Scientific American, 6/98, Dora Y. Ho and Robert M. Sapolsky, page 96
Inserting genes into brain cells may one day offer doctors a way to slow, or even reverse, the damage from degenerative neurological disease
The prospect of acquiring any chronic illness is disturbing. But for most people, the threat of neurological impairment evokes a special dread. Afflictions such as Parkinson's disease or amyotrophic lateral sclerosis (Lou Gehrig's disease) progressively rob control of the body. Damage to the spinal cord can create equal misery in just an instant. And Alzheimer's disease attacks the very essence of one's personality as it destroys the mind.
Unfortunately, physicians and medical researchers have made only limited progress in the battle against such diseases, in large part because the brain and spine are so vulnerable. Unlike many types of cells, neurons (nerve cells) in the central nervous system of adults are typically unable to divide. That fact of life creates the central tragedy of neurological illness or injury: under normal circumstances, neurons that are lost are gone for good, and injured nerve tissue of the brain and spinal cord cannot be expected to repair itself.
But scientific advances might yet change that grim situation. Some of the most ambitious research in neurology aims to replace lost cells in damaged tissue by transplanting neurons or by delivering growth factors - chemicals that can stimulate surviving neurons to ex tend their reach or that can awaken the cells' dormant ability to regenerate. Such therapies would be immensely beneficial, but it will probably take many years before they become routine. Preventing neuron loss in the first place is a more modest goal-one that is perhaps not so distant.
In the past few years, researchers have learned a great deal about how neurons die after a sudden medical insult such as a stroke, seizure or head injury, as well as during progressive diseases such as Parkinson's or Alzheimer's. Some at tempts to take advantage of these recent discoveries suggest that administering certain drugs may protect threatened neurons or even that lowering the temperature of the brain can avert the death of fragile cells during a neurological crisis. What is more, new knowledge about how neurons succumb to various diseases has raised the exciting possibility of protecting these cells by modifying their genes.
Reprogramming for Survival
Genes instruct cells to make specific proteins, such as the enzymes that catalyze various chemical reactions. Nerve cells, for example, produce enzymes that synthesize neurotransmitters - substances that carry chemical signals across the tiny gaps (synaptic spaces) between one neuron and another. Gene therapy targeted to failing neurons could potentially provide them with a gene specifying a protein that is able to shield these cells from whatever threat may loom.
To create such remedies, researchers must first decide what kinds of proteins would be most helpful. In some cases, the goal would be to augment the production of a particular brain protein when the naturally occurring version is dysfunctional or made in inadequate amounts. Or, in theory, one might want to add a novel protein, found in a different type of tissue-or even in a different organism entirely.
Another strategy, called the antisense approach, also constitutes a form of gene therapy [see "The New Genetic Medicines," by Jack S. Cohen and Michael E. Hogan; SCIENTIFIC AMERI CAN, December 1994]. Antisense tactics aim to limit the manufacture of proteins that are doing damage. Some types of amyotrophic lateral sclerosis and certain other neurological diseases result from the destructively intense activity of a normal protein or from the action of an abnormal protein that works in an injurious manner. Antisense therapies might also help when neurons synthesize proteins that (for reasons that are still in explicable) exacerbate a neurological crisis. To that end, many researchers are now trying to find ways to block the production of so-called death proteins, which induce endangered neurons to commit cellular suicide.
Once the basic understanding of a particular neurological disease is in place, it does not take great imagination to come up with a list of genes that might save neurons from destruction. The challenge is in figuring out how to deliver those genes. In principle, one can insert a gene into brain tissue by directly injecting appropriately coded segments of pure DNA. Unfortunately, this brute-force method is rarely successful, because neurons are not particularly efficient at picking up such "naked" DNA. A better technique is to encase the gene in a fatty bubble called a liposome. Because of the chemical nature of these tiny containers, liposomes easily transport DNA into target neurons by fusing with the cell membrane and re leasing their contents into the interior. The cell, for reasons not entirely under stood, will then incorporate some of this material into its nucleus, where its own DNA resides, and will use the gene as a blueprint for making the therapeutic protein [see "Nonviral Strategies for Gene Therapy," by Philip L. Felgner, on page 86).
An even better tool for putting genes into cells is a virus. In the course of a typical infection, viruses insert their genetic material into cells of the victim, where this added genetic code directs the synthesis of various molecules needed to make new viral particles. Although natural viruses can be immensely destructive, scientists can tame and convert some of them into microscopic Trojan horses, which then can carry a therapeutic gene and quietly deposit it inside a cell without causing damage. For gene therapy in the central nervous system, investigators are focusing much of their effort on just a few viral types, including adenoviruses and herpesviruses.
Combating Parkinson's
Experiments with these viral vectors, as such delivery vehicles are called, have provided the first hints that gene therapy in the nervous system can work. One promising area of research is directed against Parkinson's disease. This devastating disorder arises because a part of the brain, known as the substantia nigra, degenerates over time. This region helps to regulate motor control, and its destruction makes it hard for a person to initiate movements or exe cute complex coordinated motion. The loss also brings on the classic parkinsonian tremor [see "Understanding Parkinson's Disease," by Moussa B. H. Youdim and Peter Riederer; SCIENTIFIC AMERICAN, January].
