The Journal of Neurophysiology Vol. 81 No. 4 April 1999, pp. 1881-1888
Copyright ©1999 by the American Physiological Society

Functional Expression of Exogenous Proteins in Mammalian Sensory Hair Cells Infected With Adenoviral Vectors

 ¹Department of Neurobiology, Harvard Medical School and Massachusetts General Hospital;  ²Howard Hughes Medical Institute, Boston, Massachusetts 02114;  ³Section of Molecular and Cellular Cardiology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205; and  ⁴Centre for Research in Neuroscience, Montreal General Hospital, Montreal, Quebec H3G 1A4, Canada

Holt, Jeffrey R., David C. Johns, Sam Wang, Zheng-Yi Chen, Robert J. Dunn, Eduardo Marban, and David P. Corey. Functional expression of exogenous proteins in mammalian sensory hair cells infected with adenoviral vectors. To understand the function of specific proteins in sensory hair cells, it is necessary to add or inactivate those proteins in a system where their physiological effects can be studied. Unfortunately, the usefulness of heterologous expression systems for the study of many hair cell proteins is limited by the inherent difficulty of reconstituting the hair cell's exquisite cytoarchitecture. Expression of exogenous proteins within hair cells themselves may provide an alternative approach. Because recombinant viruses were efficient vectors for gene delivery in other systems, we screened three viral vectors for their ability to express exogenous genes in hair cells of organotypic cultures from mouse auditory and vestibular organs. We observed no expression of the genes for -galactosidase or green fluorescent protein (GFP) with either herpes simplex virus or adeno-associated virus. On the other hand, we found robust expression of GFP in hair cells exposed to a recombinant, replication-deficient adenovirus that carried the gene for GFP driven by a cytomegalovirus promoter. Titers of 4 × 10⁷ pfu/ml were sufficient for expression in 50% of the ~1,000 hair cells in the utricular epithelium; < 1% of the nonhair cells in the epithelium were GFP positive. Expression of GFP was evident as early as 12 h postinfection, was maximal at 4 days, and continued for at least 10 days. Over the first 36 h there was no evidence of toxicity. We recorded normal voltage-dependent and transduction currents from infected cells identified by GFP fluorescence. At longer times hair bundle integrity was compromised despite a cell body that appeared healthy. To assess the ability of adenovirus-mediated gene transfer to alter hair cell function we introduced the gene for the ion channel Kir2.1. We used an adenovirus vector encoding Kir2.1 fused to GFP under the control of an ecdysone promoter. Unlike the diffuse distribution within the cell body we observed with GFP, the ion channel-GFP fusion showed a pattern of fluorescence that was restricted to the cell membrane and a few extranuclear punctate regions. Patch-clamp recordings confirmed the expression of an inward rectifier with a conductance of 43 nS, over an order of magnitude larger than the endogenous inward rectifier. The zero-current potential in infected cells was shifted by 17 mV. These results demonstrate an efficient method for gene transfer into both vestibular and auditory hair cells in culture, which can be used to study the effects of gene products on hair cell function.