Background
Temporal lobe epilepsy (TLE) is one of the most common types of intractable epilepsy and is characterized by the periodic and unpredictable occurrence of seizures. Glial cell activation and proliferation, a well-described pathological feature of TLE, can alter blood-brain barrier integrity and ion and neurotransmitter homeostasis and cause an inflammatory response, resulting in neuronal hyperexcitability and the generation and spread of seizure activity [
1‐
3]. Although impairment of these functions is thought to be associated with the pathophysiology of epilepsy, the mechanisms underlying glial cell activation are complex and are not fully understood [
1].
Long non-coding RNA (lncRNA) H19, an imprinted gene, is located on human chromosome 11 and is transcribed from the maternally inherited allele [
4]. Despite being identified over 20 years ago, the function of H19 remains unclear and its pathological role as a non-coding RNA has only recently been elucidated [
5]. H19 has diverse functions depending on physiological and pathological state. In the central nervous system (CNS), H19 is overexpressed in glioblastoma tissue and promotes the proliferation, differentiation, migration, and invasion of glioma cells [
6,
7]. However, the biological function of H19 in non-neoplastic CNS diseases including epilepsy remains unknown.
We previously showed by high-throughput microarray and bioinformatics analyses that H19 is upregulated in the latent period of TLE in rat and is involved in various aspects of epileptogenesis, including cell proliferation and immune and inflammatory responses [
8]. We therefore speculated that H19 may be involved in hippocampal glial cell activation during epileptogenesis. This was investigated in the present study by gain- and loss-of-function studies in a rat model of TLE. We also examined the possible downstream targets of H19.
Methods
Animal and human samples
Male Sprague–Dawley rats weighing 200–220 g were obtained from Vital River Experimental Animal Technology Co. (Beijing, China) and were housed in a temperature-controlled room with free access to standard food and water under a 12:12-h light/dark cycle. Surgically resected hippocampus specimens were obtained from patients with intractable TLE who underwent surgical treatment at Beijing Tiantan Hospital. Control hippocampal tissue was obtained from autopsies of four patients without a history of epilepsy or other neurological diseases within 8 h after death.
H19 overexpression and knockdown
H19 was overexpressed or silencing using an adeno-associated viral (AAV) vector delivery system as previously described [
8]. Briefly, a vector harboring H19 (AAV-H19) or a short hairpin RNA targeting H19 (AAV-shRNA) was constructed by Gene Chem Co. (Shanghai, China). The negative control was an empty AAV vector (AAV-NC) or one harboring a scrambled sequence (AAV-Scr: 5′-TTCTCCGAACGTGTCACGT-3′). The titers used were 1.0 × 10
12 for AAV9-H19 and 4.0 × 10
12 for AAV9-shRNA. A total of 6 μl AAV was infused into the right dorsal hippocampus (3.12 mm posterior to the bregma, 3.0 mm lateral to the midline, and 3.4 mm ventral to the bregma) and ventral hippocampus (5.04 mm posterior to the bregma, 5.0 mm lateral to the midline, and 6.4 mm ventral to the bregma; 3 μl at each location) [
9] through a microsyringe at a speed of 0.2 μl/min.
Epilepsy model
A kainic acid (KA)-induced status epilepticus (SE) model was established by intra-amygdala microinjection of KA 14 days after AAV injection according to our previously described technique [
8]. Briefly, the rats were placed in a stereotaxic apparatus (David Kopf Instruments, Tujunga, CA, USA), and 0.7 μl KA (1 μg/μl; Sigma–Aldrich, St. Louis, MO, USA) was injected into the right amygdala (2.76 mm posterior to the bregma, 4.5 mm lateral to the midline, and 8.6 mm ventral to the bregma) [
9] at a speed of 0.2 μl/min. Sham-operated controls were injected with an equal volume of saline.
Immunofluorescence analysis
Coronal sections (25 μm) were prepared at the level of the dorsal hippocampus (2.50–3.50 mm posterior to the bregma). Frozen sections were dried, washed, permeabilized, blocked in 5% goat serum, and incubated overnight with antibodies against neuronal nuclei (NeuN) (ab177487, 1:500 and ab104224, 1:200), glial fibrillary acidic protein (GFAP) (ab7260, 1:500), and OX42 (ab1211, 1:100) (all from Abcam, Cambridge, MA, USA). Immunolabeled sections were washed and incubated with goat secondary antibodies conjugated with Alexa Fluor 594 or Alexa Fluor 488 (Merck Biosciences, Nottingham, UK). Sections were mounted with medium containing 4′,6-diamidino-2-phenylin-dole (DAPI) (Vector Laboratories, Burlingame, CA, USA), and images were captured using an inverted fluorescence microscope (Olympus, Tokyo, Japan). GFAP+ or OX42+ cells were manually counted using ImageJ software (US National Institutes of Health, Bethesda, MD, USA).
