"We all have a common objective, which is to answer the question most often asked of us: Is there a treatment for this disorder [muscular dystrophy]?" said Ronald J. Schenkenberger, MDA's director of research and patient services administration, as the third international gene therapy meeting ended.

"Those who will provide that answer are in this room," Schenkenberger continued, addressing an audience that included several whose research led to the identification of the gene for DMD in 1986.

WHICH GENE TO TEST?

It's been more than a decade since the X-chromosome gene for the muscle protein known as dystrophin was identified. That protein, when absent or very seriously flawed, leads to DMD. When it's partially functional, it leads to Becker dystrophy, a less severe disease.

Scientists like Jeffrey Chamberlain at the University of Michigan have devoted most of the decade to studying precisely what parts of the gene serve what functions, including which parts can be safely removed to make a smaller gene, suitable for gene therapy.

Dystrophin Minigenes

At the May meeting, Chamberlain presented detailed accounts of the different parts of this gene (see "Shrinking the Dystrophin Gene"). He described several dystrophin "minigenes" he thought could help, although probably not cure, a person with DMD.

"The best potential is to take out the center portion of the rod domain," Chamberlain said. One minigene, known as the delta 17-48 (because it's missing sections 17 to 48), "works well enough," he said. "It would probably ameliorate the disease course."

But some researchers worry about the dystrophin gene, even if it can be trimmed to a manageable size. Boys with DMD make hardly any normal dystrophin protein, so some researchers aren't sure how their immune systems will react to dystrophin. Immune-system cells, the researchers reason fearfully, might reject the protein as if it were a foreign invader (see "The Body at War With a Cure").

Utrophin Minigenes

Some investigators, such as Kay Davies and Jon Tinsley of Oxford University in England, suggest a better choice for treating Duchenne or Becker might be the gene for utrophin. This protein, identified by Davies some years ago, may be capable of pinch-hitting for dystrophin if there were enough of it (see related article). A key difference between the proteins from the standpoint of gene therapy is that, since people with Duchenne or Becker already make normal utrophin, their bodies are unlikely to treat it as foreign. Utrophin is also a large gene, however, and would have to be trimmed to a condensed form.

Other Genes

Still other ideas involve trying some smaller genes as a first start in muscular dystrophy gene therapy. For example, the genes that, when flawed, cause Emery-Dreifuss MD and several forms of limb-girdle MD are now known. They're much smaller than the dystrophin and utrophin genes and might make an easier test case for gene therapy of muscle.

A smaller gene can be fitted into a smaller "vector" -- delivery vehicle -- and that has its advantages. Smaller vectors may get into muscles better, and some may be well tolerated by the immune system.

No Genes At All

Still other researchers, such as Leslie Leinwand of the University of Colorado at Boulder, Arthur Burghes of Ohio State University and George Dickson of the University of London (England), are trying a new idea that's already being tested in other diseases, such as ovarian cancer and lymphoma. Instead of inserting a gene, they're trying to change the way a cell "reads" the old gene, using compounds known as "antisense oligonucleotides." These can be put inside small vectors, just as genes can be.

WHICH VECTOR?

Vectors, the delivery vehicles used to transport genes, have long vexed scientists. Viruses have a natural inclination to invade human cells and deposit their genetic cargo in a cell's nucleus (that's how they cause viral infections), so they seem a natural vector for gene therapists to try. But, they also have their drawbacks. The body's immune system isn't generally friendly to them, and some of them are dangerous in their own right, having the potential to cause disease.

The past decade of test-tube and animal experiments in gene therapy has brought vector technology a long way from the early days of using natural, or so-called "wild-type," viruses, with all their perils. But the problems still aren't completely solved.

Adenoviruses

At the May meeting, most scientists reported experiments using highly altered forms of the adenovirus, a virus that, in nature, normally causes only mild human disease (such as a chest cold). With most of its own genes removed, the adenovirus should have minimal disease-causing potential and have room to carry at least a condensed dystrophin gene. But it remains an irritant to the body's immune system.

