Dr. Paul, tell us about your early research in the metabolism of the branched-chain amino acids.
I became interested in MSUD from my interest in the metabolism of the branched-chain amino acids (BCAAs), leucine, isoleucine and valine. Since 1974, I have been involved with research on BCAA metabolism, first at the University of Pittsburgh School of Medicine and more recently at Biomed Research & Technologies, Inc., a biotech company that I founded in 1996.
In our earlier studies, we investigated the effect of nutritional and hormonal changes, such as starvation and diabetes, on the metabolism of the BCAAs. These studies were carried out largely in the liver and skeletal muscle of laboratory animals. We also studied the effect of carnitine and ketone bodies on BCAA metabolism.
As more information about the branched-chain keto acid dehydrogenase (BCKDH) - the enzyme responsible for the catabolism of the BCAAs - became available, our studies focused on the regulation of this critical enzyme. For example, we, along with others, were able to show the regulation of the BCKDH by phosphorylation and dephosphorylation cycle. We also studied factors and physiological states that play a role in the interconversion of the enzyme from an active to inactive form in various tissues.
Subsequently, we studied the gene expression of the BCKDH in cultured liver cells, as well as in animal tissues. We found that expression of the BCKDH subunit genes is affected by hormones and drugs. We also studied the expression of the BCKDH kinase gene by hormones and drugs. All the above studies provided very useful information about the BCKDH, particularly its regulation and gene expression.1
How did these studies on the metabolism of the BCAAs and the BCKDH influence your interest in MSUD?
With most of the basic research on the BCKDH completed, we then focused our attention towards clinical application. Obviously, the most important clinical problem in this area was MSUD. As you know, there are several clinical variants of the disease, the classic form being the most severe form of MSUD. We were aware of the difficulties in managing this disease. In spite of dietary intervention, the disease produces several complications, the most notable being mental retardation. Moreover, even if successful, dietary intervention does not cure the disease.
How did this lead to gene therapy as a cure for MSUD?
Because of the limitation of the dietary approach in managing MSUD, we began to think of other possible treatment options. By this time, genes encoding the BCKDH subunits were cloned and vectors to deliver genes were becoming available. Because of these advances, we began to think of the possible use of gene therapy to cure this disease. As a first step, and to establish feasibility of such an approach, we tested gene therapy in cultured cells derived from an MSUD patient. We obtained skin fibroblasts carrying a mutation in the E2 subunit of the BCKDH. These cells expressed less than 2% of the BCKDH activity as compared to normal cells. Using a retroviral vector, we were able to deliver the normal E2 gene into mutant cells. Following E2 gene delivery, the level of the BCKDH was restored to 93% of the level observed in normal fibroblasts. We also showed that the newly synthesized E2 protein was localized to mitochondria, the site of BCKDH.
The above results established the correction of the BCKDH deficiency at the cellular level by the transfer of the normal gene to cells derived from an MSUD patient. Our results with cultured cells were very encouraging and confirmed that further studies can now be undertaken to cure MSUD by gene therapy.
After you were successful in correcting the enzyme deficiency which causes MSUD at the cellular level, what was the next step?
To advance the research from the cellular level to the level of the whole animal, we now needed an animal model of this disease. Unfortunately, no animal model of MSUD was available.
At this time, I sought the help and advice of Gregg E. Homanics, Ph.D., a professor at the University of Pittsburgh School of Medicine. Dr. Homanics is very experienced in creating animal (mouse) models of human diseases by gene targeting in embryonic stem cells (ES cells). Dr. Homanics not only agreed to help me create an animal model, but also showed a strong interest and enthusiasm for the MSUD project. I was, indeed, fortunate to have him team up with me.
With Dr. Honanics help, we embarked on creating a mouse model of MSUD. This animal model was not only essential to show feasibility of gene therapy but also to establish the long-term safety of an unproven treatment. Because our previous gene therapy experience was with cells carrying a mutation in the E2 subunit, we decided to make an MSUD mouse model with a mutation in the same E2 subunit. Also, on a world-wide basis, the highest incidence of MSUD was reported to be due to a mutation in the E2 subunit. However, it is the E1 mutation that is found in the Mennonite population, as well as in other Caucasians, and in Italians, Japanese, Hispanics and several other ethnic groups.
