Pathophysiological Consequences of KATP Channel Overactivity and Pharmacological Response to Glibenclamide in Skeletal Muscle of a Murine Model of Cantù Syndrome

Cantù syndrome (CS) arises from mutations in ABCC9 and KCNJ8 genes that lead to gain of function (GOF) of ATP-sensitive potassium (KATP) channels containing SUR2A and Kir6.1 subunits, respectively, of KATP channels. Pathological consequences of CS have been reported for cardiac and smooth muscle cells but consequences in skeletal muscle are unknown. Children with CS show muscle hypotonia and adult manifest fatigability. We analyzed muscle properties of Kir6.1[V65M] CS mice, by measurements of forelimb strength and ultrasonography of hind-limb muscles, as well as assessing KATP channel properties in native Flexor digitorum brevis (FDB) and Soleus (SOL) fibers by the patch-clamp technique in parallel with histopathological, immunohistochemical and Polymerase Chain Reaction (PCR) analysis. Forelimb strength was lower in Kir6.1wt/VM mice than in WT mice. Also, a significant enhancement of echodensity was observed in hind-limb muscles of Kir6.1wt/VM mice relative to WT, suggesting the presence of fibrous tissue. There was a higher KATP channel current amplitude in Kir6.1wt/VM FDB fibers relative to WT and a reduced response to glibenclamide. The IC50 of glibenclamide to block KATP channels in FDB fibers was 1.3 ± 0.2 × 10−7 M in WT and 1.2 ± 0.1 × 10−6 M in Kir6.1wt/VM mice, respectively; and it was 1.2 ± 0.4 × 10−7 M in SOL WT fibers but not measurable in Kir6.1wt/VM fibers. The sensitivity of the KATP channel to MgATP was not modified in Kir6.1wt/VM fibers. Histopathological/immunohistochemical analysis of SOL revealed degeneration plus regressive-necrotic lesions with regeneration, and up-regulation of Atrogin-1, MuRF1, and BNIP3 mRNA/proteins in Kir6.1wt/VM mice. Kir6.1wt/VM mutation in skeletal muscle leads to changes of the KATP channel response to glibenclamide in FDB and SOL fibers, and it is associated with histopathological and gene expression changes in slow-twitch muscle, suggesting marked atrophy and autophagy.


