ACS Publications Division - Journals/Magazines
About MDD - Subscription Info
September 2001, Vol. 4
No. 9, pp 43, 45, 48.
To MDD Home Page
Focus: Molecular Modeling
Feature Article
Update on gene therapy


Viruses remain the main vehicle as other possibilities come on line and promising new results are reported.

Since the death of Jesse Gelsinger nearly two years ago, gene therapy has been under a spotlight of scrutiny by federal regulators. The Arizona teenager’s death was the first time that a subject enrolled in a gene therapy clinical trial died as a result of the treatment. In the wake of his death, officials imposed tighter regulations that seemed poised to create setbacks in a field already challenged by genetic and technological hurdles.

Despite these challenges, researchers are discovering ways to overcome the difficulties that have limited the efficacy of gene transfer. More recently, headlines have described significant strides in animal models and early clinical studies in humans. Such breakthroughs are a fortuitous turn of events at a time when ebbing public confidence and stricter regulations raise the question of whether gene therapy will show medical promise.

Researchers at the University of Pennsylvania in Philadelphia have reported a key breakthrough in the treatment of a rare, inherited genetic form of blindness. The condition, known as Leber congenital amaurosis (LCA), causes nearly total blindness in infancy. Reporting in the May issue of Nature Genetics, the research team used dogs that have a naturally occurring mutation in the gene that causes them to develop impaired vision or blindness similar to that of affected infants. Using a common gene therapy virus-based vector, they were able to deliver a healthy copy of the gene to the dogs and restore their vision.

In a report published in Science last December, Canadian scientists at the University of Alberta, in Edmonton, infected K cells of the gut with insulin-making genes. Islet cells in the pancreas normally make insulin. Mice were subsequently able to produce insulin from the gut cells, protecting them from developing diabetes. Their findings offer hope for a gene-based alternative to daily insulin injections for diabetics.

So far, a total of 532 gene therapy clinical trials have been conducted around the world, according to stats compiled by the Journal of Gene Therapy, and are at various stages that range from pending to completion. The majority of trials are being carried out in the United States.

Gene therapy pundits generally agree that the genetic medicine of the future will be based on different technology than what is currently being used. Nevertheless, present technology is already showing steady signs of overcoming the obstacles that have limited gene therapy’s success so far.

Viral delivery methods
Viruses remain the most common method for delivering genetic material to cells. After millions of years of evolution, they are little more than self-replicating genetic material surrounded by a protein coat and have mastered the art of invading cells and setting up shop.

One of the most widely used viral delivery vehicles, or vectors, is adenovirus, a common cold virus to which nearly everyone has been exposed. What makes adenovirus so attractive is the extraordinary efficiency with which it can infect a large number of either dividing or nondividing cells. Nor is it discriminating—it infects a variety of cell types.

On the downside, adenovirus is notorious for its ability to trigger the immune system on subsequent exposure. To scale down the potential for an immune assault, viral genes are chopped out to prepare it for use as a vector. The end result is a weakened, or attenuated, virus. But until recently, even that hasn’t been enough to stave off an immune attack.

Compounding the problem, the delivered gene only produces its therapeutic protein for a couple of weeks at best. “One of the major reasons is that although the virus is attenuated, it will express the viral genes at low levels,” says Frank Graham, a biologist at McMaster University (Hamilton, ON), who is regarded as the grandfather of adenovirus research. Despite the best efforts, scant traces of viral genes can remain. “The leaky expression of viral genes is quite antigenic,” he adds. “The cells behave as if viral-infected and are eliminated by the immune system.” (Jesse Gelsinger is believed to have died as a result of a severe immune reaction to the adenovirus vector being used.)

Graham’s team has devised a way around the critical problem. A few years ago, they completely gutted all of the viral genes from the virus, generating what are called fully deleted or helper-dependent vectors. Now the vectors are being put to the test in long-term animal studies to determine their safety and efficacy.

