Abstract
We sought to establish a large animal model of inherited hypertrophic cardiomyopathy (HCM) with sufficient disease severity and early penetrance for identification of novel therapeutic strategies. HCM is the most common inherited cardiac disorder affecting 1 in 250–500 people, yet few therapies for its treatment or prevention are available. A research colony of purpose-bred cats carrying the A31P mutation in MYBPC3 was founded using sperm from a single heterozygous male cat. Cardiac function in four generations was assessed by periodic echocardiography and measurement of blood biomarkers. Results showed that HCM penetrance was age-dependent, and that penetrance occurred earlier and was more severe in successive generations, especially in homozygotes. Homozygosity was also associated with progression from preclinical to clinical disease. A31P homozygous cats represent a heritable model of HCM with early disease penetrance and a severe phenotype necessary for interventional studies aimed at altering disease progression. The occurrence of a more severe phenotype in later generations of cats, and the occasional occurrence of HCM in wildtype cats suggests the presence of at least one gene modifier or a second causal variant in this research colony that exacerbates the HCM phenotype when inherited in combination with the A31P mutation.
Similar content being viewed by others
Introduction
Hypertrophic cardiomyopathy (HCM) is the most common inherited cardiomyopathy in people affecting 1:250–500 individuals1. HCM is traditionally characterized by asymmetric or symmetric left ventricular (LV) hypertrophy with impaired diastolic relaxation. However, clinical symptoms are heterogeneous, with most patients showing no clinical signs while others develop a range of consequences including congestive heart failure (CHF), malignant arrhythmias, or sudden cardiac death2.
Part of the complexity in understanding preclinical and clinical HCM phenotypes is due to heterogeneity in underlying genetic causes of the disease with more than 1,000 mutations in at least 15 genes associated with HCM1. However, up to ~ 70% of mutations occur in two genes, MYH7 and MYBPC3, encoding beta-myosin heavy chain (myosin) and cardiac myosin-binding protein-C (cMyBP-C), respectively. Myosin is the force-generating protein in muscle responsible for the power that drives cardiac contraction during each heartbeat, whereas, cMyBP-C is a regulatory protein that modifies actomyosin interactions to regulate the rate and extent of contraction and relaxation3,4. Because myosin, cMyBP-C, and most other proteins carrying causative mutations for HCM localize to sarcomeres (the basic contractile units of muscle), HCM has become known as a “disease of the sarcomere”. This basic understanding has led to intense efforts to target the sarcomere for drug therapies which have led to exciting new first-in-class drugs that directly affect muscle function by inhibiting or activating myosin5,6. While these drugs hold promise for effective treatments and potentially prevention of HCM, obstacles still remain; such as variable mutation penetrance and differences in expressivity that make it difficult to identify at-risk individuals that will benefit from treatment.
Animal models of HCM can be useful in understanding some of this variability, especially large animal (non-rodent) models that have disease etiologies and phenotypes similar to human HCM7,8. In this regard, HCM is the most common naturally occurring inherited cardiomyopathy in domestic cats with a reported incidence in one study as high as 15%, and is the most common cause of CHF9. Unlike engineered rodent models of HCM, feline HCM shows similar clinical and pathological presentations as human HCM including left ventricular outflow tract obstruction (LVOTO)10. Because LVOTO is a significant factor affecting patient quality of life in people11,12, animal models that recapitulate hemodynamic complications of HCM are especially valuable. For instance, LVOTO is not typical of genetically engineered mouse models of HCM and therefore such models are limited in the pursuit of novel therapeutics that can mitigate LVOTO gradients. HCM in cats also appears to have similar underlying genetic causes as in people; to date, two causal mutations in MYBPC3 have been identified in Maine Coon and Ragdoll cats13,14. However, thus far other genetic causes have proven difficult to identify in cats and few longitudinal studies of feline HCM with a known causative mutation have been performed, thus, exacerbating gaps in our understanding of feline disease progression15.
Here we aimed to characterize the long-term impact on cardiac function of the A31P missense mutation in cMyBP-C first identified in Maine Coon cats13. The mutation results in a single amino acid substitution causing a proline to be substituted for an alanine at the protein’s 31st amino acid position near the N’-terminus of the molecule (91G > C; p.A31P). Previous studies showed that the mutation disrupts the folded domain structure of the first immunoglobulin (Ig)-like domain of cMyBP-C and increases susceptibility to proteolytic degradation, however, the mutation does not lead to significant loss of cMyBP-C (haploinsufficiency)16. Normal levels of cMyBP-C were also reported in another cat homozygous (i.e., carrying two mutant alleles) for A31P where the mutation was reported to decrease binding to actin17.
Because the incidence of clinical HCM in cats heterozygous (i.e., carrying a single mutant allele) for the A31P mutation appears low in the general cat population18,19, we hypothesized that disease phenotype would be more penetrant in cats homozygous for the mutation. Cats were purpose bred to obtain homozygous A31P genotypes to determine whether homozygous status would constitute a useful animal model of HCM with requisite high penetrance of disease in young animals (< 1 year) ideal for preventative therapeutic studies. This study was designed to minimize bias and confounders by prospectively and blindly assessing cats housed in a controlled environment over time from kitten to adult ages. Cats homozygous for the A31P mutation develop preclinical asymmetric ventricular growth and diastolic dysfunction suggesting that continued selective breeding in this unique research colony will be valuable for identifying interventional therapies for both human and veterinary medicine.
