Protein Misfolding and Cardiac Disease

Abstract

Numerous neurodegenerative diseases are characterized by the accumulation of misfolded amyloidogenic proteins. Recent data indicate that a soluble pre-amyloid oligomer (PAO) may be the toxic entity in these diseases and the visible amyloid plaques, rather than causing the disease, may simply mark the terminal pathology. In prior studies, we observed PAO in the cardiomyocytes of many human heart failure samples. To test the hypothesis that cardiomyocyte-restricted expression of a known PAO is sufficient to cause heart failure, transgenic mice were created expressing polyglutamine repeats of 83 (PQ83) or 19 (PQ19).Long PQ repeats (>50) form PAOs and result in neurotoxicity in Huntington’s disease, whereas shorter PQ repeats are benign. PQ83 expression caused the intracellular accumulation of PAOs and aggregates leading to cardiomyocyte death and heart failure. Evidence of increased autophagy and necrosis accompanied the PQ83 cardiomyocyte pathology. The data confirm that protein misfolding resulting in intracellular PAO accumulation is sufficient to cause cardiomyocyte death and heart failure.

Background

Heart failure (HF) is a leading cause of death in the US, affecting between 2–3 million Americans. HF is a common terminal endpoint in many cardiac diseases that stem from diverse etiologies including coronary artery disease, hypertension, and idiopathic and familial cardiomyopathies.^{1, 2} Many cardiomyopathies are caused by mutations in cardiac sarcomeric, cytoskeletal, and associated proteins.¹ These mutations, as well as epigenetic stresses, can promote or cause protein misfolding, which can have diverse effects on that and other proteins’ function, interactions, localization, steady state levels and turnover.^4,5 Some myopathic mutations are associated with known chaperones. For example mutations in desmin or its chaperone αB-cystallin, a small heat shock-like protein, can cause a desmin-related myopathy (DRM).^5–7 DRMs are characterized by insoluble intracellular proteinaceous accumulations of the mutant protein and its interaction partners ultimately resulting in distil muscle weakness and in a dilated cardiomyopathy.^5–8 This protein misfolding results in intracellular dysgenesis of the cardiac sarcomeres and hence is unique and can be distinguished from the misfolded protein diseases associated with muscle and/or cardiac tissue, resulting in extracellular amyloidoses.

The accumulation of misfolded proteins into intracellular or extracellular aggregates is a common characteristic of the neurodegenerative amyloidoses including Alzheimer’s, Huntington’s and Parkinson’s diseases.^9,10 As noted above, recent data suggest that the insoluble aggregates may not be inherently pathogenic; instead the toxic entity appears to be the soluble intermediates of fibril formation known as PAO.^11–14 These soluble oligomers share a common conformation-dependent protein structure, regardless of genetic sequence^15,16 and can be detected by conformation-specific antibodies.^7,16 Staining of mouse models of DRM and human HF samples for PAOs demonstrate that cardiomyocytes from diverse cardiomyopathies accumulate PAO, whereas healthy control hearts do not.² The presence of PAOs in heart-failure patients implies that there may be common pathogenic mechanisms between the neurodegenerative amyloidoses and at least some cardiomyopathies that result in end-stage HF. However, the direct relationship between intracellular PAO generation, protein aggregate deposition and cardiac disease pathology remains unclear.

Can PAO Expressed in Cardiomyocytes Cause Heart Failure?

To test the hypothesis that PAOs are intrinsically pathogenic upon expression and intracellular accumulation in cardiomyocytes and sufficient to cause cardiomyopathy, we generated transgenic mice with cardiomyocyte autonomous expression of an ectopic peptide containing83 repeats of glutamine (PQ83), which is causative in Huntington's disease. Parallel lines of mice that expressed equal or higher amounts of a non-amyloidogenic peptide containing only 19 repeats (PQ19) were created in order to ensure that high levels of an ectopic protein were not responsible for any resulting pathology.³ If intracellular expression and accumulation of PAO is inherently toxic, expression of PQ83 should result in cardiomyocyte pathology, potential cell loss and subsequently HF.

Consistent with the hypothesis, expression of low levels of PQ83 peptide caused deposition of both PAOs and PQ83-containing aggresomes within the cardiomyocytes, leading to cardiomyocyte loss, progressive cardiac dilation and eventual death from HF by approximately 5–7 months. Further study of the observed cardiomyocyte death reveals increased autophagosomal activity and lysosomal content. Necrotic cells are apparent but apoptotic activation and markers of endoplasm reticulum stress could not be detected. Ultrastructural analysis finds a marked increase in the number of mutilamellar bodies and lysosomes as well as examples of double-membraned autophagic vacuoles engulfing the polyglutamine aggresomes. It is currently unclear whether the increases in autophagic and lysosomal content are beneficial or detrimental to cardiomyocyte survival during the development of PQ83-induced cardiomyopathy. Further studies are ongoing to determine whether autophagy is contributing to PAO-induced cardiomyocyte toxicity and death and at which stage of the disease.

Evaluating Potential Roles of Autophagy in Heart Failure

Autophagy is a critical pathway for the clearance of misfolded and aggregated proteins. Autophagy can function as a form of programmed cell death in some models but can be instrumental in cell survival during cellular starvation.^18,19 Autophagic dysregulation clearly contributes to cellular pathogenesis in many cell types, including the heart.^18,20 However, autophagic processes are also essential for basal cardiomyocyte function.^21,22 It is clear that ablation of autophagic or lysosomal function can, under some circumstances, be detrimental to the heart.^21,23,24 However, gross upregulation of autophagic/lysosomal function could be deleterious as well.^25,26 Thus, as is the case for any essential cellular function, cardiomyocyte-based autophagy likely requires finely tuned control so that homeostasis is maintained.

It remains to be determined if PAOs can be degraded by lysosomes independently from the non-soluble aggregates. Many data show that lysosomes can consume aggregates and, conversely, inhibition of lysosomal function can increase protein aggregate content in cells.^22,27 In some neurodegenerative models, autophagic induction is protective against toxic intracytosolic aggregate-prone proteins, whereas other evidence suggests that sequestering toxic amyloidogenic proteins into aggregates may prevent their cytotoxic actions.^20,28,15 These issues have not been rigorously addressed in human disease and whether the observed PAO accumulation in human HF samples is due to improper protein folding, insufficient or compromised degradative machinery, or ineffective delivery to the degradation apparatus is presently unclear.

Is protein misfolding and PAO a generalized pathway in the development of human HF? In many tissues, cellular stress and aging can cause the accumulation of non-degraded proteins, as well as generating relatively high levels of misfolded proteins.⁴ Hundreds or even thousands of proteins have the capacity to misfold and become amyloidogenic and, if not cleared, can be cytotoxic.¹⁵ Further compromising the cardiomyocyte’s ability to clear these accumulations is its post-mitotic state, a trait shared with many cells types affected in the neurodegenerative disorders. Future work will establish which catabolic process is rate-limiting, and test the effects of autophagic/lysosomal modulation on cardiomyocyte survival. The putative relationships between autophagic/lysosomal accretion and necrotic cardiomyocyte death also need to be determined. Finally, we need to define the means by which PAO accumulation induces cell death and what commonalities, if any, are present as a result of PAO generation from different proteins and stresses.