Atrogin-1 and MuRF1 regulate cardiac MyBP-C levels via different mechanisms

Abstract

Aims: Familial hypertrophic cardiomyopathy (FHC) is frequently caused by cardiac myosin-binding protein C (cMyBP-C) gene mutations, which should result in C-terminal truncated mutants. However, truncated mutants were not detected in myocardial tissue of FHC patients and were rapidly degraded by the ubiquitin-proteasome system (UPS) after gene transfer in cardiac myocytes. Since the diversity and specificity of UPS regulation lie in E3 ubiquitin ligases, we investigated whether the muscle-specific E3 ligases atrogin-1 or muscle ring finger protein-1 (MuRF1) mediate degradation of truncated cMyBP-C.

Methods and results: Human wild-type (WT) and truncated (M7t, resulting from a human mutation) cMyBP-C species were co-immunoprecipitated with atrogin-1 after adenoviral overexpression in cardiac myocytes, and WT-cMyBP-C was identified as an interaction partner of MuRF1 by yeast two-hybrid screens. Overexpression of atrogin-1 in cardiac myocytes decreased the protein level of M7t-cMyBP-C by 80% and left WT-cMyBP-C level unaffected. This was rescued by proteasome inhibition. In contrast, overexpression of MuRF1 in cardiac myocytes not only reduced the protein level of WT- and M7t-cMyBP-C by >60%, but also the level of myosin heavy chains (MHCs) by >40%, which were not rescued by proteasome inhibition. Both exogenous cMyBP-C and endogenous MHC mRNA levels were markedly reduced by MuRF1 overexpression. Similar to cardiac myocytes, MuRF1-overexpressing (TG) mice exhibited 40% lower levels of MHC mRNAs and proteins. Protein levels of cMyBP-C were 29% higher in MuRF1 knockout and 34% lower in TG than in WT, without a corresponding change in mRNA levels.

Conclusion: These data suggest that atrogin-1 specifically targets truncated M7t-cMyBP-C, but not WT-cMyBP-C, for proteasomal degradation and that MuRF1 indirectly reduces cMyBP-C levels by regulating the transcription of MHC.

Figure 1

(A) Schematic representation of the adenovirus vectors containing the cDNA of interest and EGFP cDNA under the control of distinct CMV promoters, and schematic structure of the proteins encoded by the adenovirus: human full-length cMyBP-C (myc-WT-cMyBP-C), human C-terminally truncated cMyBP-C (myc-M7t-cMyBP-C), mouse atrogin-1 (myc₆-atrogin-1), and mouse MuRF1 (myc-MuRF1). The absence of exon 6 in myc-M7t-cMyBP-C results in a frameshift, loss of the terminal 1055 residues including the MyBP-C motif (PPP), and additional 41 novel amino acids (black box). (B) Effect of proteasome inhibitors on the expression of myc-M7t in NRCM. NRCM were infected with myc-M7t adenovirus in the absence (Ctr) or presence of the proteasome inhibitors MG132 (132; 1 µM), MG262 (262; 100 nM), or epoxomicin (epo; 500 nM) for 24 h. Western blot (right panel) was stained with the anti-myc antibody and left panel shows the corresponding Ponceau. (C) Immunofluorescence staining and confocal microscopy of NRCM 24 h after infection with adenovirus. Myocytes were double-stained with an anti-myc antibody (green) and with the anti-titin (Z1 domain, red) or anti-cMyBP-C antibody (C0C1 domains; red). Arrowheads show aggregates.

