The future of gene therapy

S. THANTHONI

Interview with Dr. Nick Wood, Professor of Clinical Neurology and Neurogenetics.

Dr. Nicholas William Wood, Professor of Clinical Neurology and Neurogenetics, National Hospital, Queens Square, London, is a world-renowned expert in neurosciences. He has received a number of prestigious awards, including the Charles Symmonds Memorial Award in 1993, the Duroshiler Award of the British Medical Association in 1996, and the Linacre Medal in 1999 from the Royal College of Physicians, London. Co-editor of two books on neurology, Dr. Wood is on the review board of several international journals. Dr. Wood has a doctorate in neurogenetics and training in clinical neurology, which form a rare combination.

Dr. Wood is the European coordinator of a study on the genetic basis of Parkinson's disease. He recently mapped an autosomal recessive locus for the disease. He has done genome-wide search and homozygosity mapping in a consanguineous Italian family. Over the past three years, the group has mapped successfully four other inherited movement disorders, including autosomal dominant ataxia, dystonia gene, geniospasm and paroxysmal kinesiogenic choreathetosis. Cloning projects are under way for each of these. He collaborated with three groups investigating the genetic basis of epilepsy. He is also a member of an international interdisciplinary group to study cellular physiology.

When he was in Chennai to deliver the 17th Gopalakrishna Oration conducted by the K. Gopalakrishna Department of Neurology at the Voluntary Health Services, Dr. Wood spoke to Asha Krishnakumar on the status of the Human Genome Project, the genetic basis of neurology, the future of gene therapy and state-of-the-art research in neurogenetics. Excerpts from the interview:

What is the Human Genome Project? How will it help our understanding of biology?

The Human Genome Project was an international consortium that set out to sequence the whole genome. Everyone's genome varies, but only very slightly. You and I, despite coming from different parts of the world, are mostly alike. Most of our base-pass or code is identical. We probably vary in about 1 per cent of our genome. So, if you sequence just one person, you get 99 per cent of the information about the human genome. A small part of the basic human genome is yet to be decoded. But once done, you have something like the periodic table for chemistry. It is like how it was 100 years ago in the case of chemistry. According to me, biology is 100 years behind the physical sciences in terms of its basic understanding. The genome project is one of the major developments that would help bring rapid progress.

According to reports on the genome project, only mono-gene disorders have been sorted out. How will this help in our understanding of the causes and treatment of disorders arising from a single gene?

You are right. Only simple, mono-gene disorders have been sorted out. But even that is a major development. Such genes act in the family either you have it or you don't. If you don't, you will not get the disease and if you have the gene, you will get it. Huntington Disease, for instance, comes under this category. It is a dominant disorder coming through the generations. The genome project helps find these diseases very quickly.

From the genome project results, how does one go about finding out the existence of the gene that causes a mono-gene disorder?

You take a family with, say, 20 affected individuals and have markers scattered throughout the genome. You then apply the markers to the family and look for the marker that segregates the disease. Everyone in the family who could get the disease will have one type of marker and those who would not get it, another type of marker. Suppose you apply the markers and those who would not get the disease have the 200th marker and those who could get the disease have the 201st marker, then you know that wherever the 201st marker is must be close to the diseased gene. You then go to the database provided by the genome project, in the computer, and find out what genes are from the region where the 201st marker is. And let us say that you get a list of 20 genes. You then find out which ones of those are expressed in the central nervous system. Say, 10 of them are, and suppose you already know that five of them cause some other disease, then you are left with only five genes, which you then sequence to find out which one of those causes the disease. This is the process by which you identify very rapidly a single gene responsible for a particular disease.

How was this done prior to the genome project?

It was done by a method called `linkage'. Once you know which chromosome the gene causing the disease was on and the marker associated with it, then you had to sequence the DNA [deoxyribonucleic acid] yourself. To do that you had to clone it all and it was a huge task. Now it has been done for you. You just have to find the mutation.

So, do we now have the facility to sort out all single-gene disorders?

In the next five to 10 years almost all disorders caused by a single gene will be sorted out. This is no mean achievement.

What are the dominant mono-gene disorders that are to be sorted out by the genome project?

Huntington, some forms of Alzheimer's, epilepsy and Parkinson's disease and a lot of muscle diseases including muscular dystrophy. There is a long list of rare diseases.

