The Role of HSP70 and Its Co-chaperones in Protein Misfolding, Aggregation and Disease - The Networking of Chaperones by Co-chaperones - page 250

induce the expression of HSP70 and other chaperones (Soti et al. 2005 ). These
include the hydroxylamine derivative bimoclomol and its analogue, arimoclomol,
and the benzoquinone ansamycin antibiotic, geldanymycin and derivatives thereof.
These compounds potentiate chaperone expression by activating heat shock tran-
scription factor Hsf-1 (Soti et al. 2005 ). Hydroxylamine derivatives bind Hsf-1 to
stabilize the active phosphorylated trimer and prolong its binding to the heat shock
response element found in the heat shock gene promoters. In contrast, the benzo-
quinone ansamycins bind to the ATP site on HSP90 and block its interaction with
Hsf-1 and other clients, thereby leading to Hsf-1 trimerization and stimulating the
transcription of heat shock proteins.
These pharmacological compounds have been tested in several models of
polyQ disease. The geldanamycin derivative 17-AAG (17-allylamino-17-deme-
thoxygeldanamycin) (Tanespimycin) suppressed neurodegeneration and rescued
lethality in a Drosophila model of SCA3 and HD dependent on Hsf-1 activation
and HSP induction (Fujikake et al. 2008 ). 17-AAG and the derivative 17-DMAG
(17-(dimethylaminoethylamino)-17-demethoxygeldanamycin) (Alvespimycin), were
reported to significantly ameliorate polyQ-mediated motor neuron degeneration
in the SBMA transgenic mouse model without detectable toxicity through HSP90
chaperone complex formation and preferential proteasome-dependent degradation
of polyQ-expanded mutant AR (Waza et al. 2005 ; Tokui et al. 2009 ). Arimoclomol
has also shown very encouraging results in the transgenic mouse model of SBMA,
significantly improving motor neuron survival and rescuing the neuromuscular phe-
notype (Malik et al. 2013 ). Moreover, arimoclomol has been reported to delay disease
progression in ALS mice, resulting in a 22 % increase in lifespan (Kieran et al. 2004 ).
This correlated with a slight increase in Hsp27 (HSPB1) levels and a significant in-
crease in both HSP70 and HSP90 levels in the spinal cord of the treated ALS mice.
Arimocolomal has reached clinical testing in ALS (Cudkowicz et al. 2008 ; Lanka
et al. 2009 ). However, the complete mechanism of action of arimoclomol (and other
heat shock response activating drugs) is uncertain at present, and it is not clear that all
the beneficial effects of arimoclomol are related to the potentiation of the heat shock
response; for example, recent evidence showed that arimoclomol can also potentiate
the unfolded protein response when there is ER stress (Parfitt et al. 2014 ).
The clinical development of geldanamycin has been limited by its poor phar-
macokinetic profile, including poor solubility and blood-brain-barrier permeability.
Geldanamycin derivatives, including 17-AAG and 17- DMAG, have better pharma-
cokinetic profiles but their use is limited by poor blood-brain-barrier permeability
and toxicity, respectively (Kim et al. 2009 ; Porter et al. 2010 ). Moreover, there is
evidence that these drugs may directly affect the interaction of HSP90 with client
proteins leading to undesirable cellular effects. For example, Aquil￡ et al. ( 2014 )
recently reported that inhibition of HSP90 by 17-AAG and the HSP90 inhibitor
2-amino-7, 8-dihydro-6H-pyrido[4,3-D]pyrimidin-5-one NVPHSP990 (HSP990)
post-transcriptionally down-regulated HSP90 client proteins essential for vision,
including the G-protein-coupled receptor kinase (GRK1) and the ʲ subunit of reti-
nal phosphodiesterase (PDE6ʲ). Interestingly, the stability of retinal PDE6 requires