Results
Colony breeding and pedigree relationships
The breeding strategy used to obtain cats carrying the A31P mutation in cMyBP-C is presented in Fig. 1. The A31P mutation in exon 3 of MYBPC3 was first identified in a research colony of cats with HCM and was also identified in Maine Coon cats in general cat populations13. However, the original research colony consisted of a mixed genetic population of cats with HCM and not all cats in that colony with HCM carried the A31P mutation. Therefore, multiple genetic etiologies likely contributed to HCM in the original colony20. To focus exclusively on disease progression from a known genetic cause, we selected a single male cat from the original HCM colony that was heterozygous for the A31P mutation. Sperm from this heterozygous A31P male was used to develop a second-generation research colony using artificial insemination of three healthy, unrelated, wildtype DSH (domestic shorthair) female cats. Resulting heterozygous A31P cats were crossed to obtain cats homozygous for the A31P mutation. Homozygous A31P cats were born in Mendelian ratios and survived into adulthood. Homozygous X homozygous crosses produced healthy litters, albeit, litter sizes were frequently small for all colony breeding pairs regardless of genotype (typically two-to-five kittens).
Longitudinal echocardiographic assessments
Cardiac structure and function was assessed by echocardiography in kittens three-to-six months of age and thereafter at regular intervals of six-to-twelve months throughout the term of the study. Standard cardiac morphology and functional indices were measured via M-mode, 2-D, and color Doppler imaging for a complete echocardiographic exam at each time point. Blood samples were also collected at each exam for analysis of NT-proBNP and cardiac troponin-I (cTnI) serum biomarkers. Affected status was determined after each exam according to criteria described in “Methods”.
Analyses of longitudinal echocardiographic assessments revealed gradual progression of preclinical disease penetrance in most colony cats. The progression was evident as an increase in cats designated affected with increasing number of evaluations, increasing age at time of exam, and with A31P homozygous genotype (Fig. 1). As expected, penetrance was greatest in cats homozygous for the A31P mutation and was least for wildtype cats without the mutation; however, some wildtype cats without the A31P genotype also progressed to affected status (Fig. 2). Individual cardiac indices were next grouped and analyzed according to A31P genotype. Statistically significant differences were achieved for final measurements of interventricular septal (IVS) diameter in diastole, and absolute left atrial (LA), and normalized LA-to-aortic root (LA/Ao) dimensions between wildtype cats versus cats homozygous for the A31P mutation (Table 1). To ensure that compensatory effects of cardiac remodeling did not blunt detection of greater differences at earlier time points, we also compared highest (or lowest) individual values obtained for each index regardless of final value recorded at the last time point. Statistically significant differences were found for Ao and LA/Ao values between genotypes (Table S1).
LVOTOs were also common (present in ~ 18% of examinations) and dynamic within the colony. Although dynamic LVOTO was occasionally observed in wildtype cats, the incidence of mid-ventricular obstruction was common in cats with the A31P variant and HCM affected status. Juvenile cats with early onset and severely affected disease status also commonly demonstrated LVOTO. As seen in Table S1, cats that were heterozygous or homozygous for the A31P mutation had median aortic flow velocities at or above 2 m/sec, constituting LVOTO in these genotype groups compared to wildtype cats with a median of 1.4 m/sec21,22,23,55.
NT-proBNP values were significantly associated with HCM affected status, A31P genotype, and the presence of LVOTO in colony cats (Tables 2 and 3). This is consistent with the observed enlargement of the LA and LV thickening that predominates in HCM-affected A31P homozygous cats within this study. Increased serum NT-proBNP values were significantly associated within affected cats, specifically homozygotes when compared to other genotypes (Fig. 3). Serum cTnI values were not significantly different between affected and unaffected cats (data not shown).
Inheritance of the A31P mutation in homozygous cats leads to a statistically significant steady progression of preclinical indices of HCM (e.g., regional wall thickening and LA/Ao ratios); homozygotes are more likely to develop HCM compared to heterozygotes (OR = 8.5 and OR = 3.75, respectively) (Table 4). However, the odds ratio for heterozygous A31P status did not achieve statistical significance. Multiple logistic regression analysis was subsequently performed, and the homozygous A31P genotype remained significantly associated with the development of HCM (OR = 11.8) (Table S2).
Preclinical to clinical HCM progression of A31P homozygotes
Disease penetrance, progression, and severity in cats homozygous for the A31P mutation have not been well described due in part to the small number of homozygous cats available for study in general cat populations. Here we studied 22 cats homozygous for the A31P mutation. Of these, three cats progressed from preclinical status to CHF during the course of the study. The first cat that progressed from preclinical to clinical disease was a cat from the first generation of cats obtained from heterozygous X heterozygous crosses (cat #2011-14, Fig. 1). Compared to other homozygous cats within the same generation, cat #2011-14 had more severe, uniform, and rapidly progressive LV hypertrophy (Fig. 4). Cat #2011-14 had abnormal LV diastolic wall thickness at 23 months of age and increased LA size at 29 months of age that increased to severe enlargement by the final echocardiographic evaluation. CHF occurred at 41 months as evidenced by severe LA enlargement, pulmonary edema, and pericardial effusion.
A31P allelic interactions and association with early onset, severe, HCM phenotypes
In subsequent generations, HCM affected status occurred at earlier times either at first echocardiographic exam (typically between three-to-six months of age) or before one year-of-age. Individuals that resulted from a cross involving a heterozygous male (#2011-44) and homozygous female (#2011-45) cat presented with a more severe phenotype at earlier ages than in previous generations. For instance, the second cat that progressed to CHF occurred in later generations (cat #2011-48, Fig. 1). The initial echocardiographic exam for this cat at four months showed severe diastolic IVS and LV posterior wall (LVPW) thickness of 9 and 8 mm, respectively, and mild LA enlargement (Fig. 5). HCM progressed rapidly to CHF seen as pulmonary edema, and scant pleural and pericardial effusions at 10 months of age. A third homozygous cat that developed CHF by 30 months of age was also produced from this same breeding pair (cat #2011-50, Fig. 1).
