Deleting Full Length Titin Versus the Titin M-Band Region Leads to Differential Mechanosignaling and Cardiac Phenotypes

Abstract

Background: Titin is a giant elastic protein that spans the half-sarcomere from Z-disk to M-band. It acts as a molecular spring and mechanosensor and has been linked to striated muscle disease. The pathways that govern titin-dependent cardiac growth and contribute to disease are diverse and difficult to dissect.

Methods: To study titin deficiency versus dysfunction, the authors generated and compared striated muscle specific knockouts (KOs) with progressive postnatal loss of the complete titin protein by removing exon 2 (E2-KO) or an M-band truncation that eliminates proper sarcomeric integration, but retains all other functional domains (M-band exon 1/2 [M1/2]-KO). The authors evaluated cardiac function, cardiomyocyte mechanics, and the molecular basis of the phenotype.

Results: Skeletal muscle atrophy with reduced strength, severe sarcomere disassembly, and lethality from 2 weeks of age were shared between the models. Cardiac phenotypes differed considerably: loss of titin leads to dilated cardiomyopathy with combined systolic and diastolic dysfunction-the absence of M-band titin to cardiac atrophy and preserved function. The elastic properties of M1/2-KO cardiomyocytes are maintained, while passive stiffness is reduced in the E2-KO. In both KOs, we find an increased stress response and increased expression of proteins linked to titin-based mechanotransduction (CryAB, ANKRD1, muscle LIM protein, FHLs, p42, Camk2d, p62, and Nbr1). Among them, FHL2 and the M-band signaling proteins p62 and Nbr1 are exclusively upregulated in the E2-KO, suggesting a role in the differential pathology of titin truncation versus deficiency of the full-length protein. The differential stress response is consistent with truncated titin contributing to the mechanical properties in M1/2-KOs, while low titin levels in E2-KOs lead to reduced titin-based stiffness and increased strain on the remaining titin molecules.

Conclusions: Progressive depletion of titin leads to sarcomere disassembly and atrophy in striated muscle. In the complete knockout, remaining titin molecules experience increased strain, resulting in mechanically induced trophic signaling and eventually dilated cardiomyopathy. The truncated titin in M1/2-KO helps maintain the passive properties and thus reduces mechanically induced signaling. Together, these findings contribute to the molecular understanding of why titin mutations differentially affect cardiac growth and have implications for genotype-phenotype relations that support a personalized medicine approach to the diverse titinopathies.

Keywords: cardiomyopathies; heart diseases; hypertrophy; models, animal; muscles; myocardial contraction.

Figure 1.. Failure to thrive and early lethality of titin knockouts deficient in titin (E2) versus titin’s M-band region (M1/2).

A) Schematic illustration of titin’s first and last exons of the wildtype, exon 2 (E2) and M-band knockout (M1/2). Triangles indicate lox sites (locus of X(cross)-over). B) E2-KO and M1/2-KO mice die from 10 and 15 days of life, respectively with no survivors beyond day 40, while control mice (CTRL) survive. C) Compared to the respective control without the Cre recombinase, E2-KO and M1/2-KO mice appear smaller and weaker at 35 days. Differences in posture include kyphosis and outward extended hindlimbs. D) Body weight does not increase from week two after birth and stays at ~10 g for either knockout model. E) Muscle strength as determined in an inverted screen grid holding test is severely reduced in E2 and M1/2-KO mice and precedes the reduced weight gain and muscle atrophy. Log rank (Mantel Cox) test, *** p < 0.001, n = 23 control, n = 14 for E2, n = 9 for M1/2 knockout (B). Repeated measure 2-way ANOVA with Bonferroni post-test, ** p < 0.01; *** p < 0.001, n = 12 controls, n = 6 per knockout (D, E).

Figure 2.. Progressive loss of titin causes skeletal muscle atrophy and sarcomere disassembly in E2 and M1/2 knockout mice.

A, B) Deletion of exon 2 in E2-KO (E2) and exons 358, 359 in M1/2-KO mice (M1/2) vs control (CTRL) was confirmed by real-time qPCR on cDNA derived from left ventricular mRNA of 35-day-old mice with probe-sets on exon 2 to 3 (A) and exons 357 to 358 (B). 1-way ANOVA with Tukey post-test, *** p < 0.001. n = 5 per genotype. C) Agarose gel electrophoresis of quadriceps protein from 5-week-old E2- and M1/2-KO mice. T1-M is the M-band deficient titin. T2 is the proteolytic product of T1. D) Total titin (TTN=T1+T2) protein is reduced to <20% in the E2-KO. One-way ANOVA and Tukey post-test, *** p < 0.001, n = 6 animals for control, n=3 for each knockout. E) The truncated T1-M protein in the M1/2-knockout amounts to ~80% of total T1 titin after 5 weeks. F) Reduced quadriceps (Q) muscle weight normalized to tibia length in 35-day-old animals. G) Tibia length is unchanged between control and KOs, n = 12 for controls, n = 6 per knockout. H) Masson Trichrome staining of quadriceps muscle (size bar = 100µm). I) Quantification of fiber size. J) Number of nuclei per field of view (slides from n = 6 controls and n = 3 for each knockout strain). One-Way ANOVA with Bonferroni post-test, ** p < 0.01, *** p < 0.001. K) Disassembly of the sarcomere in both knockout strains visualized by electron microscopy with partially remaining Z-discs and disorder at M- and A/I-band (size bar =1µm).

