C57BL/6N mice carrying two point mutations in the Tcap gene to ablate telethonin phosphorylation sites (S157A/S161A) were generated by Taconic (Cologne) (Supplementary Methods). Mice were housed in pathogen-free, individually ventilated cages with a 12 h light-dark cycle, controlled humidity and temperature (20–22°C), fed standard mouse chow, and water ad libitum. The study was approved by the local Ethics Review Board and animal handling and experiments were performed according to the Home Office regulations, as detailed in the Home Office Guidance on the Operation of the Animals (Scientific Procedures) Act 1986, HMSO (London), under Directive 2010/63/EU (PPL 70/8948). Animals were euthanized by cervical dislocation under isoflurane-induced anesthesia (3–4% isoflurane via inhalation) according to Schedule 1 United Kingdom Animals (Scientific Procedures) Act 1986.

Echocardiography was performed as described previously (Puhl et al., 2016), using the Visual Sonics Vevo^® 770 imaging system (Scanhead: RMV707B, 15–45 MHz, cardiac mouse). Anesthesia was maintained via nose-cone inhalation of 1.5–2.0% isoflurane throughout the procedure. Body temperature was monitored and maintained at 37 ± 1.5°C to sustain a heart rate of approximately 400–600 bpm (average 522 ± 52.87 bpm for baseline measurements, recorded range 399–636 bpm).

Osmotic mini pumps (Model 1002, Alzet) were implanted in male and female mice at 8–9 weeks of age, as described previously (Puhl et al., 2016). Anesthesia was maintained via nose-cone inhalation of 3–4% isoflurane throughout the procedure. Vetergesic (buprenorphine hydrochloride; 0.03 mg/ml) was administered via single intraperitoneal injection during mini-pump implantation as a post-operative analgesic. Mice were randomly allocated to receive either a pump containing 0.9% saline (vehicle control) or isoprenaline hydrochloride (Sigma I5627), dissolved in 0.9% saline and delivered at a dose of 30 mg/kg/day for 14 days. Mice were imaged via echocardiography 1 day prior to the implantation of the mini-pump and additionally after 14 days of infusion.

Echocardiographic long and short axis one-dimensional M-mode projections were taken at baseline and 2–3 min after bolus intraperitoneal injections of 0.9% saline then isoprenaline hydrochloride (Sigma I5627), dissolved in 0.9% saline to deliver a dose of 0.1 mg/kg body weight. Isoprenaline injections were administered 10 min after the administration of the 0.9% saline control.

Following excision, cardiac tissues were frozen in isopentane (chilled using liquid nitrogen) and 10 μm sections were cut using a −20°C cryostat (Leica). The sections were fixed in cold acetone for 5 min at −20°C, washed in PBS and blocked in 10% normal goat serum/PBS for 1 h at RT. Sections were incubated with primary antibodies for anti-telethonin (1:250, Abcam 133646) and anti-Titin-T-12 (1:10, gift from Dr. Elisabeth Ehler, KCL) overnight at 4°C, followed by fluorescently conjugated anti-mouse Cy3 (1:500) and anti-rabbit Alexa Fluor 488 (1:500, Invitrogen) secondary antibodies. The slides were mounted with ProLong Gold Antifade reagent (Invitrogen). Sections were imaged using an inverted confocal microscope (TCS SP5 system, Leica). Alternatively (ISO study), cardiac tissues were fixed in 4% paraformaldehyde at 4°C overnight before processing and embedding in paraffin wax for tissue sectioning. Five micrometers paraffin sections were stained using Alexa Fluor 488 conjugated wheat germ agglutinin (WGA; Invitrogen) to identify myocyte cell borders or Picro Sirius Red Stain Kit (Polysciences) to detect collagen deposition. The mounted sections were imaged using a confocal microscope (WGA-fluorescent slides) or a Leica DM600B microscope (Picro Sirius Red slides). Transverse-cut myocytes with centrally located nuclei were demarcated manually to calculate myocyte-cross sectional area using NIH ImageJ software. Fibrosis was quantified by using the color-threshold method using NIH ImageJ software.