Have you been successful in developing an MSUD mouse to use for researching gene therapy for this disease?
Yes, after nearly six years of research, we finally were successful in creating a mouse model for MSUD. As I mentioned before, this model was created by targeted inactivation of the E2 subunit of the BCKDH. The BCKDH is made up of several subunits and all subunits are essential for enzyme activity. Therefore, mutation or inactivation of any subunits, including the E2, results in MSUD.
As to be expected of our MSUD mouse model, all homozygous pups were normal at birth, but as they suckled their mother's milk, within hours they became sick and eventually died. Nearly 80% of the MSUD mouse pups died within a day after birth and the remaining 20% died during the next 6 to 7 days. Such a rapid and high rate of mortality in this model was due to the fact that there is a 100% loss of enzyme activity because the E2 gene knockout is complete. The heterozygous litter mates (which have half of the enzyme activity as compared to normal controls) remained normal and healthy. Because of this neonatal lethality of the homozygous pups, we were unable to test any new treatments.
The fact that these first MSUD mice could not be kept alive long enough to permit gene therapy must have been discouraging. How did you address this issue?
While we were disappointed with this neonatal lethality, we were also reassured that we have indeed created a model which mimicked the classic form of MSUD.
To overcome this problem, we undertook to create a conditional transgenic rescue of the neonatal phenotype of the MSUD mouse. In this new transgenic line, the E2 gene is under the control of a tetracycline responsive promoter, and the investigator can control the level of transgene expression by simply adding (or removing) tetracycline from the drinking water. Since the level of gene expression can be controlled, not only can an MSUD null phenotype be created, but intermediate phenotypes with low levels of residual enzyme activity, mimicking what is observed in many MSUD humans, can also be created by partially suppressing the expression of the gene.
We have completed all the work and have now produced the transgenic rescue model. In this improved MSUD model, most of the homozygous pups do not die. As a result, we are now able to keep these animals alive for several months. Our longest surviving MSUD mouse is nearly a year old.
What is the advantage of keeping these mice alive for a longer period of time?
There are three major advantages to keeping these animals alive on a long-term basis. First, we can now use these animals to test gene therapy for MSUD. Most gene therapy protocols require the use of viral vectors as a gene delivery system. Most currently available vectors do not show the effect of the delivered gene immediately, but require 7 to 10 days to show the effect. We could not test gene therapy in our first MSUD model because the animals died within a week. The fact that we can now keep the improved MSUD mice alive for several months provides the opportunity to test gene therapy for MSUD.
Second, by keeping the animals alive, we allow the animal to develop and grow normally to adulthood before testing gene therapy.
Third, in the longer surviving animals, we can control the transgene expression by tetracycline. By titrating the dose of this drug, we can partially or fully suppress the expression of the gene. The use of an appropriate dose of tetracycline allows us to produce mice not only of the classic form of MSUD, but also of the intermediate and intermittent form of the disease, which will make it possible to test gene therapy on all variants of MSUD.
Do heterozygous mice (carriers - each having one copy of the gene) reproduce with the same incidence of MSUD in their offspring as do humans?
Yes, pups derived from heterozygous by heterozygous mating were born at the expected frequency. Genotype analysis of the pups revealed that wild types [normal], heterozygotes, and homozygotes were present at nearly the expected 1:2:1 [25%, 50%, 25%] frequency. Thus, our heterozygous mice accurately model the genetics and reproduction of the human MSUD.
Evidently there is a great deal of time and cost involved in producing and maintaining these lines?
