Genotype-specific pathogenic effects in human dilated cardiomyopathy

Abstract

Key points: Mutations in genes encoding cardiac troponin I (TNNI3) and cardiac troponin T (TNNT2) caused altered troponin protein stoichiometry in patients with dilated cardiomyopathy. TNNI3_p.98trunc resulted in haploinsufficiency, increased Ca²⁺ -sensitivity and reduced length-dependent activation. TNNT2_p.K217del caused increased passive tension. A mutation in the gene encoding Lamin A/C (LMNA_p.R331Q ) led to reduced maximal force development through secondary disease remodelling in patients suffering from dilated cardiomyopathy. Our study shows that different gene mutations induce dilated cardiomyopathy via diverse cellular pathways.

Abstract: Dilated cardiomyopathy (DCM) can be caused by mutations in sarcomeric and non-sarcomeric genes. In this study we defined the pathogenic effects of three DCM-causing mutations: the sarcomeric mutations in genes encoding cardiac troponin I (TNNI3_{p.98truncation} ) and cardiac troponin T (TNNT2_{p.K217deletion} ; also known as the p.K210del) and the non-sarcomeric gene mutation encoding lamin A/C (LMNA_p.R331Q ). We assessed sarcomeric protein expression and phosphorylation and contractile behaviour in single membrane-permeabilized cardiomyocytes in human left ventricular heart tissue. Exchange with recombinant troponin complex was used to establish the direct pathogenic effects of the mutations in TNNI3 and TNNT2. The TNNI3_p.98trunc and TNNT2_p.K217del mutation showed reduced expression of troponin I to 39% and 51%, troponin T to 64% and 53%, and troponin C to 73% and 97% of controls, respectively, and altered stoichiometry between the three cardiac troponin subunits. The TNNI3_p.98trunc showed pure haploinsufficiency, increased Ca²⁺ -sensitivity and impaired length-dependent activation. The TNNT2_p.K217del mutation showed a significant increase in passive tension that was not due to changes in titin isoform composition or phosphorylation. Exchange with wild-type troponin complex corrected troponin protein levels to 83% of controls in the TNNI3_p.98trunc sample. Moreover, upon exchange all functional deficits in the TNNI3_p.98trunc and TNNT2_p.K217del samples were normalized to control values confirming the pathogenic effects of the troponin mutations. The LMNA_p.R331Q mutation resulted in reduced maximal force development due to disease remodelling. Our study shows that different gene mutations induce DCM via diverse cellular pathways.

Keywords: dilated cardiomyopathy; heart failure; protein phosphorylation; troponin.

Figure 1. Schematic representation of the troponin complex

cTnT is shown in yellow, cTnI in blue and cTnC in red. The letters N and C indicate the N‐ and C‐terminus, respectively. The upper diagram shows the troponin complex in the presence of Ca²⁺ while the lower diagram shows the troponin complex without Ca²⁺. Location of the studied mutations are indicated with stars and an arrow in the upper panel. The letter H indicates a helix structure. IR, inhibitory region.

Figure 2. Baseline contractile properties

A, F _max, measured at pCa 4.5, was significantly decreased (P < 0.05) in LMNA_p.R331Q samples (17.9 ± 1.6 kN m⁻², N = 3, n = 19) compared to controls (28.2 ± 2.1 kN m⁻², N = 6, n = 21) while IDCM (27.5 ± 2.3 kN m⁻², N = 5, n = 18), TNNI3_p.98trunc (26.8 ± 3.2 kN m⁻², N = 1, n = 13) and TNNT2_p.K217del (31.4 ± 3.9 kN m⁻², N = 1, n = 14) samples showed similar F _max as controls. Data obtained from Hoorntje et al. (2016). B, F _pass, measured at pCa 9.0, in membrane‐permeabilized cardiomyocytes of IDCM (N = 5, n = 19), were similar compared to control (N = 4, n = 10). C, F _pass, measured at pCa 9.0, in membrane‐permeabilized cardiomyocytes of TNNI3_p.98trunc sample (N = 1, n = 6), were similar compared to control (N = 4, n = 10). D, F _pass, measured at pCa 9.0, in membrane‐permeabilized cardiomyocytes of TNNT2_p.K217del patient (N = 1, n = 8), was significantly increased (P < 0.01) compared to control (N = 4, n = 10). E, F _pass, measured at pCa 9.0, in LMNA_p.R331Q samples (N = 3, n = 12), were similar compared to control (N = 4, n = 10). F, Ca²⁺‐sensitivity was non‐significantly increased and ΔEC₅₀ was non‐significantly reduced in IDCM (N = 5, n = 11) compared to control (N = 6, n = 13). G, Ca²⁺‐sensitivity was significantly increased (P < 0.05) in TNNI3_p.98trunc patient (N = 1, n = 7) compared to control (N = 6, n = 13), ΔEC₅₀ was non‐significantly reduced. H, Ca²⁺‐sensitivity was only slightly and non‐significantly reduced compared to controls and ΔEC₅₀ was preserved in TNNT2_p.K217del sample (N = 1, n = 7) compared to control (N = 6, n = 13). I, Ca²⁺‐sensitivity was significantly increased (P < 0.01) in LMNA_p.R331Q samples (N = 3, n = 7) compared to control (N = 6, n = 13) while ΔEC₅₀ was preserved. N, number of samples; n, number of total cardiomyocytes measured.