INTRODUCTION
Cantù syndrome (CS, OMIM#23985, also known as hypertrichotic osteochondrodysplasia) is a rare autosomal dominant inheritance, multi-organ condition (Cantú et al., 1982), characterized by cardiomegaly, vascular dilation, and low blood pressure (Grange et al., 2006;Scurr et al., 2011;Brownstein et al., 2013;Nichols et al., 2013) together with hypertrichosis and skeletal malformations. Neuromuscular symptoms have been observed in some patients (Leon Guerrero et al., 2016). The molecular basis of CS is now recognized to be gain of function (GOF) mutations in the ABCC9 and KCNJ8 genes, which encode the regulatory sulfonylurea receptor SUR2 (ABCC9) and pore-forming Kir6.1 (KCNJ8) subunits of ATP-sensitive K + (KATP) channels (Harakalova et al., 2012;Cooper et al., 2014). To date, ∼70 individuals with CS, associated with >30 missense ABCC9 or KCNJ8 mutations have been reported in the literature . All characterized mutations resulted in enhanced activity of recombinant KATP channels when expressed in heterologous expression systems that arises from the reduced sensitivity to ATP (Cooper et al., 2015;McClenaghan et al., 2018).
Patients with CS have a complex phenotype from birth. Some children with CS show muscle hypotonia leading to delays in the development of motor skills such as sitting, standing, and walking (https://ghr.nlm.nih.gov/condition/cantu-syndrome#definition). Also, adult CS patients tend to have a muscular appearance, caused by an increase in muscle size, although masked by edema when older, and self-report fatigability. KATP channels are present in the membrane of skeletal muscle fibers where they are responsible for unique feedback between muscle cell metabolism and electrical activity (Flagg et al., 2010). In resting skeletal muscle fibers, these sarco-KATP channels are mostly closed and contribute little to the resting membrane potential (Tricarico et al., 1997). Activation of sarco-KATP channels occurs as a response to metabolic stress and reduced ATP/ADP ratio, leading to a reduction in action potential duration (Tricarico et al., 2016a); in so doing, they contribute to the decline in skeletal muscle excitability and force production during prolonged repetitive stimulation and fatigue (Pedersen et al., 2009). Therefore, KATP channels play a protective role in skeletal muscle, helping to preserve structural integrity, avoiding fiber damage caused by intense exercise, buffering ATP levels during fatigue and tetanus, contributing to glucose uptake regulation, and Ca 2+ handling. This interpretation is supported by the finding that pinacidil, an opener of KATP channels, increases the rate of fatigue in Extensor digitorum longus and Soleus muscles of wild-type mice (Tricarico et al., 2016b). On the other hand, reduction of sarco-KATP channel activity is one mechanism of the primary and secondary forms of hypokalemic periodic paralysis, transient weakness and hypokalemia (Tricarico et al., 1999a;Tricarico et al., 1999b). Sarco-KATP channels are predominantly composed of Kir6.2/SUR2A subunits, but we have documented the existence in skeletal muscle of hybrid assemblies of other subunits organized as heteromeric complexes (Tricarico et al., 2006). CRISPR/Cas9-modified mice, in which CS-associated single nucleotide mutations have been introduced into native KCNJ8 (Kir6.1[V65M]) or ABCC9 (SUR2[A478V]) loci provide tractable animal models in which to understand cellular mechanisms and organ consequences of CS (Huang et al., 2018), and this has led to the recognition of the role of channel overactivity in vascular smooth muscle in generating the cardiovascular phenotype CS (McClenaghan et al., 2020). The majority of studies investigating KATP channel in skeletal muscle have either used non-selective pharmacological modulators (glibenclamide/pinacidil/diaxoxide) or knock-out of Kir6.2. GOF mutation of KCNJ8 gene is found to be responsible for the most severe CS phenotype both in humans and animal models. Even so, the consequence of CS mutations in skeletal muscle and the molecular mechanisms responsible for exercise intolerance and muscle fatigability in CS patients appear so far uninvestigated. In this work, therefore, we evaluated the effects induced by the Kir6.1[V65M] CS mutation on skeletal muscle from heterozygous Kir6.1[V65M] (Kir6.1 wt/VM ) and wild type (WT) mice, by combining in vivo and ex vivo experiments. We investigated the biophysical and pharmacological properties of KATP channels in fast-twitch Flexor digitorum brevis (FDB) and slow-twitch Soleus (SOL) muscle fibers by patch-clamp technique. The primary endpoint was the change in the KATP channel current recorded in excised macro-patches from skeletal muscle fibers of Kir6.1[V65M] (Kir6.1 wt/VM ) and wild type (WT) mice, and response to glibenclamide. Secondary endpoints were forelimb strength measurements and ultrasonography evaluation of hind limb muscles to assess functional and morphological consequences. Polymerase Chain Reaction (PCR) analysis of gene expression, as well as histopathological and immunohistochemistry evaluations in different muscle types, were performed to understand the cellular origins of these features.

Animal Care
Novel knock-in Kir6.1[V65M] mice, resembling the human Cantù syndrome (CS), were generated through CRISPR/Cas9 gene editing and genotyped (Huang et al., 2018) at Washington University, Saint Louis, USA, and then transferred to Italy. Kir6.1wt/VM mice (N mice 4) and wild type (WT) mice (N mice 4) were maintained two to four per cage at the Stabulario of the Dipartimento di Farmacia-Scienze del Farmaco, University of Bari, Italy, under the supervision of the veterinary officer according to D.lgs. 26/2014. Experiments were performed on male mice since no evidence of gender differences currently exist in CS . The temperature of the laboratory was maintained at 22 ± 1°C, with a relative humidity of 50 ± 5%, and under 12:12 light/dark cycles; the animals were provided and maintained on a standard laboratory diet and water ad libitum.
O.P.B.A (Organization for Animal Health) (prot. 8515-X/10, 30-01-2019). The animal care, protocols, and the sample size (number of mice needed) were calculated at the minimum required to reach the statistical significance based on the primary endpoint according to the 3R "Replace, Reduce, Refine" rules (see below). For minimizing the risk of observer bias and other "experimenter effects," experiments were conducted "in blind," meaning that experimenters were unaware, when possible, of the genotype of the animals (Holman et al., 2015).

