Rapid and efficient electroporation-based gene transfer into primary dissociated neurons

Abstract

Non-viral gene transfer into neurons has proved to be a formidable task. Here, we describe an electroporation-based method that allows efficient and reliable DNA transfer into dissociated neural cells before they are plated and cultured. In hippocampal neural cells derived from either neonatal mouse or embryonic chicken brains, a high transfection rate was already observed 5 h after transfection, and reached 40–80% in 24 h, as monitored by expression of enhanced green fluorescent protein (eGFP). The level of eGFP expression per cell depended on the amount of DNA used in a gene transfer experiment. The survival and neuritic length of transfected cells resembled that of non-electroporated cells. The transfected neurons showed normal immunostaining for endogenous synaptic protein synaptophysin and the neural cell adhesion molecule (NCAM). Furthermore, efficient gene transfer of the NCAM isoform NCAM140 and eGFP-tagged NCAM140 could be achieved, allowing visualization of NCAM140 expression. Also, a glycosylphosphatidylinositol-anchored eGFP could be efficiently expressed, highlighting lipid rafts without altering electrophysiological properties of transfected neurons. When neurons transfected with green and red fluorescent proteins were cocultured, fine details of their interactions could be revealed in time-lapse experiments. Thus, the method provides a useful tool for elucidation of genes involved in different neuronal functions, including neurite outgrowth, synaptogenesis and synaptic transmission.

Introduction

Efficient and controlled gene delivery into neural cells has proven difficult to achieve. Non-viral methods, such as calcium phosphate/DNA co-precipitation, cationic lipids, ballistic methods or microinjection, are often toxic to neurons and rarely give a transfection efficiency exceeding 20% of the surviving cells (Berry et al., 2001, Washbourne and McAllister, 2002). Transfection methods using modified adeno-associated or semliki-forest viruses provide high transfection rates, but the construction of viral vectors is often time consuming and associated with safety concerns. Furthermore, the size of the inserted DNA is limited, and more importantly, viral infections might interfere with metabolic processes in the cell, e.g. protein processing (Wu et al., 1998, Nilsson et al., 1998). Also, after viral transduction the level of protein expression is usually very high, and this often results in abnormal protein targeting and/or intracellular signaling. The available data suggest that the electroporation method could be viable alternative to viral transfection in achieving high transfection rates (Teruel et al., 1999, Morales et al., 2000, Colicos at el., 2001). In the present study we used a novel electroporation-based method, the Nucleofection™, for efficient transfection of murine and chicken neurons. This method allows high transfection efficiencies of different primary cells and manipulation of protein expression levels can be achieved by altering the amount of DNA per gene transfer experiment (Chun et al., 2002, Buttgereit and Schmidt-Wolf, 2002, Hamm et al., 2002, Schmidt-Weber et al., 2002, Lenz et al., 2003). We demonstrate that the method allows efficient gene transfer into neural cells and that transfected neurons form normal neurites, receive synaptic contacts and exhibit normal electrophysiological properties.