The role of cardiac troponin T quantity and function in cardiac development and dilated cardiomyopathy

Abstract

Background: Hypertrophic (HCM) and dilated (DCM) cardiomyopathies result from sarcomeric protein mutations, including cardiac troponin T (cTnT, TNNT2). We determined whether TNNT2 mutations cause cardiomyopathies by altering cTnT function or quantity; whether the severity of DCM is related to the ratio of mutant to wildtype cTnT; whether Ca(2+) desensitization occurs in DCM; and whether absence of cTnT impairs early embryonic cardiogenesis.

Methods and findings: We ablated Tnnt2 to produce heterozygous Tnnt2(+/-) mice, and crossbreeding produced homozygous null Tnnt2(-/-) embryos. We also generated transgenic mice overexpressing wildtype (TG(WT)) or DCM mutant (TG(K210Delta)) Tnnt2. Crossbreeding produced mice lacking one allele of Tnnt2, but carrying wildtype (Tnnt2(+/-)/TG(WT)) or mutant (Tnnt2(+/-)/TG(K210Delta)) transgenes. Tnnt2(+/-) mice relative to wildtype had significantly reduced transcript (0.82+/-0.06[SD] vs. 1.00+/-0.12 arbitrary units; p = 0.025), but not protein (1.01+/-0.20 vs. 1.00+/-0.13 arbitrary units; p = 0.44). Tnnt2(+/-) mice had normal hearts (histology, mass, left ventricular end diastolic diameter [LVEDD], fractional shortening [FS]). Moreover, whereas Tnnt2(+/-)/TG(K210Delta) mice had severe DCM, TG(K210Delta) mice had only mild DCM (FS 18+/-4 vs. 29+/-7%; p<0.01). The difference in severity of DCM may be attributable to a greater ratio of mutant to wildtype Tnnt2 transcript in Tnnt2(+/-)/TG(K210Delta) relative to TG(K210Delta) mice (2.42+/-0.08, p = 0.03). Tnnt2(+/-)/TG(K210Delta) muscle showed Ca(2+) desensitization (pCa(50) = 5.34+/-0.08 vs. 5.58+/-0.03 at sarcomere length 1.9 microm, p<0.01), but no difference in maximum force generation. Day 9.5 Tnnt2(-/-) embryos had normally looped hearts, but thin ventricular walls, large pericardial effusions, noncontractile hearts, and severely disorganized sarcomeres.

Conclusions: Absence of one Tnnt2 allele leads to a mild deficit in transcript but not protein, leading to a normal cardiac phenotype. DCM results from abnormal function of a mutant protein, which is associated with myocyte Ca(2+) desensitization. The severity of DCM depends on the ratio of mutant to wildtype Tnnt2 transcript. cTnT is essential for sarcomere formation, but normal embryonic heart looping occurs without contractile activity.

Figure 1. Generation of Tnnt2 ^{+/ −} mice.

A. A targeting construct containing a neomycin resistance gene (neo/zeo) between loxP sites (triangles) was introduced into the murine Tnnt2 locus by homologous recombination in murine ES cells, ablating the 3′ segment of the gene, including exon 14. ES cells were microinjected into mouse blastocysts to generate chimeras, which were bred with wildtype mice for germline transmission of the Tnnt2 ablation. The neomycin resistance gene was excised using Cre-mediated excision by mating with EIIa-Cre recombinase mice. Horizontal arrows indicate PCR primers (F1, F2, R1, R2, R3) used for genotyping as described in “Materials and Methods.” B. Genomic DNA from mice prior to Cre-mediated excision of the neomycin resistance gene was digested with Hind III and Southern blotted with the probe indicated in panel A to demonstrate homologous recombination. A Hind III restriction site in the neomycin resistance gene produced a smaller restriction product in the presence of homologous recombination. WT, wildtype; TK, thymidine kinase gene.

Figure 2. Tnnt2 gene expression in mouse lines.

