Session 7: Enhancement 5: Genetic Enhancement of Muscle
H. Lee Sweeney, Ph.D.,
Professor and Chairman of Physiology,
University of Pennsylvania.
CHAIRMAN KASS: All right. I know there are some counsel members with planes to catch, and I don't want us to waste any more of Dr. Sweeney's time. We're delighted to welcome Dr. Lee Sweeney, who is Professor and Chairman of the Department of Physiology at the University of Pennsylvania, who has done just a significant amount of outstanding work on muscle physiology and who is going to speak to us today about the genetic enhancement of skeletal muscle and its performance.
Thank you very much, Dr. Sweeney, for being with us.
Push your button there. There we go.
DR. SWEENEY: Thank you.
Yeah, I'm going to just try to give you in half an hour or so background about some of the work we're doing, and then I'm hoping just to allow you to drive a lot of the discussion because I'll set up some of the issues that I see, but as a scientist, I'm afraid I don't think some of the ethical ramifications through quite to the extent of the discussions I've already heard this morning.
My interest is in disease states of both skeletal and cardiac muscle. I'm going to restrict it to skeletal muscle because I think it gives you a good example of really how gray the boundary is between therapeutics and enhancement when one is starting to think about what can be done with genetic manipulation in adults.
Skeletal muscle is a big target in a person because it makes up the majority of the mass of their body, and it's an interesting tissue in that it has built into it cells called satellite cells which are not differentiated muscle cells, but which are actually cells that are called upon to divide and differentiate and regenerate the skeletal muscle.
So they're not really a resident stem cell population because they're not pleuripotential like a true stem cell, but they're an uncommitted set of cells that can with the proper stimuli be induced to become skeletal muscle. They can become other types of tissue as well, but a fairly limited repertoire.
Now, my interest in this began with years ago the beginning of the whole promise of gene therapy, and of course, the idea at the beginning of gene therapy was really to tackle the simplest of genetic diseases, those that involved single genes and usually the genes being missing or at least defective, which was the cause of a large number of diseases, most of them fairly rare diseases.
The issue, and still the issue, the recent gene therapy really hasn't as quickly progressed as we all would have hoped it would have ten years ago, is because the problems are really finding the right vector for a given tissue, that is, a delivery device to actually get the genetic material into the adult tissue, and then also figuring out how to get it there, the delivery.
So these are still the key stumbling blocks in gene therapy today, although great inroads have been made.
Now, in terms of muscle what is sort of emerging as the best sort of ways to deliver genes to muscle are sort of listed here, and this is as of today. Muscle is actually quite good at taking up so-called naked DNA or just plasmid DNA. This is DNA that is not encapsulated in anything.
When I say quite good, I mean it does it with some efficiency. It's perhaps one or two percent of the cells if you inject DNA into a given muscle would take something up.
So this is not an inefficiency that's useful in correcting a primary genetic disease of the muscle itself. However, this would be useful in terms of getting the muscle to produce a substance that would then be secreted into the blood.
And, in fact, this technique has been shown to be successful in an agricultural setting where a colleague of mine has taken DNA that codes for the growth hormone releasing hormone. This is actually a protein which stimulates the release of growth hormone, and in doing so, he can demonstrate that the pigs will now secrete much larger much larger levels of growth hormone.
Obviously one of the other problems with it is it's transient, but in that sort of setting perhaps it's useful to be transient just to sort of give them a growth boost through some period, and then it will go away, but obviously not something one could think of for permanent sort of genetic correction, and especially not a genetic correction of the muscle itself.
Now, viruses are still the preferred gene delivery vehicles in terms of efficiency. Getting genes into a given tissue are what viruses are engineered to do, and so they can be reengineered to deliver therapeutic genes instead of viral genes.
And probably the best vector for muscle is a virus that's known as AAV, which stands for adeno-associated virus. It's a virus that's unrelated to adenovirus, which has been used, as you're probably aware, in gene therapy trials with complications, and the most severe which took place at my own institution.
But this is a different virus that actually does not cause any known disease in humans. Often it can be found in 20 percent or so of the human population that have been infected by AAV with no consequences that can be demonstrated.
And in the last maybe two years a number of different sort of variants or serotypes of the virus have been isolated, and at least two of them are extremely efficient at targeting skeletal muscle. And so there are now gene therapy trials that have begun and many more that are about to be proposed and started that use this virus to try to get genes into skeletal muscle or in the liver. It turns out that this virus is also very good at targeting the liver, which would be very useful in diseases where one wants something secreted into the blood, like in the case of hemophilia.