DNA sequencing and variant analysis
Because of the relatively high degree of inbreeding in the closed colony until this point (Fig. 1), we hypothesized that the appearance of the more severe phenotype may be due to co-inheritance of a modifier variant or an independent HCM variant introduced by the male founder or by one of the three unrelated female founder cats. We therefore used whole-genome sequencing (WGS) to identify sequence variants in 15 colony cats, resulting in a total of 89,817,458 variants identified. After quality control and elimination of variants in intergenic regions, 32,218,731 variants remained. A total of 380,695 variants were localized to genes previously or suspected to be associated with HCM or in electrophysiology genes (Table S3). Of these, 15 showed statistically significant association with disease phenotype via Chi-square testing. To determine the frequency of these variants in the general cat population, we next calculated the allele frequency of the 15 variants in 195 unrelated, unphenotyped cats from a feline genetics consortium (99Lives Cat Genome Sequencing Initiative; http://felinegenetics.missouri.edu/99lives/scientist-information). Only the previously identified MYBPC3 A31P variant was identified in the majority of HCM affected colony cats and not in any of the wildtype control cats or the general cat population (Table S4).
Discussion
This study describes a comprehensive longitudinal assessment of HCM progression in a purpose-bred research colony of cats carrying a single known deleterious missense mutation in MYBPC3, the gene encoding the cMyBP-C protein. The mutation results in a proline for alanine substitution at codon 31 (A31P) and was first identified in an inbred research HCM cat colony13. This mutation was also found in general populations of client-owned cats of the Maine Coon breed, making it the first identified, naturally occurring, causal mutation for HCM in any companion animal species24,25. Because heterozygous inheritance of the A31P allele in general cat populations was associated with low disease penetrance18,19,26, we bred cats to achieve homozygosity for the A31P allele. Our goals were to determine 1) whether homozygous inheritance of the A31P mutation was associated with higher disease penetrance and a more severe phenotype as a prerequisite for use of A31P cats as a feline model of HCM, and 2) provide a controlled, longitudinal assessment of preclinical disease progression in a population of purpose-bred cats with and without the A31P mutation. Result from this study confirm that the A31P mutation is causative for HCM, especially when inherited in the homozygous state (OR = 8.5, Table 4). We conclude that purpose-bred cats homozygous for the A31P cMyBP-C mutation provide a reliable model of inherited HCM.
Significant findings by echocardiography included characteristics commonly reported for preclinical disease in both people and cats including regional and/or global LV wall thickening, increased diameter of the LA, and increased LA/Ao dimensions in HCM-affected cats. Additional findings included frequent indicators of diastolic dysfunction as assessed by Doppler imaging. Consistent with diastolic dysfunction, NT-proBNP concentrations were also elevated in A31P cats as previously reported27. NT-proBNP concentrations were highest in homozygotes and were also frequently elevated in HCM-affected cats. However, not all severely affected cats had elevated NT-proBNP. For instance, cat #2011-44 (Fig. 1) was severely affected, but NT-proBNP concentrations measured in repeated blood samples consistently yielded values in the normal range. Singh et al28 also reported that NT-proBNP concentration did not correlate with preclinical disease and that some severely affected Maine coon cats did not show elevated NT-proBNP. Biologic variability in NT-proBNP has also been reported in healthy cats29. Because NT-proBNP concentrations were also lower in a subgroup of sarcomeric mutation negative HCM patients30, low NT-proBNP concentration alone may not exclude disease.
Systolic abnormalities included dynamic LVOTO and systolic anterior motion of the mitral valve (SAM) frequently observed in affected cats. Consistent with previous reports, LVOTO was not associated with higher incidence of adverse outcomes31, although the relatively young ages and the small number of cats here precludes definitive conclusions regarding outcome correlations with LVOTO. All other parameters, including systolic function assessed by fractional shortening (FS) and ejection fraction (EF), were not significantly different in affected cats.
Taken together, these echocardiography results establish that cats homozygous for the A31P mutation reliably develop preclinical HCM with characteristics common to both human and feline diseases. Purpose-bred cats homozygous for the A31P mutation therefore represent a unique, naturally occurring, large animal model of HCM that can offer significant advantages over existing engineered rodent models of disease. For instance, while genetically engineered rodent models of HCM are essential for confirming deleterious effects of mutations identified in people32,33, mouse models cannot recapitulate significant aspects of disease comorbidities including hemodynamic complications such as LVOTO or susceptibility to arrhythmia. Because LVOTO is a significant factor impacting human patient quality of life11,12, non-rodent models are critically needed to identify new therapeutics that reduce LVOTO gradients.
Another advantage of homozygous A31P cats as an animal model of disease is that HCM occurs naturally as a result of an inherited mutation so that additional surgical, drug, or dietary interventions are not required to induce cardiac dysfunction as in other models of cardiac dysfunction34,35,36,37. This facilitates wider implementation in laboratory or clinical settings where access to secondary interventions may be limited. Because mutations in MYBPC3 are the most common cause of HCM in people, understanding the natural history of HCM due to a cMyBP-C mutation in cats may offer insights into critical points of decompensation as well as provide a highly relevant model for testing new drug or gene therapies37,38,39.
Most missense MYBPC3 mutations identified in people lead to premature protein truncation40. However, missense mutations coding for single amino acid substitutions, like A31P, also occur in people. In this respect, the A31P mutation occurs in the cardiac-specific C0 domain of cMyBP-C and other missense mutations have been identified in the C0 domain of HCM patients41. The precise function of C0 is not yet known, but C0 binds to both F-actin and the regulatory light chain (RLC) of myosin42,43. Disruption of these interactions could therefore affect the ability of cMyBP-C to activate the thin filament or to stabilize the super-relaxed (SRX) state of myosin44,45,46, either of which would be expected to directly cause contraction or relaxation abnormalities resulting in a cardiomyopathy.