Proteins were separated by SDS/PAGE and Phos-tag^TM SDS/PAGE as previously described (Kinoshita et al., 2006; Candasamy et al., 2014). For Phos-tag SDS-PAGE, samples were resolved on 10% polyacrylamide gels supplemented with 40 μmol/L phosphate affinity reagent Phos-tag^TM (Wako) and 80 μmol/L MnCl₂. The following primary antibodies were used: anti-telethonin (1:1000, Abcam 133646), anti-α-actinin (1:2000, Abcam 50599), anti-troponin I (1:1000, CST 4002), anti-GAPDH (1:10,000, CST 2118), anti-calsequestrin (1:1000, ThermoFisher PA1-913), and anti-MyBPC (1:90000) (Gautel et al., 1995). HRP-conjugated secondary antibodies anti-rabbit IgG (1:2000, CST, 7074) and anti-rat IgG (1:2000, CST 7077) were used and protein bands were visualized by ECL Western Blotting Detection Reagents (GE Healthcare) on chemiluminescence film. Protein expression was quantified by GS-800 calibrated imaging densitometer (BioRad).

Quantitative-PCR (q-PCR) analysis of Tcap, Pnmt, Col1a1, Col3a1, p53, and p21 was performed using Power SYBR Green PCR Master mix (ThermoFisher) according to manufacturer’s instructions. Five nanograms cDNA/reaction was analyzed in triplicate to quantify gene expression using 7900HT thermal cycler (Applied Biosystems) with the following cycling conditions: 95°C for 10 min, 40 cycles of 95°C for 1 min, 60°C for 15 s. Primer sequences were obtained using PrimerBank¹ and are provided in Supplementary Table 1. Relative quantity was calculated using the 2^–ΔΔCT method, using Hprt or Gapdh as reference genes.

Cardiomyocytes were isolated from hearts of adult male mice by collagenase-based digestion, as described previously (Candasamy et al., 2014), and maintained in M119 culture medium supplemented with 2 mM L-creatine, 5 mM carnitine, 5 mM taurine, and 100 IU/ml penicillin/streptomycin. One hour post plating, myocytes were incubated with modified M199 culture medium containing 10 μmol/L MG132 (Sigma M7449) or vehicle control (DMSO). Cells were treated for 16 h prior to lysis in sample buffer.

Quantitative data are presented as mean ± SD or SE, as indicated. Statistical significance was assessed using Graph-Pad Prism 7.0 software. Data sets were analyzed by unpaired T-test or one or two-way ANOVA, followed by Tukey’s or Sidak’s post hoc test, as appropriate. Differences were considered significant when p < 0.05.

To assess the physiological significance of telethonin phosphorylation in vivo, we generated a KI mouse model in which the Tcap gene was targeted with two single nucleotide substitutions to replace TCT codons for the phospho-acceptor S157 and S161 residues with GCT codons for non-phosphorylatable alanine (Figure 1A illustrates the gene-targeting strategy). Mice heterozygous (HET) or homozygous (HOM) for the mutated Tcap allele (Tcap^S157/161A) were born in a Mendelian ratio, with no evidence of gestational mortality or impaired pre- or post-natal growth. Gravimetric and echocardiographic assessment of body weight, heart weight, cardiac function, and morphology at 16 weeks of age revealed no significant differences between HET and HOM mice and their WT littermates (Figure 1B; Supplementary Table 2). Similar observations were made in both male and female mice (female data not shown). These observations indicate that telethonin phosphorylation is not a critical determinant of viability and cardiac structure and function in mice, under basal conditions.

Telethonin expression level and phosphorylation were assessed in cardiac ventricular homogenates from WT, HET, and HOM mice at 16 weeks of age, by SDS-PAGE or Phos-tag SDS-PAGE followed by immunoblotting (Figure 2A). In WT and HET mice, cardiac telethonin existed predominantly in bis-phosphorylated (2P) form, with mono-phosphorylated (1P) and non-phosphorylated (0P) species present at much lower abundance (Figure 2A); this mirrors our earlier findings in hearts from adult WT mice and rats (Candasamy et al., 2014). In contrast, cardiac telethonin was present exclusively as the 0P species in HOM mice, as expected (Figure 2A). Interestingly, the cardiac abundance of total telethonin protein (comprising all phosphoprotein species present) was partially reduced in HET mice (67.5% expression relative to WT) and dramatically reduced in HOM mice (9.8% expression relative to WT littermates) (Figure 2A). Analysis of telethonin mRNA expression by qPCR revealed no significant difference in transcript abundance between cardiac samples from WT, HET, and HOM mice (Figure 2B). Again, similar observations were made in both male and female mice (Supplementary Figure 3). These findings indicate that ablating the telethonin phosphorylation sites at S157 and S161 has a marked impact on cardiac telethonin protein expression level and that this occurs in the absence of a change in telethonin mRNA abundance, suggesting a post-translational mechanism.