A. Northern blots of total cardiac RNA from wildtype (WT) and Tnnt2 heterozygous ablated (Tnnt2 ^+/−) mice using probes complementary to the coding sequence and the 3′ untranslated region (UTR) of the Tnnt2 transcript, and the GAPDH transcript as a loading control. A mild deficit in Tnnt2 transcript, quantified at 18% by densitometry, was apparent in Tnnt2 ^+/− mice. B. Northern blots of total cardiac RNA from mice of the indicated genotypes using a probe complementary to the coding sequence of the Tnnt2 transcript, comprising both endogenous and transgene transcript (total); and a probe complementary to the human growth hormone 3′ UTR, specific to the transgene transcript. The intensity of 18S and 28S rRNA bands from ethidium bromide stained agarose gels was used to quantify RNA loading. Significant increases in Tnnt2 transcript were apparent in mice carrying the wildtype (TG^WT) or the K210Δ Tnnt2 (TG^K210Δ) transgene. C. Reverse transcription-PCR and sequencing of cardiac Tnnt2 mRNA showing deletion of codon AAG encoding lysine at position 210 in hearts from TG^K210Δ and Tnnt2 ^+/−/TG^K210Δ, but not wildtype (WT), mice. D. Immunoblots of total protein extracts from the hearts of wildtype (WT) and Tnnt2 ^+/− mice using antibodies specific for cardiac troponin T (cTnT), troponin C (TnC), troponin I (cTnI), tropomyosin, actin, and α myosin heavy chain (MHC). GAPDH loading control immunoblots are shown corresponding to the membranes used for the immunoblots above. Levels of these sarcomeric proteins were unchanged between genotypes. E. An immunoblot of total protein extracts from the hearts of three wildtype, Tnnt2 ^+/−, Tnnt2 ^+/−/TG^K210Δ, and Tnnt2 ^+/−/TG^WT mice was performed using an antibody specific for cardiac troponin T (cTnT). cTnT protein levels were unchanged among all genotypes. Electrophoresis of these protein extracts on a polyacrylamide gel, followed by Coomassie blue staining, was used to correct for the relative quantity of protein loaded for the immunoblot.

Figure 3. Hematoxylin and eosin staining of hearts from wildtype and Tnnt2 ^{+/ −}/TG^K210Δ mice.

Hematoxylin and eosin (H&E) stained tissue is shown at low (2X) and high (10X) magnification, and Masson's trichrome (MT) stained tissue is shown at high (10X) magnification. Massive dilation of the mutant heart was apparent.

Figure 4. Electrophysiological study in a Tnnt2 ^{+/ −}/TG^K210Δ mouse.

A surface electrocardiogram (ECG) and an intracardiac electrogram are shown of a Tnnt2 ^+/−/TG^K210Δ mouse which had an inducible arrhythmia. Ventricular tachycardia (VT) was inducible by a drive train at 80 ms cycle length, which self-terminated after approximately four seconds. NSR, normal sinus rhythm.

Figure 5. Normalized force-pCa relationships in Tnnt2 ^{+/ −}/TG^WT and Tnnt2 ^{+/ −}/TG^K210Δ skinned papillary muscle fibers.

Normalized force (i.e., ratio of force at a given pCa and maximally activated force at pCa = 4.33) developed at a range of Ca²⁺ concentrations was assessed at sarcomere lengths 1.9 µm (Tnnt2 ^+/−/TG^WT •, Tnnt2 ^+/−/TG^K210Δ ▪) and 2.3 µm (Tnnt2 ^+/−/TG^WT ○, Tnnt2 ^+/−/TG^K210Δ □). There was a rightward shift of the force-pCa curve in Tnnt2 ^+/−/TG^K210Δ muscle, indicating Ca²⁺ desensitization. Values are mean±SE (n = 10 for Tnnt2 ^+/−/TG^WT and n = 7 for Tnnt2 ^+/−/TG^K210Δ).

Figure 6. Morphology of Tnnt2 ^+/+ (wildtype), Tnnt2 ^{+/ −}, and Tnnt2 ^−/− embryos.

A. Column 1, representative embryos at age 9.5 days postcoitum. Columns 2–3, whole embryos and hearts stained with H&E at low and high power respectively. Columns 4–6, transmission electron microscopy of hearts at increasing magnifications. Whereas Tnnt2 ^+/− embryos appeared normal, Tnnt2 ^−/− embryos showed pericardial effusions on gross inspection, thinning of the myocardium on H&E histology, and loss of organized sarcomeres on electron microscopy. B. A northern blot of total pooled RNA from five embryos of each genotype (Tnnt2 ^+/+, Tnnt2 ^−/−, and Tnnt2 ^+/−) hybridized with a probe complementary to the Tnnt2 transcript showed absence of transcript in Tnnt2 ^−/− embryos and a 29% deficit of transcript in Tnnt2 ^+/− relative to Tnnt2 ^+/+ embryos.