While a second missense mutation in MYBPC3 (encoding R820W) was identified in Ragdolls14, cats expressing the R820W mutation are not currently available for research studies. HCM and dilated cardiomyopathy (DCM) have been identified as naturally occurring diseases in a number of large animals including dogs, pigs, and non-human primates, but a genetic basis has yet to be established in these species. Purpose-bred cats expressing the A31P mutation in cMyBP-C represent a unique genetic model of HCM with relevance to human disease and are currently the only available naturally occurring HCM model with a defined genetic mutation.
Despite significant advantages offered by use of purpose-bred A31P cats, challenges still remain in their development as an effective model of HCM. First, HCM penetrance in cats and people is variable with different expressivity observed even in closely related individuals that carry the same mutation. Here we attempted to overcome this limitation by breeding cats to homozygosity as gene dosage effects can lead to more severe disease phenotypes. This strategy led to higher penetrance of affected status in cats with a homozygous genotype relative to heterozygous cats carrying a single copy of the mutant allele or to wildtype cats that did not express the mutation (Fig. 1). However, disease severity and time to affected status were still variable in cats homozygous for the A31P mutation, especially in the first generations of cats bred to homozygosity.
One explanation for the observation that affected status occurred earlier with successive generations of breeding is that inbreeding within the closed colony led to co-selection of another heritable modifier or causative variant resulting in an earlier, more severe phenotype when combined with the A31P mutation. A second variant could have been inadvertently introduced into the colony via one of the three apparently healthy female cats given the high incidence of asymptomatic HCM-affected individuals within client-owned populations9,31. Alternatively, a second variant could have been introduced by sperm from the founder male as this male was acquired from a research colony of cats bred for HCM with or without the A31P mutation20. However, it seems unlikely that inheritance of an unidentified variant alone can account for HCM in this colony, because WGS sequence analysis identified only the A31P mutation in the majority of colony cats with HCM (Table S4). Continued selective breeding of cats with the most severe, early, phenotypes should result in continued improvement of model traits and provide further opportunities to identify disease modifiers or other variants through sequencing of severely affected cats. The role of environmental and genetic modifiers is increasingly acknowledged as an underlying cause of disease heterogeneity in HCM patients47,48,49,50,51. Combined use of A31P cats with other interventions designed to induce cardiac dysfunction34,35,52 may also result in a more severe phenotype that progress to CHF over shorter, more defined, time frames.
There are several limitations to this study. First, it is well known that there is considerable variability in HCM disease penetrance even in closely related individuals that carry the same genetic mutation53, implying that there are significant other genetic or environmental factors that contribute to disease etiologies. Here we attempted to minimize the impact of environmental factors and genetic factors that could account for the differences in disease penetrance by maintaining a closed breeding colony with a single known genetic mutation linked to disease. However, we still observed considerable difference in disease penetrance that could not be accounted for by environment or identified genetic factors. An additional limitation of our longitudinal study design is that older generations of cats received more screening examinations than the youngest generation and thus not all cats were followed for the same amount of time. For echocardiography, cats were sedated and so potential effects of sedation on echocardiographic results cannot be excluded. Also, blood pressure measurements were not routinely performed. However, all cats with a diagnosis of HCM had a blood pressure recorded to be < 160 mmHg by Doppler Sphygmomanometry in combination with normal renal function as assessed by serum creatinine concentration within the normal reference range. Finally, histopathology analyses were not performed.
In conclusion, we describe a second-generation research colony of purpose-bred cats selected to carry a single genetic mutation in MYBPC3. Results from this study confirm deleterious effects of the A31P mutation in cMyBP-C by showing that cats homozygous for the mutation reliably develop preclinical indicators of HCM at early time points often before the first year of age. Cats carrying the A31P mutation should therefore be valuable for interventional studies designed to reverse or slow progression of HCM using novel and next-in-class therapeutic approaches.
Methods
Colony breeding and periodic assessments
A purpose-bred colony of cats was created for research studies through selective breeding of cats with the A31P mutation in cMyBP-C13. All aspects of animal care and use, including ethical usage of animals in research, were approved and performed in accordance with the Institutional Animal Care and Use Committee (IACUC) of the University of California Davis (protocol #22,376). In addition, this study was carried out in compliance with the ARRIVE guidelines (http://www.nc3rs.org.uk/page.asp?id=1357), the Animal Welfare Act, and the Institute for Laboratory Animal Research Guide for the Care and Use of Laboratory Animals. Priority at all times was given to cat comfort, health, and well-being. Briefly, cats were housed together in social groups in large enclosures with access to natural light, food and water (ad libitum), environmental enrichment including toys, scratching posts, perches, and beds, and were provided daily socialization with caretakers. Importantly, cats in the colony are maintained in a pre-clinical state, meaning they are without signs or symptoms of HCM and so experience minimal, if any, discomfort. Upon first development of CHF symptoms, cats are humanely euthanized.
The colony was founded using sperm from a single male Maine Coon/mix breed cat that was heterozygous for the A31P mutation obtained from a previously established research colony of cats with multiple causes of HCM at UC Davis13,20. Sperm from the founder cat was used to artificially inseminate three unrelated domestic short hair (DSH) females. Heterozygous A31P cats were then bred to obtain cats homozygous for the A31P mutation. A full colony pedigree chart is shown in Fig. 1 for the study period of ~ seven years (circa 2011–2018). Note that due to the finite study design, older generations of cats received more follow up screening than cats in younger generations.
All cats were genotyped and underwent periodic complete echocardiography and cardiac biomarker assessment (details provided in “Echocardiography” and “Biomarker measurements”). Examinations occurred approximately every six-to-twelve months with initial evaluation occurring between three-to-six months of age. Cardiovascular examinations were performed under sedation with a combination of butorphanol (0.2–0.3 mg/kg intramuscularly or intravenously) and acepromazine (0.05–0.1 mg/kg intramuscularly or intravenously) as needed to facilitate evaluation with minimal manual restraint.
