Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

The Metabolome in Finnish Carriers of the MYBPC3-Q1061X Mutation for Hypertrophic Cardiomyopathy

  • Benedicte Jørgenrud ,

    Contributed equally to this work with: Benedicte Jørgenrud, Mikko Jalanko

    Affiliations Hormone laboratory, Aker hospital, Oslo University Hospital, Oslo, Norway, Division of Women and Children’s Health, Department of Pediatric Research, Oslo University Hospital, Oslo, Norway

  • Mikko Jalanko ,

    Contributed equally to this work with: Benedicte Jørgenrud, Mikko Jalanko

    Affiliation Helsinki University Central Hospital, Department of Cardiology, Helsinki, Finland

  • Tiina Heliö,

    Affiliation Helsinki University Central Hospital, Department of Cardiology, Helsinki, Finland

  • Pertti Jääskeläinen,

    Affiliation Kuopio University Hospital, Heart Center, Kuopio, Finland

  • Mika Laine,

    Affiliation Helsinki University Central Hospital, Department of Cardiology, Helsinki, Finland

  • Mika Hilvo,

    Affiliation VTT Technical Research Centre of Finland, Espoo, Finland

  • Markku S. Nieminen,

    Affiliation Helsinki University Central Hospital, Department of Cardiology, Helsinki, Finland

  • Markku Laakso,

    Affiliation University of Eastern Finland and Kuopio University Hospital, Department of Medicine, Kuopio, Finland

  • Tuulia Hyötyläinen,

    Affiliations Steno Diabetes Center, 2820 Gentofte, Denmark, VTT Technical Research Centre of Finland, Espoo, Finland

  • Matej Orešič ,

    mtjo@steno.dk

    Affiliations Steno Diabetes Center, 2820 Gentofte, Denmark, VTT Technical Research Centre of Finland, Espoo, Finland

  • Johanna Kuusisto

    Affiliation University of Eastern Finland and Kuopio University Hospital, Department of Medicine, Kuopio, Finland

The Metabolome in Finnish Carriers of the MYBPC3-Q1061X Mutation for Hypertrophic Cardiomyopathy

  • Benedicte Jørgenrud, 
  • Mikko Jalanko, 
  • Tiina Heliö, 
  • Pertti Jääskeläinen, 
  • Mika Laine, 
  • Mika Hilvo, 
  • Markku S. Nieminen, 
  • Markku Laakso, 
  • Tuulia Hyötyläinen, 
  • Matej Orešič
PLOS
x

Abstract

Aims

Mutations in the cardiac myosin-binding protein C gene (MYBPC3) are the most common genetic cause of hypertrophic cardiomyopathy (HCM) worldwide. The molecular mechanisms leading to HCM are poorly understood. We investigated the metabolic profiles of mutation carriers with the HCM-causing MYBPC3-Q1061X mutation with and without left ventricular hypertrophy (LVH) and non-affected relatives, and the association of the metabolome to the echocardiographic parameters.

Methods and Results

34 hypertrophic subjects carrying the MYBPC3-Q1061X mutation, 19 non-hypertrophic mutation carriers and 20 relatives with neither mutation nor hypertrophy were examined using comprehensive echocardiography. Plasma was analyzed for molecular lipids and polar metabolites using two metabolomics platforms. Concentrations of branched chain amino acids, triglycerides and ether phospholipids were increased in mutation carriers with hypertrophy as compared to controls and non-hypertrophic mutation carriers, and correlated with echocardiographic LVH and signs of diastolic and systolic dysfunction in subjects with the MYBPC3-Q1061X mutation.

Conclusions

Our study implicates the potential role of branched chain amino acids, triglycerides and ether phospholipids in HCM, as well as suggests an association of these metabolites with remodeling and dysfunction of the left ventricle.

Introduction

Hypertrophic cardiomyopathy (HCM) is an autosomally inherited myocardial disease occurring in approximately 1 out of 500 individuals [1]. Mutations in the genes encoding mainly cardiac sarcomeric proteins result in ventricular hypertrophy and dysfunctional myocardium. HCM patients have an increased risk of sudden death, and HCM is the most common cause of sudden cardiac death attributable to fatal ventricular arrhythmias in athletes and adolescents. Although many patients remain asymptomatic through life, some develop dyspnea, arrhythmias and end-stage heart failure [1].

