In the course of studying inflammatory muscle diseases (polymyositis, dermatomyositis, and related diseases), we have encountered patients with many other muscle diseases. We have studied patients with two genetic metabolic myopathies in detail: phosphofructokinase (PFK) deficiency, and acid maltase (acid alpha-glucosidase) deficiency (also known as Glycogen Deficiency Type II or GSD II or Pompe Syndrome). For the last several years, we have focused particular attention on GSD II because of its close resemblance to myositis. It is a recessively inherited lysosomal storage disease in which glycogen accumulates in the lysosomes, particularly those in skeletal muscle. When the enzyme, known also as GAA, is completely absent, affected infants are usually sick at birth and die in infancy of heart failure, rarely living longer than a year. Apparently the enzyme is needed in the heart only in infancy since affected individuals with even a small amount of effective enzyme survive without cardiac involvement. Survivors generally develop a progressive proximal myopathy with pulmonary failure secondary to diaphragmatic involvement in later years. The long-term aim of our studies is to prevent and to treat this devastating disease, particularly the adult variety, in which the level of enzyme is only slightly (less than two-fold) below the minimum necessary for a normal life. From quite early in the project, it was driven by the belief that GSDII, being a lysosomal storage disease, should be amenable to enzyme replacement and possibly gene therapy, and since there are well over 30 diseases of abnormal lysosomal storage, our results might have wider application. There are estimated to be several thousand cases of GSD II worldwide, and hundreds in the United States, of which many are fatal in infancy (Pompe Syndrome). Our guiding plan has been to do research directed towards therapy, but without trying to move into areas likely to be covered by pharmaceutical companies.
We aim ed to develop tools that would advance the development of therapy while at the same time learning new biology or developing techniques that might be applicable to other lysosomal or other enzyme deficiency diseases, especially those in muscle. We have followed lines of research designed to provide answers necessary for the development of optimal therapy for GSD II. Starting with the knockout mice, a human GAA expression vector was put in transgenically under the control of either a skeletal muscle or a liver specific promoter controllable by doxicycline [164]. These mice were designed to determine whether the liver, a much easier organ to target, could serve as a suitable depot organ for the synthesis and secretion of GAA that could be taken up by the heart and by skeletal muscle. We discovered that the liver was a superior site to the skeletal muscle for the repair of both heart and skeletal muscle. [164] The achievement of therapeutic levels with skeletal muscle transduction required the entire muscle mass to produce high levels of enzyme of which little found its way to the plasma, whereas liver, comprising <5% of body weight, secreted 100-fold more enzyme, all of which was in the active 110 kDa precursor form. Since the transgene for human GAA was expressed during fetal development, we anticipated that the animals would be tolerant to injection of exogenous human GAA - which they were. However, we had one line able to express GAA below the level at which it could be detected by enzymatic assay (mRNA was present by RT-PCR), and this line was not only phenotypically indistinguishable from the knockout parent but was also tolerant to repeated injections of recombinant human GAA. Thus, this line has skeletal muscle and heart disease and can be repeated injected with human enzyme without developing blocking antibodies. It is thus much better suited as a test animal for preclinical testing of recombinant human GAA than the knockout animal model, and it is now the test animal of choice, and has been widely distributed. In the course of these studies, we have had the chance to study the outcome of enzyme replacement therapy either through the endogenous production of transgenic enzyme in the muscles or through the repeated injection of recombinant human enzyme obtained from several companies developing it for human therapy. Of great interest and importance is the discovery that with neither route is it possible to eradicate completely the stored glycogen from skeletal muscle. If the therapy is begun very early (transgene turned on early or injections given early), the heart accumulation can be either prevented or essentially completely reversed. Later therapy is less effective for the heart. These findings suggest that the dismal results of the early human trials cannot be ascribed solely to poor recombinant enzyme, but probably reflect the disease process. And they have pointed towards our path for future research. If the residual glycogen is not digested by exogenous or endogenously replaced GAA, the enzyme may not ever reach the affected lysosomes; the enzyme may reach the lysosomes but the conditions within the lysosome (e.g. pH) may not be optimal for enzymatic action; the residual glycogen may have developed structural features, for example, cross-links, that render it resistant to GAA; or the residual glycogen may primarily be outside lysosomes. Although some have observed glycogen outside lysosomes in patients with GSD II, we have found extremely little. We have planned and begun experiments to test these possibilities. We have also begun treating cerebral glycogen accumulation by altering transport of GAA across the blood brain barrier. Knockout mice accumulate large amounts of glycogen in neurons, and it is possible that some of the wasting and paresis in those mice is due to neurologic damage. We can anticipate similar accumulation in the brains of infants with very low levels of enzyme if enzyme replacement therapy allows them to live much longer. We have entered into a collaboration with Stanley Rapaport of NIMH, who has worked on blood brain barrier for many years. He devised a method to transiently disrupt the barrier osmotically by the injection of arabinose or mannose into the carotid artery. Several drugs that affect calcium transport prolong the osmotic effect during which time it should be possible to infuse recombinant enzyme into the carotid artery where it could cross the blood brain barrier at the capillary level. The technical problem of extending to mice the surgical techniques used in rats have rrecently been largely overcome, and definitive treatment experiments will begin in the near future. An additional project was concerned with the transcriptional control of GAA since this seemed an alternative route to therapy. We decided to pursue the identity of the silencing activity because of the hope that a method of accelerating transcription off the allele bearing g-13t might offer a simpler approach to therapy of at least the adult disease. The experiments demonstrated that a region of intron 1 in which transcriptional activity was controlled in a complex and tissue specific fashion. The identities of several of the transcription factors were determined, and others were observed but not identified. Of greatest interest was the discovery that human GAA is regulated by the Notch-1/Hes-1 pathway, an important pathway in a number of developmental processes, suggesting that this lysosomal enzyme may have still unrecognized physiological roles in development. This line of experiments has been terminated since it appears very unlikely to yield therapeutically usable information in the near future.
Showing the most recent 10 out of 17 publications
