Mutations in sarcomere genes are a common cause of familial hypertrophic cardiomyopathy (HCM), a prevalent cardiovascular genetic disorder that affects ~1 in 200–500 asymptomatic young adults in the United States (1–11). Mutations in MYBPC3, which encodes cardiac myosin-binding protein-C (cMyBP-C), a thick-filament cardiac muscle protein that regulates cardiac contractility (6, 7, 12–14), are associated with ~40% of all HCM cases (9, 15), many of which present late in life (16–22). HCM is characterized by asymmetric left ventricular (LV) thickening, diastolic dysfunction, heart failure (HF), and sudden cardiac death (SCD) (4, 23, 24). Affected individuals can remain asymptomatic or present with symptoms, such as dyspnea, angina, or palpitation (9), in early or late adulthood (11, 22). Heterogeneity in the phenotypic expression of disease from MYBPC3 genetic variation requires an understanding of early features, optimal care, and longitudinal follow-up.

The prevalence of HCM may be affected by ancestry (4, 25), but the contribution of ancestry-based genetic variants in the pathogenesis of the disease is not yet well-established. Of the South Asian (SA)-specific variants, a polymorphic 25-base pair (bp) deletion in intron 32 (Δ25bp) of MYBPC3 is present in 4–6% (26–28) of SA individuals and a risk allele for late-onset LV dysfunction, hypertrophy, and HF with multiple forms of cardiomyopathy, such as HCM with an odds ratio of ~7 (26, 29) (Figures 1A,B). Data have shown that asymptomatic MYBPC3^Δ25bp carriers are at risk of late-onset disease progression (16, 22, 26). This variant was defined as a risk allele based on a large case-control study from 6,273 individuals belonging to 107 ethnic populations across 35 Indian states and 2,085 individuals of 63 ethnic/racial groups from 26 countries, including all five continents (26, 30). Nonetheless, in recent studies, MYBPC3^Δ25bp was reported with incomplete penetrance, and the presence of additional genetic or non-genetic risk factors may predispose carriers to cardiomyopathies (26, 27, 31). Our recent investigation has revealed no significant difference in cardiac features between the carrier and non-carrier (NC) cohorts at baseline (27). However, several individual MYBPC3^Δ25bp carriers showed increased fractional shortening at baseline, and this hyperdynamic state is often seen as an early pathological finding in HCM (27). Harper et al. identified an enriched haplotype in SAs with HCM with both MYBPC3^Δ25bp and an associated variant, MYBPC3 c.1224-52G>A (31). However, in Harper's study, only 134 subjects out of 5,394 HCM cases were defined as SAs, and only 17 carried the MYBPC3^Δ25bp variant. We also recently identified a co-segregating novel variant, D389V (MYBPC3^D389V). It was observed in ~10% of MYBPC3^Δ25bp carriers and associated with hyperdynamic cardiac features on baseline echocardiography (27). These data have led to the hypothesis that the risk of HCM is not caused by the MYBPC3^Δ25bp allele alone, but rather conferred by additional rare, pathogenic variants present on the MYBPC3^Δ25bp haplotype. This possible coinheritance of additional risk alleles changes the interpretation of the role of the MYBPC3^Δ25bp in the development of LV dysfunction and HCM. It also means that the pathogenicity of the MYBPC3^Δ25bp variant alone and in the presence of additional modifying risk factors needs further investigation and correlation, considering the growing number of SA carriers with the MYBPC3^Δ25bp variant, predisposing them to severe adverse events, such as SCD, even with the occult clinical phenotype (23).

MYBPC3 gene variants, such as the MYBPC3^Δ25bp, are generally associated with late-disease onset (16, 22, 26). Herein, we continued genetic screening of the United States (US) SA general population for the presence of MYBPC3^Δ25bp and additional rare risk alleles using next-generation sequencing (NGS) technology. To determine the pathogenicity of the MYBPC3^Δ25bp variant, we conducted an exploratory pilot study and prospectively evaluated 10 asymptomatic male carriers of the MYBPC3^Δ25bp variant and 10 age-matched NCs for the changes in cardiac function under exercise stress using submaximal bicycle stress exercise echocardiography (BSE) and continuous cardiac monitoring. We hypothesized that asymptomatic MYBPC3^Δ25bp SA carriers have detectable subclinical risk factors under exercise stress conditions that predispose this group to develop LV hypercontractility and impaired relaxation. Our results suggest that MYBPC3^Δ25bp is consistent with its role as a risk allele for LV dysfunction and cardiomyopathy in SAs.