"Adenovirus is an extremely potent adjuvant [booster of the immune system]," noted Jeffrey Leiden of the University of Chicago, describing his experiments in mice exposed to the gene for erythropoietin tucked inside an adenovirus. Not only did the mice reject the new protein, he said, but they also turned against their own version of erythropoietin once their immune systems began to associate it with the viral invader. This phenomenon is called "breaking immunologic tolerance" and could mean a gene therapy disaster if not adequately planned for and averted.

Immunosuppressive drugs, especially the new drug FK506, look like they might be able to handle the adenovirus and any new genes it may carry, but many experts believe limited and careful human experiments will have to be done before anything is certain.

"Some components, such as the response to the adenoviral vectors, may need to be tried in human patients with no thought of efficacy (effectiveness)," said Bruce Smith of Auburn University in Auburn, Ala. "Responses in the dog and cat may not be the same as the response in humans."

Adeno-Associated Viruses

Mainly because of the immunologic problems, some researchers have turned away from using adenoviruses and have turned to "adeno-associated" viruses. These are much smaller than adenoviruses (see "Ranking the Vectors," below), but they have other advantages. They seem to slip past the immune system unscathed, and they cause no known human diseases, even in their natural state.

Adeno-associated viruses can't carry a large gene -- not even one of the dystrophin or utrophin minigenes now available. But the adeno-associated virus can hold a small gene, such as one of the genes for a sarcoglycan protein that, when mutated, causes a limb-girdle type of MD. A sarcoglycan-containing adeno-associated virus would make an ideal test vector, some researchers say.

"The adeno-associated virus works very well in muscle," said James Wilson of the University of Pennsylvania in Philadelphia. "It's stable for a long time without causing an inflammatory response."

RANKING THE VECTORS Each gene therapy vector has its pros and cons. The adenovirus is large enough to carry a dystrophin or utrophin minigene, but it's a potent stimulator of an immune response. The adeno-associated virus is much smaller, able to carry only a small gene, but it's less likely to provoke the immune system. A large gene can be delivered "naked" or encased in a liposome (fat-containing bubble), and these methods probably won't cause an immunologic problem. However, they aren't as effective as viruses.

No Viruses At All

Jon Wolff at the University of Wisconsin in Madison has been working on so-called "naked DNA" gene therapy strategies since about 1990.

Naked DNA doesn't home in on cells the way a virus does, and many people have questioned whether it could ever reach enough cells to make a difference. Wolff, however, hasn't given up. At this meeting, he reported injecting naked DNA (the gene for luciferase, a so-called "marker" protein because it lights up and is easily detected on routine assays) into rat arteries and getting much better entry into muscles than with intramuscular injection. He also got fairly good results using a dystrophin minigene.

"I'm starting to really believe in naked DNA," he said.

One way to use naked DNA might be as part of an indirect strategy for gene therapy. Instead of trying to put naked DNA directly into patients, it might be possible to put it into their cells in the laboratory and then inject those cells into patients.

Another way of avoiding viral vectors is the use of "liposomes," fat-containing bubbles that can deliver substances to cells. This kind of non-viral DNA delivery was presented by Ronald G. Worton of Ottawa General Hospital (Canada). This strategy also lends itself to indirect gene therapy approaches. Worton has gotten 30 percent of the cells in a lab dish to take up gene-carrying liposomes.

WHICH DELIVERY METHOD?

Correcting any disorder with gene therapy requires getting the gene into the target tissue, such as heart, brain, lung or blood cells. Few tissues pose as great a challenge as muscle. The body's musculature is a very large area, and its cells can't easily be removed for manipulation and reinserted as, for example, blood cells can be. Muscle also has plenty of physical and chemical barriers to keep out invaders.

Viruses can reach a lot of muscle with one injection, but no one's really sure how much they can reach in humans or how much muscle has to be penetrated for gene therapy to be effective.

Direct or Indirect Delivery?

There are two general approaches to gene therapy -- the direct method, in which genes, inside vectors or not, are inserted directly into a person's body.

In contrast, indirect gene therapy involves taking cells from a person, putting the new gene into those cells -- by using a virus or other technique -- and then reinserting the cells into the body.