HCM affected status was strictly defined by one of two possible scenarios: 1) a diastolic LV wall segment ≥ 6 mm identified in at least two separate echocardiographic examinations in the absence of any other systemic or metabolic disease, or 2) a diastolic LV wall segment ≥ 6 mm identified on a single echocardiographic examination in conjunction with a serum NT-proBNP concentration of > 100 pmol/L54 in the absence of any other systemic or metabolic disease. LVOTO was defined as color flow turbulence in association with a region of obstruction (LVOTO secondary to SAM or mid-ventricular) and a measured flow velocity of ≥ 2 m/sec, as previously reported21,22,23,55. All cats diagnosed with HCM were determined to have normal renal function via biochemistry, absence of any other known metabolic disease, a systolic blood pressure < 160 mmHg via Doppler sphymomanometry, and a total thyroxine level within the reference range if over the age of six years. CHF was defined as the presence of pleural and/or pericardial effusion via thoracic ultrasound and/or pulmonary edema on thoracic radiographs in conjunction with clinical signs of tachypnea or dyspnea. Cats diagnosed with CHF were euthanized at the first signs of clinical progression and no cats were maintained beyond diagnosis of CHF to ensure that there were no opportunities for cats to go untreated with progressive respiratory distress.
Genotyping
DNA for PCR genotyping was extracted from either buccal swabs or EDTA blood samples for all cats. Genotyping for the A31P mutation was carried out either by the UC Davis Veterinary Genetics Laboratory or by standard PCR and Sanger sequencing as previously reported56.
Echocardiography
Echocardiographic examinations were performed by a single, board-certified, veterinary cardiologist (J.A.S) using a 12-4 mHz sector-array transducer (Philips iE33 Ultrasound, Philips Healthcare, Andover, MA) and a standard approach55 of two-dimensional, M-mode, color, and spectral Doppler modalities. Cats were monitored via a continuous lead II ECG throughout the echocardiogram when tolerated at each evaluation. Sedated cats were first manually restrained in right then left lateral recumbencies to obtain echocardiographic images. All images were obtained and stored offline for measurement and analysis by a single, board-certified, veterinary cardiologist (J.A.S) using commercially available software and workstation (Syngo Dynamics, Siemens Medical Solutions, Malvern, PA). Cat images were identified by six-digit numbers such that the investigator was blinded to cat age, gender, and A31P genotype during measurement. When possible, the average value of three consecutive cardiac cycles was obtained for each measurement for all two-dimensional or M-mode measures. Spectral Doppler measures were recorded as the peak modal velocity. Measurements were taken to avoid ectopic cardiac cycles or cycles immediately following ectopy. LV wall measurements were obtained from both 2D and M-mode imaging as previously described with care to avoid the insertion sites of observed moderator bands9. LA diameter and LA/Ao measured by 2D echocardiogram in right parasternal short axis was obtained as previously described57. The left atrium was considered enlarged if the maximal left atrial diameter was ≥ 16 mm or the left atrium to aortic root ratio (LA:Ao) was ≥ 1.658,59. Peak LV outflow tract velocity was obtained from the left apical imaging plane with parallel alignment to blood flow and noted turbulence by color-flow Doppler imaging.
Biomarker measurements
Cardiac biomarkers were assessed in venous blood samples obtained from each cat at each cardiovascular examination when possible. NT-proBNP (Cardiopet® proBNP) and high-sensitivity cardiac troponin-I (Hs-cTnI, Beckman Access AccuTnI test) were measured in serum by a diagnostic laboratory according to sample handling and preparation guidelines (IDEXX Research & Development, IDEXX Laboratories Inc, Westbrook, ME). All cats had an indirect ophthalmic examination performed during one of their final two examinations to rule out retinal lesions consistent with systemic hypertension.
Statistical evaluation
Statistical analyses were performed using commercially available software packages (Prism v8.2.1, GraphPad Software, La Jolla, CA; STATA v15, Stata Corporation, College Station, TX). Normality testing for continuous echocardiographic and patient characteristic variables was done using D’Agostino-Pearson test. Descriptive statistics were performed and reported as mean ± SD for parametric values and median and interquartile range (IQR) for nonparametric values.
Echocardiographic parameters obtained during the last examination were compared among cats with different A31P genotypes using Kruskal–Wallis test with Dunn’s post-hoc comparison. The highest (IVS diastole, IVS systole, LVPW diastole, LVPW systole, LA diastole, Ao systole, LA/Ao, A Vmax, Septal bulge, AoV, and PV) or lowest values (LV diastole, LV systole, FS, E Vmax, E/A, and Lar Flow) of echocardiographic parameters were also compared among cats with different A31P genotypes using Kruskal–Wallis test with Dunn’s post-hoc comparison.
The highest NT-proBNP values in the repeated measurements in the various categorical variables (sex, HCM status, A31P genotype, and presence or absence of LVOTO) were compared using Mann–Whitney U test or Kruskal–Wallis test with Dunn’s post-hoc comparisons. Correlations between the highest NT-proBNP values in the repeated measurement with continuous variables including age and body weight were assessed using the Spearman correlation method. Simple and multiple regression analysis after standard assumptions for linear regression were performed. Backward and forward selection techniques were performed by sequentially deleting or adding the variables with P < 0.15 for multiple regression analysis.
Logistic regression analysis was performed to determine the odds ratio of characteristic variables (age, sex, body weight, and A31P genotype) of developing disease. The odd ratios of developing HCM was also obtained by multiple regression analysis by including all variables used for the simple regression analysis.
