IgGs from patients with amyotrophic lateral sclerosis and diabetes target CaVα2δ1 subunits impairing islet cell function and survival

Significance We provide evidence of a mechanistic link between ALS and T2DM. Our data show that a subgroup of ALS-T2DM patients have sera that enhance CaV1 channel-mediated Ca2+ influx and exaggerate [Ca2+]i. These effects occur because the sera accommodate cytotoxic IgG autoantibodies that immunocapture CaVα2δ1 subunits. As a consequence, impairments in [Ca2+]i dynamics, mitochondrial function, insulin secretion, and cell viability appear. We could clarify not only the identity of this serum factor but also the molecular mechanisms underlying its effects on the islet cells. Our findings may lay the foundation for a treatment strategy for this complex and severe group of diabetic patients.

Patients with amyotrophic lateral sclerosis (ALS) often show hallmarks of type 2 diabetes mellitus (T2DM). However, the causal link between ALS and T2DM has remained a mystery. We now demonstrate that 60% of ALS patients with T2DM (ALS-T2DM) have sera that exaggerated K + -induced increases in cytosolic free Ca 2+ concentration ([Ca 2+ ] i ) in mouse islet cells. The effect was attributed to the presence of pathogenic immunoglobulin Gs (IgGs) in ALS-T2DM sera. The pathogenic IgGs immunocaptured the voltagedependent Ca 2+ (Ca V ) channel subunit Ca V α 2 δ1 in the plasma membrane enhancing Ca V 1 channel-mediated Ca 2+ influx and [Ca 2+ ] i , resulting in impaired mitochondrial function. Consequently, impairments in [Ca 2+ ] i dynamics, insulin secretion, and cell viability occurred. These data reveal that patients with ALS-T2DM carry cytotoxic ALS-T2DM-IgG autoantibodies that serve as a causal link between ALS and T2DM by immunoattacking Ca V α 2 δ1 subunits. Our findings may lay the foundation for a pharmacological treatment strategy for patients suffering from a combination of these diseases. amyotrophic lateral sclerosis | calcium channel | cytosolic free Ca 2+ concentration | diabetes | immunoglobulin P atients with amyotrophic lateral sclerosis (ALS) show progressive dysfunction and degeneration of motor neurons in the brainstem and spinal cord (1). There are no effective therapeutics available for ALS (1). Clinically, ALS patients manifest advanced muscular weakness and paralysis and die within 2 to 5 y from the onset of the disease (1). Although the exact pathogenic mechanisms of ALS are not clarified, dysregulation of voltagedependent Ca 2+ (Ca V ) channels, cytosolic free Ca 2+ concentration ([Ca 2+ ] i ), and synaptic plasticity induced by altered humoral immunity has been proposed to participate in the development of the disease (2,3). Exposure to purified immunoglobulin G (IgG) from the serum of ALS patients produced ultrastructural abnormalities with Ca 2+ accumulation in and increased transmitter release from rodent motor neurons (4). It has long been recognized that intracellular accumulation of Ca 2+ is cytotoxic and causes mitochondrial dysfunction, free radical damage, and Ca 2+ -dependent cell death (5)(6)(7).
A series of neurodegenerative diseases, like ALS, is known to be associated with hallmarks of type 2 diabetes mellitus (T2DM), such as impaired glucose homeostasis, but their causal links are not known (8)(9)(10). Blood glucose homeostasis is under strict control of hormone release from pancreatic islet cells. T2DM and its characteristic symptom hyperglycemia occur in people with inadequate islet cell mass and function. Islet cells and neurons share a series of physiological and pathological mechanisms, such as Ca 2+dependent exocytosis and Ca 2+ -triggered cell death, for their function/dysfunction and survival/death (11)(12)(13)(14). Therefore, the present study hypothesized that IgG from ALS patients with T2DM (ALS-T2DM) may recognize similar targets in islet cells, as revealed in motor neurons, and thereby impair islet cell function and viability by disturbing Ca 2+ signaling. Indeed, the present study reveals that a subgroup of ALS-T2DM patients has sera that enhance K + -induced [Ca 2+ ] i responses in islet cells via cytotoxic IgGs. Moreover, it demonstrates that ALS-T2DM-IgGs immunocapture Ca V α 2 δ1 subunits and thereby enhance Ca V 1 channel-mediated Ca 2+ influx, resulting in altered [Ca 2+ ] i dynamics and, consequently, impaired mitochondrial function, insulin secretion, and cell viability.