In-vivo Parameters of Muscle Strength
Evaluation of forelimb strength was performed using a grip strength meter (Columbus Instruments, Columbus, Ohio), according to TREAT-NMD SOPs (De Luca, 2019), by an investigator blinded to mouse genotype. Mice were allowed to grasp a triangular ring connected to a force transducer and then gently pulled away until the grip was broken. The force applied by the animal at this point represents the maximal resistance the animal can use with its forelimbs. For each animal, at least five separate measurements were made within 2 min and averaged. Both the absolute and normalized (to body weight) medium forelimb force values were used for statistical analysis (De Luca et al., 2003).

Ultrasound Evaluations
Ultrasonography experiments were conducted using the ultrahigh frequency ultrasound bio-microscopy system Vevo 2100 (VisualSonics, Toronto, ON, Canada), by an investigator blinded to mouse genotype. Each animal was anesthetized via inhalation (induction with 3% isoflurane and 1.5% O 2 l/min, then constantly maintained via nose cone at 2-1.5% isoflurane and 1.5% O 2 ) and placed on a thermo-statically controlled table (kept at 37°C) equipped with four copper leads which allowed monitoring of heart and respiratory rate. A rectal probe was used to monitor body temperature. The mice were prepared as previously described (Mele et al., 2016;Whitehead et al., 2016;Conte et al., 2017).
Ultrasound acquisitions of the hind limb were performed to evaluate total hind limb volume (in mm 3 ), percentage of vascularization (PV%) and hindlimb echodensity by using the MS250 probe. A 3-dimensional (3D) volume scan of the hind limbs was acquired by translating the ultrasound probe parallel to the long axis of the hind limb. Multiple 2D images were acquired at regular intervals in Power Doppler mode. At the end of the procedure, 3D images were reconstructed from previously collected multiple 2D frames and visualized with VisualSonics 3D software (Mele et al., 2019). The software provides us the hindlimb volume value and PV%. Three 2D frames were selected for each mouse and used for echodensity evaluation. In particular, the images were analyzed by using ImageJ® software by creating a gray-scale analysis histogram on the entire outlined hind limb section of constant dimensions of 7392.12 ± 18.45 pixel. For each mouse, hindlimb echodensity was obtained as the main value obtained from 3 images analysis. Echodensity differences were expressed as the percentage change of the mean echodensity of the pixels included in the selected area.

Animal Sacrifice and Tissue Collection
At the end of the in vivo evaluations, the ex vivo experimental phase started. No more than 2 animals were sacrificed per week, thus a window of 4-5 weeks was necessary to complete these procedures. Animal sacrifice was made by cervical dislocation under ZOLETIL 50/50 (40 mg/kg i.p.) profound anesthesia (Maqoud et al., 2020). Skeletal muscles and hearts were extracted under profound anesthesia. Blot-dried organs were weighed and weights were normalized to the tibia length. All experimental surgical procedures were performed under a sterile cell culture hood in which all the necessary equipment was sterilized to prevent contamination.

Drugs and Solutions
The normal Ringer solution used during muscles and organ biopsy contained:145 mM NaCl, 5 mM KCl, 1 mM MgCl2, 0.5 mM CaCl2, 5 mM glucose and 10 mM 3-(N-morpholino) propanesulfonate (Mops) sodium salt and was adjusted to pH 7.2 with Mops acid. For inside-out patch experiments on FDB/SOL fibers, the patch-pipette solution contained: 150 mM KCl, 2 mM CaCl2 and 1 mM Mops (pH 7.2); the bath solution contained 150 mM KCl, 5 mM EGTA and 10 mM Mops (pH 7.2). Stock solutions of glibenclamide (Glib) (118.6 mM) were prepared by dissolving the drug in DMSO . DMSO applied at the maximal concentration tested, which was 0.05%, did not affect the channel currents in the absence or the presence of ATP (solvent control).

Patch-Clamp Experiments on Flexor digitorum brevis and Soleus Fibers
Experiments on FDB/SOL fibers were performed in inside-out configurations by using the standard patch-clamp technique (Tricarico et al., 2008a). Isolated fibers were obtained from FDB and SOL muscles by enzymatic digestion with collagenase (C9697 Sigma, ≈0.5 mg/ml). Channel currents were recorded in excised macro-patches (R 0.91 ± 0.07 MΩ) during voltage steps from a holding potential of 0 mV to −60 mV (Vm) immediately after excision, at 20-22°C. Currents were recorded at a 1-kHz sampling rate (filter 0.2 kHz) by using an Axopatch-1D amplifier equipped with a CV-4 headstage (Molecular Devices, CA) (Scala et al., 2019). The current amplitude was measured using Clampfit 10.0. Patch pipettes were pulled from borosilicate glass capillaries (Glass type 8250, King, USA) and fire-polished. Macro-patches containing significant voltage-dependent K+ channels or other Kir channels or showing marked loss of channel currents during the time of observation were excluded from the analysis. No correction for liquid junction potential was made, estimated to be <1.9 mV in our experimental conditions.