The helper-dependent adenovirus vector is unable to replicate without a helper because its replication machinery is gutted, along with nearly everything else—save its ends, the therapeutic DNA, and the DNA sequence that enables it to package the newly replicated DNA into new virus particles. The helper virus contains all of the features necessary for viruses to replicate and produce millions of copies.

Both viruses are then used to infect a special cell line that Graham’s team developed, which prevents the helper virus from getting packaged into new virus particles. As a result, the helper virus assists with reproduction of the therapeutic vector, but itself is left behind.

The results, so far, are encouraging. Because the adenovirus is missing all of its viral genes, very little, if any, immune response occurs. Instead of the therapeutic gene being expressed for only one to two weeks, gene expression has been observed to last as long as two to three years, Graham says.

Brendan Lee, a physician and geneticist in the department of molecular and human genetics at Baylor College of Medicine in Houston, and colleagues have had success using helper-dependent vectors from the McMaster group for treating a genetic form of high cholesterol. Familial hypercholesterolemia is a condition that results from a missing receptor called the very low density lipoprotein receptor (VLDLR).

Using a helper-dependent vector carrying the VLDLR gene, Lee’s group supplied the gene that makes the missing receptor to mice that lacked it. As a result, the cholesterol levels in the mice dropped significantly and their arteries were nearly free of plaque. A year and half later, the VLDLR gene is still being expressed in the mice. “It offers new hope for getting around host immune responses,” he says.

Adeno-associated virus (AAV) is also commonly used for gene therapy. Unlike adenovirus, AAV doesn’t trigger an immune response because it doesn’t cause any known human disease. But its tiny size relative to other viral vectors means it can only fit roughly half the amount of foreign DNA that other viruses can accommodate. On the plus side, however, AAV doesn’t pose the risks unique to other viruses. Because it doesn’t trigger the immune system, gene expression lasts longer.

Phil Laipis, a biochemist at the University of Florida (Gainesville), and colleagues are using AAV to treat phenylketonuria (PKU), an inherited metabolic disorder that causes severe neurological damage and lack of pigmentation due to a shutdown of melanin biosynthesis. Laipis and his team inserted the phenylalanine hydroxylase gene in an AAV vector and used it in mouse models of PKU. “We have dropped serum phenylalanine levels down to a normal level and reversed melanin biosynthesis,” says Laipis. “We’re hoping within several years to be at a stage where it is accepted by the FDA [for testing in humans].” Their findings were presented at the annual meeting of the American Society for Gene Therapy in Seattle in June 2001.

Similarly, other viruses are being vigorously studied for treating other conditions. Retroviruses have the advantage of being able to integrate into the host chromosome—a desirable feature if permanent gene expression is required. Herpes simplex virus, with its natural tendency to target the central nervous system, offers potential for treating nerve disorders.

In the not-so-distant future, it may become possible to create the best of all possible scenarios. “My guess,” says Laipis, “is that in 10 years, people will have made artificial viruses.”

Liposome delivery
Nonviral delivery of gene therapy using liposomes overcomes the intrinsic problems associated with viruses. Liposomes are synthetic, noninfectious, and won’t spark any adverse immune reaction, says John Park, a physician and researcher in the division of hematology and oncology at the University of California in San Francisco. When coalesced, their small internal diameter can restrict the amount of DNA that can be squeezed in. Consequently, cationic lipids have been developed that are capable of forming complexes with DNA through electrostatic interactions, overcoming earlier size limitations. The resultant liposome–DNA complex is called a lipoplex.

The downside to liposomes, however, can be their instability, he says. “Results have been exciting in vitro but unstable in vivo,” he says. “They break down immediately upon delivery and are not suitable for intravenous delivery. They’re very reactive and start binding to any membranes in their vicinity.”

Researchers are devising ways to get around this problem. Because some treatments only require that the patient’s cells be removed, they can be treated with drug-containing liposomes or gene-carrying lipoplexes and returned to the body. Other groups are attempting lipoplex gene delivery by aerosol spray, such as for treating cystic fibrosis.

