Prelamin A mediates myocardial inflammation in dilated and HIV-associated cardiomyopathies

Abstract

Cardiomyopathies are complex heart muscle diseases that can be inherited or acquired. Dilated cardiomyopathy can result from mutations in LMNA, encoding the nuclear intermediate filament proteins lamin A/C. Some LMNA mutations lead to accumulation of the lamin A precursor, prelamin A, which is disease causing in a number of tissues, yet its impact upon the heart is unknown. Here, we discovered myocardial prelamin A accumulation occurred in a case of dilated cardiomyopathy, and we show that a potentially novel mouse model of cardiac-specific prelamin A accumulation exhibited a phenotype consistent with inflammatory cardiomyopathy, which we observed to be similar to HIV-associated cardiomyopathy, an acquired disease state. Numerous HIV protease therapies are known to inhibit ZMPSTE24, the enzyme responsible for prelamin A processing, and we confirmed that accumulation of prelamin A occurred in HIV+ patient cardiac biopsies. These findings (a) confirm a unifying pathological role for prelamin A common to genetic and acquired cardiomyopathies; (b) have implications for the management of HIV patients with cardiac disease, suggesting protease inhibitors should be replaced with alternative therapies (i.e., nonnucleoside reverse transcriptase inhibitors); and (c) suggest that targeting inflammation may be a useful treatment strategy for certain forms of inherited cardiomyopathy.

Keywords: AIDS/HIV; Cardiology; Cardiovascular disease.

Figure 1. Prelamin A accumulated in a heart of a patient with dilated cardiomyopathy (DCM).

(A) Confocal micrographs of human heart sections from DCM patients and nonfailing (NF) controls subjected to immunofluorescence staining to detect prelamin A (green), myomesin (red), and DAPI (blue). Arrows point to prelamin A⁺ CM nuclei, many of which exhibit nuclear morphology defects. Scale bar: 10 μm. (B) The number of nuclei that stained positively for prelamin A were quantified as a percentage of CM nuclei for nonfailing (blue circles, n = 10 nonfailing donor samples) and DCM (red squares, n = 21 samples from explanted DCM hearts) myocardial sections (mean ± SD). (C) Western blotting did not detect prelamin A in a selection of DCM patient samples. There was not enough sample remaining from DCM05 to run on a Western blot.

Figure 2. Targeted transgenesis of prelamin A led to nuclear accumulation in CMs and resulted in premature death in mice.

(A) Schematic representation showing the site of prelamin A (LMNA-L647R) cDNA insertion and the modifications required for conditional expression. SA, splice acceptor site; neo^R, neomycin resistance; pA, polyadenylation signal. (B) Western blotting for prelamin A showing expression was restricted to heart tissue. (C) Confocal micrographs of myocardium stained for prelamin A showing nuclear rim localization in csPLA-Tg hearts. Scale bar: 10 μm. Arrows indicate prelamin A expressing nuclei. (C) Growth curves showing that csPLA-Tg mice stop growing after 30 days. n = 3 males/group. Two-way ANOVA with repeated measures with Sidak’s post hoc test for multiple comparisons was performed. *P < 0.05, ***P < 0.001. (D) Kaplan-Meier survival analysis showing that csPLA-Tg male and female mice exhibited attenuated survival compared with FLctrl counterparts. n = 7 FLctrl males, 8 FLctrl females, 9 csPLA-Tg males, 8 csPLA-Tg females. Log-rank Mantel-Cox test was performed. P < 0.0001.

Figure 3. Cardiac function was attenuated in 4-week-old csPLA-Tg mice.

(A) Representative images of echocardiographs and corresponding graphs of analysis performed on movies acquired in B-mode showing severely compromised cardiac function in 4-week-old mice (n = 12/group [6 females, 6 males], except csPLA-Tg 2 weeks, which was n = 9 [7 females, 2 males]). Values are mean ± SD. Two-way ANOVA, no repeated measures, was performed with Sidak’s post hoc test for multiple comparisons. *P < 0.05, ****P < 0.0001. (B) Representative cardiac MRI images of myocardium in end-systole and end-diastole and corresponding graphs displaying a decrease in ejection fraction alongside increases in left ventricle end-diastolic (LVEDV) and LV end-systolic volume (LVESV). Mass was statistically unchanged, but with increased variation, and concurs with postmortem heart weight measurements. (C) Increased relaxation time (R1) of gadolinium contrast in 4-week-old csPLA-Tg myocardium, indicative of fibrosis remodeling. n = 6 males/group. Values are mean ± SD. Student’s 2-tailed t test was performed. **P < 0.01, ***P < 0.001, ****P < 0.0001. IVS;d, intraventricular septal thickness in diastole; LVPW;d, left ventricle posterior wall thickness in diastole.