There are a few ideas about which cells could be used. An old candidate cell is the so-called "myoblast," an immature muscle cell that's been used in several types of research experiments. Johnny Huard at the University of Pittsburgh described his moderately successful experiments using it as a gene therapy vector in mice, after altering it to carry a new gene.

(In the early 1990s, there were six clinical trials in which boys with DMD received donated immature muscle cells from their fathers or brothers. The trials failed to show any functional effect, and few of the cells survived, probably because of the body's immune response to the donated cells and characteristics of the cells themselves. Research on this organ transplant strategy, known as "myoblast transfer," continues in animals.)

The myoblast is relatively young and flexible, but some experts say it's still too mature, too far along in its development, to be an ideal gene therapy vector.

They say a muscle cell that's still earlier in its development -- called a muscle "stem" cell or "precursor" cell -- might make a better vector. Jennifer Morgan of Hammersmith Hospital in London said her group may have identified such a cell.

Another idea for cell-borne, or indirect, gene therapy came from Giulio Cossu of the University of Rome (Italy), who described how immature skin cells -- fibroblasts -- can be readily converted into immature muscle cells by adding a chemical called MyoD to them in the lab. Skin cells are far easier to extract than muscle cells, so the idea is intriguing.

Immune-system cells known as macrophages and monocytes could also be used to the gene therapist's advantage, said George Dickson. These cells normally gather around degenerating muscle in MD. Why not make them the bearers of a new gene, Dickson suggested. They're easy to extract, since they circulate in the blood, and they're easy to grow in the lab, where they can be made to carry a therapeutic gene.

SHRINKING THE DYSTROPHIN GENE The extremely large dystrophin gene has four sections: the actin-binding section (blue), the rodlike section (green), the cysteine-rich section (purple) and the C terminus (yellow/orange). The only part that seems safe to condense is the rod section. Some people with very mild muscular dystrophy have parts of this section missing.

Genes Under Pressure

Hansell Stedman of the University of Pennsylvania School of Medicine reported on his experiments in delivering genes under pressure to muscle groups in animals. Stedman and postdoctoral student James Greelish use a unique gene-delivery system, which requires shutting off parts of an animal's circulation while opening others.

Stedman is a surgeon who lost two brothers to DMD, and Greelish also lost a brother to this disease. Stedman said he believes his system can deliver an effective concentration of gene therapy vectors to the desired muscles. His method has worked well in rats, where it delivered vectors to the majority of hind limb fibers from a single injection into an artery. Without this approach, most of the vectors put into the bloodstream ended up in the animal's liver.

Realizing his technique sounded a bit drastic, Stedman quoted the ancient Greek physician Hippocrates, who said, "For extreme illnesses, extreme treatments are most fitting."

The high-pressure system could make direct delivery of genes without viral vectors more realistic, some believe.

To The Clinic?

A spirited discussion of the nature and timing of human trials in gene therapy followed the meeting. The consensus in the hours and weeks following was that limited aspects of gene transfer strategies will soon be tested, probably in a single muscle, and probably in more than one medical center.

Currently, MDA is reviewing proposals for this type of testing, with the goal of determining the safety of various vectors and possibly to look at persistence of a therapeutic or marker gene in muscle. No therapeutic effects on strength or function are expected in these early experiments.

At the meeting, Jeffrey Chamberlain seemed to speak for many when he said, "We're probably going to have to go into the clinic for more information, then back to the lab, then back to the clinic." George Karpati of Montreal Neurological Institute emphasized that the first tests should focus on safety, rather than efficacy.

But no one can say that taking gene therapy for MD beyond the clinic door is a trivial matter. It's a long-awaited first step on a path that now looks cleared of many of its early obstacles.

"We can look ahead to the future with a great deal of optimism," said MDA's Schenkenberger in late June as MDA's Research Department began holding telephone conferences to plan the first human tests. "It now appears, based on recent advances in molecular genetics, that gene therapy will have a much broader application as a potential therapy for neuromuscular disorders than was originally thought."