Figure 2

Effect of atrogin-1 on the levels of WT/M7t-cMyBP-C after gene transfer in NRCM. NRCM were co-infected with myc₆-atrogin-1 and with either myc-M7t-cMyBP-C or myc-WT-cMyBP-C in the presence or absence of 1 µM MG132 for 24 h. (A) Immunoprecipitation (IP) of the NRCM lysates treated with MG132 was performed with 0.8 µg anti-atrogin-1 antibody (+) or without antibody (−), and western blot (WB) was revealed with anti-atrogin-1 and anti-myc antibodies. P0 and SN correspond to the crude fraction and the supernatant after IP. (B) Western blot of cell lysates (20 µg) was stained with the anti-ubiquitin antibody that recognizes a smear of ubiquitinated proteins (between 50 and 250 kDa), the anti-myc antibody for detection of exogenous myc-M7t-cMyBP-C (32 kDa), myc-WT-cMyBP-C (150 kDa), and myc₆-atrogin-1 (52 kDa), and the anti-GFP antibody for loading control (GFP, 29 kDa). (C) Quantification of M7t and WT-cMyBP-C levels normalized to Ponceau in the absence (upper bars) or presence (lower bars) of MG132. Values are expressed as mean ± SEM. ***P < 0.001 vs. absence of atrogin-1, Student's t-test. The number of experiments is indicated in the bars.

Figure 3

Effect of MuRF1 on the levels of WT/M7t-cMyBP-C after gene transfer in NRCM. (A) Localization of the region of overlap in the four cMyBP-C prey clones identified by yeast two-hybrid screens with MuRF1 bait. (B) NRCM were co-infected with myc-MuRF1 and with either myc-M7t-cMyBP-C or myc-WT-cMyBP-C in the presence or absence of 1 µM MG132 for 24 h. Western blot of cell lysates (15 µg) stained with the anti-ubiquitin antibody, the anti-myc antibody for detection of exogenous myc-M7t-cMyBP-C (32 kDa), myc-WT-cMyBP-C (150 kDa), and myc-MuRF1 (44 kDa), and the anti-GFP antibody for loading control (GFP, 29 kDa). (C) Quantification of M7t- and WT-cMyBP-C levels normalized to Ponceau in the absence (upper bars) or presence (lower bars) of MG132. Values are expressed as mean ± SEM. ***P < 0.001 vs. absence of MuRF1, Student's t-test. The number of experiments is indicated in the bars.

Figure 4

Investigation of the UPS in MuRF1-KO mice. (A) Representative western blot stained with the anti-ubiquitin antibody, corresponding Ponceau and steady-state level of ubiquitinated proteins (normalized to Ponceau) in ventricular tissue of 24-week-old WT and MuRF1-KO mice. (B) Chymotrypsin-like activity of the proteasome in WT and KO ventricular tissue. (C) mRNA levels of MuRF2, MuRF3, and atrogin-1 determined by RT–qPCR with specific Taqman probes and normalized to GAPDH. Bars represent the mean ± SEM. The number of animals is indicated in the bars.

Figure 5

Investigation of the UPS in MuRF1-TG mice. (A) Representative western blot stained with the anti-ubiquitin antibody, corresponding Ponceau and steady-state level of ubiquitinated proteins (normalized to Ponceau) in ventricular tissue from 8- to 12-week-old WT and MuRF1-TG mice. (B) Chymotrypsin-like activity of the proteasome in WT and TG ventricular tissue. (C) mRNA levels of MuRF2, MuRF3, and atrogin-1 determined by RT–qPCR with specific Taqman probes and normalized to GAPDH. Bars represent the mean ± SEM. *P < 0.05 vs. WT, Student's t-test. The number of animals is indicated in the bars.

Figure 6

Determination of the level of cMyBP-C in MuRF1-KO and MuRF1-TG mice. (A) Representative western blot, corresponding Ponceau and quantification of ventricular cMyBP-C levels in 24-week-old WT and MuRF1-KO mice. (B) Representative western blot, corresponding Ponceau and quantification of ventricular cMyBP-C levels in 8- to 12-week-old WT and MuRF1-TG mice. (C) cMyBP-C mRNA levels determined by real-time RT–PCR in ventricular RNA from WT/KO mice and WT/TG mice. Bars represent the mean ± SEM. **P < 0.01, ***P < 0.001 vs. WT, Student's t-test. The number of animals is indicated in the bars.