Significance
We provide evidence of a mechanistic link between ALS and T2DM. Our data show that a subgroup of ALS-T2DM patients have sera that enhance Ca V 1 channel-mediated Ca 2+ influx and exaggerate [Ca 2+ ] i . These effects occur because the sera accommodate cytotoxic IgG autoantibodies that immunocapture Ca V α 2 δ1 subunits. As a consequence, impairments in [Ca 2+ ] i dynamics, mitochondrial function, insulin secretion, and cell viability appear. We could clarify not only the identity of this serum factor but also the molecular mechanisms underlying its effects on the islet cells. Our findings may lay the foundation for a treatment strategy for this complex and severe group of diabetic patients.
in motor neurons, thereby damaging these cells in a Ca 2+ -dependent manner (2)(3)(4). This prompted us to explore if ALS-T2DM serum drives similar pathological events in mouse islet cells. To implement such an exploration, we collected 4 types of sera from healthy human subjects (HSs) and patients with ALS,  T2DM, and ALS-T2DM (SI Appendix, Table S1). During the course of the present study, 2 separate batches of sera were collected. The first batch of sera was obtained from 12 patients with ALS-T2DM as well as 12 HSs, 8 patients with ALS, and 8 patients with T2DM as controls (SI Appendix, Table S1). With each individual serum we treated dissociated islet cells and conducted high-throughput measurements of K + depolarizationinduced [Ca 2+ ] i responses. General analysis of pooled data revealed that average [Ca 2+ ] i response to stimulation with 25 mM KCl in the ALS-T2DM serum-treated group was significantly greater than that in groups subjected to treatment with HS, ALS, or T2DM sera (Fig. 1A). Furthermore, the 3 latter groups did not significantly differ in this parameter (Fig. 1A). Interestingly, thorough analysis of individual data showed that 7 out of 12 ALS-T2DM sera produced significant increases in [Ca 2+ ] i in response to 25 mM KCl, whereas the rest did not, in comparison to HS, ALS, and T2DM sera ( Fig. 1 B and C). To corroborate the results obtained with the first batch of sera, we repeated [Ca 2+ ] i measurements with the second batch of sera donated by 5 patients with ALS-T2DM and 6 T2MD patients (SI Appendix, Table S1 and Fig. 1 D and E). Consistent with the first batch of sera, the ALS-T2DM group displayed a significant increase in mean [Ca 2+ ] i response to KCl depolarization compared to the T2DM group, and 3 out of 5 ALS-T2DM sera gave rise to significant elevations in K + -induced [Ca 2+ ] i responses in comparison to T2DM sera ( Fig. 1 D and E). Moreover, the specific Ca V 1 channel blocker nifedipine almost completely ablated K +induced [Ca 2+ ] i responses in mouse islet cells (SI Appendix, Fig.  S1). In addition, there was no significant difference in basal [Ca 2+ ] i between T2DM and ALS-T2DM groups (Fura-2 F340/F380 ratio for the T2DM group of 0.537 ± 0.028 vs. Fura-2 F340/F380 ratio for the ALS-T2DM group of 0.541 ± 0.017, P > 0.05). Taken together, 60% of ALS-T2DM patients have sera that authentically exaggerate K + -induced [Ca 2+ ] i responses in islet cells. These ALS-T2DM sera were defined as positive ALS-T2DM sera and randomly chosen for subsequent experiments.
Pathogenic IgGs Present in Positive ALS-T2DM Sera Enhance K + -Induced [Ca 2+ ] i Responses in Mouse Islet Cells. It has been demonstrated that pathogenic IgGs reside in sera of ALS patients and account for a great deal of Ca 2+ -dependent destruction of motor neurons and skeletal muscle cells (4,(15)(16)(17)(18). This raised the question of whether IgGs in positive ALS-T2DM sera (ALS-T2DM-IgGs) also serve as molecular pathogenic factors to impair islet cell function and survival by perturbing [Ca 2+ ] i homeostasis. To tackle this question, we purified IgGs from positive ALS-T2DM sera and T2DM sera. Thereafter, we measured the effects of these purified IgGs on K + -induced [Ca 2+ ] i responses in islet cells. Incubation with these individual ALS-T2DM-IgGs induced significantly stronger [Ca 2+ ] i responses to 25 mM KCl in comparison to exposure to IgGs from T2DM sera (T2DM-IgGs) in mouse islet cells ( Fig. 2 A and B). Furthermore, the effect of ALS-T2DM-IgGs was lost when boiled (Fig. 2C). Cells exposed to either boiled ALS-T2DM-IgGs or T2DM-IgGs responded similarly to KCl stimulation with regard to increases in [Ca 2+ ] i (Fig. 2C). These data demonstrate that ALS-T2DM-IgG in ALS-T2DM sera enhances K + -induced [Ca 2+ ] i responses.