Whole-genome sequencing and variant detection
Whole-blood or tissue samples were collected for 15 research colony cats following protocols approved by the University of California Davis IACUC. Of the 15 cats, three cats had HCM but were wildtype for the A31P mutation. An additional five cats had severe or early HCM and were either heterozygous (three cats) or homozygous (two cats) for the A31P mutation. Seven cats were healthy and were wildtype for the A31P mutation. DNA was extracted from EDTA whole blood or frozen tissue samples using a commercially available kit following the manufacturers protocol (Qiagen Puregene, Germantown MD).
Ten DNA samples were submitted for WGS (McDonnell Genome Institute, Washington University, Saint Louis, MO). To ensure DNA integrity and sufficient concentration, DNA samples were run on a 2% electrophoresis agarose gel and Qubit Fluorometer (ThermoFisher, Waltham MA) assessments were performed. DNA samples that passed quality control were selected, a 450 bp KAPA PCR-free library was constructed, and 150 bp pair-ended sequencing was performed on an Illumina NovaSeq S4 instrument (Illumina, San Diego, CA). The additional five cat samples in this study previously underwent WGS using 350 bp and 550 bp PCR-free Illumina TruSeq libraries and sequenced on the Illumina HiSeq2500 sequencing platform as described56. All cats were sequenced to ~ 30X coverage. The generated sequences were processed as previously described56.
Variant call files were uploaded and analyzed via SNP & Variation Suite (SVS) commercial software (Golden Helix, Bozeman MT). Variants that had a call rate less than 80% were removed from the analysis. The remaining variants were filtered with the Ensembl FelCat Genes 94 track using SVS software; intergenic variants were removed from further analysis. A total of 142 genes were analyzed which included genes that were previously associated with HCM, suspected to be associated with HCM, as well as electrophysiology genes (e.g., genes encoding membrane transport proteins). To identify statistically significant variants, a Chi-square test was completed. Variants located in exons or in the 3’ and 5’ untranslated regions (UTR) that had a P-value less than 0.05 were further analyzed for the following effects: 5’—UTR premature start codon variants, disruptive inframe indels, frameshift variants, initiator codon variants, missense variants, splice acceptor/donor variants, splice region variants, and stop gain/loss variants. Variants were further filtered based on Ensembl IMPACT rating60. Variants that were modifiers or had a high, moderate, or low predicted effect on protein structure based on IMPACT rating and, when applicable, a SIFT (Sorting Intolerant From Tolerant) score less than or equal to 0.05 were further analyzed in a large cohort (n = 180) of unphenotyped whole-genome sequenced cats (http://felinegenetics.missouri.edu/99lives). The allele frequency of statistically significant variants was calculated for the initial 15 cats and the additional 180 cat population consisting of 43 different breeds.
Data availability
Restrictions apply to the availability of these data. Data are available from the corresponding author upon request.
References
Yotti, R., Seidman, C. & Seidman, J. Advances in the genetic basis and pathogenesis of sarcomere cardiomyopathies. Annu. Rev. Genom. Hum. G 20, 1–25 (2019).
Ho, C. Y. et al. Genotype and lifetime burden of disease in hypertrophic cardiomyopathy. Circulation 138, 1387–1398 (2018).
Rosas, P. C. et al. Phosphorylation of cardiac Myosin-binding protein-C is a critical mediator of diastolic function. Circ. Hear Fail. 8, 582–594 (2015).
Palmer, B. M. et al. Role of cardiac myosin binding protein C in sustaining left ventricular systolic stiffening. Circ. Res. 94, 1249–1255 (2004).
Green, E. M. et al. A small-molecule inhibitor of sarcomere contractility suppresses hypertrophic cardiomyopathy in mice. Sci. New York N Y 351, 617–621 (2016).
Malik, F. I. et al. Cardiac myosin activation: A potential therapeutic approach for systolic heart failure. Science 331, 1439–1443 (2011).
Freeman, L. M., Rush, J. E., Stern, J. A., Huggins, G. S. & Maron, M. S. Feline hypertrophic cardiomyopathy: A spontaneous large animal model of human HCM. Cardiol Res 8, 139–142 (2017).
Maron, B. J. & Fox, P. R. Hypertrophic cardiomyopathy in man and cats. J. Vet. Cardiol. 17, S6–S9 (2015).
Payne, J. R., Brodbelt, D. C. & Fuentes, V. L. Cardiomyopathy prevalence in 780 apparently healthy cats in rehoming centres (the CatScan study). J. Vet. Cardiol. 17(Suppl 1), S244–S257 (2015).
Kittleson, M. D., Meurs, K. M. & Harris, S. P. The genetic basis of hypertrophic cardiomyopathy in cats and humans. J. Vet. Cardiol. 17, S53–S73 (2015).
Ho, C. Y. et al. Study design and rationale of EXPLORER-HCM. Circulation Hear Fail 13, CIRCHEARTFAILURE120006853 (2020).
Pelliccia, F. et al. Multidisciplinary evaluation and management of obstructive hypertrophic cardiomyopathy in 2020: Towards the HCM heart team. Int. J. Cardiol. 304, 86–92 (2020).
Meurs, K. M. et al. A cardiac myosin binding protein C mutation in the Maine Coon cat with familial hypertrophic cardiomyopathy. Hum. Mol. Genet. 14, 3587–3593 (2005).
Meurs, K. M., Norgard, M. M., Ederer, M. M., Hendrix, K. P. & Kittleson, M. D. A substitution mutation in the myosin binding protein C gene in ragdoll hypertrophic cardiomyopathy. Genomics 90, 261–264 (2007).
Borgeat, K., Casamian-Sorrosal, D., Helps, C., Fuentes, V. L. & Connolly, D. J. Association of the myosin binding protein C3 mutation (MYBPC3 R820W) with cardiac death in a survey of 236 Ragdoll cats. J. Vet. Cardiol. 16, 73–80 (2014).
van Dijk, S. J. et al. The A31P missense mutation in cardiac myosin binding protein C alters protein structure but does not cause haploinsufficiency. Arch Biochem. Biophys. 601, 133–140 (2016).