The present study evaluated the exercise response in asymptomatic MYBPC3^Δ25bp SA carriers to determine the presence of any subclinical features. Overall, our findings determined the presence of hyperdynamic manifestations under exercise conditions in asymptomatic MYBPC3^Δ25bp variant carriers of US SA ancestry as compared to NCs who were presented with significantly higher EF, LVOT, and AV peak velocities, and impaired relaxation presented with a significant difference in the LVID_dia and E/A ratio, but the non-significant difference in the average E/e' ratio. Hyperdynamic features and evidence of cardiac dysfunction were detected by echocardiography in MYBPC3^Δ25bp carriers. These findings are also in line with the phenotype of genotype-positive individuals at stage 1 or those with non-hypertrophic HCM. These individuals can remain asymptomatic or present with subtle echocardiographic phenotype, such as diastolic dysfunction (34–36). In stage 2, however, individuals present with hypertrophy and LV hyperdynamic status (36), whereas MYBPC3^Δ25bp variant carriers in the current study were non-hypertrophic, but did show hyperdynamic phenotype under exercise conditions. As shown in the regression lines, NCs and carriers responded differently to the stress stage, the independent variable, as resistance increased. Starting at 75 W equiv. 4.9 METs, the regression line showed no increase in EF% in NCs after 15 min of exercise. This finding can be explained by the Frank-Starling law, which holds that an increase in workload and venous returns stretches myocardial muscle fibers and increases preload (end-diastolic volume), stroke volume, and, ultimately, cardiac output until maximum capacity is reached. However, the weak increase in LVID_dia from the baseline, as indicated in the regression line for carriers (Figure 4A), could be a contributing factor in maintaining an upward trend in EF% in carriers after 75 w since lower LVID_dia can contribute to lower LV diastolic volume (LVEDV) as resistance augments. However, further studies with an adequate number of subjects are needed to decisively explain this observed phenotypic feature. Collectively, then, these data suggest that exercise stress itself, as a secondary risk factor, potentially triggered hyperdynamic phenotype in asymptomatic MYBPC3^Δ25bp carriers, as compared to NCs, considering that the hemodynamic parameters of MAP and HR did not differ between the two groups (MYBPC3^Δ25bp variant carriers and NCs).

Previous studies have examined the distribution (26, 27, 37, 38) and clinical correlations (26, 27, 31, 39) of the MYBPC3^Δ25bp variant, noting, importantly, that most of these studies were conducted in individuals of SA ancestry. In the first report (2003), the MYBPC3^Δ25bp variant was initially detected in two affected Indian families. Then, it was reported in 16 of 229 unrelated healthy SA individuals and present in 3.8% of the general population, confirming variability in disease penetrance (29). In the next case-control study with a multi-ethnic population of 6,273 individuals in the SA Diaspora, the carrier frequency of the MYBPC3^Δ25bp variant was reported to be 4% (26). Simonson et al. (38) and Bashyam et al. (37) reported 8 and 6%, respectively, in their studies. In 2018, we determined 6% prevalence in a sample taken from individuals of SA ancestry living in the United States (27). However, MYBPC3^Δ25bp is largely associated with incomplete penetrance and delayed onset (26, 27, 29). In the current study, we identified a slightly higher prevalence at 6.75%, but given the nature of our study, this result may have been influenced by ascertainment. Collectively, these data support that MYBPC3^Δ25bp is a common variant and that its distribution is associated with ethnicity (4, 25). Furthermore, our interpolation indicates that a prevalent SA-specific MYBPC3^Δ25bp variant predisposes an estimated ~100 million people of SA ancestry worldwide (26) to such adverse cardiac events as cardiomyopathies, arrhythmias, HF, and SCD.