To clarify if ALS-T2DM serum affects β cell Ca V channels, we examined the effect of positive ALS-T2DM serum on β cell Ca V channel currents. Indeed, whole-cell patch-clamp analysis revealed that treatment with individual positive ALS-T2DM sera obtained in the first and second batches significantly elevated whole-cell Ca V channel currents in mouse β cells, as manifested by representative whole-cell Ca 2+ current traces and average Ca V channel current density, in comparison to T2DM serum exposure ( Fig. 3 A-C). These data verify that ALS-T2DM serum up-regulates Ca V 1 channels, causing pathologically exaggerated [Ca 2+ ] i responses.
The up-regulation of Ca V 1 channels by ALS-T2DM serum raises the possibility that IgGs in ALS-T2DM serum may target Ca V channel subunits in β cells. We chose the most important pore-forming subunit Ca V 1.2 as a starting point. Immunoprecipitation assays followed by immunoblot analysis showed that antibodies against Ca V 1.2 subunits efficiently pulled down Ca V 1.2 The Ca V α 2 δ1 subunit, an important constituent of Ca V channel complexes, including the β cell Ca V 1.2 channel complex, is critical for the surface expression of functional Ca V channels (12,(25)(26)(27). Moreover, the entire Ca V α 2 δ1 is exposed extracellularly (28). Among all β cell Ca V 1.2 channel components, they have the highest likelihood of serving as targets for factors, such as ALS-T2DM-IgGs, of positive ALS-T2DM sera. In addition, polyclonal anti-Ca V α 2 δ1 antibodies selectively recognize extracellular Ca V α 2 δ1 subunits associated with the plasma membrane of living cells bathed in a physiological solution (28,29). This prompted us to clarify if ALS-T2DM-IgGs interacts with Ca V α 2 δ1 subunits under physiological conditions by using anti-Ca V α 2 δ1 antibodies. We carried out 4-(2-[6-(Dioctylamino)-2-naphthalenyl]ethenyl)-1-(3-sulfopropyl)pyridinium inner salt (di-8-ANEPPS) labeling of the plasma membrane and immunofluorescence staining of Ca V α 2 δ1 subunits in intact living mouse islet cells and tsA-201 cells stably expressing Ca V α 2 δ1 subunits coincubated with antibodies against Ca V α 2 δ1 subunits and ALS-T2DM-IgGs. Ca V α 2 δ1-specific immunofluorescence was clearly localized in the di-8-ANEPPS-labeled plasma membrane (Fig. 3 D-R and SI  Appendix, Fig. S3). Interestingly, ALS-T2DM-IgGs effectively competed with the anti-Ca V α 2 δ1 antibodies for the extracellular Ca V α 2 δ1 subunits, resulting in a significant reduction in the immunofluorescence intensity of the anti-Ca V α 2 δ1 antibodies in comparison to HS-IgGs or T2DM-IgGs (Fig. 3 S and T). This verifies that ALS-T2DM-IgGs are capable of directly interacting with Ca V α 2 δ1 subunits in living cells in the absence of interferences from detergents, high ionic strengths, and substantial rinsing, which are unavoidable in immunoprecipitation and immunoblot analyses of association between ALS-T2DM-IgGs and Ca V 1.2 subunits.
