By David N. Leff

Have you ever noticed that bystanders are invariably "innocent?" That's because they're always in harm's way at an event - but never participate in it.

Now a team of neuroscientists reports populations of bystander brain cells, which are fortunate rather than innocent. They are the lucky beneficiaries of a novel gene therapy experiment that employs HIV-1 - the AIDS virus - as a vector delivering transgenes to the brain.

"When you do gene therapy," observed neurogeneticist Roscoe Brady, "ordinarily you may have to introduce the new or therapeutic gene into each cell - unless you can make the gene-encoded protein come out of the cell entered by the new transgene, then go to bystander cells, and correct them at the same time for the protein deficiency. We have developed a lentiviral vector," he added, "that looks as if it can do that."

Brady is chief of the Developmental and Metabolic Neurology Branch at the National Institute of Neurological Disorders and Stroke (NINDS). He is co-author of a paper in the current issue of the Proceedings of the National Academy of Sciences (PNAS), dated Oct. 10, 2000. Its title: "Intercellular delivery of a herpes simplex virus VP22 fusion protein from cells infected with lentiviral vectors."

"In early gene therapy work," Brady told BioWorld Today, "people always went for a dividing cell because it was easy to integrate the new gene - the transgene - into dividing cells. As they divide, the DNA opens up and a new gene can get in and be incorporated into the daughter cells. However," he pointed out, "the neurons in the brain, unlike malignant tumors or bone-marrow stem cells, do not divide. If we were going to do any successful gene therapy, we had to get our gene of interest into the non-dividing cell. And we showed that HIV lentiviral vectors can do that.

"Our approach does not necessarily exclude dividing cells." Brady said. "But the benefit we seek is for genetic disorders, using gene therapy to cure the neurons that are not dividing."

In One Package: HIV, Herpes, Green-Glow Protein

At the heart of that approach was a recombinant gene-delivery vector embodying non-infectious sequences of the HIV-1 lentivirus. "It carried nucleotide fragments," Brady recounted, "that contain two fusion genes of interest. The first encoded the herpes simplex virus type-1 tegument protein VP22; the second, the demonstrator marker gene EGFP - enhanced green fluorescent protein.

"Herpes VP22," Brady explained, "can take different proteins across cell membranes, and get them into other uncorrected or bystander cells that don't have the new gene. It facilitates delivery of the transgene product into lots of cells, even though they don't have the correcting gene in them. That bystander mechanism," he continued, "has to do with the structure of the VP22 protein. It has a short sequence of amino acids, which can move proteins across cell membranes.

"So if we don't have VP22," he explained, "the EGFP just sticks in the cell, with a little bit of movement. But if we put the VP22 tegument protein into the construct with the marker, then it will move out of the transduced cell and go to nearby cells, or even farther. We don't know how far it's going to go. In one experiment, it went into the cortex."

That in vivo experiment involved injecting the gene therapy construct into the brain ventricles - the fluid-filled canals that crisscross the brain's solid cerebral substance - of anesthetized mice. "The construct," Brady recounted, "went from the site of injection and spread out from the ventricle throughout the region of the brain into neighboring tissues, which is what we needed to do. For beneficial effect, you can't just correct a few cells," he pointed out. "You've got to correct all of them - globally.

"Whether your gene gets into every neuron that has to be corrected is problematic," he observed. "It probably won't. So what you want to do is have the transgene product move from the corrected cells to the nearby cells and even distant ones.

"The question is," he went on, "How do you do it? Do you aim into the ventricles, or do you have to inject it directly into the substance of the brain? It's very hard at this point to give a concise decision. We did both - intraventricular and brain substance as well. The results showed us that these transgenic fusion proteins would go from the gene-corrected cells to non-corrected bystander cells and presumably correct their metabolic defect - or whatever defect they have."

Brady made the point that this successful outcome is strictly a proof of principle, rather than a preclinical experiment. Conceptually, he suggested that the principle might eventually provide gene therapy for a neurological disorder such as Tay-Sachs disease. This horrendous, inherited pediatric affliction results from the miscoding of the gene for the enzyme hexosaminidase A.

Future Tay-Sachs Therapy? Animals First

"For gene therapy some day of Tay-Sachs disease," Brady foresees, "you would put the hexosaminidase gene into a viral construct with VP22, and introduce it into the brain - first in animal models. If it looks good, then you can do it in humans. We haven't gone to the animal models yet."

Meanwhile, he noted, "The federal government may want to patent this construct. We have submitted it for consideration of a patent application." Its three inventors are the PNAS paper's first and last authors, Zhennan Lai and Jakob Reiser, along with Brady.

One besetting hobgoblin of gene therapists is whether their recipient patient's immune system won't do what comes naturally to it - attack and reject the injected foreign transgene's protein product. "That immune backlash," Brady speculated, "may or may not be a problem for us. Some of these recipients, such as Tay-Sachs patients, do not make any hexosaminidase at all. When you have a cell producing this enzyme, would this new protein elicit an immune response - like a foreign protein? The answer," he suggested, "is maybe, maybe not.

"Particularly within the central nervous system," he continued, "we don't know whether there would be an immune response to the new protein or not. That is," Brady concluded, "we don't know what the immunological response to gene therapy is in the brain. We will be looking at that, that's for sure."