Godiksen, M. T. N. et al. Feline hypertrophic cardiomyopathy associated with the pA31P mutation in cMyBP-C is caused by production of mutated cMyBP-C with reduced binding to actin. Open J. Vet. Med. 2013, 95–103 (2013).
Godiksen, M. T. N., Granstrøm, S., Koch, J. & Christiansen, M. Hypertrophic cardiomyopathy in young Maine Coon cats caused by the p.A31P cMyBP-C mutation–the clinical significance of having the mutation. Acta Vet. Scand. 53, 7 (2011).
Granström, S. et al. Genotype–phenotype correlation between the cardiac myosin binding protein C mutation A31P and hypertrophic cardiomyopathy in a cohort of Maine Coon cats: A longitudinal study. J. Vet. Cardiol. 17, S268–S281 (2015).
Kittleson, M. D. et al. Familial hypertrophic cardiomyopathy in maine coon cats: An animal model of human disease. Circulation 99, 3172–3180 (1999).
Chetboul, V. et al. Quantitative assessment of velocities of the annulus of the left atrioventricular valve and left ventricular free wall in healthy cats by use of two-dimensional color tissue Doppler imaging. Am. J. Vet. Res. 67, 250–258 (2006).
Ward, J. L. et al. Retrospective evaluation of the safety and tolerability of pimobendan in cats with obstructive vs nonobstructive cardiomyopathy. J. Vet. Intern. Med. 34, 2211–2222 (2020).
Oldach, M. S. et al. Cardiac effects of a single dose of pimobendan in cats with hypertrophic cardiomyopathy; A randomized, placebo-controlled. Crossover Study. Front. Vet. Sci. 6, 15 (2019).
Mary, J. et al. Prevalence of the MYBPC3-A31P mutation in a large European feline population and association with hypertrophic cardiomyopathy in the Maine Coon breed. J Vet Cardiol 12, 155–161 (2010).
Longeri, M. et al. Myosin-binding protein C DNA variants in domestic cats (A31P, A74T, R820W) and their association with hypertrophic cardiomyopathy. J. Vet. Intern. Med. 27 (2013).
Sampedrano, C. C. et al. Prospective echocardiographic and tissue doppler imaging screening of a population of maine coon cats tested for the A31P Mutation in the myosin-binding protein C gene: A specific analysis of the heterozygous status. J. Vet. Intern. Med. 23, 91–99 (2009).
Hsu, A., Kittleson, M. D. & Paling, A. Investigation into the use of plasma NT-proBNP concentration to screen for feline hypertrophic cardiomyopathy. J. Vet. Cardiol. 11, S63–S70 (2009).
Singh, M. K., Cocchiaro, M. F. & Kittleson, M. D. NT-proBNP measurement fails to reliably identify subclinical hypertrophic cardiomyopathy in Maine Coon cats. J. Feline Med. Surg. 12, 942–947 (2010).
Harris, A. N. et al. Biologic variability of N-terminal pro-brain natriuretic peptide in adult healthy cats. J. Feline Med. Surg. 19, 216–223 (2017).
Neubauer, S. et al. Distinct subgroups in hypertrophic cardiomyopathy in the NHLBI HCM registry. J. Am. Coll. Cardiol. 74, 2333–2345 (2019).
Fox, P. R. et al. International collaborative study to assess cardiovascular risk and evaluate long-term health in cats with preclinical hypertrophic cardiomyopathy and apparently healthy cats: The REVEAL Study. J. Vet. Intern. Med. 32, 930–943 (2018).
Geisterfer-Lowrance, A. A. T. et al. A mouse model of familial hypertrophic cardiomyopathy. Science 272, 731–734 (1996).
Gannon, M. P. & Link, M. S. Phenotypic variation and targeted therapy of hypertrophic cardiomyopathy using genetic animal models. Trends Cardiovasc. Med. https://doi.org/10.1016/j.tcm.2019.11.009 (2019).
Wallner, M. et al. A feline HFpEF model with pulmonary hypertension and compromised pulmonary function. Sci. Rep-uk 7, 16587 (2017).
Pogwizd, S. M., Qi, M., Yuan, W., Samarel, A. M. & Bers, D. M. Upregulation of Na+/Ca2+ exchanger expression and function in an arrhythmogenic rabbit model of heart failure. Circ. Res. 85, 1009–1019 (1999).
Hála, P. et al. Tachycardia-induced cardiomyopathy as a chronic heart failure model in swine. J. Vis. Exp. https://doi.org/10.3791/57030 (2018).
Merkulov, S., Chen, X., Chandler, M. P. & Stelzer, J. E. In vivo cardiac myosin binding protein C gene transfer rescues myofilament contractile dysfunction in cardiac myosin binding protein C null mice. Circul. Hear Fail 5, 635–644 (2012).
Stern, J. A. et al. A small molecule inhibitor of sarcomere contractility acutely relieves left ventricular outflow tract obstruction in feline hypertrophic cardiomyopathy. PLoS ONE 11, e0168407 (2016).
Ma, H. et al. Correction of a pathogenic gene mutation in human embryos. Nature 548, 413–419 (2017).
Helms, A. S. et al. Sarcomere mutation-specific expression patterns in human hypertrophic cardiomyopathy. Circul. Cardiovasc. Genetics 7, 434–443 (2014).
Carrier, L., Mearini, G., Stathopoulou, K. & Cuello, F. Cardiac myosin-binding protein C (MYBPC3) in cardiac pathophysiology. Gene 573, 188–197 (2015).
