Introduction: Dysferlin is a transmembrane protein with Ca2+ sensitive C2 domains, localized in the sarcolemma and in vesicle membranes of striated muscle cells. Upon Ca2+ entry, Dysferlin is supposed to mediate vesicle fusion and membrane repair events, while Dysferlin loss-of-function causes progressive forms of muscular dystrophies and dilated cardiomyopathy. However, the role of Dysferlin in protection of subcellular membrane domains in ventricular cardiomyocytes (VMs) after myocardial infarction (MI) is poorly understood. Here, we hypothesize that Dysferlin prevents VMs from cell death in the MI border zone by stabilizing the transverse-axial tubule (TAT) system and the intercalated disc (ICD) cell-cell contact sites.

Results: Using RNA sequencing, we showed that among six isoforms of the Ferlin family only Dysferlin is substantially expressed in VMs. STimulated Emission Depletion (STED) nanoscopy of immunolabelled VMs localized punctate Dysferlin signals in close proximity to junctions of the TAT system with the sarcoplasmic reticulum (SR) and to the ICDs. Complexome profiling with Blue Native PAGE prior to mass spectrometry (LC-MS/MS) and co-immunoprecipitation experiments identified a protein interaction of Dysferlin with the SR Ca2+ release channel Ryanodine Receptor type 2 (RyR2), as well as with the gap junction hemichannel Connexin-43 (Cx43). Importantly, after MI surgery, Dysferlin knockout (KO) mice presented an increased mortality compared to wildtype (WT) animals (34% KO vs. 15% WT, n = 32 vs. 39 mice), which showed an upregulated myocardial Dysferlin protein expression of 148% compared to sham-operated mice 1 week post-MI. In the MI border zone 1 and 4 weeks post-surgery, facing the highest ischemic and mechanical stress, local Dysferlin expression is increased to 230% and 190%, respectively. Here, immunohistology showed VMs with disrupted TAT networks and significantly extended ICD regions. On the subcellular level, STED nanoscopy revealed Dysferlin clusters that highly decorate residual TAT structures in the MI border zone; indicating a role for tubular membrane repair that may be key to preserve functional excitation-contraction coupling in VMs post-MI. Moreover, extended ICD regions in the MI border zone were associated with 227% increased Dysferlin signals in close proximity to Cx43 plaques. Interestingly, VMs from Dysferlin KO mice showed more internalized Cx43 protein in the MI border zone compared to WT mice. Hence, increased Dysferlin expression at ICDs may compensate for the detrimental mechanical stress of VMs next to the MI infarction scar, and protect from pathophysiological Cx43 internalisation. In contrast, hypertrophied VMs of the MI remote zone express Dysferlin clusters decorating newly shaped TAT structures, implying a role of Dysferlin for tubular membrane proliferation and remodelling in cell hypertrophy induced by left ventricular volume overload.

Conclusion: Our data demonstrate that Dysferlin plays an important role in stabilizing VMs of the MI border zone. Dysferlin not only prevents from loss of TAT structures, but also compensates for increased mechanical stress at the ICDs and may protect from Cx43 internalization. In addition, Dysferlin mediates TAT membrane proliferation in VM hypertrophy induced by volume overload. Hence, Dysferlin emerges as a novel therapeutic target to control membrane nanodomain stability in MI, and to regulate TAT network proliferation necessary for VM hypertrophy.