Figure 3.. Complete titin loss leads to DCM while loss of M-band titin results in cardiac atrophy at 5 weeks.

A) Agarose protein gel electrophoresis of the left ventricle (LV) from control, E2- and M1/2-KO vs. control (E2, M1/2, CTRL) at 5-weeks. B) The total titin (TTN)/MHC protein ratio is reduced in the E2-KO; n = 6 control, n = 3 per KO. C) Titin N2BA and N2B isoform expression is unchanged in KO mice. In the M1/2-KO, the corresponding truncated isoforms amount to >50%; n = 6 for control n = 3 per KO. D) The heart of 5-weeks-old E2-KO mice is enlarged, while the deletion of titin’s M-band leads to cardiac atrophy (size bar = 1 mm). E, F) Heart weight/ body weight ratio (HW/BW) is increased in E2-KO, but body weight (BW) is reduced in both knockout strains. G) Normalization of the heart weight to the tibia length reflect the atrophy of the M1/2-KO; n = 13 control, n = 7 per KO. H) Masson trichrome staining of the heart from 5-weeks-old animals with DCM in the E2 knockout and atrophy of the M1/2 knockout heart (arrows: area of higher magnification in I; size bar = 1 mm). I) Higher magnification from H with increased number of nuclei (arrowheads indicate nuclei; size bar = 100 µm). J) Disassembly of the sarcomere in both strains. Electron microscopy of left ventricular tissue from 3-week-old animals (size bar = 1µm). Data in B, E, F and G were analyzed for statistical significance with one way ANOVA and Tukey post-test; *** p < 0.001

Figure 4.. Systolic and diastolic dysfunction predominantly affects the E2-KO at 4 weeks of age.

A) Left ventricular (LV) diameter in diastole (dia) is increased in E2- but decreased in M1/2-KO mice vs. control (E2, M1/2, CTRL). B, C) LV diameter in systole (sys) is increased and ejection fraction (EF) is reduced only in the E2-KO. D) E to A ratio is increased in E2-KO only. E) Doppler traces with similar E and A waves in control and M1/2 vs. E2 with a blunted A wave. F, G) Total passive stress of multicellular myocyte preparations is decreased in E2-KO, but not in M1/2-KO. The individual contribution of titin vs. collagen based stress is provided in Supplemental Figure 4B, C. Data in A-E were analyzed for statistical significance with Student’s T-test, n = 7–9. F and G were analyzed for significance with a mono-exponential curve fit and an extra sum of squares F-test, n = 7 animals per group; * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Figure 5.. Expression analysis of titin related kinases, chaperone, and anchoring proteins.

Western blot analysis of heart lysate from control (CTRL) and E2-KO (left) and control and M1/2-KO (right) at 4 weeks of age. A, B) The tinin kinase PKCα is upregulated only in the E2-KO. C-F) The stress response proteins CryAB and ANKRD1 are upregulated in both KO strains, but CryAB to a lesser degree in the M1/2-KO. G-L) At 4 weeks of age, there is no significant deregulation of proteins that anchor titin at the Z-disc (T-cap) or the M-band (myomesin 1), but the Z-disc signaling protein MLP is upregulated in both KO strains. M, N). Expression of the extra-sarcomeric protein desmin is not affected in either KO. Western blot of left ventricular tissue lysate was quantified by densitometry; n = 8-11 animals per group. Statistical significance was calculated using Student’s T-test, * p < 0.05; ** p < 0.01; *** p < 0.001;**** p < 0.0001. f.c. fold change

Figure 6.. Expression of titin binding and stretch signaling proteins.

Western blot analysis of heart lysate from control (CTRL) and E2-KO (left) and control and M1/2-KO (right) at 4 weeks of age. A-D) FHL1 is upregulated in both strains, while FHL2 is only upregulated in the E2-KO. E, F) ERK2 is upregulated in the E2 and to a lesser degree in the M1/2-KO, while ERK1 is not changed. G, H) CamKIIδ is significant upregulated in the E2 and to a lesser degree in the M1/2-KO. I, J) The titin binding protein p62 is upregulated in the E2-, but unchanged in the M1/2-KO. K, L) Nbr1 is unchanged between genotypes, except for the 120 kDa isoform, which is induced almost 8-fold in the E2-KO only. Student’s T-test, * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001; n = 8-11 animals per group. f.c. fold change

Figure 7.. Titin based mechanotransduction.

A) Schematic representation of the sarcomeric and molecular phenotypes of wildtype (WT), E2-KO, and M1/2-KO mice with titin in black, actin in red, and myosin in green. There are fewer titin molecules in the E2 knockout, while truncated titin does not properly integrate into the M1/2 knockout sarcomere. This leads to destabilization of the sarcomere and altered expression of titin binding proteins. Proteins which are co-regulated in both knockouts compared to the wildtype are depicted on the left, differentially regulated proteins on the right. Arrows indicate regulation of expression. B) Model of FHL2 regulation dependent on mechanical strain. Wildtype titin (WT) binds FHL2 and stabilizes the protein. Upon loss of titin in the E2-KO, we suggest that the remaining titin molecules are stretched extensively providing additional FHL2 binding sites to translate strain into a trophic stress signal. E2-KO hearts thus develop dilated cardiomyopathy (DCM). In the M-band KO, the truncated titin protein is less well integrated and should thus experience less strain resulting in reduced binding of FHL2. The remaining titin is stretched less than the remaining titin in E2 knockouts based on the mechanical contribution of truncated titin. The resulting reduction in stress signaling causes atrophy in M1/2-KO hearts.