In the course of studying inflammatory muscle diseases, we have encountered patients with 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). This failure of skeletal muscle response to recombinant enzyme has become a major focus of our work. The problem is likely to reside in properties of the target tissue itself: for example, the inability of the lysosomes to maintain low pH, or the inability of muscle cells to uptake and deliver the enzyme to those lysosomes. To study these possibilities, we have used myoblasts from normal and knockout mice. To define the subcellular compartments in Pompe myoblasts the cells were transfected and/or immunostained with markers for different organelle: Golgi complex (GM130); early endosomes (EA1 and Rab5); late endosomes (mannose-6-phosphate receptor or LAMP) and lysosomes (LAMP); autophagic vacuoles (MAP-LC3). We have solved a number of technical problems to allow these multi-marker experiments to be done, and we have developed a new technique to allow the determination of pH within late endocytic compartments in living cells that are doubly labeled with a mixture of Oregon green (ORG488) and tetramethylrhodamine (TMR) - conjugated dextrans or with fluorescein (FL)/tetramethylrhodamine (TMR)- double conjugated dextran. The non-digestible dextran enters the cells by a fluid-phase endocytosis, and it is then delivered through endosomal compartments to lysosomes, where it accumulates. The confocal microscopic late endosomes/lysosomes pH assay is based on measuring the ratio of pH-sensitive fluorescein (FL or ORG488) to pH-insensitive tetramethylrhodamine (TMR) fluorescence emissions. What we have found is that in Pompe cells the abnormalities go beyond the lysosome. Vacuoles of the endocytic pathway (the route of the hGAA) ? early and late endosomes, as well as the vacuoles of the autophagic pathway, which converges with the endocytic pathway - are all dramatically expanded in the diseased cells. The expanded early endosomes and late endosome/lysosomes appeared much closer together in the knockout compared to the wild type cell.Because we see enlargement of all the vacuoles along the lysosomal degradative pathway in the diseased cells, the question of pH becomes one not limited to the lysosomes but rather a broader issue of the acidification of the endosomal compartments. It has been well documented that efficient targeting and processing of lysosomal enzymes require a proper pH gradient along the endocytic pathway. Despite a significant expansion of the endocytic vesicles, the changes in the vacuolar pH in KO myoblasts were not large, and the majority of late endosomes/lysosomes maintained pH within the normal range. There was, however, an increased population of vesicles with pH above the normal lysosomal range and even above the normal late endosomal range, suggesting a defective acidification of a subset of the late endosomes/lysosomes that may be of significant pathogenetic importance in multinucleated muscle fibers. This should reduce the enzymatic activity of the acid alpha-glucosidase in lysosomes and may also diminish its release from the receptor in late endosomes. In addition, the dextran labeling experiments have shown that the transport from late endosomes to lysosomes may pose an additional problem in the diseased cells. Indeed, slow movement of the enlarged late endocytic compartments in Pompe cells was shown by imaging of living normal and Pompe myoblasts transfected with GFP-Lamp1. In the control cells vesicles have more freedom of movement, while in overcrowded Pompe cells their movement is limited. The immobile late endosomes/lysosomes occupied significantly higher percentage of the total late endosomal/lysosomal area in the diseased cells compared to normal cells. Thus, a global secondary impairment of vacuolar membrane trafficking of the endocytic-autophagic pathway in Pompe cells is likely to have a profound effect on the delivery of the recombinant enzyme to the glycogen-filled vacuoles. We have also demonstrated that not all muscle cells respond equally to ERT. Both ERT and transgenic studies clearly established that muscles rich in slow-twitch oxidative type 1 muscle fibers clear glycogen more efficiently compared to muscle rich in fast-twitch glycolytic type II fibers despite their higher level of glycogen accumulation. Thus, we have tried to identify the differences between the fiber types that may account for the various responses to therapy. To explore possible mechanisms for the uneven pattern of glycogen removal from different muscles fibers, we looked at the expression levels of proteins critical for the internalization, sorting and delivery of the therapeutic enzyme to the lysosomes: the recycling proteins such as mannose-6-phosphate or transferrin receptors, clathrin, AP-2 clathrin adaptor protein, and GGAs (Golgi-localized gamma-ear-containing, Arf-binding proteins). Remarkably, all these proteins were much less abundant in type II fibers in both wild type and knockout mice, suggesting that intrinsic properties of different fiber types might play a role: the entry of the rhGAA into cells and trafficking to lysosomes may be much slower and less efficient in the therapy-resistant type II fibers. To further understand the mechanisms of type II muscle fiber resistance to therapy we have now moved to studying single muscle fibers isolated from type I or type II ?rich muscles. Confocal immunohistochemical studies revealed the most striking difference between the fiber types: the presence of large autophagic vacuoles in type II, but not in type I fibers. In all eukaryotic cells, the endocytic pathway converges with the autophagic pathway, which transports cytoplasmic material to the lysosomes. Autophagy is a highly conserved mechanism for degrading most long-lived proteins and cytoplasmic organelles. It was originally believed that the endocytic and autophagic pathways merge at the lysosomes. However, recent biochemical and cytochemical studies clearly indicate that these two membrane trafficking pathways can converge before reaching lysosomes when the autophagosomes fuse with early and late endosomes. Since we see increased autophagy and a global expansion of endocytic vesicles, it is highly likely that the autophagic vacuoles fuse with late or early endosomes in the therapy-resistant type II fibers. The autophagic changes, reflecting the ongoing process of degradation of intracellular proteins and organelles, are seen in virtually every type II fiber from the knockout mice. The areas of autophagic build up could be found along the entire length of the fiber. Importantly, the autophagic build-up was also abundant in the ERT-treated type II muscle fibers. Another difference between the fiber types, most likely related to the autophagy was a dramatic reduction in the size of type II fibers in the knockout compared to the wild type. In contrast, type I fibers showed a tendency for hypertrophy rather than atrophy. Thus, on top of the global vacuolar dysfunction several factors may contribute to type II fiber resistance: first, increased autophagy; and second, intrinsic properties of the fiber types, such as low levels of trafficking proteins and structural organization of the lysosomes. We have now moved to experiments designed to directly demonstrate the localization of the therapeutic enzyme in isolated single type II fibers of the knockout mice. In parallel, we have been analyzing and verifying by real-time PCR the results of gene expression studies of the heart and type I and type II muscles from normal and knockout mice.
Showing the most recent 10 out of 17 publications