Previously, we reported no significant difference in clinical characteristics, ECG, or echocardiographic parameters between US SA carriers of the MYBPC3^Δ25bp variant and NCs at rest, except for LV fractional shortening (27), which was slightly higher in carriers as compared to NCs (p = 0.04), suggesting a minimal effect of the MYBPC3^Δ25bp variant on the development of HCM phenotype. Compared to the current study, prior analyses included more individuals (n = 47 carriers) who were younger (47.6 years) and, importantly, included both men and women (27). Both sex inclusion and age may have reduced the ability to detect a hyperdynamic state, which we discovered upon exercise stress. Many affected individuals could also remain asymptomatic until late adulthood (16, 22). Therefore, early detection of any subtle clinical phenotype and/or secondary contributing risk factors triggering symptoms in the carrier population, as we have suggested in our findings, is potentially relevant clinically. In the prior study, we also identified a second co-segregating novel variant (D389V) identified in ~10% of MYBPC3^Δ25bp carriers associated with hyperdynamic cardiac features on echocardiogram as a potential secondary modifying factor for HCM (27). In this current investigation, however, no MYBPC3^Δ25bp subjects with the D389V variant were included, meaning that the findings herein cannot be attributed to D398V. As previously noted, Harper et al. separately identified a haplotype in SAs with HCM with both the MYBPC3^Δ25bp variant and a potentially pathogenic variant, MYBPC3 c.1224-52G>A (31). None of the members of our MYBPC3^Δ25bp cohort were, in fact, found to carry the variant identified by Harper et al. Altogether, we propose that MYBPC3^Δ25bp is still a valid risk variant in the etiology of HCM and that it arises from a secondary risk factor, namely, exercise stress, and manifests as hypercontraction and impaired relaxation associated with late-onset LV dysfunction.

Exercise stress echocardiography is a standard clinical assessment and diagnostic armamentarium of HCM to assess exercise-induced LVOT obstruction in patients with resting LVOT gradient <50 mm Hg (40). The only limitation is that not all patients with HCM and controls are able to perform stress exercises. Patients with HCM usually present with dynamic LVOT gradient, arrhythmias, and mitral regurgitation. However, for asymptomatic carriers with pathogenic sarcomere mutations, the exercise stress test is potentially an advantageous method for unmasking subclinical diseases, such as impairment in diastolic function and dynamic outflow tract obstruction in the absence of a pathologically hypertrophied LV. The exercise stress test results presented here seem to have to accurately detect early changes in cardiac function, and the testing procedure was shown to be safe for enhancing and visualizing mid-range changes in cardiac function in variant carriers (41). Based on the results, asymptomatic MYBPC3^Δ25bp carriers do have sublevel phenotype during exercise, which is masked under baseline conditions. These data further indicate that exercise testing may assist in risk stratification for the presence of any cardiac events as previously described (41–43).

Hypertension, diabetes mellitus, coronary artery disease, and other comorbid conditions were known to confound the phenotypic effects of cardiomyopathy and cardiac dysfunction. Initial studies have shown that the MYBPC3^Δ25bp variant is associated with HCM (29), but then it was also determined to be associated with dilated and restrictive cardiomyopathies and HF (26). Later, it was determined that MYBPC3^Δ25bp is significantly associated with LV dysfunction secondary to coronary artery disease (39, 44). After myocardial infarction, results from these studies have suggested that patients with MYBPC3^Δ25bp show poor cardiac remodeling and recovery from the damage. This could be explained in several ways. First, cMyBP-C is sensitive to proteolysis during myocardial infarction and could be a potential earlier biomarker for heart attack (45–47). Thus, patients with MYBPC3^Δ25bp may not have enough wild-type cMyBP-C turnover, resulting in haploinsufficiency (48). The failure to compensate for the loss of wild-type during myocardial infarction finally results in poor cardiac remodeling. Second, when MYBPC3^Δ25bp carriers undergo myocardial infarction, it is possible that initial damage hinders quick recovery by the poison polypeptide effect (49). In the present study, we provided subjects with a questionnaire asking for information pertaining to their demographics, family history, medical history, and current and past medications. Using this as our guide, we then excluded any MYBPC3^Δ25bp carriers who were presented with such secondary risk factors, thus eliminating them from participation in the stress echo study. This means that the outcome of the stress echo was based solely on the presence of MYBPC3^Δ25bp and its effects on cardiac function. Collectively, results from the present study suggest LV hypercontractility under exercise stress conditions with evidence of diastolic stiffening and impairment among male asymptomatic MYBPC3^Δ25bp carriers. As such, these results also suggest the presence of some subclinical pathology in MYBPC3^Δ25bp carriers under exercise stress conditions. Based on the presence of HCM in transgenic mice expressing MYBPC3^Δ25bp mRNA (49) and subclinical cardiomyopathy phenotype in carriers (26), MYBPC3^Δ25bp can, therefore, be considered a potential risk allele for abnormal diastolic function and cardiomyopathy.