Fiber Survival Evaluation
Evaluation of the morphological parameters of FDB fibers was performed seeding fibers in the culture medium (DMEM supplemented with 10% fetal bovine serum, 1% L-glutamine and 1% penicillin-streptomycin), at 37°C , by an investigator blinded to mouse genotype. Before analysis, isolated fibers were equilibrated in the culture medium for at least 30 min at 37°C. Fiber morphology was evaluated using a Nikon TMS Inverted Microscope 4x magnification. Dead fibers were defined as cells showing marked changes of ≥40% in morphological parameters such as length and diameter within 24 h from the excision. The appearance of multiple sarcolemma blebs preceded cellular death.

Histopathological Analysis
After animal sacrifice, tissue samples were embedded in OCT tissue-freezing medium, frozen in liquid nitrogen, and stored at −80°C until analysis. Samples were fixed with 10% animal buffered formalin for a minimum period of 48 h. Tissues were incorporated into the paraffin, sections were cut to 5-10 μm and stained with standard techniques with hematoxylin and eosin (H.E.), Mallory trichrome stain, and periodic acid-Schiff (PAS). Other specimens stored at −80 were used for ATPase and Succinodehydrogenase stains in cryo-section. Muscles were cross-sectioned using a microtome. Digital images were taken from the cross-section at 10-100x magnification to evaluate muscle fiber morphology and to determine fiber cross-sectional area (CSA) measures. Images from 20 random fields were acquired for the ten stained sections of each specimen using a D 4000 Leica DMLS microscope equipped with a camera and image analyzer NIS elementes-BR-Nikon. Section analysis and the CSA evaluation of the fibers was performed by QWin software (Leica). Of the subjects sacrificed, a gross necropsy of all organs was also carried out (Trapani et al., 2017). Examination of cellular morphology as well as of intracellular structures was conducted and the severity of the observed lesions was assessed (Gibson-Corley et al., 2013), by animal pathologists blinded to mouse genotype.

Polymerase Chain Reaction
Total RNA was isolated and purified from entire FDB and SOL muscles with Trizol reagent (Invitrogen Life Technologies) and quantified using a spectrophotometer (ND-1000 Nano-Drop, Thermo Scientific). PCR amplification was achieved using PCR Master Mix (Promega) (Dinardo et al., 2012). PCR cycles consisted of denaturation at 95°C for 1 min, annealing segment at 58°C for 1 min, and extension at 72°C for 1 min, repeated for 30 cycles. Amplified PCR products were separated on 1% agarose gel. Primer sequences are reported in Table 1. The data collection and analysis were performed according to the MIQE (Bustin et al., 2009). All experiments were performed in duplicates per muscle and duplicates per genotypes.

Data Analysis and Statistics
Data were collected and analyzed using Excel software (Microsoft Office 2010), Clampfit 10.5 (Molecular Devices), and SigmaPlot 10.0, and statistical results are presented as mean ± SEM unless otherwise indicated. The number of replicates relative to each experimental dataset was reported in the results paragraph and the figure description.
The primary endpoint was the change in the current amplitude recorded in excised patch experiments. The comparisons between groups were performed by using the t-test, two tails between two independent means to evaluate variance between groups. The p-value was considered statistically significant if <0.05. The number of experimental groups: 2, WT, and Kir6.1 wt/VM mice. The calculation of the sample size was made considering a delta change of the KATP current amplitude 412 pA and SD of 100 in patch-clamp experiments performed in previous experiments on smooth muscle cells in the transgenic mice Kir6.1 wt/VM mice vs WT mice since no data on skeletal muscle cells are so far available for the calculation (Huang et al., 2018). The calculated sample size per genotype was 4, the critical t value of 2.44 with an effect size of 4. The theoretical power of the study for the calculation of the sample was 0.95 in input and remains high in output 0.99 (G * Power 3.1.9.7).
The Student t-test was also used to evaluate the significance of differences between the means of two groups for secondary endpoints, p values < 0.05 were considered to indicate statistical significance unless otherwise indicated. One WAY ANOVA was used to evaluate intergroup and intragroup Frontiers in Pharmacology | www.frontiersin.org November 2020 | Volume 11 | Article 604885 variability (p < 0.05). The percentage of KATP current inhibition induced by glibenclamide (Glib) was calculated as -(I CTRL-I drug)/(I CTRL-I leak) x 100, where I leak was the current recorded after the application of 5 × 10 −3 M MgATP on skeletal muscle fibers. The frequency of myofibers positivity was calculated on 50 cells per section (Zizzo et al., 2019).