Park and his colleagues have developed liposomes linked to monoclonal antibodies. The antibody acts as a homing device, targeting its drug-containing liposome to a tumor cell that is displaying a protein that the antibody specifically recognizes. Once antibody and receptor protein meet, the liposome delivers its toxic payload.

Targeted gene delivery
Experimental strategies geared at targeting tumor cells are making the most significant inroads, given that more than half of all gene therapy clinical trials are for cancer.

One common approach involves using viral or nonviral vectors to deliver antigens or cytokines to tumor cells, either in vivo or ex vivo. In this case, the delivered immune-stimulating molecules are then expressed on the tumor cells in the hopes of boosting the immune system, much like a red cape waving before a bull.

Gene replacement therapy using tumor suppressor genes, such as p53, is being explored as another method of stopping aberrant tumor cells from running amok. Many cells go on to form tumors because a tumor suppressor gene is mutated. Restoring a normal copy of the gene can arrest the cell cycle or even trigger cell death.

Suicide gene delivery is similarly showing early promise. A gene is used that encodes a protein that is either toxic to the cell itself or activates a prodrug into its cell-killing form. Tumor cells that produce the protein are killed off as a result.

Designing tumor-killing cells may be even more effective than drugs in some cases. “Unlike drugs, which have a half-life and go away, the virus goes on and makes multiple copies of itself and spreads from one tumor cell to another,” says E. Antonio Chiocca, a neurosurgeon and molecular pharmacologist at Massachusetts General Hospital and Harvard Medical School in Boston. “At least potentially, we may have one approach where the virus replicates and eventually goes away when there is no other tumor tissue left.”

“The hope is to create new classes of cancer agents combined with chemotherapy, radiation, and surgery,” he adds. “Cancer cells are smarter than anything we throw at it [cancer], so no one magic bullet will succeed . . . we have to throw multiple things at it.”

Although we may be years away from fixing the “broken” gene, cancer will likely be the first disease to show benefit from gene-based medicine, with infectious diseases not far behind.

“Over the past 20 years, we’ve learned a lot and made a lot of progress,” says Lee, dismissing critics who argue that gene therapy won’t live up to the hype. “Inevitably, this is going to be the future of medicine.”

Challenges facing gene therapy
Controlled gene expression. For some diseases, the correct amount of protein has to be made for the right amount of time. Genetic diseases like cystic fibrosis need a continual supply of the therapeutic protein throughout a patient’s lifetime to keep the disease in check. Other diseases don’t require such tight control. For different reasons, gene expression also sometimes works poorly or shuts off altogether shortly after it has been introduced. Scientists are not yet able to control gene activity after it’s been introduced into the body.

Getting genes to their proper targets. One big problem is getting the corrected gene into the right cells and functioning at the desired site. Often, cells other than the intended targets take up the gene as well.

Preventing destruction of the gene. Some enzymes will chew up DNA that is not protected, such as with the “naked” DNA approach or DNA released from a lipoplex. In other cases, the immune system will recognize a viral vector and destroy both it and the inserted gene.

Delivery methods. One challenge is how to most effectively administer the gene so that it ends up where you want it. In some cases, it’s possible to take the desired cells out of the body and insert the gene. Others inject the gene directly into specific sites, such as heart muscle.

Condition of the host. For some genetic diseases, irreparable damage occurs early in life. In cystic fibrosis, for example, the lungs are damaged during childhood. Treating some of these diseases will mean having to intervene before permanent damage has already occurred.

Host immune response. Another concern is how the person’s immune system will react to a foreign protein for the first time. The immune system may not react adversely the first time the vector or gene product (the protein) is encountered but can mount a severe response on subsequent exposures.

Nicole Johnston is a freelance writer based in Hamilton, ON. Send your comments or questions regarding this article to or the Editorial Office by fax at 202-776-8166 or by post at 1155 16th Street, NW; Washington, DC 20036.

Return to Top || Table of Contents

Copyright © 2001 American Chemical Society
CASChemPortChemCenterPubs Page