We make up about 30,000 genes; over half of them are expected to be in the central nervous system. Random mutations go on across those genes. So, over half of the diseases that may occur are going to be neurological. It is thus not surprising that the long list of genetic diseases would express itself on the nervous system.

What are poly-gene or complex disorders? Is there a possibility that they will be sorted out in the near future?

For a single-gene disorder, everyone within a family who has a particular genetic disease is likely to have the same genetic abnormality as there is a very strong genetic factor that is causing the disease. This is easy to find out as it stands out.

But take, for example, epilepsy, which is mostly not transmitted through generations. You may just have one or two people in a family with epilepsy and that does not give you enough information. The disease may be a result of a complex interplay of genetic and environmental factors. Thus in the case of complex disorders it is just not the relative abnormality in the genes that causes the disease. By itself the gene does not tell you anything. It is a combination of genes and environmental factors that causes complex disorders such as epilepsy.

To find out the cause of such diseases it is not enough to study one person or a few families, you need to study hundreds of people as you cannot separate them to start with as you do not know what comparisons to make. It is thus best to start the study with a large population and do the mapping. Then go back and say that this type of epilepsy is mostly because of these factors and so on. Even in this case we are only guessing. But, surely, a homogenous approach where one lumps them to start with and splits them later is good. In the past, what was done was to split patients into disease categories and then say you have got this or that type of epilepsy. There is some basis in that but I think one should not get too fixed on that.

Are there genetic differences across ethnic groups? And would that make the identification of genetic disorders easier?

Yes, undoubtedly there are ethnic variations in the genetic make-up. Some common diseases vary in a particular frequency throughout the world. For example, in Singapore, brain haemorrhage, a common cause for stroke, is more common than in the West. The reason for that is not very clear as yet. It may be because of differences in diet, environment and so on. But, as is being increasingly found out, it is to a large extent genetically driven. You will have to take into account what the frequencies of the disease are in different populations. Alzheimer's is a big problem in the United Kingdom. But in some other parts of the world, where the life expectancy at birth is low, people die before they can even get it. Thus, there are diseases such as Alzheimer's, Parkinson's and stroke, as also cancer, that are becoming major problems because people are living longer now.

How can genetic studies give clues to environmental factors that cause diseases?

For example, in the case of Parkinson's disease, an idea that has arisen over the years (though not proved as yet) is that probably some environmental poison or toxin causes it. In such a case if a person is exposed to such an environment it may be useful to study how his genetic basis would react to the exposure in order to find out whether he is going to get the disease or not. There is a biological possibility to it. For instance, if you are a slow metaboliser at a number of different genes, then, it would be only modestly harmful and you know you can cope with it.

So, the idea is that if we study those genetic factors and find them to react to toxins that cause such diseases, then we can probably identify the environmental factors. But this is painting a rosy picture of the future. There are obviously more complex and difficult problems. I am sure in another five to 10 years' time, we shall still be left with a lot of questions about the common diseases. But we shall surely be at least a bit further along in understanding them.

How does one determine the inheritability of diseases?

We can measure inheritability by twin studies. For example, the chance of identical twins being epileptic is quite high. Even a disease like TB [tuberculosis], which is infectious, has a genetic basis. It is quite clear from studies that not everyone who was exposed to TB in a family got the disease. The chance of identical twins getting TB is more than non-identical twins exposed to the same environment. With the emergence of TB in the West due to AIDS [Acquired Immune Deficiency Syndrome] or lower resistance levels, people have turned attention to why certain people are more susceptible to TB than some others. They have found out that it is genetically driven.

Probably, the only problem that is not genetically driven is trauma - being knocked over by a bus or a train. That is plain bad luck.

How can gene defects be corrected? Is gene therapy a possibility in the near future?

Gene therapy is definitely something for the future. I am sure something useful will come out of gene therapy. But there are several problems. Suppose you detect a genetic defect and have to correct it, there are ways of doing it. One way is getting a new gene into the cell. Most cells are receptive of new genetic material when they are dividing. So it is relatively easy to introduce new genes when the cells are dividing. Since liver, lungs and blood cells keep dividing all the time, it is easy to introduce new cells to correct diseases connected with these organs. But this is not possible with the nervous system as its cells keep dying, becoming fewer over the years as in the case of muscles. So a vector system needs to be developed that can transfer genetic material into non-dividing tissues. That is a technological issue and is being worked out. There are some good vectors now. That is one problem.

