Theresa M. Harter

(1) Background: Cantu syndrome (CS) arises from gain-of-function (GOF) mutations in the ABCC9 and KCNJ8 genes, which encode ATP-sensitive K⁺ (KATP) channel subunits SUR2 and Kir6.1, respectively. Most CS patients have mutations in SUR2, the major component of skeletal muscle KATP, but the consequences of SUR2 GOF in skeletal muscle are unknown. (2) Methods: We performed in vivo and ex vivo characterization of skeletal muscle in heterozygous SUR2[A478V] (SUR2^wt/AV) and homozygous SUR2[A478V] (SUR2^AV/AV) CS mice. (3) Results: In SUR2^wt/AV and SUR2^AV/AV mice, forelimb strength and diaphragm amplitude movement were reduced; muscle echodensity was enhanced. KATP channel currents recorded in Flexor digitorum brevis fibers showed reduced MgATP-sensitivity in SUR2^wt/AV, dramatically so in SUR2^AV/AV mice; IC₅₀ for MgATP inhibition of KATP currents were 1.9 ± 0.5 × 10⁻⁵ M in SUR2^wt/AV and 8.6 ± 0.4 × 10⁻⁶ M in WT mice and was not measurable in SUR2^AV/AV. A slight rightward shift of sensitivity to inhibition by glibenclamide was detected in SUR2^AV/AV mice. Histopathological and qPCR analysis revealed atrophy of soleus and tibialis anterior muscles and up-regulation of atrogin-1 and MuRF1 mRNA in CS mice. (4) Conclusions: SUR2[A478V] “knock-in” mutation in mice impairs KATP channel modulation by MgATP, markedly so in SUR2^AV/AV, with atrophy and non-inflammatory edema in different skeletal muscle phenotypes.

Cantu syndrome (CS) is caused by gain-of-function (GOF) mutations in pore-forming (Kir6.1, KCNJ8 ) and accessory (SUR2, ABCC9 ) ATP-sensitive potassium (K _ATP ) channel subunits, the most common mutations being SUR2[R1154Q] and SUR2[R1154W], carried by approximately 30% of patients. We used CRISPR/Cas9 genome engineering to introduce the equivalent of the human SUR2[R1154Q] mutation into the mouse ABCC9 gene. Along with minimal CS disease features, R1154Q cardiomyocytes and vascular smooth muscle showed much lower K _ATP current density and pinacidil activation than WT cells. Almost complete loss of SUR2-dependent protein and K _ATP in homozygous R1154Q ventricles revealed underlying diazoxide-sensitive SUR1-dependent K _ATP channel activity. Surprisingly, sequencing of SUR2 cDNA revealed 2 distinct transcripts, one encoding full-length SUR2 protein; and the other with an in-frame deletion of 93 bases (corresponding to 31 amino acids encoded by exon 28) that was present in approximately 40% and approximately 90% of transcripts from hetero- and homozygous R1154Q tissues, respectively. Recombinant expression of SUR2A protein lacking exon 28 resulted in nonfunctional channels. CS tissue from SUR2[R1154Q] mice and human induced pluripotent stem cell–derived (hiPSC-derived) cardiomyocytes showed only full-length SUR2 transcripts, although further studies will be required in order to fully test whether SUR2[R1154Q] or other CS mutations might result in aberrant splicing and variable expressivity of disease features in human CS.

Aldose reductase (AKR1B1) is an NADPH-dependent aldo-keto reductase best known as the rate-limiting enzyme of the polyol pathway. Accelerated glucose metabolism through this pathway has been implicated in diabetic cataract and retinopathy. Some human tissues contain AKR1B1 as well as AKR1B10, a closely related member of the aldo-keto reductase gene superfamily. This opens the possibility that AKR1B10 may also contribute to diabetic complications. The goal of the current study was to characterize the expression profiles of AKR1B1 and AKR1B10 in the human eye. Using quantitative reverse transcriptase-PCR and immunohistochemical staining, we observed expression of both AKR genes in cornea, iris, ciliary body, lens, and retina. Expression of AKR1B1 was the highest in lens and retina, whereas AKR1B10 was the highest in cornea. Lenses from transgenic mice designed for overexpression of AKR1B10 were not significantly different from nontransgenic controls, although a significant number developed a focal defect in the anterior lens epithelium following 6 months of experimentally induced diabetes. However, lenses from AKR1B10 mice remained largely transparent following longterm diabetes. These results indicate that AKR1B1 and AKR1B10 may have different functional properties in the lens and suggest that AKR1B10 does not contribute to the pathogenesis of diabetic cataract in humans.

Murine fibroblasts cultured in the presence of fibroblast growth factor-1 express relatively high levels of FR-1, a ∼36 kDa protein related to the aldo-keto reductase superfamily [Donohue, P. J., Alberts, G. F., Hampton, B. S., Winkles, J. A. (1994) J. Biol. Chem. 269, 8604−8609]. While the crystal structure of FR-1 shows striking homology with human aldose reductase [Wilson, D. K., Nakano, T., Petrash, J. M., Quiocho, F. A. (1995) Biochemistry 34, 14323−14330], an enzyme linked to the pathogenesis of diabetic complications, the physiological role of FR-1 is not known. We show that FR-1 is capable of reducing a broad range of aromatic and aliphatic aldehydes, including the abundant and highly reactive lipid-derived aldehyde 4-hydroxy-2-nonenal (HNE; K_m ≈ 9 μM). However, in the absence of coenzyme, HNE caused a time-dependent inactivation of FR-1. Results from electrospray ionization-mass spectrometry and Edman-degradation of peptides derived from HNE-modified FR-1 were consistent with formation of a Michael adduct at Cys298. This was confirmed with a C298S mutant, which was resistant to HNE-induced inactivation. Since steady-state K_m values determined with alkanals, α,β-unsaturated alkenals, alkadienals, and 4-hydroxyalkenals fall within their physiological concentrations, lipid-derived aldehydes appear to be potential in vivo substrates for FR-1.