Functional Glycomics homepage

Article navigation

Featured Articles

Gene therapy: Removing the GAGs

Functional Glycomics (13 January 2011) | doi:10.1038/fg.2011.1

Delivery of glycosaminoglycan-degrading enzymes to the brain relieves neuropathology in dogs with lysosomal storage diseases.

Vectors based on adeno associated viruses (AAV) are used for gene therapy. Image from GrahamColm at en.wikipedia

Lysosomal storage diseases are the most common cause of childhood neurodegeneration. Among these diseases are the mucopolysaccharidoses (MPS), hereditary conditions caused by deficiencies in lysosomal enzymes that normally degrade glycosaminoglycans (GAGs). Writing in Molecular Therapy, Jean-Michel Heard and colleagues now report promising results using gene therapy to replace deficient enzymes in dog models of two of these diseases, MPS type IIIB (MPSIIIB) and MPS type I (MPSI).

MPSIIIB (also called Sanfilippo syndrome) is caused by a deficiency in α-N-acetylglucosaminidase (NAGLU), whereas MPSI (Hurler syndrome) is caused by a defect in another lysosomal enzyme, α-iduronidase (IDUA). Both diseases are characterized by the abnormal accumulation of GAGs in the brain, which triggers inappropriate build-up of the gangliosides GM2 and GM3 and of storage vacuoles. Studies in MPSIII mouse models suggested that using adeno-associated viral (AAV) vectors to replace missing or defective enzymes in the brain might be a viable route for treating MPS, an idea that was supported by the authors' earlier work on AAV treatment in an MPSI dog model (see Further reading).

In the current study, the authors expanded their initial work with an MPSI dog model, and they also tried the technique in a more recently available MSPIIIB dog model. AAV vectors encoding either IDUA or NAGLU (for MPSI and MPSIIIB dogs, respectively) were delivered using stereotactic tracks to eight different spots in the brains of young adult animals. As previously observed in MPSI dogs, follow-up treatments with immunosuppressants were necessary to prevent problematic inflammatory responses in treated MPSIIIB dogs. After several months (about 4 months for MPSIIIB dogs, and ranging from 0.7–8.3 months for MPSI dogs), the animals were killed and their brains analyzed to assess vector delivery and neuropathology. The vector seemed to have spread efficiently throughout the brain in most animals; however, using quantitative PCR (qPCR) the authors observed large between-animal variations in vector genome copy number. As shown previously for IDUA delivery in MPSI dogs, enzyme assays detected active NAGLU in large areas of the brains of NAGLU-treated dogs that had also received immunosuppressants. Therefore, the enzyme delivery system seemed to work well.

Encouragingly, the authors observed striking reductions in neuropathology in almost all of the 21 MPS dogs that received gene therapy and immunosuppressants. Compared to untreated controls, both MPSI and MPSIIIB animals treated with enzyme-encoding vectors showed significantly decreased GM2 and GM3 accumulation and fewer storage lesions in stained histological sections, although these markers never declined to normal levels. Interestingly, whereas treatment between 7–14 months of age worked well in MSPIIIB dogs, treatment was less effective in MSPI dogs of more than 7 months of age, pointing to potential differences in the possible treatment windows for the two diseases.

Whereas hematopoietic stem cell–based treatments are currently used to treat MPSI in humans, there is no such treatment for MPSIII, a devastating disease that often causes death in the mid-teens. These results suggest that AAV-mediated gene therapy might offer a relatively safe and effective way to relieve neuropathology in MPS patients.

Carole Weaver

Original research paper

  1. Ellinwood, N.M. et al. Safe, efficient, and reproducible gene therapy of the brain in the dog models of Sanfilippo and Hurler syndromes. Mol Ther. epub ahead of print (7 December 2010) doi:10.1038/mt.2010.265  | Article

Further reading

  1. Ciron, C. et al. Gene therapy of the brain in the dog model of Hurler's syndrome. Ann. Neurol. 60, 204–213 (2010) doi: 10.1002/ana.20870 | Article