Cardiac-specific ablation of Cypher leads to a severe form of dilated cardiomyopathy with premature death

Abstract

Accumulating data suggest a link between alterations/deficiencies in cytoskeletal proteins and the progression of cardiomyopathy and heart failure, although the molecular basis for this link remains unclear. Cypher/ZASP is a cytoskeletal protein localized in the sarcomeric Z-line. Mutations in its encoding gene have been identified in patients with isolated non-compaction of the left ventricular myocardium, dilated cardiomyopathy (DCM) and hypertrophic cardiomyopathy. To explore the role of Cypher in myocardium and to better understand molecular mechanisms by which mutations in cypher cause cardiomyopathy, we utilized a conditional approach to knockout Cypher, specially in either developing or adult myocardium. Cardiac-specific Cypher knockout (CKO) mice developed a severe form of DCM with disrupted cardiomyocyte ultrastructure and decreased cardiac function, which eventually led to death before 23 weeks of age. A similar phenotype was observed in inducible cardiac-specific CKO mice in which Cypher was specifically ablated in adult myocardium. In both cardiac-specific CKO models, ERK and Stat3 signaling pathways were augmented. Finally, we demonstrate the specific binding of Cypher's PDZ domain to the C-terminal region of both calsarcin-1 and myotilin within the Z-line. In conclusion, our studies suggest that (i) Cypher plays a pivotal role in maintaining adult cardiac structure and cardiac function through protein-protein interactions with other Z-line proteins, (ii) myocardial ablation of Cypher results in DCM with premature death and (iii) specific signaling pathways participate in Cypher mutant-mediated dysfunction of the heart, and may in concert facilitate the progression to heart failure.

Figure 1.

Generation of gene-targeted mice. (A) Targeting construct for the generation of floxed Cypher mice. In the targeting vector, cypher gene exon 1 is flanked by LoxP sites as indicated with a black triangle. The neomyosin resistant gene is surrounded by FLP cassettes as shown by black rectangles. (B) Southern blot analysis shows a 6.9 kb band in heterozyogous gene-targeted embryonic stem cells when compared with a 6.2 kb band in the wild-type allele. (C) Representative western blot results showing the effective deletion of Cypher in CKO mouse heart at 1 month of age.

Figure 2.

Survival curve and enlarged Cypher knockout (CKO) mouse hearts. (A) Cumulative survival curve of CKO and control littermate mice. CKO mice die from 16 weeks of age and are all dead before 23 weeks of age (n = 29 for CKO mice and n = 33 for control littermates). (B) Hematoxylin and eosin staining showing dilated ventricular chambers in 3-month-old CKO mouse heart compared with control littermate. (C) Heart weight to body weight ratio of CKO mice versus control littermates (n = 15 for CKO mice and n = 28 for control littermates). (D) Heart weight to tibia bone length ratio of CKO mice versus control littermates (n = 12 for CKO mice and n = 12 for control littermates). HW/BW, heart weight to body weight ratio; HW/BL, heart weight to tibia bone length ratio; *P < 0.05.

Figure 3.

Echocardiography. (A) Typical example of echocardiography data showing LVIDd, LVIDs and FS in Cypher knockout (CKO) and control littermate hearts. (B) LVIDd, (C) LVIDs, (D) FS and (E) EDD/PWD results in CKO mice and control littermates at various ages as indicated (n = 4) for each group. LVIDd, end-diastolic left ventricular internal dimension at end-diastole; LVIDs, end-systolic left ventricular internal dimension; %FS, percentage of fractional shortening; EDD/PWD, end-diastolic dimension/diastolic posterior wall thickness; ^#P < 0.01.

Figure 4.

Inducible cardiac-specific deletion of Cypher. (A) Western blot results showing inducible deletion of Cypher in the heart of Cypher^f/f/Cre mice 1 month after injection with tamoxifen [ICKO (inducible cardiac-specific CKO)] or Cypher^f/f/Cre and Cypher^f/f littermate mouse injected with peanut oil (Con/P) and tamoxifen (Con/T), respectively, as controls. (B) Accumulative survival curve showing that ICKO mice die from 1 week after tamoxifen injection, and all die before 15 weeks after injection, compared with 100% survival of both the control groups (n = 27 for ICKO group, n = 20 for tamoxifen control group and n = 22 for peanut oil control group). (C) Hematoxylin and eosin staining showing severe dilated ventricules in ICKO mouse heart 1 month after injection, compared with tamoxifen control mouse heart (Con/T). Echocardiography data showing (D) LVIDd, (E) LVIDs, (F) LVPWs and (G) FS results in ICKO and control littermate mice 1 month after tamoxifen or peanut oil injection (n = 4 for each group). LVIDd, end-diastolic left ventricular internal dimension; LVIDs, end-systolic left ventricular internal dimension; LVPWs, systolic left ventricular posterior wall thickness; % FS, percentage of fractional shortening; ^#P < 0.01.

Figure 5.

Disrupted ultrastructure in Cypher knockout (CKO) mouse heart. Transmission electron microscopy showing some disorganized Z-lines in cardiomyocytes from CKO mice at 1 month of age while most Z-line structures remain intact (B) compared with the well-organized and condensed Z-line structures in control littermates (A). At 3 months of age, Z-lines are completely disrupted in cardiomyocytes from CKO mice (D) when compared with normal Z-lines in control littermates (C). Magnified figures to show the dispersed and punctured Z-lines better in a 3-month-old CKO mouse (inducible cardiac-specific CKO mice) compared with intact Z-lines in the control littermate (E).

Figure 6.

Cypher's PDZ domain interacts with calsarcin-1 and myotilin. (A) Deletion of the last residue leucine in calsarcin-1 (DL) and myotilin (DL) results in nearly complete ablation of their interaction with Cypher's PDZ domain in the yeast two-hybrid system. The strength of the interactions was quantified in a liquid α-galactosidase assay using PNP-α-gal as a substrate. (B) Schematic presentation of HA-tagged calsarcin-1 (HA-calsarcin) and myotilin (HA-myotilin) and Flag-tagged fragments encoding Cypher2C (Flag-Cypher), truncated Cypher2C containing the PDZ domain (Flag-PDZ), and PDZ-less Cypher2C (Flag-ΔPDZ). (C) In vivo immunoprecipitation of Cypher with calsarcin-1 and myotilin showing the interaction of Cypher with calsarcin-1 and myotilin through its PDZ domain. (D) Western blot analysis showing protein levels of Cypher, calsarcin-1 and myotilin in hearts from wild-type and conventional CKO mice at embryonic day 17.5.

Figure 7.

Alterations of signals in Cypher-deficient heart. (A) Dot-blot assay showing altered expression of fetal genes in Cypher knockout (CKO) mouse compared with control littermate mouse hearts at the indicated ages. (B) Western blot analyses for activation of proteins involved in intracellular signaling in mouse hearts from CKO and control littermates at 2 months of age as well as in (C) Inducible cardiac-specific CKO mice (ICKO) and control littermates 1 month after tamoxifen injection.