Positive ALS-T2DM Sera Interfere with Mitochondrial Function in Mouse Islet Cells. Translation of an excessive elevation of [Ca 2+ ] i into mitochondrial Ca 2+ overload results in mitochondrial membrane depolarization, concomitant mitochondrial dysfunction, and eventual apoptosis, thus playing an important role in driving Ca 2+dependent cell death (30). This made us wonder whether such a mitochondrial mechanism is able to convert the ALS-T2DM serum-induced exaggeration of [Ca 2+ ] i to mitochondrial dysfunction in mouse islet cells. Therefore, we measured mitochondrial membrane potential in mouse islet cells using rhodamine 123. As shown in Fig. 4 A-C, ALS-T2DM serum-treated cells displayed a significant decrease not only in basal fluorescence intensity of rhodamine 123 (Fig. 4 A and B) but also in the glucose-induced quenching of rhodamine 123 (Fig. 4 A and C), compared to those cells incubated with T2DM serum. The effects are attributed to reduced amounts of rhodamine 123 loaded into mitochondria due to less negative mitochondrial membrane potential, i.e., mitochondrial dysfunction, induced by ALS-T2DM serum treatment (31). Our results demonstrate that mouse islet cells insulted by exaggerated [Ca 2+ ] i resulting from exposure to positive ALS-T2DM sera undergo mitochondrial dysfunction and suggest that ALS-T2DM serum-induced mitochondrial dysfunction is most likely to drive islet cell death.
Positive ALS-T2DM Sera Impair [Ca 2+ ] i Dynamics and Insulin Secretion in Mouse Islets. Normal glucose homeostasis critically relies on adequately functioning β cells (32). The function of β cells is under the control of exquisitely fine-tuned [Ca 2+ ] i dynamics that serves as fingerprints for β cell well-being (12,13,24,33,34). This made us wonder if ALS-T2DM serum drives disorganized [Ca 2+ ] i dynamics and impaired insulin secretion in islets, thereby accounting for aberrant glucose homeostasis often observed in ALS patients.
We characterized [Ca 2+ ] i dynamics in β cells situated within intact islets during glucose stimulation. As shown in photomicrographs of Fura-2-loaded mouse islets, ALS-T2DM serum treatment made islets become irregular and disintegrated (Fig. 5 A,  Lower). In striking contrast, incubation with T2DM serum did not alter the morphology of islets that were intact with spherical shapes and smooth boundaries (Fig. 5 A, Upper). Indeed, ALS-T2DM serum-treated islets showed chaotic [Ca 2+ ] i dynamics manifested as a relatively steady increase in [Ca 2+ ] i with tiny amplitude oscillations in response to 11.1 mM glucose (Fig. 5 B, Lower and SI Appendix, Fig. S4, Lower). However, islets exposed to T2DM serum displayed a normal [Ca 2+ ] i profile, characterized by fast oscillations superimposed on slow oscillations, following stimulation with 11.1 mM glucose (Fig. 5 B, Upper and SI Appendix, Fig. S4, Upper). These results reveal that ALS-T2DM serum does indeed potently derange [Ca 2+ ] i handling in β cells.
The primary function of β cells is glucose-stimulated insulin secretion that crucially depends on Ca V channel-mediated Ca 2+ influx and complex [Ca 2+ ] i dynamics (12, 13, 24, 33-35).  2+ ] i dynamics should cause impaired glucose-stimulated insulin secretion. T2DM serum-and ALS-T2DM serum-treated islets released a similar amount of insulin following incubation with 11.1 mM glucose (Fig. 5 C and D). However, insulin secreted from T2DM serumtreated islets at 11.1 mM glucose was significantly greater than that at 3.3 mM glucose, whereas insulin released from ALS-T2DM serum-treated islets at 11.1 mM glucose did not significantly differ from that at 3.3 mM glucose due to increased basal insulin release ( Fig. 5 C and D). In addition, the insulin content of ALS-T2DM serum-treated islets was significantly lower than that of islets exposed to T2DM serum (Fig. 5E). These data suggest that exposure to ALS-T2DM serum interferes with the ability of the β cell to maintain adequate insulin release.

