skeletal muscle glycogen content in association with aγ3 mutation
homologous to a pathogenicγ2-variant [19].
These observations have led to the conceptualization of
PRKAG2cardiomyopathy as a glycogen storage disorder, with its
disease manifestations proposed as being largely attributable to
bulk glycogen accumulation, including glycogen-associated alter-
ation in electrophysiological conduction properties [13, 45]. How-
ever, passive cellular engorgement by glycogen is unlikely sufficient
to fully account for the magnitude of cardiac hypertrophy observed
in the transgenic murine models. Exemplifying this, in the TGT400N
model, glycogen is estimated to account for only ~4% of the
increase in cardiac mass [46], suggesting a substantial contribution
from true cellular hypertrophy [34].
Findings from the inducible Asn488Ile murine transgenic
model (seeSubheading3.1), where a period of early mutant trans-
gene suppression was sufficient to prevent ventricular pre-excita-
tion, suggest an important role for glycogen excess in its
pathogenesis [16] and, by inference, in traditional glycogen storage
diseases [32]. More direct evidence to a support a causal role for
glycogen in the development of pre-excitation comes from a study
of TGN488Imice designed to co-harbor a homozygous knock-in
mutation (Arg582Ala) in muscle glycogen synthase 1 (Gys1), ren-
dering the latter insensitive to normal allosteric stimulation by
glucose-6-phosphate (G6P) [47], a metabolic intermediary found
elevated in hearts from mutant PRKAG2 transgenic mice
[44]. Substantiating inferences from metabolic flux analysis sug-
gesting a key role for G6P-mediated glycogen synthase activation in
mutantPRKAG2-associated cardiac glycogen accumulation [44],
these mice (TGN488I/GSR582A/R582A) have substantially reduced
(~6-fold) myocardial glycogen content [48]. Strikingly this was
associated with normalization of the PR interval (Fig.5), consistent
with elimination of ventricular pre-excitation, and a corresponding
improvement in annulus fibrosus appearance [48]. However,
genetic inhibition of G6P-stimulated glycogen synthase activity
had no significant effect on cardiac hypertrophy—suggesting a
glycogen storage-independent mechanism for LVH—and only
partly rescued cardiac function (mean echocardiography-derived
LV fractional shortening 32%, 42%, and 58%, in TGN488I,
TGN488I/GSR582A/R582A, and non-TG mice, respectively)
[48]. Further investigation of myocardial signalling in TGN488I
and TGN488I/GSR582A/R582Amice has identified enhanced insulin
sensitivity (including increased P-AKTSer473) in conjunction with
mTOR pathway activation [48], the latter finding also observed in
TGT400Nmice as early as 2 weeks of age [46]. As a corollary,
inhibition of mTOR in TGN488Imice, using rapamycin from age
2–6 weeks, substantially (~45%) ameliorates their cardiac hypertro-
phy [48]. A further consequence of AKT hyperactivity identified
was that of forkhead box O (FoxO) transcription factor pathway594 Arash Yavari et al.