Figure 4. csPLA-Tg mice displayed signs of heart failure and necrotic cell death.

(A) Fetal gene expression is dysregulated in csPLA-Tg hearts at 4 weeks, indicating heart failure; n = 3 females/group. Values are mean ± SD. Two-Way ANVOA, no repeated measures, was performed with Sidak’s post hoc test for multiple comparisons. *P < 0.05. (B) Low-magnification micrographs showing cross-sectional view of the heart stained with H&E, indicating 4-week-old csPLA-Tg mice possessed dilated cardiac chambers. Scale bar: 2 mm. (C) Photographs showing csPLA-Tg appeared enlarged at 4 weeks. Scale bar: 5 mm. (D) No significant difference in heart weight to tibia length ratio was observed. Values are mean ± SD. n = 3/group. (E) Dissection of chest cavities showing transudative pleural effusions in csPLA-Tg mice. (F) Blood plasma subjected to ELISA showed elevated levels of circulating cardiac Troponin T (TnT), indicative of cardiac damage in 4-week-old mice. n = 3 females/group. Values are mean ± SD. Two-Way ANOVA, no repeated measures, with Sidak’s post hoc test for multiple comparisons was performed. *P < 0.05. (G) TUNEL staining was performed on 4-week-old csPLA-Tg heart sections, and image quantification showed a significant increase in TUNEL⁺ nuclei in csPLA-Tg heart sections at 4 weeks. n = 3 females/group. Values are mean ± SD. Student’s 2-tailed t test was performed. **P < 0.01. Arrows indicate TUNEL positive nuclei. Scale = 30 µm. (H) Western blotting showed that cleavage of caspase 3 and lamins A/C, which occurs at ~37 kDa and occurs during apoptosis, did not occur in 4-week-old csPLA-Tg myocardium.

Figure 5. Fibrotic remodeling of csPLA-Tg myocardium occurred in tandem with inflammation and senescence.

(A) Light micrographs showing myocardial disarray in 4-week-old csPLA-Tg myocardium stained with H&E. Scale bar: 30 μm. (B) Light micrographs showing Picrosirius red–stained myocardium to indicate fibrosis in 4-week-old csPLA-Tg myocardium shown by excessive red staining. Scale bar: 30 μm. (C and D) Quantitative fluorescence immunostaining for CD45 shows presence of CD45⁺ cells in 2- and 4-week-old csPLA-Tg myocardium. Scale bar: 10 μm. Values are mean ± SD. n = 3 females/group. Two-way ANOVA with Sidak’s post hoc test for multiple comparisons was performed. *P < 0.05, ***P < 0.001. (E) CD45⁺ immunostaining of Lmna^–/–myocardium showed no evidence of infiltration by CD45⁺ leukocyte populations. (F) qPCR showing the cytokine profile of csPLA-Tg myocardial mRNA. n = 4 females/group. (G) Immunohistochemical staining showing expression of p16 in 4-week-old csPLA-Tg myocardium. (H) Senescence-associated β-galactosidase was expressed in 4 week csPLA-Tg myocardium. Scale = 30 µm (G and H).

Figure 6. NF-κB signaling was activated in 4-week-old csPLA-Tg CMs and mediated by persistent DNA damage.

(A) Subcellular localization of p65 subunit of NF-κB in csPLA-Tg myocardium, highlighted by white arrows. (B) Quantification of p65 micrographs showing increases in the number of nuclei expressing p65 for 2- and 4-week-old hearts counted as a percentage of total nuclei. n = 3 females/group. Values are mean ± SD. *P < 0.05. Two-way ANOVA, no repeated measures, with Sidak’s test for multiple comparisons was performed. (C) Western blot showing increase of NF-κB subunit p65 in 4-week-old csPLA-Tg myocardium. (D) Confocal micrographs of fluorescence immunostaining showing DNA damage marker γ-H2AX (white arrows). Scale bar: 10 μm. (E) Quantification of γ-H2AX micrographs. n = 3 females/group. Values are mean ± SD. Two-way ANOVA, no repeated measures, with Sidak’s post hoc test for multiple comparisons was performed. (F) Western blots of 4-week-old myocardial lysates showing phosphorylation status of ATM and IκBα and graphs showing corresponding densitometry analyses. Values are mean ± SD. n = 3 females/group. Unpaired 2-tailed t test was performed. *P < 0.05, **P < 0.001.