Figure 3. Expression of troponin in troponin mutants

A and B, cTnI levels measured with an antibody directed to the N‐terminal of cTnI and normalized to GAPDH were decreased to 46% in the TNNI3_p.98trunc (0.28) and to 40% in TNNT2_p.K217del (0.24) samples compared to controls (N = 8, mean = 0.60, CI = 0.43–0.77). A, corresponding gel image showed no additional bands indicative of a truncated cTnI protein. C and D, cTnI levels were decreased to 39% in TNNI3_p.98trunc (0.34) and to 51% in TNNT2_p.K217del (0.44) samples compared to controls (N = 8, mean = 0.87, CI = 0.59–1.15) when normalized for α‐actinin. E and F, cTnT levels normalized to α‐actinin were also decreased to 64% in TNNI3_p.98trunc (0.70) and to 53% in TNNT2_p.K217del (0.59) samples compared to controls (N = 8, mean = 1.11, CI = 0.83–1.38). G and H, cTnC levels normalized to α‐actinin were slightly decreased to 73% in TNNI3_p.98trunc sample (0.66) but still within the 95% CI of controls (N = 8, mean = 0.91, CI = 0.67–1.15). TNNT2_p.K217del showed normal (0.88, 97% of controls) cTnC levels. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 4. Secondary disease remodelling and direct mutation effects

A, phos‐tag analysis showed separation of non‐ (0P), mono‐ (1P) and bis‐ (2P) phosphorylated cTnI. B, phosphorylation of cTnI was increased in TNNI3_p.98trunc and TNNT2_p.K217del samples compared to controls (N = 7) while cTnI phosphorylation in LMNA_p.R331Q (N = 3) and IDCM (N = 3) was decreased compared to controls. C, Ca²⁺‐sensitivity was normalized in IDCM cardiomyocytes (N = 5, n = 12) compared to control cardiomyocytes (N = 6, n = 14) after incubation with exogenous PKA. D, Ca²⁺‐sensitivity was normalized in LMNA_p.R331Q cardiomyocytes (N = 3, n = 7) compared to control cardiomyocytes (N = 6, n = 14) after incubation with exogenous PKA. E, after incubation with exogenous PKA, Ca²⁺‐sensitivity in TNNI3_p.98trunc cardiomyocytes (N = 1, n = 7) remained significantly increased (P < 0.01) compared to control cardiomyocytes (N = 6, n = 14). F, exchange with WT troponin complex restored cTnI levels in the TNNI3_p.98trunc sample to 83% of that of controls exchanged with WT troponin complex (H). G, phos‐tag gel analysis showed high phosphorylation of native troponin complex prior to exchange (NE) and incorporation of unphosphorylated recombinant protein after exchange. H, the 83% was composed of 46% recombinant troponin and 37% native troponin in the TNNI3_p.98trunc sample compared with 43% recombinant troponin in the control exchanged with WT troponin complex. I, Ca²⁺‐sensitivity and LDA were restored in TNNI3_p.98trunc cardiomyocytes (N = 1, n = 9) compared to control (N = 2, n = 11) after exchange with WT troponin complex and incubation with exogenous PKA. N, number of samples; n, number of total cardiomyocytes measured.

