Anticonvulsant and antiepileptogenic effects mediated by adeno-associated virus vector neuropeptide Y expression in the rat hippocampus

Abstract

Neuropeptide Y (NPY) inhibits seizures in experimental models and reduces excitability in human epileptic tissue. We studied the effect of long-lasting NPY overexpression in the rat hippocampus with local application of recombinant adeno-associated viral (AAV) vectors on acute kainate seizures and kindling epileptogenesis. Transgene expression was significantly increased by 7 d, reached maximal expression by 2 weeks, and persisted for at least 3 months. Serotype 2 AAV vector increased NPY expression in hilar interneurons, whereas the chimeric serotype 1/2 vector caused far more widespread expression, also including mossy fibers, pyramidal cells, and the subiculum. EEG seizures induced by intrahippocampal kainate were reduced by 50-75%, depending on the vector serotype, and seizure onset was markedly delayed. In rats injected with the chimeric serotype 1/2 vector, status epilepticus was abolished, and kindling acquisition was significantly delayed. Thus, targeted NPY gene transfer provides a potential therapeutic principle for the treatment of drug-resistant partial epilepsies.

Figure 1.

Anteroposterior spread of NPY expression in the injected hippocampus of representative rats 8 weeks after infusion of serotype 2 (a1–a5) or serotype 1/2 rAAV–NSE–NPY (b1–b5, c1–c5) or in empty vector-injected rats (d1–d5). For serotype 2 rAAV–NSE–NPY, a1–a5 depict the hybridization signal in anteroposterior representative slices encompassing the injection site. In a3 (marked with an asterisk), the signal was higher in the dentate gyrus compared with a matched slice from empty vector-injected rats (d2). a1 and a5 show a hybridization signal in the dentate gyrus similar to matched slices from controls (d1, d4, respectively). High-magnification micrographs showing the hybridization signal in the dentate gyrus of representative experimental and control rats are shown in Figure 2. The spread of the vector around the injected site was ∼1.5 mm. For serotype 1/2 rAAV–NSE–NPY, b1–c5 depict the hybridization signal in anteroposterior representative slices encompassing the dorsal (b1–b5) or ventral (c1–c5) injected hippocampus. b2 and c4 (marked by asterisks) depict the slices closest to the site of the injection. High-magnification pictures of NPY mRNA expression in representative rats injected with serotype 1/2 are shown in Figure 3. The spread of this vector around the injection site was ∼2.5 mm. e–h show Nissl-stained sections from rats injected 8 weeks previously with rAAV–NSE–NPY serotype 2 (e, f) or serotype 1/2 (g, h). e and g depict the dentate gyrus and high magnification of the polymorphic cell layer of the dentate hilus close to the injection site (f, h, arrowheads). We observed only mechanical damage attributable to insertion of the needle track similar to that found in empty vector- or saline-injected rats (data not shown). Scale bars: a1–d5, 800 μm; e–h, 500 μm.

Figure 2.

Expression of NPY in the injected (b, d, f, h, j, l) and contralateral (a, c, e, g, i, k) hippocampus of a representative rat injected 8 weeks previously with serotype 2 rAAV–NSE–NPY or rAAV–NSE–GFP (k–m). b shows enhanced NPY immunoreactivity in fibers in the hilus of the injected hippocampus and in the inner and outer molecular layers where hilar interneurons project (b, arrows). A sharp band of NPY immunoreactivity is seen in the inner molecular layer contralateral to the injection site where commissural fibers originating from hilar mossy cells terminate (a, arrow). Except for NPY staining in the inner molecular layer, peptide expression in the contralateral hippocampus was similar to that observed in empty vector-injected rats (data not shown). e–h show high magnification of corresponding areas in b and a, respectively. f and h show a lack of NPY staining in CA3 pyramidal neurons, mossy fiber terminal fields, and CA1 pyramidal cells, although stratum oriens and radiatum interneurons are visible; this was similar to what was observed in e and g from the contralateral hippocampus and in empty vector-injected rats (data not shown). i and j show dark-field photomicrographs of hilar interneurons from slices hybridized with the ³⁵S-labeled NPY oligoprobe. Note the enhanced NPY mRNA transcript level in the polymorphic cell layer (j) compared with the corresponding control section from empty vector-injected rats (i). k and l represent enlargements of the corresponding boxed areas in i and j. The granular signal over the granule cell layer in j represents background staining. m–o depict GFP distribution in interneurons in the dorsal-injected hippocampus 8 weeks after rAAV–NSE–GFP infusion. n and o show high magnification of boxed areas in m. The spotty expression of GFP in the granule cell layer in m was observed sporadically and was attributable to cell bodies likely representing GABA-containing pyramidal-shaped basket cells laying close to or within the granule cell layer (Ribak and Seress, 1983). GC, Granule cell; h, hilus; CA1 and CA3, pyramidal cells; iml, inner molecular layer; oml, outer molecular layer. Scale bars: a, b, 500 μm; c–j, m, 200 μm; k, l, n, o, 50 μm.

