydroquinone and 1.5 ml 17% silver nitrate. The distribution of Timm granules in the dentate gyrus (DG) supragranular regions and the stratum oriens of CA3 was rated on a scale of 0 to 5 according to previously described criteria (14).

Immunohistochemistry and in situ hybridization

Rats (n=60) were anesthetized with 10% chloral hydrate (i.p.) and perfused intracardially with saline and 4% paraformaldehyde. The brains were removed and coronally divided into two sections, 3.5 mm posterior to the bregma. The anterior section was placed in 4% paraformaldehyde overnight for immunohistochemistry, while the posterior section was placed in 4% paraformaldehyde with 0.1% diethylpyrocarbonate for 30 min, for in situ hybridization. Subsequent to this, tissues were sliced, routinely processed and embedded in paraffin wax. Serial sections were cut at 5-μm thickness. Immunohistochemistry was performed using the SABC method (polyclonal antibody against GSK-3β; Boster Biological Tech Ltd., Fremont, CA, USA). In situ hybridization was conducted using the in situ hybridization kit (MK1750; Boster, Wuhan, China) according to the manufacturer’s instructions. The gray value was measured and analyzed using the image analysis system HPIAS-1000 (Championimages, Wuhan, China).

GSK-3β activity assay

Rabbit monoclonal anti-GSK-3β antibody (2 μg/reaction; Abcam, Cambridge, MA, USA) was added to rat brain tissue lysate (hippocampus and frontal region) and incubated for 1 h with gentle agitation. Protein G-agarose (40 μl; Sigma-Aldrich) was added to the lysate and incubated for another 6 h, followed by high-speed centrifugation (12,000 × g) for 10 min. Immune complexes were collected with protein G-agarose and washed three times with cell dissociation buffer. The washed beads were incubated with 15 μl kinase buffer [250 μM ATP, 1.4 μCi (γ-32P) ATP, 20 mM Tris-HCl pH 7.5, 5 mM MgCl2, 1 mM dithiothreitol and 100 μM phosphate glycogen synthase peptide 2] at 30°C for 30 min. The reaction was terminated by the addition of 12.5 μl 300 mmol/l phosphate. GSK-3β kinase activity was assayed using the Liquid Scintillation Counter (Beckman Coulter, Miami, FL, USA).

Statistical analysis

Intragroup differences in Timm scores were analyzed using the Kruskal-Wallis H test and intergroup differences were analyzed with a Mann-Whitney U test, followed by a Nemenyi test for pairwise comparison. The remaining indices for the difference between the PTZ and control group were performed with Student’s t-test. The means of each group were compared using a one-way analysis of variance and then Fisher’s least significant difference test for pairwise comparison. α=0.05 was selected as the level of significance and P<0.05 was considered to indicate a statistically significant difference. All statistical analyses were two-sided and were performed using the Statistical Package for the Social Sciences (SPSS), version 16.0 (SPSS, Inc., Chicago, IL, USA).

Results

Behavioral outcomes

In each experimental group, prior to kindling, with the exception of one fatality in the 1 week group due to persistent generalized tonic-clonic seizure, seizure activity in the rats was only observed at grades 0–3. Following kindling in the experimental group, with the exception of one fatality in the 4-week group due to frequent seizures, the rats developed seizure activity at grades 3–5 following continuous PTZ injections for 18–22 days. The PTZ-treated rats were in accordance with the kindling criterion 26.1±1.6 days following PTZ injection. The PTZ-induced seizure activity usually occurred 5–10 min following injection, with a duration of 5–30 min. SRSs at grades 2–3 were detected in kindled rats as early as 23 days following PTZ injection. However, no epileptiform behavior was observed in the control rats.

Timm scores

Timm staining patterns in PTZ-treated rats differed significantly from those of the control rats. There was an increased distribution of Timm granules in the stratum pyramidale of the CA3 region at each time point in the PTZ group (P<0.05) (Fig. 1A and C). In addition, the degree of MFS in the CA3 area progressed with the development of behavioral kindled seizures and peaked at 6 weeks. However, the Timm scores for the DG supragranular regions were 0–1 throughout the experiment in the PTZ group, with no significant difference compared with those of the control group (P>0.05) (Fig. 1B and D). In addition, no significant difference was identified in the Timm scores for the CA3 area and the supragranular regions at each time-point within the control rats (P>0.05).

