Background
Epilepsy is associated with precocious development of Alzheimer-type neuropathological changes, and the
APOE ε4 genotype has been associated with further risk of development of such changes [
1,
2]. A role for glial activation with excess expression of cytokines in epilepsy pathogenesis was first recognized as enlargement of microglia and astrocytes with overexpression of IL-1 and S100B, respectively [
3‐
5]. Such findings gave rise to a new understanding of the role of glial activation and overexpression of cytokines as potential precursors of neurodegenerative change, including Aβ plaques and neurofibrillary tangles [
6]. These findings are consistent with the idea that glia-related neuroinflammatory events are early contributors to epilepsy pathogenesis.
Neuronal stress, such as the hyperexcitability induced by glutamate in epilepsy, elevates neuronal expression of βAPP and release of sAPP, which activates microglia and induces excess IL-1 production. This elevation in IL-1 production is attenuated by ApoE 3, but not ApoE 4 [
7]. In turn, IL-1 induces further neuronal expression of βAPP and sAPP leading to further microglial activation and further release of IL-1 [
8]. IL-1 also induces neuronal expression of ApoE [
9], which in turn induces further expression of βAPP in an ApoE isoform-dependent manner; with ApoE3 more effective than ApoE4 [
10].
A great deal of research has been dedicated to understanding how and why the presence of an
APOE ε4 allele(s) is so strongly associated with negative outcomes in neurological conditions, such as head injury [
11]. Here, rather than taking this tack, we chose to investigate the potential for beneficial effects conferred by
APOE ε3 alleles due to their neuroprotective potential. Tissue samples from temporal lobes resected from epilepsy patients carrying two
APOE ε3 alleles were examined regarding an association between inheritance of these alleles and determinants of neuronal resilience. These determinants included the ability of neurons to mount appropriate acute phase responses, including increases in βAPP and ApoE, as well as management of DNA damage, maintenance of morphological integrity and glial activation. Our findings indicate that the
APOE ε3,3 genotype confers a neuroprotective advantage over the
APOE ε4,4 genotype, in the setting of intractable epilepsy with its accompanying hyperexcitability-induced neuronal damage, glial activation and excessive expression of the proinflammatory cytokine IL-1α.
Methods
Patients and specimens
Resected temporal lobe tissues were obtained from 95 epilepsy patients; of those 59 were included in this study (39 males and 20 females; 52 APOE ε3,3 and 7 APOE ε4,4) with an age at surgery ranging from 0.25 to 71 years. Analyses of surgical waste remains from temporal lobectomy surgeries to treat intractable, drug-resistant epilepsy were compared to those of autopsy samples from neurologically normal individuals brought to autopsy for reasons other than this study. Both surgical waste and autopsy tissue are exempt under 46.101 5(b) and approved by our University of Arkansas Institutional Review Board.
All patients underwent anterior temporal lobectomy for treatment of medication-resistant intractable epilepsy. Tissue was sectioned at 4 mm intervals and alternate sections were preserved by flash freezing for molecular analyses and by formalin fixation for histological evaluation. Preliminary immunohistochemical analysis was performed on all epilepsy cases, and a smaller group was selected for further investigation. Six APOE ε3,3 cases (five males and one female, ages 18, 24, 38, 44, 67 and 57 years, respectively) and four APOE ε4,4 cases (three males and one female, ages 10, 22, 50 and 34, respectively) were selected for more extensive analyses, based on age in the case of APOE ε3,3 patients, and with regard to availability of sufficient frozen tissue for molecular analyses among APOE ε4,4 patients. Sufficient frozen tissue and fixed tissue was available for both immunohistochemical and molecular analyses of four APOE ε4,4 patients (three males and one female, ages as above). For uniformity, immunohistochemical examination was restricted to cortical layers III, IV, V and VI of the superior temporal lobe. For comparison of results from our APOE ε3,3 and APOE ε4,4 genotype patients, analogous temporal lobe tissues from neurologically normal individuals of varying APOE genotype and at older ages (four males and one female, ages 71, 97, 59, 50 and 93 years) were assessed. This selection was based on the premise that individuals with pre-AD (Alzheimer's disease) or with AD at these ages would have plaques.