Cantù Kir6.1 wt/VM Mice Show Enlarged, but Weaker Skeletal Muscles and Higher Echodensity
We first assessed the macroscopic properties of skeletal muscle in terms of strength and integrity. At the beginning of the experimental protocol, 33-week-old Kir6.1 wt/VM and agematched WT mice were not significantly different in weight (30.6 ± 1.7 g n 4 Kir6.1 wt/VM mice vs 29.5 ± 1.3 g n 4 WT mice, p 0.321), but forelimb grip strength (both absolute and normalized to body weight) was significantly lower in Kir6.1 wt/VM mice than in WT mice ( Figures 1A,B).
Hind limb muscle morphology was assessed by 3D ultrasound evaluation. Interestingly, hind limb echodensity was ∼20% higher in Kir6.1 wt/VM than in WT ( Figures 1C,D), reflecting lower muscle mass and fibrotic tissue/fat deposition. Although not significant, there was a trend towards increased hind limb volume and percentage of vascularization (PV%) in Kir6.1 wt/VM mice (Figures 2A,B).
Following sacrifice, individual muscles were isolated and weighed. As previously reported (Huang et al., 2018) there was dramatic cardiomegaly in Kir6.1 wt/VM mice, which showed a ∼1.7-fold increase in heart weight as evaluated by one WAY ANOVA (Figure 3). There was also a tendency toward increased skeletal muscle weights in Kir6.1 wt/VM mice, including Soleus (SOL), and significant increases in Extensor digitorum longus (EDL) and Tibialis anterior (TA) muscles as evaluated by one WAY ANOVA (Figure 3).

KATP Channels Show Reduced Sensitivity to Glibenclamide in Cantù Kir6.1 wt/VM Muscles
We confirmed the expression of KATP channel subunits in fasttwitch FDB fibers and slow-twitch SOL fibers by PCR analysis of KCNJ8, KCNJ11, ABCC8, and ABBC9 genes (encoding Kir6.1, Kir6.2, SUR1, and SUR2, respectively), of FDB (n. muscles 2) and SOL (n. muscles 2) whole muscle samples. Specific bands were detected for KCNJ11 and ABCC9 in both FDB and SOL ( Figure 4A), which is consistent with the prevailing evidence that skeletal muscle KATP channels are composed predominantly of Kir6.2/SUR2 subunits (Tricarico et al., 2006). However, a prominent band for KCNJ8 was also present in both fast-twitch and slow-twitch muscles, with low expression level in the FDB muscle of Kir6.1 wt/VM mice ( Figure 4A). KATP channel activity was recorded in macro-patches excised from acutely dissociated FDB fibers (n. muscles 4) and SOL fibers (n. muscles 4) using the patch-clamp technique. Mean current density (i.e. maximum current in zero ATP) was ∼1.5fold higher in Kir6.1 wt/VM than WT FDB patches but was not different between genotypes in SOL patches ( Figures 4B-D). However, we note that Kir6.1 wt/VM SOL fibers showed sudden changes in length and diameters and surface blebs appearance within 2 h after acute isolation, resulting in a low percentage of successful experiments, and potentially reflecting membrane damages and muscle pathology. We assessed the sensitivity of KATP to MgATP, which provides a compound measure of both inhibition by ATP itself and the activator effect of MgATP at the SUR subunit (Reimann et al., 2000;Proks et al., 2014;McClenaghan et al., 2018). There was no clear shift of MgATP-sensitivity of the KATP channels in SOL and FDB patches ( Figures 5A,B; Table 2). We tested the KATP channel sensitivity to the KATP channel inhibitor glibenclamide (Glib). In both FDB and SOL patches, there was a significant reduction of Glib-sensitivity, with ∼50% reduction in the inhibitory effect of 10 −7 M Glib in each muscle (Figures 5C-E; Table 2).
Together, these results indicate that Kir6.1 subunit is functionally present in KATP channels in both SOL and FDB muscles resulting in a marked loss of channel sensitivity to Glib in Kir6.1[V65M] mice.
Atrophic phenomena were particularly evident in type I fibers of Kir6.1 wt/VM mice, but less in type II fibers, as observed with high myosin-ATPase activity at acidic pH and high mitochondrial SDH activity evidenced in muscle cryo-sections of GA muscle (Figures 6E,F). CSA was only reduced by 13.7% in fast-twitch GA muscle ( Figure 6G); TA and EDL muscles were also less affected than SOL muscle.
Besides, following isolation, ex-vivo survival over 24 h was worse for Kir6.1 wt/VM FDB fibers than for WT FDB fibers ( Figures 9A,B). Although long-term incubation for 24 h in the presence of 10 −7 M Glib significantly reduced fiber viability in both phenotypes ( Figure 9C), Glib only caused a marked short- term loss of viability of WT fibers; consistently with the lack of inhibitory effect of Glib on Kir6.1 wt/VM mutant channels, no effect of Glib was observed on Kir6.1 wt/VM fiber survival ( Figure 9D).

