P62‐positive aggregates are homogenously distributed in the myocardium and associated with the type of mutation in genetic cardiomyopathy

Abstract Genetic cardiomyopathy is caused by mutations in various genes. The accumulation of potentially proteotoxic mutant protein aggregates due to insufficient autophagy is a possible mechanism of disease development. The objective of this study was to investigate the distribution in the myocardium of such aggregates in relation to specific pathogenic genetic mutations in cardiomyopathy hearts. Hearts from 32 genetic cardiomyopathy patients, 4 non‐genetic cardiomyopathy patients and 5 controls were studied. Microscopic slices from an entire midventricular heart slice were stained for p62 (sequestosome‐1, marker for aggregated proteins destined for autophagy). The percentage of cardiomyocytes with p62 accumulation was higher in cardiomyopathy hearts (median 3.3%) than in healthy controls (0.3%; P < .0001). p62 accumulation was highest in the desmin (15.6%) and phospholamban (7.2%) groups. P62 accumulation was homogeneously distributed in the myocardium. Fibrosis was not associated with p62 accumulation in subgroup analysis of phospholamban hearts. In conclusion, accumulation of p62‐positive protein aggregates is homogeneously distributed in the myocardium independently of fibrosis distribution and associated with desmin and phospholamban cardiomyopathy. Proteotoxic protein accumulation is a diffuse process in the myocardium while a more localized second hit, such as local strain during exercise, might determine whether this leads to regional myocyte decay.


| INTRODUC TI ON
The intracellular accumulation of misfolded proteins and damaged cell organelles is increasingly reported in heart failure. In the normal situation, protein degradation pathways, like autophagy, prevent this accumulation of misfolded proteins. Recent studies suggest an important role for impaired autophagic processes in myocardial diseases such as myocardial infarction, pressure overload hypertrophy and various forms of cardiomyopathies. 1 Autophagy regulates cellular homoeostasis by transporting intracellular cargo to the lysosomes for bulk degradation and recycling of macromolecules. 1 This process reduces proteotoxic stress and delivers new building blocks and energy for the cell. In macroautophagy, intracellular vesicles, also known as autophagosomes, are formed containing damaged macromolecules (eg mutant, misfolded or damaged proteins) and organelles (eg mitophagy). These autophagosomes travel along microtubules and then fuse with lysosomes which are then termed autophagolysosomes. After fusion with lysosomes, the intracellular cargo is degraded by lysosomal enzymes and reused. 2 P62 recognizes ubiquitinated proteins and promotes aggregation and sequestration (synonym sequestosome-1) marking protein aggregates for degradation by autophagy or the ubiquitin proteasome system. 3 After fusion of the autophagosome with the lysosome, p62 is efficiently degraded by lysosomal enzymes. When autophagy is unable to remove the p62 bound proteins, p62 accumulates in the cell. Accumulation of p62-positive aggregates in the cell may therefore be a measure of the amount of proteotoxic aggregates. These aggregates may in turn activate pathways leading to cardiomyocyte dysfunction, including cellular senescence. 4 In hearts of cardiomyopathy patients with a pathogenic variant in the gene encoding phospholamban (PLN), we have recently demonstrated aggregates that are both p62 and phospholamban positive. 5 This suggests that the autophagy system is not able to remove all the mutant phospholamban protein, resulting in accumulation of p62positive proteotoxic perinuclear aggregates in the cardiomyocytes. p62 accumulation has also been described in some other genetic cardiomyopathies. 6 In the present study, we investigated the presence of p62 proteotoxic aggregates in genetic cardiomyopathies caused by pathogenic mutations in different cardiomyopathy-related genes (DES, CRYAB, PLN, PKP2, DSP, LMNA, MYBPC3, MYH7, TNNI3 and   TNNT2). In addition, we studied the distribution of these aggregates throughout the myocardium.

| ME THODS
The study met the criteria of the code of proper use of human tissue that is used in the Netherlands. The collection of the human heart tissue was approved by the scientific advisory board of the biobank of the University Medical Center Utrecht, Utrecht, the Netherlands (protocol no. 12/387). Written informed consent for collection and biobanking of tissue samples was obtained prior to transplantation or, in certain cases, approved by the ethics committee when obtaining informed consent was not possible because of death of the patient.
Hearts from 32 patients with a genetic cardiomyopathy and 4 patients without known pathogenic mutations were included in the study that were obtained during heart transplantation or at autopsy. Genetic testing was performed in a clinical setting. Patients were classified in functional protein groups encoded by (likely) pathogenic variants (mutations) in genes encoding desmin (DES) and the desmin-related αB- A complete heart slice halfway between the apex and the atrioventricular valves was formalin-fixed and divided into smaller pieces as described previously. 8,9 The formalin-fixed heart tissue was embedded in paraffin and cut into 4 µm sections. Hematoxylin and eosin (H&E) and Masson's trichrome stains were done to evaluate general morphology and fibrosis.
A mouse anti-p62 Ick ligand antibody (BD Biosciences) was used to stain for p62 using the Ventana BenchMark Ultraplatform (Roche). For p62 quantification, the heart slice was divided into five regions for the free wall of the left ventricle (LV; anterior, anterolateral, lateral, posterolateral and posterior), one region for the septum and two regions for the right ventricle (RV; posterior and anterior) as we have described previously ( Figure 1). 8,9 Every region was then split into three areas: trabeculated, inner compact (middle layer) and outer compact (epicardial) myocardium. In all cases, a hot spot count was done in the different areas: 250 cells were counted per area (8 regions x 3 areas x 250 cells = 6000 counted cells per heart). In F I G U R E 1 Example of heart slice of PLN cardiomyopathy heart divided into 8 regions: 5 regions for the left ventricular free wall (anterior, anterolateral, lateral, posterolateral, posterior), 1 region for the septum and 2 regions for the right ventricle (posterior and anterior). Bottom of figure is anterior part of the heart the phospholamban group, the amount of fibrosis was also digitally quantified in the 6 regions of the heart in the three areas (trabeculated, inner and outer compact) as described previously. 8,9