Ratti, J., Rostkova, E., Gautel, M. & Pfuhl, M. Structure and interactions of myosin-binding protein C domain C0: Cardiac-specific regulation of myosin at its neck?. J. Biol. Chem. 286, 12650–12658 (2011).
Harris, S. P., Belknap, B., Sciver, R. E. V., White, H. D. & Galkin, V. E. C0 and C1 N-terminal Ig domains of myosin binding protein C exert different effects on thin filament activation. Proc. Natl. Acad. Sci. USA 113, 1558–1563 (2016).
Stewart, M. A., Franks-Skiba, K., Chen, S. & Cooke, R. Myosin ATP turnover rate is a mechanism involved in thermogenesis in resting skeletal muscle fibers. Proc. Natl. Acad. Sci. 107, 430–435 (2010).
McNamara, J. W. et al. Ablation of cardiac myosin binding protein-C disrupts the super-relaxed state of myosin in murine cardiomyocytes. J. Mol. Cell Cardiol. 94, 65–71 (2016).
Nag, S. et al. The myosin mesa and the basis of hypercontractility caused by hypertrophic cardiomyopathy mutations. Nat. Struct. Mol. Biol. 24, 525–533 (2017).
Wijnker, P. J. M., Sequeira, V., Kuster, D. W. D. & van der Velden, J. Hypertrophic cardiomyopathy: A vicious cycle triggered by sarcomere mutations and secondary disease hits. Antioxid. Redox Sign 31, 318–358 (2019).
Maron, B. J., Maron, M. S., Maron, B. A. & Loscalzo, J. Moving beyond the sarcomere to explain heterogeneity in hypertrophic cardiomyopathy JACC review topic of the week. J. Am. Coll. Cardiol. 73, 1978–1986 (2019).
Su, M. et al. Rare variants in genes encoding MuRF1 and MuRF2 are modifiers of hypertrophic cardiomyopathy. Int. J. Mol. Sci. 15, 9302–9313 (2014).
Christodoulou, D. C. et al. 5′RNA-Seq identifies Fhl1 as a genetic modifier in cardiomyopathy. J. Clin. Invest. 124, 1364–1370 (2014).
Harper, A. R. et al. Common genetic variants and modifiable risk factors underpin hypertrophic cardiomyopathy susceptibility and expressivity. Nat. Genet. 53, 135–142 (2021).
Riehle, C. & Bauersachs, J. Small animal models of heart failure. Cardiovasc. Res. 115, 1838–1849 (2019).
Chou, C. & Chin, M. T. Pathogenic mechanisms of hypertrophic cardiomyopathy beyond sarcomere dysfunction. Int. J. Mol. Sci. 22, 8933 (2021).
Fox, P. R. et al. Multicenter evaluation of plasma n-terminal probrain natriuretic peptide (NT-pro BNP) as a biochemical screening test for asymptomatic (occult) cardiomyopathy in cats. J. Vet. Intern. Med. 25, 1010–1016 (2011).
Côté, E. et al. Feline Cardiology. 51–67 (2011) https://doi.org/10.1002/9781118785782.ch7.
Ontiveros, E. S., Ueda, Y., Harris, S. P. & Stern, J. A. Consortium, 99. Lives Precision medicine validation: identifying the MYBPC3 A31P variant with whole-genome sequencing in two Maine Coon cats with hypertrophic cardiomyopathy. J. Feline Med. Surg. 21, 1086–1093 (2018).
Abbott, J. A. & MacLean, H. N. Two-dimensional echocardiographic assessment of the feline left atrium. J. Vet. Intern. Med. 20, 111 (2006).
Karsten, S., Stephanie, S. & Vedat, Y. Reference intervals and allometric scaling of two-dimensional echocardiographic measurements in 150 healthy cats. J. Vet. Med. Sci. 79, 17–0250 (2017).
Boon, J. A. Veterinary Echocardiography. (2011).
McLaren, W. et al. The ensembl variant effect predictor. Genome Biol 17, 122 (2016).
Funding
This research was supported by NIH R21 HL093603 to S.P.H. MyoKardia, Inc. (South San Francisco, CA), Cytokinetics, Inc (South San Francisco, CA), and Edgewise Therapeutics (Boulder, CO) provided funding support for colony development and maintenance. Support for V.N.R. was provided in part by NIH T32 HL086350 and TL1 TR001861. Support for J.L.K. was provided in part by NIH T32 OD011147 and Morris Animal Foundation D22FE-027. Y.U. was provided with post-doctoral fellowship support from Morris Animal Foundation D18FE-402. IDEXX Laboratories, Inc. (Westbrook, ME) provided support for cardiac biomarker assessments. We thank Dr. William Swanson for artificial insemination procedures at the Center for Conservation and Research of Endangered Wildlife at the Cincinnati Zoo. Lastly, we thank the colony cats and are indebted to Maria Montano and Monica Lowery for their expert daily care of the cats and to the countless undergraduate, graduate and veterinary students, veterinary staff, and research scientists who participated in the care of the colony cats over the course of this study.
Author information
Authors and Affiliations
Contributions
S.P.H. conceived of, initiated, and funded the first generations of colony cats. J.A.S. and S.P.H. conceived the experiment(s). J.A.S., V.N.R., J.L.K., Y.U., M.S.O., E.S.O., K.B.K., S.J.vD., and S.P.H conducted the experiment(s). J.A.S., Y.U., V.N.R. and S.P.H. analyzed the results. J.A.S., V.N.R., and S.P.H wrote the original manuscript draft. All authors reviewed and revised the final manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Stern, J.A., Rivas, V.N., Kaplan, J.L. et al. Hypertrophic cardiomyopathy in purpose-bred cats with the A31P mutation in cardiac myosin binding protein-C. Sci Rep 13, 10319 (2023). https://doi.org/10.1038/s41598-023-36932-5
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-023-36932-5
- Springer Nature Limited