Transverse tubule remodelling: a cellular pathology driven by both sides of the plasmalemma?

Abstract

Transverse (t)-tubules are invaginations of the plasma membrane that form a complex network of ducts, 200-400 nm in diameter depending on the animal species, that penetrates deep within the cardiac myocyte, where they facilitate a fast and synchronous contraction across the entire cell volume. There is now a large body of evidence in animal models and humans demonstrating that pathological distortion of the t-tubule structure has a causative role in the loss of myocyte contractility that underpins many forms of heart failure. Investigations into the molecular mechanisms of pathological t-tubule remodelling to date have focused on proteins residing in the intracellular aspect of t-tubule membrane that form linkages between the membrane and myocyte cytoskeleton. In this review, we shed light on the mechanisms of t-tubule remodelling which are not limited to the intracellular side. Our recent data have demonstrated that collagen is an integral part of the t-tubule network and that it increases within the tubules in heart failure, suggesting that a fibrotic mechanism could drive cardiac junctional remodelling. We examine the evidence that the linkages between the extracellular matrix, t-tubule membrane and cellular cytoskeleton should be considered as a whole when investigating the mechanisms of t-tubule pathology in the failing heart.

Keywords: Collagen; Excitation–contraction coupling; Extracellular matrix; Heart failure; Transverse tubules.

Fig. 1

Loss of contractility in human heart failure is strongly correlated to loss of the transverse elements of the t-system. a and b show exemplar confocal micrographs of WGA-stained t-tubules from the strongly contracting region (−12% fractional shortening) and the weakly contracting region (−2% fractional shortening) in a failing heart. The grey arrow in a indicates a t-tubule. Note the loss of the transverse elements of the t-system in the poorly contracting tissue region. c demonstrates the strong correlation (p < 0.001) between the percent transverse elements (of the t-system) and fractional shortening of failing heart tissue regions. Five failing hearts were analysed, with coloured circles of the same colour indicating different regions from the same heart. Note how the trend is consistent within the individual failing hearts. The black circles are normal donor hearts shown for comparison and are not part of the regression analysis. Figure adapted from Crossman et al. (2015)

Fig. 2

Costamere complexes are present within the t-tubules of the human heart. a Vinculin (green) labels both the surface and t-tubule sarcolemma, seen as radial finger-like projections situated between myofibrils, labelled with TRITC-conjugated phalloidin (red). b Dystrophin (green) labels both the surface and t-tubule sarcolemma. Nuclei are stained red with 7-AAD. Scale bars = 10 μm. Figure adapted from Kostin et al. (1998)

Fig. 3

Increased WGA labelling of t-tubules is due to collagen VI in idiopathic dilated cardiomyopathy in humans. a WGA Western blotting identifies an increase in the 140-kDa band in disease. b Western blotting (and mass spectrometry) identifies this band as collagen VI. c, i and d, i Super-resolution microscopy of normal and diseased t-tubules labelled for collagen VI (red) and dystrophin (green). c, ii and d, ii Zoom of box in i. c, iii and d, iii Confocal microscopy of equivalent t-tubules. e t-Tubule diameter measured from collagen VI labelling. f Ratio of WGA intensity over collagen VI intensity. L = ladder, N = normal, D = heart failure. For Western blotting, n = 7 normal and 11 diseased hearts. For super-resolution microscopy, n = 15 t-tubules from five normal hearts and 15 t-tubules from five diseased hearts. ***p < 0.001. **p < 0.01. *p < 0.05.

Fig. 4

t-Tubules are dilated in myocytes from the coronary artery ligation rat model of myocardial ischaemic heart failure. Sections of rat heart were labelled with WGA-488 and imaged on a Zeiss LSM 880 Airyscan with 63 × 1.4 NA oil lens. a Exemplar of t-tubule labelling of a myocyte from a sham operated animal. b Exemplar of t-tubule labelling in myocyte from an ischaemic heart failure (MI) animal. c The frequency distribution of the full width at half maximum (FWHM) of the transverse component of t-tubules were measured in 150 sham t-tubules from five hearts and 150 MI t-tubules from five hearts. The mean diameter of WGA t-tubules in sham myocytes was 222 ± 3 nm and was significantly smaller compared to those in MI myocytes, being 253 ± 4 nm (*p = 0.02; two-factor nested ANOVA)