There's an ongoing trial now with AAV targeting the liver to get the liver to secrete in this case Factor 9 for patients with Factor 9 deficiency.
Adult stem cells obviously are useful in sort of the idea of helping regenerate the tissue, although I must say that although there are a number of types of adult stem cells that can become muscle, again, efficiency becomes the problem, whether they're muscle or bone marrow derived stem cells. It's very difficult to get a large percentage of the muscle rebuilt using this approach in animal models where it's been attempted.
Nonetheless, we and others see the ability to use adult stem cells now as not in the stem cell sense of rebuilding the tissue, but perhaps in the sense of viewing them as a vector where one would take the adult stem cells in the laboratory, put new genes into the stem cells. That would then allow the tissue that they've incorporated in to secrete a substance that would then affect the surrounding tissue or if it's designed to go into the blood.
So you could sort of put adult stem cells then not only in the tissue regenerating sense, but in the sense of being a vector to carry genes into a tissue, to incorporate into the tissue, then to produce something else in that tissue.
And the advantage that they may have over viruses is one of the easiest systemic delivery. In many disease situations it may be that one could simply put the adult stem cells into the blood and they would home in on the tissue. They would home to the tissue that was actually being damaged by whatever the disease process is, and so it would be a very efficient way of targeting the tissue, which with viruses is more difficult at least at this point in time.
So I'm going to give you an example then of using the adeno-associated viral mediated gene transfer, but the example is actually one that we're now trying to do with adult stem cells, which is why I brought this up, because we think we can accomplish the same thing and deliver it much more simply using stem cells that are bone marrow derived from in this case we're working with both adult human cells, as well as adult rodent derived cells.
Adeno-associated virus, as I said, has the huge advantage for skeletal muscle in that it readily infects it. In fact, this may be the preferred tissue target of this virus at least for various forms of this virus.
It's limiting in that the size of its entire genome is only on the order of about 4.7 kilobases. So this is very small and really smaller than most genes in the human, and so one can only make synthetic genes that code for relatively small proteins. And so one has to be judicious in the choice of what one can attempt to do with this virus.
Delayed onset of expression, perhaps more so than some of the other viruses, but nevertheless, with the more robust infection that one gets with serotype one and five, expression commences within a week of injection of the virus into the tissue or systemic delivery of the virus into the tissue.
There's no viral gene expression. This is one of the big advantages of using this type of vector. You can make the synthetic vector without any viral genes, and because of that, there's no immune response against the virus itself.
Obviously there could still be immune response against whatever the product that you're causing it to make, but an example I'll give you, which is one of the advantages of it, what we're making is not something that's missing from the body, but something that we're just trying to get the body to make more of. And so there's nothing for the immune surveillance to pick up on, and so no possibility of immune response.
It integrates at a low frequency, which is both useful, but also a point of some concern in sort of the regulatory and side effect case. The integration, with any integrating virus there are two things to worry about. One is the possibility of oncogenesis being initiated by an integration event.
We have not seen this in our animal models, and other people have not seen this. So the possibility with this virus seems relatively low because there's some evidence that the integration events of AAV are somewhat site specific, and so not very likely to induce oncogenesis.
The other problem, of course, is whether or not one could get germ line transmission, and compared to something like lentivirus, the ability to do germ line transmission is not zero, but it's fairly low probability of germ line transmission, but again, this depends on the route of administration.
It would be more likely that you might get germ line transmission if you're doing a vascular delivery than of direct injection into the tissue, and the duration of expression, because of the integration, essentially is the life of the nucleus that you infected. And so as long as that cell in that nucleus exists, one will get expression. And so for the animal models that we look at, it's the life of the animal essentially.
Just to show you what efficient means, here's a cross-section through a muscle. So these are now -- if you think of the muscle fibers and muscle cells as very long cylinders, this is now slicing through them so that you basically just see their circumference and not their length.
And as you can see in this example, this is now using AAV-1. Essentially every muscle, well, every muscle in the field is now producing a protein that gives a color and an enzymatic reaction that's developed in the laboratory, a bacterial protein that can be used to give this colormetric readout, and you can see that every muscle fiber in the field is blue, and in fact, one can do vascular injection with this virus, and every fiber and every muscle virtually in the leg of the animal will be blue after vascular administration of a large enough dose.
So the efficiency is extremely high if one puts in enough of the virus.