KATP Channels and Skeletal Muscle
KATP channels are present in all muscle types and are abundant in skeletal muscle-sarcolemma (Tricario and Conte Camerino, 1994).
Although gene knockout studies have implicated Kir6.2 and SUR2A as primary constituent subunits in most muscles (Spruce et al., 1985;Tricarico et al., 2006;Tricarico et al., 2016a), expression of all four KATP channel subunits has been demonstrated at the transcript level; and prior electrophysiological and pharmacological analyses provide evidence for expression of channels with single-channel conductance and drug sensitivities consistent with Kir6.1 and SUR1 expression contributing to channels with variable properties in different muscle types to overcame different muscle function. A straightforward hypothesis is that muscle KATP channels should    (Cifelli et al., 2007), indicating that fatigue, and ultimately necrosis (Thabet et al., 2005), resulting from [Ca]-overload and energy depletion, as a consequence of the failure of KATP to limit excitability (Scott et al., 2016), is the dominant outcome. Loss-of-function mutations in Kir6.2 and SUR1 are associated with congenital hyperinsulinism (Nestorowicz et al., 1996;Thomas et al., 1996;Dunne et al., 1997;Nestorowicz et al., 1997). Interestingly, there does not appear to be any obvious skeletal muscle defect, although Albaqumi et al. (2014) described a family with congenital hyperinsulinism and rhabdomyolysis associated with Kir6.2 mutations (Albaqumi et al., 2014), andFlanagan et al. (2017) recently reported a patient with persistent hyperinsulinaemic hypoglycaemia and severe hypotonia resulting from an activating mutation in a calcium channel subunit, which etiologically may have the same basis (Flanagan et al., 2017). Recently, we described a novel ABCC9-related Intellectual disability Myopathy Syndrome (AIMS) resulting from loss-offunction mutations in ABCC9 (SUR2) (Smeland et al., 2019), in which patients also exhibit muscular pain and fatigue, and evidence of muscle fiber damage.
Conversely, increased KATP channel activity would be predicted to limit action potential generation, resulting in 'electrical fatigue' before energetic fatigue. Gain-of-function mutations in Kir6.2 and SUR1 are associated with developmental delay, epilepsy, and neonatal diabetes (DEND) syndrome, which is accompanied by muscle flaccidity and motor impairment. Transgenic mice with tissue-specific expression of Kir6.2 GOF in skeletal muscle have not revealed any obvious muscle phenotype (Clark et al., 2010), but there are no knock-in models of DEND with the targeting of the endogenous locus. Gainof-function mutations in the other canonical KATP channel gene pair, Kir6.1,and SUR2, are associated with Cantù Syndrome (CS), with a distinct, unique, set of features. Among them, CS patients tend to have a muscular appearance, accompanied by hyperextensible joints and self-reported fatigue-ability.
In the present work, we studied the effects of the Kir6.1[V65M] mutation in slow-twitch and fast-twitch muscles of a new murine model of CS, in which patient-specific disease mutation is introduced to the identical locus in the mouse genome (Huang et al., 2018). In vivo experiments showed that Kir6.1 wt/VM mice generate significantly lower forelimb forces than WT mice. This is consistent with prior studies showing that abnormal activation of skeletal muscle KATP channels is associated with early fatigue during tetanus and that in vivo treatment of mice with KATP openers causes fatigue (Cifelli et al., 2008;Tricarico et al., 2016b), and also it resembles the fatigability and exercises intolerance observed in CS patients. Despite this weakness, isolated Kir6.1 wt/VM muscles tended to be heavier and, also, ultrasonography has revealed a significant increase of echodensity in Kir6.1 wt/VM hind limb muscles, associated with a tendency toward increased hind limb volume and vascularization. Additionally, histological analysis has revealed marked myofiber atrophy (see below). Taken together, these data suggest the presence of damage in skeletal muscle integrity along with fibrous tissue deposition (Pillen et al., 2009), explaining the increase in echodensity, the decrease of muscle strength, and the pseudo-hypertrophy.