Reagents
The antibodies used were as follows: rabbit anti-human IL-1α (Peprotech, Rocky Hill, NJ, USA, 4:1,000); goat anti-human APOE (Life Technology, Grand Island, NY, USA, 1:50); mouse anti-human Aβ/βAPP (Covance, Denver, CO, USA, 1:1,000); rabbit anti-synaptophysin (Abcam, Cambridge, MA, USA, 1:1,000); rabbit anti-phosphorylated tau (Abcam 1:3,000); rabbit anti-actin (Santa Cruz Biotechnology, Santa Cruz, CA, USA, 1:5,000) were diluted in antibody diluent (DAKO, Carpenteria, CA, USA), and Iba-1 (WAKO, Richmond, VA, USA,1:400). Mounting media containing Prolong Gold antifade reagent with DAPI (Life Technologies) was used to stain nuclei.
Immunohistochemistry
Paraffin-embedded tissue was sectioned at 7 μm, deparaffinized in xylene and rehydrated in graduated ethanol solutions to deionized water. Sections for IL-1α immunoreaction were placed in boiling sodium citrate buffer (0.01 M, pH 6.0) for 20 minutes; sections for βAPP and ApoE were placed in trypsin solution for 10 minutes at 37°C, and all were blocked using protein block (DAKO), and incubated overnight at room temperature. Secondary antibodies, Alexa Fluor donkey anti-goat and donkey anti-rabbit were diluted in antibody diluent (DAKO) and sections were incubated for 60 minutes, washed three times for 5 minutes each in distilled H2O, and coverslipped with prolong Gold with DAPI.
Plaque analysis
Plaques were identified by the simultaneous presence of ApoE and Aβ immunoreactivity. The number of plaques in 10 consecutive 20X images (0.37 mm
2) from sections of tissue from each patient was enumerated. Plaque phase was based on Braak and Braak staging of Aβ plaques [
12] and estimated with regard to our experience with such estimation in Alzheimer tissue.
Image analysis
Similar to a previous study [
10], a quantitative approach was used to examine the number of glia and neurons. Three images per slide (40X magnification) were captured at identical exposure settings, using a Nikon Eclipse E600 microscope (Melville, NY, USA) equipped with a Coolsnap ES monochrome camera (Photometrics, Tucson, AZ, USA). Each of the three images, spanning 37,241.5 μm
2, was acquired and analyzed using NIS-Elements BR3 software
http://Nikon.com and thresholded. Only microglia immediately adjacent to neuron somas were counted. Data were analyzed by ANOVA to assess difference among groups. Significance was provided by
P ≤ 0.05.
Reverse transcription (RT) reaction and polymerase chain reaction (PCR) amplification
Total RNA was extracted from brain tissue using TriReagent™ RNA (Molecular Research Center, Cincinnati, OH, USA) according to the manufacturer's instructions. RT-PCR was performed as previously described [
9]. Briefly, for comparisons of mRNA levels among different RNA samples, RT reactions were performed simultaneously using reagents from a single master mix. PCR was performed using reagents from Clontech (Mountain View, CA, USA). The sequences of primers for human IL-1α and GAPDH, amplification cycles and annealing temperature are provided in Table
1. PCR reactions were stopped by incubation for 10 minutes at 72°C. Equal volumes of reaction mixture from each sample were loaded onto 1.2% agarose gels, and fluorescent images were digitally captured for analysis of intensity with NIH Image software 1.60 version
http://rsbweb.nih.gov/nih-image/. Levels of IL-1α were normalized relative to GAPDH in the same sample.
