PROJECT SUMMARY/ABSTRACT
Charcot-Marie-Tooth disease (CMT) is a collection of inherited peripheral neuropathies with a cumulative
incidence of ~1:2500 people. There is no approved treatment for any of the 100 genetic subtypes of CMT,
presenting a large unmet clinical need. Charcot-Marie-Tooth type 2D is caused by dominant mutations in glycyl
tRNA synthetase (GARS), encoding the enzyme that charges glycine onto its cognate tRNAs during translation.
Our working model for the disease mechanism is that the mutant enzyme binds its tRNA substrate, but does not
release it to the ribosome, effectively sequestering the substrate. This results in ribosome stalling at glycine
codons and activation of the integrated stress response. Support for this mechanism comes from genetic studies
in Drosophila and mouse models of CMT2D, in which transgenic overexpression of tRNAGlyGCC was able to
effectively suppress the neuropathy phenotype. In preliminary studies, we have reproduced this result using
AAV9 to deliver tRNAGly genes to three different CMT2D mouse models. Glycine has four codons (GGC, GGG,
GGA, and GGU), and therefore four potential anticodons (GCC, CCC, UCC, and ACC respectively, though ACC
is likely a nonfunctional tRNA). We made four AAV9 vectors expressing each tRNAGly anticodon driven by a
PolIII U6 promoter. We found that GCC was highly effective, almost completely suppressing the neuropathy
phenotype even in mice with a severe allele of Gars. Vectors expressing CCC and UCC were intermediate in
efficacy, and ACC was ineffective (as anticipated). This profile of efficacy correlates with tRNA abundance and
codon usage, and suggests we are replacing the sequestered substrates of GARS with the AAVs. In the R61
phase of this proposal we will optimize the vector payload (Aim 1) and capsid (Aim 2), and in the R33 phase
(Aim 3), we will use this optimized vector in rigorous preclinical studies in mouse models of CMT2D. In Aim 1
(R61), we will construct an AAV9 vector that carries all three effective tRNAGly genes (GCC, CCC, UCC) in a
single vector. We will compare this against GCC alone, which was very effective. In Aim 2 (R61), we will
recreate the U6-GCC vector in a MACPNS capsid in an attempt to create a vector that is effective with systemic
delivery, rather than dosing directly into the nervous system. We will compare the MACPNS-GCC vector to
AAV9-GCC. In Aim 3 (R33), we will test the optimized vector (GCC or combined tRNAGlys, AAV9 or MACPNS)
in two mouse models of CMT2D. We will also allow treated mice to age to show the perdurance of the effect,
and we will examine the effects of treating after the onset of neuropathy. The successful completion of these
aims will show the in vivo efficacy of an optimized gene therapy treatment for CMT2D in preclinical studies in
mouse models. This will position us for further translational research and IND-enabling studies through
mechanisms such as the Blueprint Neurotherapeutics Network for Biologicals.