Figure 1 Progression of MFS in the hippocampus. (A) The intensity of Timm granules increased with the development of kindling in the CA3 region in the PTZ group (P<0.05). (B) No obvious MFS was observed in the DG supragranular regions in the control and PTZ groups at any time-point (P>0.05). Time-dependent changes of Timm scores in the (C) CA3 and (D) DG supragranular regions of the hippocampus, respectively. (Scale bars, 100 μm) MFS, mossy fibre sprouting; PTZ, pentylenetetrazole; con, control; DG, dentate gyrus. Original magnification, ×100.

Expression and activity of GSK-3β

The expression of GSK-3β mRNA and protein in the PTZ group increased markedly from 3 days to 2 weeks (P<0.01) and declined to control levels at 6 weeks (Fig. 2), whereas the expression of GSK-3β mRNA and protein was observed at various regions in the hippocampus of control rats, with no significant difference among the time-points. GSK-3β kinase activity increased in parallel with GSK-3β protein expression from day 3 in the PTZ group, reaching a peak at 2 weeks, declining from 4 weeks and finally returning to the control levels at 6 weeks. Conversely, there was no significant difference in GSK-3β activity in the frontal region between the PTZ and control groups. Furthermore, no significant changes were identified in the GSK-3β activity in the hippocampi of the control rats (Table I).

Figure 2 Quantitative analyses of GSK-3β protein and mRNA expression. GSK-3β protein expression increased from 3 days in the PTZ group, peaking significantly at 2 weeks, declining from 4 weeks and returning to control levels at 6 weeks in the (A) CA1, (C) CA3 and (E) DG regions. GSK-3β mRNA expression increased in parallel with GSK-3β protein in the (B) CA1, (D) CA3 and (F) DG regions. Data are presented as the mean ± SEM. *P<0.01 and ΔP<0.05 compared with the control. GSK-3β, glycogen synthase kinase-3β; PTZ, pentylenetetrazole; con, control; DG, dentate gyrus.

Table I Quantitative analyses of GSK-3β activity (P-32 counts per minute value).

Table I

Quantitative analyses of GSK-3β activity (P-32 counts per minute value).

	GSK-3β activity
	Hippocampus		Frontal region
Group	PTZ	Control	PTZ	Control
3 day	5886.47±252.71a	2013.40±130.31	2074.90±214.80	2073.55±158.26
1 week	7789.87±334.51a	2011.53±163.37	2158.38±196.83	2015.13±156.15
2 weeks	11503.55±1527.55a	2061.42±203.47	2250.27±266.08	2018.53±156.66
4 weeks	5822.15±488.93a	2065.76±191.51	2115.87±250.96	1967.23±136.75
6 weeks	2142.85±129.34	2101.73±195.04	2078.28±221.22	2014.38±105.88

{ label (or @symbol) needed for fn[@id='tfn1-mmr-08-05-1337'] } Quantitative analyses of GSK-3β activity in the hippocampus and frontal region. Data are presented as the mean ± SEM. Groups were separated by the time of sacrifice following the initial injection of PTZ.

a P<0.01, vs. the control.

{ label (or @symbol) needed for fn[@id='tfn3-mmr-08-05-1337'] } PTZ, pentylenetatrazole.

Subcellular localization of GSK-3β

The subcellular localization of GSK-3β changed markedly during the kindling process. The expression of GSK-3β mRNA and protein was observed in pyramidal cells in the CA3 area, CA1 area and in the granular cells in the DG of the control rat hippocampus, predominantly in the neuronal membranes and processes. However, in the PTZ group, GSK-3β translocated to the soma throughout kindling development, which was particularly notable 2 weeks following the initial PTZ injection, and the expression decreased to a low level in the soma at 6 weeks (Fig. 3).

Figure 3 Immunohistochemistry and in situ hybridization localization of GSK-3β (A) protein and (B) mRNA in the CA3 area of control rats and PTZ-treated rats. GSK-3β protein and mRNA translocated from neuronal membranes and processes to the soma throughout kindling development, which was particularly notable in the 2 weeks following the initial PTZ injection. The expression became low in the soma at 6 weeks. (Scale bar=50 μm). GSK-3β, glycogen synthase kinase; PTZ, pentylenetetrazole. Original magnification, ×200.