So potential applications, which is what got me interested in using this in the first place, obviously the initial goal of all gene therapies of this sort was primarily genetic diseases, and for muscle that would mean Duchenne and Becker muscular dystrophy is the most common, but also others, such as the limb-girdle muscular dystrophies, myotonic muscular dystrophy, and whatnot, where one can point to a genetic defect in a single gene as the cause of the disease.
A more difficult problem, but actually in some ways, I mean, biologically a more difficult problem, but in fact, the problem that we focused on initially, which is a very real problem in this society where the society is living to be older and older, is the fact that as we get older our muscle function, our skeletal muscle function diminishes both in size of the muscle as well as the relative strength of the muscle and this is a big problem not only from an ambulatory standpoint, but also from a whole body metabolic standpoint.
If the mass of skeletal muscle drops below a critical threshold, then the whole body metabolism is no longer supported properly because the muscle actually functions not only to move the body, but as an important metabolic organ within the body.
Then the last sort of issue, which is actually a trial, trials have been ongoing in this area, is the use of gene transfer into muscle to get therapeutic proteins in the blood, such as Factor 9 deficiency.
So the initial hemophilia trials with Factor 9 were trying to actually get muscle to secrete Factor 9, but now they've shifted to liver because the liver is just a better organ for secretion into the blood than muscle is, although the muscle is capable of it.
So I want to first tell you about where we started some five years ago, which was looking at this problem that the NIH has coined sarcopenia. I think they coined a term to sort of make it sound more like a disease, probably for congressional purposes, but basically what they're really talking about is this progressive loss of muscle mass in force that essentially begins in the fourth decade life in humans and then progresses throughout.
It's slowed, but it's not prevented by exercise, and obviously as I've already mentioned has negative impacts on health and quality of life. It occurs in all mammals, which is useful because that means all of the laboratory animals one works with undergo the same process, and since they live for a much shorter period of time than humans, their life spans are much contracted.
The whole sort of progression occurs on a time scale that one can approach in the laboratory, and our hypothesis back in '96 or seven when we began this was that in large part we thought that what it was really due to was not inactivity. There had been a lot of discussion that as people got older they were just inactive and that's really the main thing that drove it, but we really thought there was a more fundamental cause, especially since there were studies showing that exercise could slow it down but not stop it.
And that was the fact that the repair mechanisms of skeletal muscle decline as you get older, and this causes the muscle to lose function because it's essentially not being repaired properly, and this goes back to what I said at the beginning, that it has within it a resident population of cells, these so-called satellite cells, that when the muscle is damaged -- and muscle is always damaged as you're using it -- are called upon to repair the muscle and rebuild it.
So this sort of rebuilding process involves some sort of damage signal coming out of the muscle which then activates the satellite cells to begin to proliferate, and they proliferate, then they make the commitment to be muscle, and then they either fuse with the existing muscle to repair it, or if the muscle has been severely damaged, they form new muscle.
So what is involved in this are a number of growth factors, some that drive the process, some that inhibit the process in sort of a yin and yang, but the one that we felt was really the most critical and the one that might be the candidate for what's going wrong in aging is a growth factor called IGF-1, which stands for insulin-like growth factor-1, which in normal muscle is involved in growth. It drives protein synthesis, and it decreases protein degradation, and importantly form the repair standpoint, it stimulates this population of satellite cells to both proliferate and differentiate.
And this is an important fact that it can do both because many of the growth factors will drive proliferation but block differentiation, and so increasing their levels could actually interfere with repair, which has been shown in some cases, but here you have one that has a little built in clock. It will drive proliferation for a while through one pathway, and then it will drive through this pathway, and then it will drive differentiation through another pathway, which it turns on with the delay.
So just the sort of thing you might want to try to drive more successful growth and repair. And the reason we thought it might be a problem in aging is because it's really part of the whole growth hormone IGF-1 axis, which as you know, the signals from the hypothalamus, the growth hormone releasing hormone, that are then taken to the anterior pituitary to stimulate it to produce growth hormone go down with aging.
The levels of growth hormone in the blood go down. The levels of IGF-1 produced by the liver; the liver produces all of the IGF-1 that circulates in the body. All of these levels go down with aging.
And what that means is that the IGF-1 levels in the various tissues of the body will also be diminished with aging.
Note: Nadia Rosenthal's presentation at the IABG indicates that it is the tissue specific form of IGF-1 which is mostly responsible for sattelite differentiaton and proliferation -KP
Now, tissues like muscle and other tissues of the body have two sources of IGF-1. They make it themselves under conditions of either injury or rapid growth, but also they have an IGF-1 input that comes from the liver that's ongoing throughout their life.