Parkinson's disease ensues after the death of nigral neurons that secrete the neurotransmitter dopamine. For complex reasons, these neurons also generate oxygen radicals, rogue chemical groups that cause damaging reactions within the cell. Hence, there is a fair amount of ongoing destruction in the substantia nigra as a normal part of aging. (This process contributes to the mild tremor typical of senescence.) Sometimes Parkinson's disease appears to strike people who are predisposed to having an excess of oxygen radicals in their brain tissue or who have been exposed to environ mental toxins that cause these oxygen radicals to form. Other cases seem to involve people who have normal amounts of these chemicals but who have impaired antioxidant defenses.
Whatever the underlying cause, it is clear that the symptoms of Parkinson's disease result primarily from the absence of dopamine after too many neurons in the substantia nigra die. Thus, a straight forward way to correct this deficit, at least temporarily, would be to boost the amount of dopamine where it is in short supply. Dopamine is not itself a protein, but the enzymes that synthesize this neurotransmitter are. So increasing the manufacture of one enzyme critical to that process (tyrosine hydroxylase) should enhance the synthesis of this much needed brain chemical for as long as the do pamine-producing cells of the substantia nigra survive.
Although administering a chemical precursor to dopamine - a substance called L-dopa - also works to augment levels of this neurotransmitter, the drug reacts throughout the brain, causing serious side effects. The lure of gene therapy in this context is that corrective changes would take effect just within the substantia nigra.
Several scientists have been working hard to exploit this possibility. In a pair of recent collaborative studies, five re search teams reported success using herpesviruses as gene vectors to correct symptoms in rats that were surgically treated in a way that caused them to exhibit some of the manifestations of Parkinson's disease. Application of gene therapy increased the production of the corrective enzyme, raised the level of dopamine near the cells that had been deprived of this neurotransmitter and partially eliminated the movement disorders in these animals.
Dale E. Bredesen and his colleagues at the Burnham Institute in La Jolla, Calif., recently explored an even more sophisticated scheme. Investigators had shown previously that transplanting neurons from the substantia nigra of fetal rats corrected some of the parkinsonian defects that were surgically induced in adult rats. This strategy worked be cause the robust young neurons were able to grow and produce dopamine for the nearby cells in need. A problem emerged, however. For some reason, the grafted neurons tended to activate an internal suicide program (a process termed apoptosis) and died after a while. So Bredesen and his co-workers carried out gene therapy on the fetal neurons before transplanting them; the researchers hoped to coax these cells to produce large quantities of a protein called bcl-2, which suppresses cell suicide.
The result was dramatic: four weeks later the rats that had received standard grafts were only marginally better, whereas the creatures that obtained the added gene in their grafts were substantially improved. Treatment for Parkinson's disease would require a longer period of effectiveness still-one would want the grafts to survive for years. Physicians have already carried out human fetal cell transplants to help patients with severe Parkinson's disease, but these at tempts have met with mixed results. Perhaps one or two clever gene modifications to the human fetal cells before they are transplanted would make that procedure work much better.
Battling Stroke
The success of current research with animals indeed sparks hope that new treatments will eventually emerge for Parkinson's and other progressive degenerative diseases of the brain. Gene therapy also offers the prospect of stemming tissue damage during such acute The result was dramatic: four weeks later the rats that had received standard grafts were only marginally better, whereas the creatures that obtained the added gene in their grafts were substantially improved. Treatment for Parkinson's disease would require a longer period of effectiveness still-one would want the grafts to survive for years. Physicians have already carried out human fetal cell transplants to help patients with severe Parkinson's disease, but these at tempts have met with mixed results. Perhaps one or two clever gene modifications to the human fetal cells before they are transplanted would make that procedure work much better. Battling Stroke he success of current research with animals indeed sparks hope that new treatments will eventually emerge for Parkinson's and other progressive degenerative diseases of the brain. Gene therapy also offers the prospect of stemming tissue damage during such acute neurological crises as the overstimulation of a seizure or the loss of oxygen and nutrients that occurs during a stroke.
Under these conditions, the most vulnerable cells in the brain are the many neurons that respond to an extremely powerful neurotransmitter called glutamate. Glutamate normally induces recipient neurons to take up calcium, which causes long-lasting changes in the excitability of synapses stimulated by this neurotransmitter. This process may, in fact, be the cellular basis of memory.
But during seizure or stroke, neurons are unable to mop up glutamate from synapses or clear the tidal wave of calcium that floods into many brain cells. Instead of fostering mild changes in the synapses, the glutamate and calcium do serious damage: the cellular architecture of the affected neurons crumbles, and newly generated oxygen radicals create further havoc. This destruction then kills cells directly or signals the initiation of internal suicide programs that will cause the swift demise of the flagging neurons.