The other problem is getting the genes in the appropriate place. It need not be in the same place but should be capable of regulating the defect. Often you have to put into the gene something that will listen to the other messages around and gets switched on and off at the appropriate times. That is not trivial, particularly because you do not want the gene and protein to be expressed all the time; only when it is meant to. That is not straightforward. The vector might be very good. But you may want to send in 25 copies. But that would be very difficult to control. So, you may want to send in one copy to the right cell at the right time. That would lead to significant technological problems.

The therapy also depends on the nature of the defect. For some defects such as tetanus disorder, which is autosomal recessive that is, both parents are well but carry the mutation and the two mutations come together in the offspring giving rise to the disease if you can put in the right gene you can correct it. And since the parents are well with 50 per cent activity, all you have to do is get 50 per cent of the level. You will probably be fine even with 5 or 10 per cent levels.

But many dominant diseases such as Huntington are not owing to loss of function; it is the gain of function that leads to the problem. So what you have got is a new mutant gene that has a toxic new function poisoning the nerve cells. Putting in a new copy of the gene would not help in this case. You have got to put in genes that stop the mutant copy. You have to go to another level of sophistication there. This will come in the future, but is not yet around the corner.

By understanding the genetic processes you will think of ways of either modifying the genes or persuading them to turn themselves on or off, as the case may be. In the case of, say, a nasty disease such as muscular dystrophy, people are now looking for small molecules that can cross the barrier into the muscle and switch on new trophyn to do the same job. Rather than trying to correct what is wrong, people are working at persuading the body to mimic in correcting the problem in some other way. I think this kind of approach has a greater chance of correcting gene defects than genetic therapy.

Among the therapies that are to be found, only some would be based on genetics; more would be molecular based.

Is there any relation between drug response and the genetic base?

The main issue is drug response. Not everyone responds to drugs the same way. And not all drugs work on everybody. Now it is done in a hit-and-miss manner. In future, the responsiveness to a drug will be genetically determined.

Our group has found the first genetically determined drug responsiveness in epilepsy. We found that we can label a gene called `multiple drug resistance gene' (which has been thought of in cancer genes for a while). If you have a certain form of this gene you have a high expression of the protein that stops foreign substances from getting into the cells. It is good if you do not want foreign substances to get in. But it is bad if you want drugs to get in. This leads to low level of drug inside the cell. We know the genetic make-up of that and we have shown that if you have a high expression of the protein there is a high epileptic drug resistance. Epilepsy, for instance, is difficult to treat in such cases. With such studies we shall be able to design drug trials better in the future. Instead of conducting trials in, say, 10,000 people, it can be done with just 500 people who do not get side-effects and respond better to drugs. This will bring down trial costs tremendously.

That is the real potence of drug responsiveness and the genetic applications to it. The drug companies are very interested in this as they can save a lot of money.

There are some ethical issues involved in this, particularly when people are genetically resistant to drugs and are kept out of drug trials. How does one tackle such issues?

That is a real problem as results of drug trials will not be applicable to this group of people. There is a significant ethical dimension to some of this work. It is complex.

What are the state-of-the-art research efforts in pharmacology for neurological disorders?

As I mentioned earlier, pharmacogenetics is one the genetic basis for drug responsiveness. And the other is geneticular molecular technology. In this, they take one element of protein known to be good at binding neurons or getting into neural cells and so on. You can design a friendlier compound using molecular technology. That is going to get a lot more sophisticated.

We know the periodic table (from the genome project) but know very little about the protein structures. There is going to be layers and layers of sophistication in this.

What is going to happen in the post-genomic sequencing era?

First is population genetics and what is called `linkage' or equilibrium or haphazard mapping. The NIH [National Institutes of Health] in the United States has put in several millions of dollars in what is called the `Hap Map Project'. The other thing is to understand what regulates genes because we only produce 30,000 genes, far fewer than what we expected. The explanation for this is that we probably had one gene to make several proteins by splicing itself in different ways. It is probably regulated at different times and so forth. Understanding all those subtleties of the genetic structure can be very complex. That is called genomics.

On top of that is to understand what proteins the genes make. We know the very basic structure from the Human Genome Project. But how they are folded, how are proteins made, how they get modified and so on are issues that have to be worked out. So it is going to be layers of structures of work genetic, genomic, proteomic and so on.