Figure 3.

Expression of NPY in the injected hippocampus of a representative rat 8 weeks after infusion of serotype 1/2 rAAV–NSE–NPY. a–j depict immunocytochemical distribution of NPY. NPY immunoreactivity was predominantly expressed in the mossy fibers of the injected dorsal hippocampus (b, d, f). It was present in the inner and outer molecular layers of the injected hippocampus (b, d, arrows) and forms a dense immunoreactive band in the inner molecular layer of the contralateral hippocampus (a, c, arrow), reflecting NPY contained in terminals of commissural fibers presumably arising from excitatory mossy cells near the injection site. NPY was also increased in CA3 (f) and CA1 (h) pyramidal neurons and in the subiculum (j). e, g, and i represent sections of the contralateral hippocampus corresponding to f, h, and j, respectively. NPY staining in these sections was similar to that observed in empty vector-injected rats (data not shown). k–n depict dark-field photomicrographs of NPY mRNA in coronal sections of the injected hippocampus hybridized with ³⁵S-labeled NPY oligoprobe; an intense hybridization signal is observed in granule cells (k, arrow), hilar interneurons (k), CA3 (l), CA1 (m), and the subiculum (n, arrows). Arrowheads in the various panels depict labeled interneurons. o–q show the anteroposterior pattern of distribution of GFP in the injected dorsal hippocampus. GC, Granule cells; h, hilus; CA1 and CA3, pyramidal cells; iml, inner molecular layer; oml, outer molecular layer; Sub, subiculum. The arrow in a depicts the inner molecular layer enlarged in c; the arrows in b depict inner and outer molecular layers. Scale bars: a, b, o–q, 500 μm; c–n, 200 μm.

Figure 4.

EEG tracing depicting seizure activity induced by 250 ng of intracerebroventricular kainic acid in rats injected 8 weeks previously with rAAV–NSE–NPY serotype 1/2 (d, f) or empty vector (a–c). a and d represent baseline recordings in the right cortex (RCTX) and left cortex (LCTX) or right hippocampi (RHP) and left hippocampi (LHP). Traces depict discrete (b, e) and prolonged (c) seizure episodes, respectively. Only discrete seizure episodes (e) were observed in rAAV–NSE–NPY-injected rats. Calibration, 5 sec; time elapsed from kainic acid injection is reported above the traces. The annexed table reports quantification of EEG seizure activity, as described in Materials and Methods.

Figure 5.

Inhibition of intrahippocampal kainate-induced seizures in serotype 2 versus serotype 1/2 rAAV–NSE–NPY-injected rats. Error bars indicate means ± SE (n = 5–10) of the number of seizures, time spent in seizures (ictal activity), and time to onset of first seizure expressed as a percentage of values in respective controls (rats injected with rAAV–NSE-empty: onset, 6.5 ± 0.3 min; number of seizures, 28 ± 5; time in seizures, 92.8 ± 14.3 min; n = 10). Rats were injected with the respective vectors bilaterally into the dorsal hippocampus, and seizures were induced 8 weeks later by unilateral intrahippocampal application of 40 ng of kainic acid. *p < 0.05; **p < 0.01 versus respective controls; °p < 0.05 versus serotype 2 by Tukey's test.