The Cellular and Pharmacological Consequence in Cantù syndrome Muscle
We, therefore, found an enhancement of the KATP current density in Kir6.1 wt/VM FDB fibers concerning WT and no significant enhancement of this parameter in SOL fibers of Kir6.1 wt/VM . Patch clamp data allow statement that the Glib data indicates that mutant channels are present in the muscle.
The relative expression of different KATP channel subunits varies between muscles, and may be altered in different disease states (Tricarico et al., 2006;Tricarico et al., 2008b), though as yet unknown mechanisms. We do not have evidence of an increase of the Kir6.1 or Kir6.2 gene expression in the transgenic mice muscles, but our PCR analysis revealed expression of Kir6.1 subunit in WT SOL and FDB muscle but, unexpectedly, lower levels of Kir6.1 in Kir6.1 wt/VM FDB than in WT, despite higher KATP currents in the former. Reduction in mRNA expression, and of protein expression, has been observed for mutations linked to Kir6.2 GOF in neonatal diabetes (Lin et al., 2006;Lin et al., 2013), and we speculate that down-regulation of the overactivated Kir6.1 subunit in fast-twitch FDB muscle fibers could be a protective mechanism which preserves the function and the morphology of these tissues, with reduction of current density acting as a compensatory effect for the nucleotide insensitivity caused by mutation. Interestingly, there was no down-regulation of Kir6.1 transcript in Kir6.1 wt/VM SOL muscle, which additionally manifested worse atrophy, the low survival rate of the fibers, and lack of response to glibenclamide.
We however failed to show a significant rightward shift of MgATP-sensitivity in FDB and SOL patches. Based on experiments with recombinant channels, the loss of nucleotide inhibitory sensitivity is expected to be relatively subtle (Cooper et al., 2017) and can be masked in our experiment in native tissues due to the high expression levels of Kir6.2 and SUR2 subunits. The modest reduction of Glib-sensitivity observed in both Kir6.1 wt/VM FDB and SOL patches, as is reported for vascular smooth muscle cells in these animals (Huang et al., 2018), supports the presence of the Kir6.1 subunit in the muscle. It has been reported that the reduction in ATP sensitivity of Kir6.2/Kir6.1 (V65M) channels is ∼5 fold but the reduction of Glib sensitivity is ∼1000 fold heterologous expression system (Cooper et al., 2017). The 10 fold change in Glib sensitivity in muscle suggests is significantly less than that shown in heterologous expression. This also suggests that the reduction in ATP sensitivity of the KATP channel of the skeletal muscle is also drastically less than that the 5 fold change shown in heterologous expression (Cooper et al., 2017), and questions the pathological significance of its presence in the skeletal muscle. Additional experiments at single-channel levels are needed to clarify these issues.
The functional presence of a KATP channel current component uncoupled to the nucleotide metabolism as is expected for the Kir6.1 wt/VM mice may have deleterious effects specifically in the slow-twitch oxidative SOL fibers rather than in FDB or other fast-twitch muscles.
In WT FDB fibers, Glib induced cytotoxic effects within the first few hours of incubation but was without effect in Kir6.1wt/ VM cells, consistent with loss of sensitivity to the drug. After long-term (24 h) incubation Glib had similar toxicity in both genotypes. Previous studies have demonstrated that ex vivo longterm exposure to KATP channel inhibitors is coupled to apoptosis and atrophic signaling in muscle fibers (Tricarico et al., 2010;Mele et al., 2014a;Mele et al., 2014b;Cetrone et al., 2014), although this may be due to non-KATP channeldependent actions of the drug (de Sant'Anna et al., 2015;Subramaniyam et al., 2018). We have recently shown that Glib treatment can effectively reverse many of the cardiovascular complications of CS (Ma et al., 2019;McClenaghan et al., 2020). However, reduced Glib sensitivity in recombinant Kir6.1[V65M] channels, and the markedly reduced sensitivity that we show here, raises the potential that Glib treatment may be ineffective for treating muscle impairments in certain patients and that other therapeutic approaches may be needed.
Atrophy was more marked in Type I fibers than Type II, and there was a markedly lower survival rate in Kir6.1 wt/VM SOL vs. FDB fibers. This raises the possibility that SOL muscle, and slowtwitch fibers more generally, maybe more severely affected in CS than fast-twitch fibers, such as FDB. Markers of atrophy and autophagy, including Atrogin-1, MuRF1, and BNIP3 genes were all upregulated in Kir6.1wt/VM SOL, and expression was not detected in WT muscles (Foletta et al., 2011). Relative to WT, muscle diameter was more markedly reduced in Kir6.1 wt/VM SOL than fast-twitch GA muscle, further suggesting a major pathological involvement of slow-twitch fibers, a possibility that should be examined in future studies of CS patients.
An additional hypothesis can be drawn explaining at least in part the observed changes of the muscle function in the Kir6.1 wt/VM mice. We have shown that changes in heart size/function in the CS mouse model are secondary to changes in the KATP channel function in Vascular Smooth Muscle and these changes are reversed by glibenclamide (McClenaghan et al., 2020). The recorded changes in muscle function may be therefore secondary to skeletal muscle remodeling rather than an altered function of the KATP channel in the SOL muscle as observed in the cardiovascular apparatus in these mice. It should be of note, however, that we failed to evidence any significant change of % of vascularization of the hind limb muscles in the Kir6.1 wt/VM mice vs WT mice as evaluated by ultrasonography thereby not supporting this hypothesis. Inflammatory cells were also observed in CS muscle. The evaluation of the relative contribution of the direct and indirect effects of the Kir6.1 wt/VM mutation in skeletal muscle requires further experiments. This is the first report showing a direct genotype-phenotype correlation in CS skeletal muscle, revealing how Kir6.1 wt/VM mutation in the KATP channel Kir6.1 subunit is associated with a reduction of limb strength, skeletal muscle atrophy, autophagy, and connective tissue replacement of myofibers in this animal model of CS. It should of note that several factors can mediate the muscle-specific degeneration that we found in Kir6.1 wt/VM SOL muscle. For instance, lactate accumulation into the muscle following the overactive KATP channels may be involved in atrophy and degeneration in slow-twitch muscle.
In conclusion, these data suggest that Kir6.1 wt/VM mutation affects directly and/or indirectly skeletal muscle through vascular dysfunction and this could be a significant issue, particularly for slow-twitch muscle, in CS patients suffering from this mutation and warrants future investigation.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation, to any qualified researcher.

