Dr Mann’s talk focused on his investigations on the role of haemodynamic load in left ventricular (LV) remodelling and reverse remodelling. The biology of cardiac remodelling involves three main components in succession: myocyte and myocardial alterations resulting in alterations to LV geometry. His early work in 1989 provided the first evidence that an increase in load on adult mammalian cardiomyocyte can activate hypertrophy.1 Numerous studies have since shown that cardiomyocyte stretch alone is sufficient to activate multiple growth pathways, although the exact stretch receptors are not known.2 D’Angelo et al. provided evidence of recapitulation of the cardiac hypertrophy gene expression by the overexpression of a single component of the stretch-activated pathway (Gαq-protein signalling pathway).3 Stretching of cardiomyocytes induces alterations in multiple pathways in myocyte biology in parallel, including sarcomeric changes, cytoskeletal proteins and mitochondria.
LV remodelling is classified based on the patterns of structural changes relative to normal (defined as normal LV mass [LVM] with normal mass-to-cavity [M-C] ratio): concentric remodelling (normal LVM with high M-C ratio); eccentric hypertrophy (high LVM with normal M-C ratio), typically observed with volume overload; and concentric hypertrophy (high LVM with high M-C ratio), typically observed with pressure overload.4 Studies have proposed that a transition may exist between early ‘compensatory’ cardiac hypertrophy in the setting of prolonged and continuous pressure overload to a ‘decompensated state’ that leads to heart failure (HF).5,6 However, emerging epidemiological studies suggest that this transition from concentric hypertrophy to HF does not occur in humans.7,8 Also, the transition from concentric LV geometry to eccentric hypertrophy occurs in <10% of patients during long-term follow-up, suggesting that the concept of using extreme pressure overloads to produce eccentric hypertrophy to understand the biology of cardiac remodelling is not clinically relevant.
Drazner et al. proposed multiple potential pathways for the progression from hypertension (classic disease of pressure overload) to HF. He showed that concentric hypertrophy progresses to dilated cardiac failure most commonly via an interval MI.9 Interestingly, clinical evidence shows that the concomitant occurrence of hypertensive and ischaemic heart disease may lead to the development of HF.10,11 Hence, Weinheimer et al. developed a surgical approach that combined transverse aortic constriction (TAC; mimicking pressure overload) and distal left anterior coronary ligation (inducing MI) to produce a gradual and predictable progression of maladaptive LV remodelling that leads to HF.12 Importantly, in this surgical mouse model, a small apical infarct (<25% infarct size) or moderate TAC alone did not lead to LV remodelling. In essence, this novel mouse model demonstrated that pressure overload acts synergistically with tissue injury to provoke LV remodelling. Dr Mann briefly discussed how tissue injury leads to LV remodelling. The inflammation initiated by tissue injury activates matrix metalloproteases, which prime the extracellular matrix to receive the increased haemodynamic load signal, leading to LV remodelling.13
This led to the question of how reverse LV remodelling happens. Dr Mann’s research team developed a murine model wherein mice develop LV remodelling after TAC and a small apical MI undergo reverse LV remodelling after removal of the aortic band at 2 weeks post-TAC/MI.14 De-banding normalised LV volumes, LV mass and cardiac myocyte hypertrophy at 6 weeks with no difference in myofibrillar collagen with or without de-banding. LV ejection fraction (LVEF) and radial strain improved after de-banding; however, both remained decreased in the de-banded mice relative to sham and were not different from non-de-banded mice at 6 weeks. Haemodynamic unloading in the de-banded mice was accompanied by a 35% normalisation of the HF genes at 2 weeks and 80% of the HF genes at 4 weeks.
Further, bioinformatic analyses showed that the reversal of the LV HF phenotype is accompanied by significant changes in the expression of multiple genes residing within each of the five different cardiac myocyte gene modules: extracellular matrix, integrin/cytoskeleton, sarcomere, excitation-contraction coupling and metabolism. These analyses also suggested that the changes in myocyte function precede changes in the integrin/cytoskeleton during reverse LV remodelling. Interestingly, reverse LV remodelling was not merely a reversal of the functional pathways that become dysregulated during HF. Instead, reverse LV remodelling represented the summation of the complex interactions between multiple biological networks that adopt a novel less-pathological configuration when the inciting stress is removed. This observation raises the possibility that some of the changes that occur during reverse LV remodelling confer vulnerability to a subsequent stress. This concept of ‘robust yet fragile’ may explain, at least in part, the observation that stable patients whose hearts undergo reverse LV remodelling with normalisation of LVEF continue to experience recurrent HF events.15,16
Future studies will determine whether the re-tuning of gene networks during reverse LV remodelling represents a ‘good enough solution’ to accommodate biological function, or whether it represents an example of a ‘robust yet fragile’ biological system that has been optimised to maintain robustness (i.e. homeostasis following loss of cardiac myocytes) at the expense of increased fragility (i.e. increased myocardial fibrosis).17