Abstracts

Rationale: Hypothalamic hamartomas (HH) are associated with refractory epilepsy, and are intrinsically epileptogenic. The basic mechanisms responsible for intrinsic epileptogenesis of HH tissue are unknown. Small neurons occurring in clusters predominate in HH, although larger neurons are also present (Coons, et al, J Neuropathol Exp Neurol 2007;66:131-141). The study of acutely dissociated neurons from surgically resected HH tissue has recently shown that small HH neurons possess intrinsic spontaneous pacemaker-like firing activity and are immunoreactive for glutamic-acid decarboxylase-67 (GAD67) (Wu, et al, Ann Neurol 2005;58:371-382). Here we have studied HH tissue with immunohistochemistry for GAD and a panel of calcium-binding proteins.Methods: Four-micron thick sections from surgically resected HH tissue were studied with immunohistochemistry for GAD67 (n=7 cases), GAD65/67 (n=18), calbindin (n=12), calretinin (n=10), and neuropeptide Y (NPY) (n=10). Tissue for immunostaining was obtained from a total of 34 patients with HH and refractory epilepsy (mean age at time of surgery 13.4 years, range 0.7 to 55.8 years).Results: Immunostaining with GAD67 and GAD65/67 showed robust punctate staining of the neuropil, suggesting the presence of GABAergic processes and abundant axodendritic synapses. Punctate staining immediately adjacent to neuron cell bodies was also seen, suggesting the presence of axosomatic GABAergic terminals. These were most prominent surrounding large neurons, but were also seen around small neurons. There was a striking degree of regional variation of GAD immunoreactivity within individual sections at low magnification. This variation may conform, in part, to the nodular microarchitecture seen with most HH cases. A minority of small neuronal (~10 micron diameter) cell bodies was immunopositive for GAD. Few large neurons (~20 micron diameter) were also positive, and some rare very large cells (~50 micron diameter) could also be identified as positive. Calbindin and NPY immunostaining was quite limited, with positive labeling of a few isolated processes and/or large neurons in only half of the cases studied. Conversely, calretinin immunostaining showed consistently robust reactivity in all cases studied, with labeling of multiple cell bodies and excellent delineation of randomly oriented processes.Conclusions: HH tissue shows robust GAD immunoreactivity of neuropil, presumably reflecting the localization of GAD protein in axonal processes and synaptic structures. Small and large neurons are also immunopositive for GAD. Analysis of possible subgroups of GABAergic neurons by immunostaining with calcium-binding proteins shows abundant reactivity to calretinin, but relatively limited immunostaining with calbindin and NPY. GABAergic cells with a calretinin-expressing phenotype may be a prominent cell type within HH tissue. Further study of this “interneuron” phenotype in HH tissue is underway, and may help explain the mechanism for intrinsic epileptogenesis of HH tissue. (This work supported by a grant from the Barrow Neurological Foundation [JFK]).