ALS-T2DM serum-induced defects in [Ca
Positive ALS-T2DM Sera Reduce Mouse Islet Cell Viability in an IgGand Ca V 1 Channel-Dependent Manner. The exaggerated Ca V 1 channel-mediated Ca 2+ influx, increased [Ca 2+ ] i , and disturbed [Ca 2+ ] i dynamics in islet cells exposed to ALS-T2DM serum might explain the destructive action of this serum on islet integrity and islet insulin content. Therefore, we examined the possible effects of ALS-T2DM serum on islet cell survival. WST-1 assay showed that ALS-T2DM serum exposure significantly decreased islet cell viability, as reflected by significantly reduced WST-1 absorbance, in comparison to treatment with T2DM serum (Fig. 6A). Furthermore, cell death imaging with SYTOX Orange nucleic acid stain revealed that SYTOX Orange-positive profiles, representing dead nuclei, were significantly greater in dissociated islet cells incubated with ALS-T2DM serum compared to T2DM serumtreated ones (Fig. 6 B and C). Furthermore, the selective Ca V 1 channel blocker nifedipine fully ablated ALS-T2DM seruminduced reduction of islet cell viability (Fig. 6D). In addition, treatment with ALS-T2DM-IgGs significantly reduced mouse islet cell viability in comparison to incubation with T2DM-IgGs. The effects of ALS-T2DM-IgG on islet cell viability were effectively ablated by boiling (Fig. 6E). The results demonstrate that positive ALS-T2DM sera interfere with islet cell survival in an IgG-and Ca V 1 channel-dependent manner. Taken together, our results suggest that ALS-T2DM serum treatment destroys islet cells by excessively increasing Ca V 1 channel-mediated Ca 2+ influx and [Ca 2+ ] i and then pathologically translating the exaggerated [Ca 2+ ] i to eventual mitochondrial dysfunction.

Discussion
We have identified a subgroup of ALS-T2DM patients who have positive sera that exaggerate [Ca 2+ ] i in pancreatic islet cells upon depolarization. This suggests that these positive sera are likely to interfere with Ca V channels via serum molecular constituent(s). Indeed, we demonstrate that pathogenic IgG is accommodated in the sera of this subgroup of ALS-T2DM patients. This not only establishes a mechanistic link between ALS and T2DM but also suggests a potential role of altered humoral immunity in the development of ALS-associated T2DM.
Importantly, we reveal that ALS-T2DM serum significantly increases whole-cell Ca 2+ currents predominantly passing through Ca V 1.2 channels in mouse β cells (24). This mechanistically explains how ALS-T2DM serum and ALS-T2DM-IgG promote K +evoked [Ca 2+ ] i responses and pinpoints that Ca V 1.2 channels most likely serve as downstream targets of ALS-T2DM serum and ALS-T2DM-IgG. This finding is intriguing since ALS-T2DM serum is verified to functionally interfere with mouse β cell Ca V 1.2 channels, which almost exclusively mediate the nifedipinesensitive Ca 2+ currents (11-13, 36, 37).
Interestingly, we found that ALS-T2DM-IgGs are strong enough to compete with an IgG rabbit polyclonal antibody specific to the extracellular epitope of Ca V α 2 δ1 subunits in living islet cells and in tsA-201 cells stably expressing Ca V α 2 δ1 subunits. This is in accordance with the fact that the Ca V α 2 δ1 subunits are entirely exposed to the extracellular space and thereby the most accessible to serum components among all β cell Ca V 1.2 channel subunits. In addition, the Ca V α 2 δ1 subunit serves as an indispensable building element of the β cell Ca V 1.2 channel complex to up-regulate the conductivity and surface expression of functional Ca V channels (12,(25)(26)(27). Based on our results, we propose that ALS-T2DM-IgGs serve as autoantibodies that immunocapture Ca V α 2 δ1 subunits in the plasma membrane, thereby enhancing Ca V 1 channelmediated Ca 2+ influx and [Ca 2+ ] i in islet cells. This process likely occurs through allosteric activation and/or gradual accumulation of Ca V 1 channels in the β cell plasma membrane since antibodies in some cases activate and accumulate rather than inhibit and neutralize their binding partners (38,39). This autoimmune mechanism is particularly interesting since ALS autoantibodies have been shown to target only Ca V 1.1, Ca V 2.1, and Ca V 2.2 channels prior to the present work (2,18,(20)(21)(22)(23). Now the immunocapture of β cell Ca V α 2 δ1 subunits by ALS-T2DM-IgGs and consequent up-regulation of β cell Ca V 1.2 channels come into the picture. Importantly, the present work reveals that humoral autoimmunity arises as the pathogenic machinery leading to this subset of diabetes. It is intriguing to explore if such a humoral mechanism not only operates in the pathogenesis of T2DM but also in that of type 1 diabetes in addition to a T cell-mediated autoimmune destruction of β cells (40). Moreover, our findings offer a causal link between ALS and T2DM and shed light on potential therapeutic targets for prevention and treatment of a subgroup of ALS-T2DM patients.