Figure 5. Alterations in titin isoform composition and phosphorylation in DCM mutants

A, titin isoforms, N2BA and N2B, separated by agarose gel electrophoresis. B, titin N2BA/N2B ratios were increased in IDCM (0.99 ± 0.20, N = 5), TNNI3_p.98trunc (0.78, N = 1), TNNT2_p.K217del (0.82, N = 1) and LMNA_p.R331Q (0.99 ± 0.38, N = 3) samples compared to controls (0.50 ± 0.02, N = 12, CI = 0.49–0.55). C, phosphorylated Ser4010 compared to total titin levels. D, titin phosphorylation at Ser4010 was decreased in IDCM (N = 5), TNNI3_p.98trunc (N = 1) and LMNA_p.R331Q (N = 3) compared to controls (N = 10, CI = 0.93–1.17), while TNNT2_p.K217del (N = 1) showed slight increased phosphorylation of Ser4010 compared to control. E, phosphorylated Ser12022 compared to total titin levels. F, titin phosphorylation at Ser12022 was decreased in TNNI3_p.98trunc (N = 1), TNNT2_p.K217del (N = 1) and LMNA_p.R331Q (N = 3), compared to control (N = 9, CI = 0.58–1.66), while phosphorylation at Ser12022 was within the 95% CI of controls in IDCM (N = 5). G, phosphorylated Ser11878 compared to total titin levels. H, titin phosphorylation at Ser11878 was decreased in TNNI3_p.98trunc (N = 1), TNNT2_p.K217del (N = 1) and LMNA_p.R331Q (N = 3), compared to control (N = 11, CI = 0.62–1.49) while phosphorylation at Ser11878 was within the 95% CI of controls in IDCM (N = 5).

Figure 6. TNNT2_p.K217del increases passive tension

A, F _pass remained significantly increased (P < 0.0001) in TNNT2_p.K217del cardiomyocytes (N = 1, n = 9) compared to controls (N = 4, n = 13) after incubation with exogenous PKA. B, phos‐tag gel analysis showed high phosphorylation of native troponin complex prior to exchange (NE) and incorporation of unphosphorylated recombinant protein after exchange. C, after exchange 43% of present troponin complex in controls was recombinant WT cTnI while in the TNNT2_p.K217del sample this was 59% and in control exchanged with TNNT2_p.K217del mutant troponin complex this was 34%. D, upon exchange with WT troponin complex, cardiomyocytes of TNNT2_p.K217del (N = 1, n = 11) showed restoration of F _pass compared to controls exchanged with WT troponin complex (N = 2, n = 8) while F _pass was significantly increased (P = 0.001) in control cardiomyocytes exchanged with mutant TNNT2_p.K217del troponin complex (N = 2, n = 7). E, after incubation with exogenous PKA, cardiomyocytes of TNNT2_p.K217del exchanged with recombinant WT troponin complex (N = 1, n = 13) showed normalization of F _pass compared to control cardiomyocytes exchanged with WT troponin complex (N = 2, n = 9) while F _pass was significantly increased (P < 0.0001) in control cardiomyocytes exchanged with mutant TNNT2_p.K217del troponin complex (N = 2, n = 7). N, number of samples; n, number of total cardiomyocytes measured.

Figure 7. Overview of pathogenic effects of TNNI3_p.98trunc, TNNT2_p.K217del and LMNA_p.R331Q

The TNNI3_p.98trunc mutation did not result in a truncated protein and instead caused haploinsufficiency leading to increased Ca²⁺‐sensitivity and impaired LDA. The TNNT2_p.K217del mutation might act as a poison peptide and caused decreased Ca²⁺‐sensitivity as shown by others. We showed that the sample with TNNT2_p.K217del mutation resulted in decreased expression of the troponin proteins and in addition has a poison peptide effect. Since the decreased expression of the troponin proteins increased Ca²⁺‐sensitivity and the poison peptide decreased Ca²⁺‐sensitivity, there was no significant change in Ca²⁺‐sensitivity in the TNNT2_p.K217del sample. In addition, F _pass was increased. The LMNA_p.R331Q mutation caused decreased myofibril density and subsequent impaired contractility.