And so it's this component in particular that's being lost in the aging animal. And so we sought to essentially replace it by supplementing the amount of IGF-1 that the muscle itself could make.
So the strategy was quite simple that we took. We would use gene delivery into the muscle to give it a synthetic gene to have it produce more IGF-1 so it would not be particularly dependent on the liver for a source of IGF-1.
And then the question was: would that then in muscle promote growth and regenerative pathways and would that, in turn, allow the muscle to function throughout the life of the animal without the aging related loss?
So it's a very simple synthetic gene that we put together using a muscle specific promoter driving the rodent IGF-1, and then it's flanked by the viral ITRs and packaged into the AAV viral capsid, and then just to inject into the animals either a vascular delivery into the leg or a direct injection into specific muscles.
And we could show -- this is just using PCR to detect the existence now of our synthetic gene that four months, nine months, even two years after injection the synthetic gene is present and producing IGF-1 messenger RNA.
So then we asked the question with it: is this going to increase the rate of muscle regeneration and maintain mass and old age? And this is the paper that I included in your packet.
What we showed was that if we injected mice, essentially middle aged mice or late middle age in mice -- mice that live to be about 27 to 30 months in age at least in our colonies, and so we injected them about halfway through their life where they were all just beginning to start losing muscle mass, and then we asked, you know, what would their muscle mass in force and force for cross-section look like when they became old?
And so looking at them at 27 months, which was nearing the end of their life, they normally would have experienced in terms of mass about a 15 to 18 percent drop over that age period compared to a six month old mouse when they're sort of at their peak.
Whereas if we had injected them in middle age, they maintained the same mass or even a little greater than they had when they were younger. The same with the amount of force they were able to produce. Their muscles were able to produce normal force instead of showing the decline in force that they would normally see, and the force for cross-sectional area was maintained, as well.
But the speed of the muscle was maintained and the power output was maintained to an even greater extent because one of the other things that happens as the animals get old, as mammals get old, is they selectively lose their fast and most powerful fiber type.
So skeletal muscle is a heterogeneous tissue in terms of it has some of the fibers in it that are small. Some are big; some are slow; some are fast. We lose the very fastest ones as we get older and preferentially replace them with slower fibers. That's one reason why some of the first athletic things that go are your ability to compete in power events or speed events, because that's the first loss that you experience before you really lose muscle endurance or any of those sorts of properties.
And we were able to prevent that totally. The mice did not lose any of their fast fibers, and they had the same speed and power output when they were 27 month old muscles as they did as animals that were only six months of age.
So from that we were able to conclude that IGF-1 over-expression could prevent all of the hallmarks of age related atrophy and loss of skeletal muscle function in mammalian aging, at least based on the rodent model, and now we're hoping to pursue this in larger animal models.
The skeletal muscle regeneration rate is diminished in old animals, and we showed that in another paper other than the one I showed you, and that seems to be the primary problem, that even if you injure the old muscle, it cannot mount a normal regenerative response, but if you maintain IGF-1 expression, it can maintain a normal repair response, and this also, of course, is this hypothesis that we were looking at.
And also it suggests that one could go about this whole pathway as a therapeutic means of maintaining muscle mass either through the strategy that we used so far in these animals, which is to give them an IGF-1 gene supplement or, as I'll mention at the end, one can think about doing it in other ways that might actually be a little simpler to achieve that we're still evaluating.
This also suggested that maybe in dystrophic muscle where the rate of muscle degeneration or the rate of muscle damage is so high that it exceeds the rate of the muscle to repair it, we wanted to ask the question: if you actually increase the rate of muscle repair by up regulating IGF-1 production, could you slow down the damage, the cumulative damage in these dystrophic animals and maintain their muscle mass?
So that's what we looked at, and here is just now comparing a dystrophic muscle where now we've taken a transgenic approach, but we've also done this with virus, showing that -- this isn't projecting very well, at least not from where I am, but this, the IGF-1 producing tissue shows a lot less of degenerative signs than the dystrophic muscle where there's lots of fragmentation of fibers, lots of clumping, lots of regeneration.
There's infiltration from macrophages because there's an ongoing destruction and inflammatory response in the muscles. Even in the diaphragm, which is virtually destroyed by the time these animals reach about 20 months of age, there's been massive sort of hypertrophy and hyperplasia in the diaphragm. So the diaphragm has become much larger and stronger, and interestingly, the amount of connective tissue.