We explored the potential impact of phosphorylation-site ablation on the subcellular distribution of telethonin protein by fluorescence confocal microscopy imaging and biochemical fractionation of cardiac samples from WT and HOM mice. Immunolabelling and confocal microscopy revealed that in ventricular myocardium of WT mice, telethonin was localized at the sarcomeric Z-disc, where it co-localized with the titin N-terminus, as expected (Figure 3A). Although telethonin signal intensity was reduced in sections from HOM hearts, reflecting the marked reduction in protein abundance (Figure 2A), Z-disc localization was not disrupted (Figure 3A). The specificity of immunolabelling with the anti-telethonin antibody was confirmed by negative control experiments with concentration-matched non-immune rabbit IgG (Supplementary Figure 4). Fractionation of myocardial samples revealed that in WT hearts telethonin was present exclusively in the Triton-insoluble fraction, where another Z-disc protein α-actinin was also exclusively localized (Figure 3B). The distribution of telethonin in HOM hearts was similar, although once again the reduced abundance of the protein in these samples was readily apparent (Figure 3B). These data show that the subcellular distribution of telethonin is not influenced by ablation of its phosphorylation sites at S157 and S161.

A potential post-translational mechanism for the reduced telethonin expression level in the hearts of HOM mice is a greater susceptibility of non-phosphorylated telethonin protein to degradation by intracellular pathways. Previous studies indicate that telethonin protein is subject to degradation by the UPS in multiple cell models (Heng et al., 2008; Polge et al., 2018; Jiang et al., 2021). To explore the potential role of the UPS system in regulating telethonin protein expression level in the presence and absence of phosphorylation at S157/S161, we determined the effects of the UPS inhibitor MG132 (Rock et al., 1994) in ventricular myocytes that were isolated from the hearts of WT and HOM mice and maintained in short-term culture. As shown in Figure 4A, MG132 treatment increased the abundance of ubiquitylated proteins, indicating effective inhibition of the proteasomal degradation of such proteins. Importantly, MG132 treatment significantly increased the expression level of S157/161A telethonin protein in myocytes from HOM mice, but not the predominantly bis-phosphorylated endogenous protein in myocytes from WT mice (Figure 4B). These findings indicate that in cardiac myocytes, non-phosphorylated telethonin protein is prone to UPS-mediated clearance at baseline, and by corollary that phosphorylation at S157/S161 may shield telethonin protein from degradation through the UPS.

To explore the potential physiological importance of telethonin phosphorylation under conditions of a sustained increase in hemodynamic load, we explored the responses of WT and HOM mice to a 2-week infusion of the β-adrenergic agonist isoprenaline (ISO), delivered through a subcutaneously implanted osmotic mini-pump (control mice received vehicle). Echocardiographic analysis was performed 1 day prior to mini-pump implantation (Pre) and repeated 2 weeks later (Post). As shown in Figure 5, left ventricular (LV) fractional shortening and ejection fraction were unaltered between the Pre and Post time points in WT and HOM mice receiving vehicle and in WT mice receiving ISO. In contrast, there was a significant decline in both fractional shortening (Figure 5A) and ejection fraction (Supplementary Table 3) in HOM mice receiving ISO. Importantly, there was a comparable chronotropic response to ISO infusion in WT and HOM mice, which displayed significant elevations in heart rate of similar magnitude (Figure 5B). As a result, a significant increase in cardiac output was observed in WT mice in response to ISO treatment, whilst cardiac output was not augmented in the HOM group (Supplementary Table 3). An increase in LV posterior wall thickness was observed in WT mice receiving ISO, however, this was not significant in HOM mice (Figure 5C). Other echocardiographic parameters did not differ significantly between groups (Supplementary Table 3). These findings suggest that ablation of telethonin phosphorylation sites impairs LV systolic function following sustained β-adrenergic stress.