Our group has examined the possibility that gene therapy could interrupt this calamitous sequence of events. For our first experiments, we extracted some brain cells from a rat and cultured them in a petri dish. We then subjected these neurons to a modified herpesvirus engineered to carry a gene for a protein that transports energy-rich glucose molecules across the cell membrane. In a patient suffering a neurological crisis, a similar type of therapy might increase the influx of glucose just when the beleaguered neurons would benefit most from extra energy (which is needed, among other tasks, to pump the excess calcium out of these cells).
Early experiments showed that our treatment enhanced the uptake of glucose and helped to maintain proper metabolism in neurons subjected to the test-tube equivalent of seizure or stroke. We later found that we could lessen the damage from stroke in rats by injecting the viral vector into the vulnerable region of the brain before an injurious event occurred. It is obviously not possible for a person to forecast when a seizure or a stroke will happen. But, as Matthew S. Lawrence and Rajesh Dash discovered when they worked in our laboratory, there is a window of a few hours after a seizure when the gene treatment to these rats still helps to protect neurons from additional damage which suggests that humans, too, might one day benefit from a similar kind of therapy.
Another form of potential gene therapy for stroke and trauma targets the activation of suicide pro grams. Howard J. Federoff and his colleagues at the University of Rochester, and Lawrence, working with our group, independently constructed herpes virus vectors that included the suicide-suppressing bcl-2 gene. Application of this vector tends to shield the brain cells of rats from dam age when crisis conditions occur, even if the treatment begins only after the insult.
Investigators also report advances that might one day prevent the so-called lipid storage diseases-genetic disorders that cause defects in certain enzymes and lead to a fatal accumulation of fatty molecules in the brain. To explore such therapies, researchers have produced mice that carry a mutation in the gene encoding the enzyme beta-glucoronidase. (People suffering from a rare malady called Sly syndrome have the same gene mutation.) John H. Wolfe and his colleagues at the University of Pennsylvania transplanted into an afflicted mouse fetal neurons engineered to produce beta-glucoronidase. These researchers found that the implants were able to dispose of damaging lipids throughout the animal's brain.
Despite the many encouraging first steps in applying gene therapy to the nervous system, sundry hurdles remain. For example, significant problems persist in engineering viral vectors. Cripple the virus too much, and it becomes difficult to maintain sufficient potency to infect cells. Strip away too little, and the virus will damage the host neurons. Be cause the viral vectors now available each suffer from one or the other of these shortfalls, a great deal of refinement will be needed before scientists can safely begin testing gene therapies in people with neurological disease.
Another difficulty emerges simply be cause the brain-a vital but delicate organ - is encased in a relatively impregnable skull. Thus, injecting a therapeutic drug directly into the affected tissue is rather difficult. Most researchers con ducting animal studies resort to neurosurgery: drilling a hole in the skull and injecting the vector directly into the endangered part of the brain. But, clearly, human patients would require some thing less invasive for routine treatment. Although one could give a vector intravenously (if it could be designed to enter only nerve tissue), the virus would be unlikely to get past the blood-brain barrier, a specialized network of capillaries that lets only small molecules pass into brain tissue. So, without further special measures, virtually all of the viral vector would wastefully end up in places other than the brain.
Even if these obstacles could be over come, some final stumbling blocks would still stand in the way. After a viral vector reaches a patch of neurons, it does not go far (Viruses that are able to replicate can spread readily in brain tissue, but these agents cannot be used for gene therapy, because they invariably provoke a damaging immune response. ) A safe viral vector traverses a limited area, where it infects only a small percentage of neurons. Hence, these viruses are not particularly effective in reaching diseased tissue. Furthermore, for most of the vectors tried so far, activity persists for a few weeks at most-too brief a period to combat slow but relentless degenerative illnesses. So researchers will need to find ways to improve the spread, efficiency and duration of these engineered infections.
Future Shocks?
Despite the many challenges, studies of gene therapy for diseases of the nervous system-like most ambitious efforts that have shown some initial successes - have generated an aura of optimism. Perhaps with adequate effort, gene therapy for the brain will eventually be commonplace.
A glimpse of the possibilities ahead comes from the work of Anders Bjorklund and his colleagues at the University of Lund in Sweden. These researchers, who pioneered methods to transplant fetal neurons, have engineered grafts to produce large quantities of a nerve growth factor. They implanted some of these engineered cells into mature rats, targeting a region of the brain that is critical for learning and memory-an area that, not surprisingly, slowly degenerates during normal aging. Remarkably, this maneuver re versed cognitive decline in aged rats.
That success suggests that gene therapy might serve not just to blunt the edges of disease but to improve memory, sensation and coordination in older people. At present, scientists have enormous strides to make before they can hope to aid geriatric patients in this way. But ultimately, gene therapists may be able to offer powerful medicines for rejuvenating aging brains.
Such treatments might also be able to make younger people's minds work "better than well," to borrow a now popular phrase describing the effects of Prozac. Few areas of medical research pry at such a Pandora's box as does work on improving normal brain function. But this prospect-and its possible abuses-will be difficult to sidestep if scientists are to continue to pursue genetic treatments directed at specific neurological diseases. So further re search on applying gene therapy to the nervous system, like some other swiftly moving currents in the flow of biomedical inquiry, will surely force vexing ethical questions to float to the surface.