The mitochondrion is a vulnerable target downstream of excessive accumulation of [Ca 2+ ] i to mediate Ca 2+ -dependent impairments in cell function and viability (30,41). Consequently, islet cells exposed to positive ALS-T2DM sera not only display exaggerated [Ca 2+ ] i but also mitochondrial dysfunction. These findings give a strong rationale for the ALS-T2DM serum-induced islet cell dysfunction and death. We found that positive ALS-T2DM sera derange [Ca 2+ ] i dynamics, impair insulin secretion, and drive islet cell death in a Ca V 1 channel-and IgG-dependent manner. The occurrence of these pathological phenotypes is well accounted for by impaired mitochondrial function. Our data thus suggest that IgG autoantibodies in ALS-T2DM sera immunocapture Ca V α 2 δ1 subunits in the plasma membrane, resulting in a destructive exaggeration of [Ca 2+ ] i followed by its pathological translation into mitochondrial dysfunction, subsequent impairment of insulin secretion, eventual islet cell death, and diabetes. Of note, in vivo ALS-T2DM-IgGs may target peripheral insulin-sensitive tissues of patients, leading to insulin resistance (8).
The exact reason why only a fraction of ALS patients develop diabetes is unclear but is most likely due to a heterogeneous β cell sensitivity to the evoked [Ca 2+ ] i challenges. In addition, both ALS and T2DM represent an etiologically heterogeneous group of disorders where patients are likely to experience particular environmental challenges on top of their specific genetic  (Upper) or ALS-T2DM serum (Lower), followed by perifusion with 3.3 and then 11.1 mM glucose (G), respectively. (C and D) Insulin secretion from islets treated with T2DM or ALS-T2DM serum for 10 h followed by exposure to 3.3 or 11.1 mM G for 30 min. *P < 0.05 vs. the T2DM/3.3 mM G group (C). Net insulin secretion induced by 11 mM G in T2DM and ALS-T2DM groups. **P < 0.01 vs. the T2DM group (D). Experiments were done with 6 T2DM and 6 ALS-T2DM sera in triplicate. (E) Insulin content in islets exposed to T2DM or ALS-T2DM serum for 10 h. *P < 0.05 vs. the T2DM group. Experiments were done with 4 T2DM and 3 ALS-T2DM sera in triplicate.
predisposition. In this context, a specific combination of genetic and environmental factors may trigger the production of autoantibodies to the autoantigen Ca V α 2 δ1 subunit. This leads to a pathological Ca V channel conductivity with resulting damage to motor neurons and islet cells.
In conclusion, the present work demonstrates that a subgroup of ALS-T2DM patients have sera that enhance Ca V 1 channelmediated Ca 2+ influx, resulting in exaggerated [Ca 2+ ] i . These effects are attributed to the fact that the sera accommodate cytotoxic IgG autoantibodies that immunocapture Ca V α 2 δ1 subunits. As a consequence, impairments in [Ca 2+ ] i dynamics, mitochondrial function, insulin secretion, and cell viability occur. This suggests that cytotoxic ALS-T2DM-IgG autoantibodies serve as a causal link between ALS and T2DM by interacting with and modulating the Ca V α 2 δ1-Ca V 1 channel complex, which may lay the foundation for a pharmacological treatment strategy for patients suffering from a combination of these severe diseases.

Methods
Animals. Male C57BL/6J mice aged from 8 to 10 wk were purchased from The Jackson Laboratory (Bar Harbor, ME). All animal experiments were conducted according to the guidelines of the Ethics Committee at Pohang University of Science and Technology (2011-0001) and the Animal Experiment Ethics Committee at Karolinska Institutet (N183/13).

Enrollment of Patients.
Seventeen patients with ALS-T2DM (11 males and 6 females, age: 55.0 ± 2.1), 12 HSs, 9 patients with ALS, and 14 patients with T2DM (7 males and 7 females, age: 60.4 ± 2.5) were randomly selected (SI Appendix, Table S1). One ALS patient carrying a SOD1 mutation was excluded from the study (SI Appendix, Table S1). Ethical approval to use serum from patients was obtained from Hanyang University Hospital in Seoul, Korea (HYUH 2006-04-001-004). Blood samples were collected after obtaining written informed consent. Actual patients had no known family history of neuromuscular disease or wasting. All subjects diagnosed with T2DM were well controlled with hypoglycemic agents or insulin.
Additional experimental procedures are presented in the SI Appendix.
Data Availability. All of the data, associated protocols, code, and materials for this study are available within the paper and its SI Appendix.