Table 1
Human gene sequences for IL-1α and GAPDH, PCR annealing temperatures and number of amplification cycles
Gel-based PCR |
IL-1α | F: AAG CCT TCC TGC CGC AAC | 57 | 32 |
| R: CTG CAC CTA CCA AAC ACG G | | |
GAPDH | F: AGG TCG GAG TCA ACG GAT TTG | 57 | 32 |
| R: TGG CAG GTT TTT CTA GAC GGC | | |
Western immunoblot assay
Proteins were extracted from brain tissue in a lysis buffer comprising 20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1% Nonidet P40, 1 mM EGTA, 1 mM EDTA and 1% sodium deoxycholate; lysates were quantified using a Micro BCA assay reagent kit (Pierce, Rockford, IL, USA) as described previously [
13]. Aliquots (50 μg each) were loaded onto 4 to 12% Criterion
XT precast Gels (Biorad, Hercules, CA, USA, Catalog # 345-0123), subjected to electrophoresis at 90 V for 1.5 h, and transferred to nitrocellulose membranes. Blots were blocked in I-Block Buffer (Applied Biosystem Inc., Bedford, MA, USA) for 60 minutes, then incubated overnight at 4°C with either goat polyclonal antibody anti-IL-α (Santa Cruz Biotechnology 1:500), mouse anti-human Aβ/βAPP (Covance 1:1,000), rabbit anti-synaptophysin (Abcam 1:1,000), rabbit anti-phosphorylated tau (Abcam 1:3,000), or rabbit anti-actin (Santa Cruz Biotechnology 1:5,000); the latter of which was used here for calculating the relative levels of the other proteins assessed by western blot analyses. Membranes were then incubated for 1 h at room temperature with alkaline phosphatase-conjugated secondary antibody and developed using the Western-Light™ Chemiluminescent Detection System (Applied Biosystem Inc., Bedford, MA, USA). Autoradiographs were digitized and analyzed using NIH Image software, version 1.60.
TUNEL staining procedure
For terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) (NeuroTacs Kit, 4823-30-K, Trevigen, Gaithersburg, MD, USA) reactions, rehydrated sections were permeablized with NeuroPore® for 30 minutes at room temperature, washed in PBS buffer, placed in TdT labeling buffer for 5 minutes, treated with the labeling reaction mix (TdT dNTP, 50 × Mn+2, and TdT Enzyme) for 60 minutes at 37°C followed by stop buffer for 5 minutes, then streptavidin AF 594 conjugate (Invitrogen, S32356) for 10 minutes at room temperature. The sections were then treated with 0.1% Sudan black B in 70% ethanol for two minutes to block lipofuscin autofluorescence, washed in three changes of distilled H2O, five minutes each; and coverslipped with Prolong Gold with DAPI.
Statistical analysis
Data were analyzed using an unpaired t-test, and values were considered significantly different when the P-value was ≤ 0.05. Results are expressed as mean ± SD.
Discussion
Tissues from patients undergoing temporal lobectomies for drug-resistant epilepsy reveal
APOE genotype-specific links between glial and neuronal stress responses. This influence of
APOE genotype in epilepsy appears to occur without regard to gender or age at the time of surgery. Glial activation with overexpression of IL-1 is well known to induce neuronal expression of two AD-associated, stress-related proteins ApoE and βAPP [
10,
15]. Connections among
APOE genotype, epilepsy and AD have been drawn, but mechanisms by which the
APOE ε4,4 genotype heightens intensity of neuronal damage or, conversely, how the
APOE ε3,3 genotype may act to promote neuronal resilience remains unclear.
The numbers of neurons in temporal lobe tissue of our epilepsy patients who were either
APOE ε3,3 or
APOE ε4,4 genotype were similar, but there were striking differences in the indicators of degeneration in neurons, as neurons from patients with
APOE ε3,3 were larger, appeared more normal morphologically, and had less DNA damage. These findings suggest that neurons from individuals with the
APOE ε3,3 genotype are better able to mount appropriate and more liberal repair responses to the damaging hyperexcitability of epilepsy than are their
APOE ε4,4 counterparts, suggesting that
APOE ε3, but not
APOE ε4, alleles confer resilience to host neurons no matter the type of injury. This might be inferred from studies reporting earlier onset of epilepsy, especially following traumatic brain injury in patients with
APOE ε4 alleles [
16,
17].
Our finding of elevated synthesis of IL-1α in the temporal lobe of epilepsy patients compared to that in neurologically normal controls confirms an earlier report [
4] of elevated IL-1α protein and accompanying glial activation and other neuroinflammatory changes. However, the association made here between this overexpression of IL-1α and beneficial effects toward enhancing neuronal resilience may help to explain, at least in part, why IL-1α elevation is necessary for neuronal survival in dorsal root ganglion cell cultures [
18]. Moreover, evidence of greater neuron sparing in epilepsy patients with
APOE ε3,3 than
APOE ε4,4 genotype may be a case in point for genetic variation favoring typical, evolutionarily old, acute phase responses [
19] of neurons to adverse stimuli, which includes elevation of IL-1α, βAPP and ApoE expression [
10] and protection against DNA fragmentation.