In making the improved and non-lethal MSUD model, we first created several tetracycline regulated E2 transgenic mouse lines. We then crossed some of these lines to the original MSUD knockout mouse in order to produce the non-lethal improved MSUD model. All this work requires maintaining a large colony of both types of mice. As an example, the cost of maintaining 200 mouse cages exceeds $70,000 per year. Also, considerable time and breeding are required to produce improved MSUD mice. For example, only 14% of the offspring are expected to be homozygous for both genes. This requires a great deal of time and cost to carry out such a project. Finally, genotyping all the newborn pups and their further characterization, such as testing their response to tetracycline by blood amino acid analysis, requires additional cost and time.
You mentioned the mutation in the E2 subunit. What about the other mutations in persons with MSUD? Will it be necessary to create a mouse model for each mutation of MSUD in order to test gene therapy for that mutation?
Our current model represents only one variant of MSUD, namely a mutation in the E2 subunit. Additional mouse models with a mutation in the E1 subunit will have to be created to cover all the MSUD phenotypes.
Which tissues are you planning to target for gene therapy?
Our initial thinking was to deliver the gene to the liver. Recent studies with human tissues, however, have shown that a significant amount of BCKDH is present also in skeletal muscle. Therefore we are now planning to target both the liver and the skeletal muscle. We believe delivering the gene at these two sites would be more effective in curing MSUD than delivering it to a single tissue such as the liver. Moreover, skeletal muscle is
a much bigger tissue than liver, and restoration of the enzyme activity in this tissue is expected to be very effective in the disposal of the BCAAs in MSUD mice and in human patients.
A project of this magnitude must require a significant amount of funds. How is a small company such as yours able to do this research?
We have been fortunate to obtain funding in the past from the NIH to support this project. In addition, the founders of the company have continued to use their own family resources for the project. We are very appreciative of the financial help provided by the MSUD Family Support Group. We are hopeful for the continued support of the project by the NIH in the future. As we continue to make progress, we anticipate drawing the interest of one or more mid to large size companies for this project that could lead to research or a business alliance with one of these companies.
What is your long-term goal in creating these MSUD mice.
Our long-term goal is to develop novel therapies through research using these mice. We believe our MSUD mouse will serve as a very practical resource to test gene therapy, and possibly other therapies, to cure MSUD. These studies will establish the long-term safety and efficacy of gene therapy for this disease. The experience gained from animal studies will pave the way for subsequent gene therapy for MSUD in humans.
A successful gene therapy model for MSUD could be applicable and transferable to other similar disorders. Furthermore, the defects of MSUD cross boundaries into other diseases. For example, neurologic and myelin disorders, GABA-related disorders, and neural cell apoptosis are also prevalent in other diseases. Thus, the MSUD mouse could provide a model for an expanded clinical disease base far beyond MSUD.
Is there a possibility of clinical trials on humans with MSUD in the near future?
Clinical trials on humans will depend on the outcome of gene therapy studies with animals. Successful demonstration of gene therapy in animals, including its long-term safety, would establish the "proof-of-principle" and would reduce the risk for similar studies in humans. Other issues that must be considered before embarking with clinical trials are related to adverse reactions, such as an immune response to vectors used to deliver normal genes. Therefore, a lot more needs to be done in a pre-clinical settings before gene therapy could be attempted in humans.
Besides gene therapy, other approaches to correct genetic diseases are being developed. For example, experiments are being attempted that target genetic repair of the defective gene. Whether this new approach at gene repair would be better than gene therapy remains to be seen. Even for these new approaches, our animal model will remain a very valuable resource.
1. Dr. Paul and colleagues have published 25 papers related to BCAA metabolism between 1976 and 1999.
A rich man once asked a friend, "Why am I criticized for being so miserly? Everyone knows I will leave everything to charity when I die." "Well," said the friend, "let me tell you about the pig and the cow. The pig was lamenting to the cow one day about how unpopular he was, "People are always talking about your gentleness and your kind eyes," said the pig. "Sure you give milk and cream, but I give more. I give bacon, ham and bristles. They even pickle my feet. Still nobody likes me. Why is this?" The cow thought a minute, then replied, "Well, maybe it's because I give while I'm still living."