ETHICS STATEMENT
The animal study was reviewed and approved by Animal care and all experimental protocols are in accordance with the European Directive 2010/63/EU on Animal Protection Used for Scientific Experiments, and the Washington University School of Medicine Institutional Animal Care and use Committee, and were approved by the Italian Ministry of Health and by the Committee of the University of Bari O.P.B.A (Organization for Animal Health) (prot. 8515-X/10, 30-01-2019).

AUTHOR CONTRIBUTIONS
CM and TH was responsible for animal husbandry, DT and CN wrote the manuscript, RS contributes significantly to data presentation and analysis. GC and RS performed PR experiments. RS and FM performed patch-clamp and in vitro experiments on native fibers, and organs and tissues isolation. AM was responsible for ultrasonography of hind-limb muscle evaluation. NZ was responsible for mice health. NZ and GR performed histochemistry and immunohistochemistry experiments. FZ was responsible for immunohistochemistry data. All authors critically reviewed the manuscript.

FUNDING
This work was supported by NIH R35 HL140024 (to CN). CM was supported by American Heart Association Postdoctoral Fellowship 19POST34380407. This research was also supported by M.I.U.R. PhD program to RS (tutor DT) and in part funded by Regione Puglia (Italy) project "Cluster in Bioimaging" code QZYCUM0, through FSC 2017-2013/ Programma regionale a sostegno della specializzazione intelligente e della sostenibilita ambientale. Intervento "Cluster Tecnologici Regionali (DT).