Elsevier

Brain Research

Volume 793, Issues 1–2, 18 May 1998, Pages 169-175
Brain Research

Research report
Recombinant adeno-associated virus vector: use for transgene expression and anterograde tract tracing in the CNS

https://doi.org/10.1016/S0006-8993(98)00169-3Get rights and content

Abstract

We used a recombinant adeno-associated virus vector (AAV) to deliver a foreign gene, green fluorescent protein (GFP), into mature neurons in adult rat CNS in vivo. Microinjections of AAV as small as 50 nl transduced hundreds of neurons at the injection site. There was virtually no retrograde transport as fewer than one neuron per brain was found distant from the injection site that exhibited GFP immunoreactivity. The gene product, GFP, filled the entire neuronal cytoplasmic compartment; GFP immunoreactivity was robust in cell bodies, axons, and nerve terminals. There was no tissue damage at the injection sites or pathogenicity indicated by changes in astrocytic or microglial markers. There was no inflammatory response as judged by leukocytic invasion. Gene expression in transduced cells was robust and apparently permanent: there was no evidence of phenotypic reversion up to 12 weeks following infection. AAV is an excellent vector for introducing foreign genes into mature CNS neurons. Not only might it be an ideal vehicle for gene therapy, but also the GFP-containing AAV presents a new strategy for tracing long axonal pathways in the CNS, which is difficult with current tracers (PHAL, biotinylated dextrans).

Introduction

The ability to induce the expression of identified genes by specific populations of neurons in the brain has long been a goal of molecular neurobiology. A variety of means has been used to accomplish this goal, including liposomes and recombinant retro- and adenoviral vectors [12]. Each of these methods has specific disadvantages, so that none has yet emerged which provides reliable and continuous gene expression in CNS neurons, in the absence of local inflammation or tissue damage.

Recently, an adeno-associated virus (AAV) vector has been introduced which may circumvent many of these problems (see Ref. [14]for review). As the AAV vector DNA does not contain any viral coding sequences, there is no expression of viral proteins. As a result, the only exposure to viral proteins is the capsid, which is degraded soon after uptake. AAV then can potentially provide a means for introduction of foreign DNA to postmitotic cells, without inducing an immune reaction.

In tissue culture and in the CNS in vivo AAV has been capable of inducing long-lived, continuous expression of foreign genes by postmitotic neurons. If the injected AAV only transfects neurons at the injection site, and infection is without pathogenic consequences, the AAV vector might provide a long-sought method for CNS gene delivery. This vector could be used not only clinically for gene therapy, but also as a gene transfer tool for neuroscientists. However, the usefulness of AAV would be limited if the vector spreads throughout the brain in an unpredictable manner. In this study, we examined potential pathogenic responses to injection of a recombinant AAV containing the marker gene, green fluorescent protein (AAV-GFP). To assess transduction stability, we quantified the number of GFP immunoreactive neurons over time. Furthermore, we examined the spread of GFP through CNS pathways. Our observations suggest that AAV-GFP may prove to be an excellent anterograde tracer for long axonal pathways.

Section snippets

Materials and methods

Construction of rAAV vector plasmids. Recombinant AAV vector plasmids were constructed containing a synthetic form of the jellyfish green fluorescent protein gene, the so-called `humanized' form, which is termed UF5 [15]. The recombinant pAcp-UF5 was derived from pSSV9 by excising all of the AAV coding sequences flanked by two Xba1 sites, leaving only the viral inverted terminal repeats (ITRs), and inserting an UF5 expression cassette in its place (Fig. 1). The UF5 expression cassette was

Stable, efficient neuronal transduction

At each injection site, clusters of neurons were found that expressed GFP (Fig. 2A, Fig. 3). In the first four cases, sections were initially viewed under fluorescence optics, and green fluorescent neuronal profiles were seen at each injection site. Subsequently, immunocytochemical staining for GFP was found to identify larger numbers of neurons, and did not fade, and so this method was used to quantify gene expression in all cases.

Because the GFP filled out the cell bodies and proximal

Discussion

Our results indicate that AAV can be used as a vector to safely transduce cells in the CNS without causing gliosis or cellular inflammatory reactions. The expression seems to be limited to neurons at the site of injection, and is both stable and reliable. The expressed GFP filled the cytoplasm and all of the neuronal processes, down to the most distal axon terminals. Thus, this method is potentially suitable as a very powerful anterograde tracer of long pathways in the CNS.

We have confirmed and

Acknowledgements

This work was supported by USPHS grants NS22835, NS33987 and AG12856. We would like to thank Quan Hue Ha and Minh Ha for excellent technical work. We are grateful to Dr. RJ Samulski (The University of North Carolina) for donating the pAd8 and pSSV9 plasmids and Dr. N Muzyczka (University of Florida) for graciously donating the pTRUF5 plasmid.

References (15)

There are more references available in the full text version of this article.

Cited by (191)

  • Routes of the thalamus through the history of neuroanatomy

    2021, Neuroscience and Biobehavioral Reviews
View all citing articles on Scopus
View full text