Transverse (t‑tubule) membranes enable rapid, synchronous Ca2+ release by bringing L‑type calcium channels into close proximity with ryanodine receptors, a critical arrangement for efficient cardiac contraction. In heart failure with reduced ejection fraction (HFrEF), loss and disorganisation of t‑tubules accompany remodelling of the cardiac microtubule cytoskeleton and are linked to impaired calcium handling and contractile dysfunction.
The canonical role of t‑tubules in excitation–contraction coupling is well established, but the cellular mechanisms by which these membrane invaginations develop and are maintained have been less clear. Amphiphysin‑II/BIN1 has been implicated in delivery of L‑type channels to t‑tubules, yet whether microtubules directly regulate the formation and homeostasis of t‑tubules remained uncertain.
This study aimed to determine whether cardiac microtubules actively regulate t‑tubule biogenesis and stability, and to identify key microtubule‑associated factors involved. The authors used a reductionist model to separate membrane tubule formation from the complex architecture of mature adult cardiomyocytes.
Neonatal rat ventricular myocytes (NRVMs) lack endogenous t‑tubules, providing a simplified setting in which exogenous BIN1 overexpression induces nascent membrane tubules. The authors used this system to test whether disrupting microtubules, microtubule motors, or microtubule plus‑end tracking proteins alters BIN1‑driven tubule initiation, elongation, and maintenance.
When microtubules were depolymerised pharmacologically with nocodazole prior to BIN1 overexpression, BIN1‑driven membrane tubule formation was impaired. Specifically, microtubule depolymerisation reduced both the density and the length of BIN1‑driven tubules in NRVMs, indicating that an intact microtubule network is required for efficient tubule initiation and/or elongation in this model.
Inhibition of the minus‑end directed motor dynein using EHNA produced effects similar to microtubule depolymerisation, reducing tubule formation and length during BIN1‑driven tubulogenesis. These observations support a role for microtubule‑based motor activity in the elongation phase of membrane tubule growth.
Knockdown of the microtubule plus‑end tracking protein CLIP‑170 reduced BIN1‑driven tubule density in NRVMs. This implicates CLIP‑170‑dependent capture of microtubule plus ends at membrane sites as an important mechanistic step for initiating BIN1‑mediated tubule formation.
Microtubules were not only required for the formation of nascent tubules but also necessary to maintain existing structures. In NRVMs with established BIN1‑driven tubules, interventions that disturbed microtubule dynamics—either depolymerisation, artificial stabilisation, or dynein inhibition—each reduced tubule density and length. These data indicate that ongoing microtubule dynamics and motor activity support the persistence and structural integrity of membrane tubules once formed.
To assess relevance beyond the reductionist NRVM model, the authors performed acute microtubule depolymerisation and stabilisation experiments in isolated adult sheep left ventricular myocytes that possess native t‑tubule networks. Acute perturbation of microtubules in these cells disrupted native t‑tubule architecture, consistent with the requirement for microtubule dynamics in maintaining mature tubule networks.
Based on the combined observations, the authors propose a model in which BIN1‑dependent membrane tubule formation requires microtubule plus‑end capture mediated by CLIP‑170, followed by dynein‑dependent elongation along microtubule tracks. After tubules are established, continuing microtubule dynamics and motor activity are necessary to preserve mature t‑tubule structure and organization. Disturbance of any of these components—microtubule integrity, plus‑end capture, motor function, or dynamic turnover—compromises both tubule formation and maintenance.
The study links two features observed in HFrEF—t‑tubule disruption and microtubule remodelling—by demonstrating that microtubules actively shape t‑tubule architecture. These findings suggest that microtubule‑dependent trafficking and dynamics are potential contributors to pathological loss of t‑tubules and impaired calcium handling in disease. Future work should explore how disease‑associated microtubule remodelling alters BIN1, CLIP‑170, or dynein function in vivo and whether targeting these interactions can preserve t‑tubule networks in heart failure.
Competing interests: The authors declared no competing interest. Funding: British Heart Foundation support was declared by the authors.