Figure 7. Prelamin A accumulation led to disorganization of molecular structure at 4 weeks and loss of chromatin and histone marks at 2 weeks.

(A and B) Western blot analysis showing the protein expression changes occurring at 4 weeks in structural proteins of the nuclear envelope-lamin A/C, emerin, SUN2, nesprin 2α, and also the cytoskeleton-desmin, with corresponding semiquantitative densitometry analysis. n = 3 females/group. Values are mean ± SD. Unpaired 2 tailed t test was performed on age-matched groups. Welch’s correction was applied to data for lamin A, lamin C, emerin, SUN2, and nesprin 2α at 4 weeks. *P < 0.05, **P < 0.01, ***P < 0.001. (C) Electron micrographs showing nuclear shape and size changes in csPLA-Tg myocardium; red arrows point to regions of nuclear in-folding characteristic of nuclear morphology defects. Scale bar: 1 μm.

Figure 8. Prelamin A accumulation resulted in loss of heterochromatin and H3K9me3 repressive histone marks in the myocardium of 2-week-old mice.

(A) Representative electron micrographs show heterochromatin displacement and loss of chromocentres. Scale bar: 500 nm. Arrows indicate regions of condensed chromatin. (B) Quantitative IHC showed a profound loss of H3K9me3 staining as a percentage of total Hematoxylin stain in 2-week-old csPLA myocardium. Scale bar: 30 μm. Values are mean ± SD. n = 3 females/group. Two-way ANOVA, no repeated measures, with Sidak’s test for multiple comparisons was performed.*P < 0.05.

Figure 9. Prelamin A accumulated in hearts of patients with HIV-associated cardiomyopathy under retroviral therapy.

(A) The DCM patient sample in which prelamin A accumulated showed profound mononuclear infiltration unlike other DCM samples. Scale bar: 40 μm. (B) H&E and CD3⁺ IHC showing inflammation in HIV⁺ myocardium consistent with the csPLA-Tg model. Scale bar: 30 μm. (C) IHC showing focal prelamin A accumulation in CM nuclei (black arrows) and non-CM populations (green arrows) of HIV⁺ myocardium supported by Western blotting, which detected accumulation of prelamin A in hearts of HIV patients. (D) Electron micrographs showing nuclear morphology defects in HIV⁺ myocardium (red arrow; scale bar: 3 μm) and nuclear pore complexes surrounded by evenly spread heterochromatin in nondiseased myocardium (large white arrows) and heterochromatin displacement in HIV⁺ myocardium (small white arrows). Arrows are located at the inner nuclear membrane of lower right panel of D. Scale bar: 1 μm. N, nucleus; L, lipid bodies.

Figure 10. Induction of prelamin A expression in adulthood leads to a progressive and fatal decline in cardiac function in inducible csPLA-Tg (icsPLA-Tg) mice.

(A) Schematic showing the protocol for assessing the effect of prelamin A acquired in adult mouse myocardium via utilization of a tamoxifen inducible MerCreMer promoter. (B) Immunofluorescence staining of heart sections showing the expression of prelamin A in cardiomyocytes of icsPLA-Tg mice. Scale bar: 20 μm. (C) Kaplan-Meier survival analysis showing that csPLA-Tg male and female mice died early compared with FLctrl counterparts. Log-rank Mantel-Cox. P < 0.0001. n = 9/group. (D) Body weights remained constant during the course of the protocol. (E) Cardiac function declined progressively over the course of the protocol. Values are mean ± SD. n = 9–8 males and 1 female/group. Two-way ANOVA with repeated measures and Sidak’s test for multiple comparisons was performed. ****P < 0.0001.

Figure 11. Induction of prelamin A in mouse hearts induces cardiac remodeling and inflammation.

(A) H&E and (B) Picrosirius red staining showed myocardial disarray and fibrosis, respectively. (C) Immunoflourescence micrographs showing CD45⁺ cells were evident in icsPLA-Tg myocardium. Scale bars: 30 μm. (D) Quantification of CD45⁺ cells showing an increase in leukocyte population in icsPLA-Tg myocardium. Values are mean ± SD. n = 4 males/group. Unpaired student’s t test was performed. ***P < 0.001.