| Statistics
Data are expressed as median [interquartile range]. A Mann-Whitney test was used to compare percentages between two groups and a Kruskal-Wallis test to compare multiple groups. Dunn's multiple comparison test was used to correct for multiple comparisons.
Linear regression was used to test the association between two continues variables.

| Patient characteristics
Patient characteristics are summarized in Table 1

| Differences between mutation groups
p62 showed a globular staining pattern of small or larger perinuclear aggregates or diffuse staining of the sarcolemma in the majority of cases, which is in concordance with our previous observation ( Figure 2). 5 In some cases, the aggregates were present in large parts of the sarcolemma with very limited remaining sarcomeres.
In the healthy control group, a median of 0.3% [0.  Figure 3A). P62 accumulation in the latter two groups was comparable with hearts without known mutations (1.6% [1.2-2.4]). Both the desmin and phospholamban hearts showed significantly higher percentages of cardiomyocytes with p62 accumulation as compared to the other groups. Within the desmosomal group, percentages of p62-positive cells were comparable for the PKP2 and DSP gene mutations ( Figure 3B).

| Differences between RV, septum and LV
In the whole group of cardiomyopathy hearts, no significant differ- In the subgroup analysis of the other mutation groups, no significant difference was observed between RV, septum and LV ( Figure 3C).

| Difference between trabecular, inner and outer compact myocardium in RV and LV
Both in the cardiomyopathy hearts pooled group and in the different mutation groups, no differences were observed between the trabecular, inner and outer compact myocardium for RV, LV, RV and LV combined and posterolateral wall of the LV ( Figure 3D).

| P62 is not associated with the amount of fibrosis in phospholamban hearts
Given the relatively high numbers of phospholamban hearts, we decided to perform a subgroup analysis to study the association between p62 and fibrosis. The percentage of p62-positive cardiomyocytes was not associated with the percentage fibrosis in the TA B L E 1 Patient characteristics per pathogenic genetic mutation group Age ( myocardium of the whole heart (P = .8), RV (P = .9), LV (P = .9) and the three separate layers of the left ventricular wall (P = .7, 0.8 and 0.6 for trabulated, inner and outer compact myocardium, respectively).

| D ISCUSS I ON
The accumulation of potentially proteotoxic protein aggregates due to insufficient autophagy is a possible contributing mechanism of disease development in genetic cardiomyopathy. 1,4 We studied the in a heterozygous co-expression experiment, that is more comparable to the human situation, mutant and wild-type desmin molecules were incorporated into the same filamentous structures without aggregate formation. 12 The present study shows that also  5,19,20 In the present study, we show that proteotoxic stress in phospholamban cardiomyopathy is present in a relatively large percentage of cardiomyocytes in the myocardium as compared to most other genetic cardiomyopathies. Strikingly we did not find a correlation between p62-positive aggregates and fibrosis in phospholamban cardiomyopathy. p62 was more diffusely present in the myocardium, whereas we have previously shown that fibrosis is predominantly observed in the RV wall and the LV posterolateral wall, 8,9 areas with high strain during exercise. The combination of these findings suggests that the accumulation of proteotoxic aggregates that replace parts of the contractile elements in the sarcolemma is the first hit in the pathophysiology of phospholamban cardiomyopathy, leading to instable cells which are vulnerable for a second hit such as local mechanical stress (Figure 4). Another disease mechanism that has been suggested to play a role in phospholamban cardiomyopathy is activation of calmodulin-dependent kinase II and calcineurin A due to disturbed calcium handling, which F I G U R E 4 Hypothesis of pathophysiological mechanism. Proteotoxic stress in desmin and phospholamban cardiomyopathy is present in a relatively high percentage of cardiomyocytes in the myocardium as compared to most other groups of genetic cardiomyopathies. p62 was diffusely present in the myocardium, not related to fibrosis. These findings might suggest that the accumulation of proteotoxic aggregates in the sarcolemma is the first hit in the pathophysiology of desmin and phospholamban cardiomyopathy, making the cell less stable and vulnerable for a second hit such as local mechanical stress may lead to maladaptive remodelling of the macromolecular protein complex that forms the intercalated disc. only twofold compared to controls. 23 Overall, in cardiomyopathy hearts an increase of p62-positive aggregates was observed as compared to controls, also in hearts without known pathogenic mutation. Therefore, a slightly diminished autophagic degradation due to increased proteotoxic stress might be a more general phenomenon of diseased cardiomyocytes.
A possible limitation of this study is the relatively low number of hearts per mutation group. We advise that in the near future cardiovas- In conclusion, accumulation of p62-positive protein aggregates is homogeneously distributed in the myocardium independently of fibrosis distribution and strongly associated with desmin and phospholamban cardiomyopathy. Proteotoxic protein accumulation probably is a diffuse process in the myocardium where a second hit, such as local strain during exercise, might determine whether this leads to regional myocyte decay and fibrosis.

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.