The original report of a role for IL-1α in induction, maintenance and propagation of axonal sprouting in an experimental model of neurodegeneration [
20] and an association between glial activation and sprouting of mossy fibers in epilepsy [
21] is supported by our finding of somewhat elevated synaptophysin levels in combination with high numbers of neuron-associated, IL-1α immunoreactive microglia and elevation of IL-α mRNA and protein levels. In addition, the apparent elevation of synaptophysin expression noted here in immunoblots of neural tissue proteins from our epilepsy patients compared to that from our neurologically and neuropathologically normal controls may be explained if, as previously noted in animal models of epilepsy, [
21,
22] there is neuronal sprouting in epilepsy patients.
Amyloid-β plaques are obligatory for the diagnosis of AD and are most prominent in the elderly. In contrast, Aβ plaques in epilepsy, as shown here and as reported in about 10% of cases [
2], are evident at young ages. For instance among our patients, a 10-year-old patient had Aβ/ApoE immunoreactive plaques in a distribution similar to that noted in temporal lobes of Alzheimer patients. The presence of plaques at such early ages suggests that they are harbingers of impending neurodegeneration and AD. Although the number of plaques was similar in tissue from our patients without regard to
APOE genotype, in our one
APOE ε4,4 patient the developmental phase of Aβ plaques appeared to be advanced relative to those observed in our
APOE ε3,3 patients -- our
APOE ε4,4 patient had dense core neuritic Aβ plaques, while such dense core plaques were not found among the plaques observed in our
APOE ε3,3 patients. This observation is consistent with the possibility that the phase of Aβ plaque progression is accelerated in those with
APOE ε4,4 genotype and supports the findings of Marz
et al., regarding the role of
APOE genotype in the onset of Aβ plaque pathology and the presence of dense core plaques [
23].
Alzheimer's patients are more likely to have seizures than are those in the general population [
24]. This, together with our findings and the previously reported preferential occurrence of seizures in younger Alzheimer patients [
25], supports a suggested relationship between the high levels of Aβ in the brains of epilepsy patients [
26] and increased risk for development of AD. These findings are consistent with the idea that AD-related neuronal stress and its sequelae, including excess neuronal βAPP and ApoE expression and glial activation with elevated cytokine expression, combined with known IL-1-driven elevation of neuronal and glial glutamate production contribute to the hyperexcitability of epilepsy [
9]. Moreover, these findings, together with evidence from our epilepsy patients, suggest that ApoE genotype, in particular
APOE ε4,4 may favor rapidity of disease progression as well as risk for associated memory disturbances. Conversely, a better understanding of mechanisms by which
APOE ε3 alleles confer the neuronal protection shown here may facilitate development of therapeutic strategies toward improving outcomes for epilepsy patients, as well as patients with other neuronal distresses.
Conclusion
The most striking aspect of this work is that our findings illuminate the "other" side of the
APOE genotypic equation in showing ways in which
APOE ε3 alleles may act to preserve important aspects of neuronal abilities to mount appropriate, beneficial stress responses to hyperexcitability, neuroinflammation and neuronal DNA damage. In addition, our findings are consistent with the idea that as neurons with
APOE ε4 alleles are less resilient to the chronic excitation of epilepsy and more susceptible to DNA damage, patients who carry
APOE ε4 alleles are at greater risk of developing AD than are those with
APOE ε3 alleles. Moreover, our findings are in accord with the possibility that epilepsy-related neuropathological changes, such as increases in the levels of Aβ peptides, contribute to propagation of epileptiform activity in adjacent neurons and furtherance of neuropathological changes and the risk of AD [
26].
Acknowledgements
The authors are especially grateful to the patients who shared with us; without them, this work could not have been done. We would also like to thank Dr. John L. Greenfield and Dr. Steven W. Barger for their helpful advice, and Dr. Ling Liu, JoAnn Biedermann and Richard A. Jones for their skillful technical assistance and advice. This work was supported in part by NIH-NIA AG12411, the Windgate Foundation, the Donald W. Reynolds Foundation, and the Grand Aerie Fraternal Order of the Eagles, Auxiliary #60.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
OA conducted and helped with the design of all experiments, with interpretation of the data and writing of the manuscript. REM conducted all neuropathological evaluations and contributed to interpretation of results and writing of the manuscript. FB was the neurosurgeon who provided the tissue and reviewed the writing. WSTG designed the study with OA, verified and helped with interpretation of the data, and contributed to the writing. All authors read and approved the final manuscript.