Skip to Content
Merck
  • Lack of CAR impacts neuronal function and cerebrovascular integrity in vivo.

Lack of CAR impacts neuronal function and cerebrovascular integrity in vivo.

Experimental neurology (2016-06-01)
Baddreddine Boussadia, Giuseppe Gangarossa, Laila Mselli-Lakhal, Marie-Claude Rousset, Frederic de Bock, Frederic Lassere, Chaitali Ghosh, Jean-Marc Pascussi, Damir Janigro, Nicola Marchi
ABSTRACT

Nuclear receptors (NRs) are a group of transcription factors emerging as players in normal and pathological CNS development. Clinically, an association between the constitutive androstane NR (CAR) and cognitive impairment was proposed, however never experimentally investigated. We wished to test the hypothesis that the impact of CAR on neurophysiology and behavior is underlined by cerebrovascular-neuronal modifications. We have used CAR(-/-) C57BL/6 and wild type mice and performed a battery of behavioral tests (recognition, memory, motor coordination, learning and anxiety) as well as longitudinal video-electroencephalographic recordings (EEG). Brain cell morphology was assessed using 2-photon or electron microscopy and fluorescent immunohistochemistry. We observed recognition memory impairment and increased anxiety-like behavior in CAR(-/-) mice, while locomotor activity was not affected. Concomitantly to memory deficits, EEG monitoring revealed a decrease in 3.5-7Hz waves during the awake/exploration and sleep periods. Behavioral and EEG abnormalities in CAR(-/-) mice mirrored structural changes, including tortuous fronto-parietal penetrating vessels. At the cellular level we found reduced ZO-1, but not CLDN5, tight junction protein expression in cortical and hippocampal isolated microvessel preparations. Interestingly, the neurotoxin kainic acid, when injected peripherally, provoked a rapid onset of generalized convulsions in CAR(-/-) as compared to WT mice, supporting the hypothesis of vascular permeability. The morphological phenotype of CAR(-/-) mice also included some modifications of GFAP/IBA1 glial cells in the parenchymal or adjacent to collagen-IV(+) or FITC(+) microvessels. Neuronal defects were also observed including increased cortical NEUN(+) cell density, hippocampal granule cell dispersion and increased NPY immunoreactivity in the CA1 region in CAR(-/-) mice. The latter may contribute to the in vivo phenotype. Our results indicate that behavioral and electroencephalographic changes in adult CAR(-/-) mice are concomitant to discrete developmental or structural brain defects. The latter could increase the vulnerability to neurotoxins. The possibility that interfering with nuclear receptors during development could contribute to adulthood brain changes is proposed.