Post-mortem analysis of hearts after the 2-week infusion period revealed a significantly greater heart weight/tibia length ratio in HOM mice that received ISO relative to HOM mice that received vehicle (Figure 6A), indicating ISO-induced cardiac hypertrophy. In contrast, there was no significant difference in heart weight/tibia length ratio between WT mice that received vehicle and those that received ISO (Figure 6A), suggesting that the hypertrophic response to sustained β-adrenergic stress was exaggerated in HOM mice. A similar trend was observed with respect to ISO-induced increases in myocyte cross sectional area, which tended to be greater in HOM mice than in WT mice (Figure 6B), although the inter-group differences were not statistically significant. To explore whether there was any indication of pathological cardiac remodeling in response to ISO infusion in mice of either genotype, we performed histological analysis of ventricular sections using Sirius Red to label collagen. This revealed minimal labeling in WT and HOM mice infused with either vehicle or ISO, indicating the absence of significant cardiac fibrosis (a hallmark of pathological remodeling) in all four groups (Figure 6C). Nevertheless, qPCR analysis revealed small but significant ISO-induced increases in the expression of the collagen genes Col1a1 and Col3a1 in both WT and HOM mice, with no significant difference between the genotypes in the ISO response (Figure 6D). There was no induction of the pro-apoptotic gene p53, or the p53 target gene p21, in response to ISO infusion in either WT or HOM mice (Figure 6D). Taken together, these data indicate that in response to a sustained β-adrenergic stress protocol that induces a mild phenotype in WT mice, HOM mice exhibit a more pronounced hypertrophic response, but without evidence of pathological remodeling.

We also determined the expression level and phosphorylation status of telethonin in ventricular myocardium from WT and HOM mice that received a 2-week infusion of vehicle or ISO. Interestingly, ISO infusion led to a large increase in telethonin protein expression levels in the hearts of WT mice but not those of HOM mice (Figure 7). In contrast, the expression levels of two other sarcomeric proteins, the inhibitory subunit of cardiac troponin (cTnI) and cardiac myosin-binding protein C (cMyBP-C) were unchanged by ISO infusion in both WT and HOM mice (Supplementary Figure 7B). Cardiac telethonin was expressed predominantly in bis-phosphorylated form in WT mice and exclusively in non-phosphorylated form in HOM mice, regardless of the infusion of vehicle or ISO; these findings indicate that phosphorylation at Ser157 and Ser161 is maintained during sustained β-adrenergic stimulation and that such stimulation does not induce telethonin phosphorylation at other sites. Despite the ISO-induced increase in telethonin protein expression level in ventricular myocardium from WT mice (Figure 7), there was no difference in telethonin mRNA expression between the groups (Supplementary Figure 7A). Furthermore, subcellular distribution of telethonin, as assessed by immunolabelling and myocardial fractionation, is not affected despite an increase in telethonin protein expression level in response to ISO infusion (Supplementary Figures 5A,B). These findings reveal that the cardiac response to ISO infusion in WT mice includes a marked increase in the expression level of telethonin protein, but not telethonin mRNA, and that this increase requires that Ser157 and Ser161 are available for phosphorylation.

The differential cardiac responses of WT and HOM mice to sustained β-adrenergic stress led us to determine whether ablation of telethonin phosphorylation sites impacts also on cardiac responses to acute β-adrenergic stimulation. To that end, serial echocardiographic assessment of cardiac function was undertaken in WT and HOM mice at baseline, and 2–3 min after vehicle administration (VEH), then ISO administration (ISO), with both VEH and ISO administered by intraperitoneal injection in an identical volume. As illustrated in Supplementary Figure 6, ISO injection induced comparable increases in fractional shortening, ejection fraction, and heart rate in both WT and HOM mice. These findings indicate that ablation of telethonin phosphorylation sites does not impact on the positive inotropic and positive chronotropic responses of the mouse heart to acute β-adrenergic stimulation in vivo.