Discussion

The present study demonstrated that as the degree of aberrant MFS increased, the severity of seizures and the amplitude and frequency of excitatory postsynaptic currents increased. MFS was observed prior to the occurrence of TLE and was increased following kindling, which suggested that repeated seizures aggravated MFS; however, it did not indicate that MFS was a consequence of SRS. Furthermore, MFS may not be the sole determinant of the onset of TLE, and other mechanisms, such as hilar ectopic dentate granule cells, may also be involved in epileptogenesis (15). An increased distribution of Timm granules was observed at the CA3 region, but not at the supragranular regions, which differs from that in the pilocarpine and kainic acid (KA) kindling models (16,17) and is consistent with the results of a previous study (18). The difference in the distribution of Timm granules among the PTZ-kindling, pilocarpine and KA models may be attributed to the differences in the type and dose of the different convulsants and the severity of the induced seizures. Pilocarpine and KA may induce status epilepticus (SE), which leads to hippocampal neuronal death, particularly in the CA3 area. This study used moderate doses of PTZ to establish a kindling model in which SE rarely occurred. The extent of neuronal injuries caused by generalized tonic-clonic convulsions and SRS was less severe than that by SE. Regardless of the marginal injuries in the observed CA3 region and the hilar neurons, sprouted mossy fibers were able to locate sufficient post-synaptic target cells without returning to the DG supragranular regions.

GSK-3β is involved in numerous neuronal functions and affects neuronal polarity through undermining the structural stability of axons, regulating microtubule dynamics in growing axons, which leads to synaptic reorganization and the formation of recurrent excitatory circuits (19). Moreover, a study demonstrated that GSK-3β was able to hyperphosphorylate tau protein and may contribute to the pathogenesis of focal cortical dysplasia (FCD) in older patients. FCD is an important neurodevelopmental cause of refractory human epilepsy. However, thus far there have been no studies regarding a correlation between GSK-3β and MFS (20). In the present study, the expression of GSK-3β mRNA and protein was increased in the DG, however, no MFS was observed in this area. This does not indicate that GSK-3β does not exhibit a correlation with MFS, as a previous study determined that in the PTZ-kindling model, hippocampal neuronal death in the DG area was more severe than in the CA3 area (13), therefore sprouted mossy fibers were not able to locate sufficient post-synaptic target cells and turned back to the inner molecular layer of the DG. In a previous study, it was demonstrated that tau protein and its phosphorylation at Ser202 increased significantly in the PTZ group (21). Furthermore, in the present study, the expression of GSK-3β mRNA and protein, as well as its activity, also increased significantly from 3 days to 4 weeks in the PTZ group, which was correlated with an increase in phosphorylated tau protein. These results suggested that GSK-3β may be involved in MFS in the kindling rat hippocampus through the phosphorylation of tau.

The mechanisms underlying the correlation between GSK-3β and MFS remain unknown. A study demonstrated that GSK-3β is abundant in soma when its activity is high, whereas it is expressed in axons when the activity is low (19). Results also suggest that in the PTZ group, GSK-3β translocated from the axon to soma during the progression of MFS. Tau is synthesized and subsequently phosphorylated within the cell body; however, in the proximal axonal process, such as within growing neurons, it is dephosphorylated by phosphatases during axonal transport, resulting in the stabilization of the microtubule cytoskeleton. Furthermore, excessive aggregation of GSK-3β and phosphorylated tau in soma exhibited high cytotoxicity and resulted in apoptosis and necrosis (22,23). This may be due to an alternate mechanism of MFS caused by GSK-3β. Increased expression and activity of GSK-3β led to the phosphorylation of tau protein, which subsequently induced MFS through reducing its microtubule binding ability and promoting microtubule assembly function.

In conclusion, the present study demonstrated a correlation between MFS and GSK-3β. GSK-3β may be involved in the progression of MFS through abnormal phosphorylation of tau protein. However, the importance and specific mechanisms of GSK in the formation of MFS were not determined. GSK is not the only enzyme that regulates tau protein phosphorylation and MFS may not be the sole determinant of the onset of TLE. Further studies are required that modulate the expression and activity of GSK-3β in mossy fiber sprouting in the PTZ-kindling model, and assess whether there are corresponding changes in the level of tau protein phosphorylation, MFS formation and the severity of epilepsy. Understanding the involvement of GSK-3β in MFS may lead to novel therapeutic interventions, and therefore ameliorate epileptogenic network dysfunction and associated morbidities.

References