Complementary DNA sequencing of mRNA isolated from the biopsy of an MYBPC3^Δ25bp-positive patient confirmed the presence of exon 33 skipping in vivo (26, 29). Exon 33 skipping results in the loss of 62 amino acids, resulting in a modified C10 domain of cMyBP-C (13). Exon 33 skipping also moves the stop codon to the 3′ untranslated region (UTR), adding a novel 55 amino acids in this newly modified C10 domain at the carboxyl terminus (cMyBP-C^ΔC10) (49). However, if the C10 domain is modified, or truncated, then cMyBP-C will not properly localize in the sarcomere (50–52). The repercussions of this event can involve pathogenicity with the onset of HCM.

In fact, we have previously demonstrated that the presence of the MYBPC3^Δ25bp variant could lead to exon 33 skipping during transcription (26, 29). We next determined that the MYBPC3^Δ25bp variant in the presence of exon 33 skipping is pathogenic, both in vitro (46) and in vivo (49). Using a transgenic mouse model, we demonstrated that the expression of cMyBP-C^ΔC10 is sufficient to cause HCM. Based on this observation, we propose that haploinsufficiency increases the disordered relaxed state (DRX) and decreases the super relaxed state (SRX) of myosin, leading to hypercontraction and HCM (53, 54). However, the molecular mechanism underlying these phenomena is yet to be validated in a humanized knock-in MYBPC3^Δ25bp mouse model that shows normal cardiac function under baseline conditions (55). Thus, future studies will involve defining the pathogenicity of MYBPC3^Δ25bp in a humanized knock-in mouse model in the presence of a comorbidity or acute insult.

We presented the current data at the American Heart Association Scientific Sessions on November 12, 2020 (abstract # 13481) entitled “South Asian-specific MYBPC3^Δ25bp Intronic Deletion Carriers Demonstrate Hypercontractility and Impaired Diastolic Function Under Exercise Stress” (Circulation. 2020;142:A13481; https://doi.org/10.1161/circ.142.suppl_3.13481).

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author/s.

The studies involving human participants were reviewed and approved by Prevalence study: Loyola University Chicago IRB# LU207815 and 207359, Chicago, Illinois and University of Cincinnati IRB# 2016-7580, Cincinnati, Ohio. Genotype-phenotype study: University of Cincinnati IRB# 2016-4948, Cincinnati, Ohio. The patients/participants provided their written informed consent to participate in this study.

SB, SV, EM, and SS: conceptualization. SB, RS, SG, MP, KS, RRS, MK, DH, JR, and EM: data curation. SB, RS, SG, KS, MK, RJ, DH, JR, and EM: formal analysis. RB and SS: funding acquisition. SB, RS, SG, MP, MK, DH, JR, EM, and SS: investigation. SB, SV, MP, RJ, DH, JR, EM, and SS: methodology. DH, EM, and SS: project administration. RS, KS, RRS, RJ, DH, JR, RB, EM, and SS: resources. DH, EM, and SS: supervision. RJ, JR, and RB: validation. SB, MP, and SS: writing the original draft. SB, SS, MP, RJ, DH, JR, RB, EM, and SS: writing, reviewing, and editing. All authors contributed to the article and approved the submitted version.

This research was supported by Heart, Lung, and Vascular Institute startup funding at the College of Medicine, University of Cincinnati. The investigators were further supported by the National Institutes of Health grants R01 AR078001 (SS), R01 HL130356 (SS), R01 HL105826 (SS), R38 HL155775 (SS), R01 HL143490 (SS), U01 HL131914 (EM), and R01 HL128075 (EM), and the American Heart Association 2019 Institutional Undergraduate Student (19UFEL34380251, SS), Cardiovascular Genome-Phenome Study (15CVGPSD27020012, SS, MP, and EM), Catalyst (17CCRG33671128, SS, MP, and EM), Transformation (19TPA34830084, SS), Career Development (189CDA34110460, MP), and Postdoctoral Training Fellowship (17POST33630157, SV) awards, and the PLN Foundation (PLN crazy idea, SS). RRS was supported by an Amgen Postdoctoral Fellowship.

SS provides consulting and collaborative research studies to the Leducq Foundation (CURE-PLAN), Red Saree Inc., Greater Cincinnati Tamil Sangam, Pfizer, Novo Nordisk, AstraZeneca, MyoKardia, Merck, and Amgen. EM serves as a consultant to AstraZeneca, Amgen, Pfizer, Tenaya Therapeutics, and Invitae. RB serves on scientific advisory boards for Janssen and Basking Biosciences and DSMB Committees for Ionis Pharmaceuticals, Akcea Therapeutics, and Novartis. These activities are unrelated to the content of this work. RRS has been a postdoctoral fellow of Amgen, starting from June 2019, and performs research at the University of Cincinnati. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.