Western blotting
Western blot analysis was performed as previously described [
10] using the following primary antibodies: rabbit polyclonal anti-GFAP (ab7260, 1:1000), mouse monoclonal anti-OX42 (ab1211, 1:500), rabbit polyclonal anti-interleukin (IL)-1β (ab9722; 1:500), mouse monoclonal anti-IL-6 (ab9324; 1:500), rabbit polyclonal anti-tumor necrosis factor (TNF)-α (ab6671; 1:500), rabbit monoclonal anti-p-Stat3 (ab76315; 1:500), and rabbit polyclonal anti-c-Myc (ab39688; 1:500) (all from Abcam, Cambridge, MA, USA). Rabbit monoclonal anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH) antibody (Abcam, ab181602, 1:3000) was used as a control. Protein band density was quantified using an Epson V330 Photo scanner (Seiko Epson Co., Nagano, Japan) and Quantity One software (Bio-Rad, Hercules, CA, USA).
Statistical analysis
Data are presented as mean ± standard error of the mean. Two-group comparisons were made with the unpaired Student’s t test, and multi-group comparisons were made by one-way analysis of variance followed by Sidak’s multiple comparison tests using Prism 5 software (GraphPad Inc., San Diego, CA, USA). Significance was accepted at P < 0.05.
Discussion
Most previous studies on H19 function focused on tumorigenesis. H19 was initially proposed as a tumor suppressor due to its capacity to suppress clonogenicity and tumorigenicity in tumor cells [
14,
15]. However, recent studies showed that H19 acts as an oncogene by promoting cell proliferation, migration, invasion, and metastasis in various malignancies including glioblastoma [
16‐
18]. Apart from these functions, H19 is also implicated in several other physiological conditions or diseases, such as cartilage degeneration in osteoarthritis [
19], skeletal muscle differentiation and regeneration [
20], and glucose metabolism in muscle cells [
21]. However, the role of H19 in non-neoplastic CNS diseases including epilepsy remains unclear. In the present study, we provide the first evidence that H19 promotes glial cell activation and stimulates inflammation in the hippocampus of rats with TLE.
Astrocyte activation is a continuum that includes changes in molecular expression, progressive cellular hypertrophy, and, in severe cases, proliferation and scar formation [
22]. In mild or moderate astrocyte activation, GFAP expression is slightly upregulated and the cell body and processes undergo hypertrophy, with little or no astrocyte proliferation. However, in severe diffuse reactive astrogliosis, GFAP expression is markedly increased, which is accompanied by extensive hypertrophy of the cell body and processes and astrocyte proliferation [
23]. In the present study, astrocytes in the hippocampus of epileptic rats were activated in the latent and chronic phases of TLE, which is consistent with previous reports [
13,
24]. Mild to moderate gliosis, which typically does not cause astrocyte proliferation, is usually observed in the early stages after SE [
25]. We also found that astrocyte activation was mild or moderate, as evidenced by the upregulation of GFAP expression and cellular hypertrophy without an obvious increase in cell number in the latent period of TLE (7 days after SE). Furthermore, H19 induced an increase in GFAP expression and hypertrophy of astrocytes rather than cell proliferation in the latent period of SE. In the chronic period (30 days after SE), astrocyte activation was extreme with severe hypertrophy of the cell body and processes; moreover, astrocyte proliferation in areas of pyramidal neuron loss was comparable to the hippocampal sclerosis observed in TLE patients [
26]. Unlike astrocytes, high expression of OX42 protein and cellular hypertrophy as well as proliferation of microglia was observed in both the latent and chronic periods of TLE, as previously reported [
27]. Furthermore, H19 induced the upregulation of OX42 and cellular hypertrophy and increased the number of microglia in the latent period of SE. Recent studies have shown that inflammatory cytokines are produced both by microglia and astrocytes [
28]; the increased levels of proinflammatory cytokines in epileptic rats observed here is in agreement with these findings. Moreover, H19 stimulated the release of proinflammatory cytokines. However, compared to sham rats, H19 knockdown did not inhibit proinflammatory cytokine expression, possibly because under normal conditions, proinflammatory cytokine levels in the hippocampus are too low to result in an observable difference upon H19 knockdown.
Molecular triggers that lead to glial cell activation and proliferation have not been fully characterized. There is increasing evidence to suggest that H19 has a growth-promoting function, since it enhances cell proliferation in tumors [
29,
30] and other diseases [
20,
31]. In the present study, we showed that H19 promotes astrocyte and microglia activation and proliferation under both epileptic and normal conditions. This is consistent with earlier reports as well as with our previous research [
8]. Various intracellular signaling pathways associated with Stat3, nuclear factor κB, and nuclear respiratory factor (Nrf) mediate cell hypertrophy, proliferation, and pro- or anti-inflammatory effects in astrocytes [
23]. The transcription factor Stat3, a key component of the JAK/STAT pathway, is important for astrocyte proliferation in CNS diseases [
11,
32,
33]. P-Stat3 is highly expressed in the rat hippocampus during different phases of epilepsy and in the temporal lobe of TLE patients [
13]. Astrocyte activation can be suppressed by inhibiting JAK/STAT signaling, indicating that Stat3 activation induces GFAP expression [
13]. In our study, Stat3 as well as its downstream effector c-Myc in the JAK/STAT pathway were upregulated after SE in the rat hippocampus and in TLE patients. Furthermore, H19 overexpression induced whereas its knockdown inhibited Stat3 and c-Myc protein expression in both normal and epileptic rats. Thus, H19 can itself promote gliosis via JAK/STAT signaling in addition to its role in astrocyte activation.