Mutations of the myosin-binding protein C (MYBPC) gene often result in truncated proteins, and are the most common genetic cause of HCM worldwide [2]. In Finland, a novel founder mutation of the cardiac myosin-binding protein C gene (MYBPC3-Q1061X) has been identified in Finnish HCM patients, which leads to a truncated protein lacking the binding sites for myosin and titin [3,4] (S1 Fig). This mutation accounts for approximately 11% of HCM in Finnish patients, and is the most prevalent HCM-causing mutation in Finland with a penetrance of approximately 78% [5]. Because MYBPC3 mutations in previous studies have been reported to have a rather favorable prognosis, it may be even more common [6,7]. Generally, the phenotype of subjects with MYBPC3 mutations is variable, and cardiac hypertrophy usually does not manifest until early adolescence with penetrance increasing with age and peaking at approximately 60 years of age [810]. The molecular mechanisms leading to HCM in mutation carriers, such as asymmetric hypertrophy, myofibrillar disarray and fibrosis, are poorly understood.

Metabolomics is the global study of metabolites, such as lipids, sugars and amino acids, in cells, tissues and biofluids. The metabolome is sensitive to various pathogenically relevant factors including genetic variation, diet, age, immune system status and gut microbiota [1115]. Metabolomics is therefore a powerful tool for the characterization of complex phenotypes affected by genetic and environmental factors as well as the interactions thereof [16]. Metabolomics studies in HCM and other cardiovascular diseases are of particular interest, both to better understand the pathophysiology and to identify novel biomarkers [17]. Since cardiovascular diseases are often characterized by long prodromal periods, with clinical symptoms occurring rather late in the pathogenesis, it is important to identify individuals at high risk in the prodromal phase when intervention would be most optimal. Several different metabolite profiling methods have been developed, which allow for extensive and quantitative investigation of a large amount of different metabolites [18]. To our knowledge, no metabolomics studies on HCM in humans have so far been reported.

We aimed to investigate the metabolic profiles of carriers of the MYBPC3-Q1061X mutation with and without left ventricular hypertrophy (LVH), and compare these profiles to those of subjects with neither the mutation nor LVH. All subjects were assessed with comprehensive echocardiography including tissue Doppler imaging. Plasma samples were analyzed with two analytical platforms for metabolomics, with broad analytical coverage of molecular lipids and polar metabolites.

Methods

Subjects

A total of 73 genotyped subjects from families carrying the MYBPC3-Q1061X mutation were recruited from the Helsinki and Kuopio University Hospital regions. Mutation carriers diagnosed with clinical HCM (maximum wall thickness of left ventricle (MWT) ≥ 13 mm on echocardiography) numbering 34 patients were included in the G+/LVH+ group. The G+/LVH- group included 19 mutation carriers without hypertrophy (MWT < 13 mm), and the control group included 20 subjects from the same families with neither the mutation nor LVH. Exclusion criteria were age (<18 years) and pregnancy. The cardiac phenotype of the subjects was characterized by anthropometrics, blood samples and echocardiography. The blood samples were collected in the fasting state (12 hours) and immediately centrifuged at 3200 G for 10 minutes at 4°C. Plasma was separated and stored at -70°C (Kuopio) and -80°C (Helsinki). The local ethics committees of the University Hospitals of Helsinki and Kuopio in Finland approved the study protocol (96/2008). The study is in agreement with the principles outlined in the Declaration of Helsinki. All patients gave written consent.

Echocardiography

To assess cardiac structural changes and cardiac function, GE Vivid 7 ultrasound equipment with an M4S probe at the University Hospitals of Helsinki and Kuopio were used. Echocardiography was performed by experienced clinical cardiologists (MJ, PJ, JK). Measurements were analyzed offline with EchoPac software (GE Vingmed, version 10.0.1, Norway) by the same observer (MJ) blinded to clinical and genetic data at the Helsinki University Central Hospital.

Conventional measurements of cardiac anatomy and function were performed with M-mode, 2D and pulsed-wave Doppler according to guidelines [19]. Biplane Simpson method in the apical four- and two-chamber views was used to measure left ventricular ejection fraction (LVEF). Parasternal short axis views were used to measure maximum wall thickness (MWT). If apical hypertrophy was suggested by imaging an apical view was employed. Presence of systolic anterior motion (SAM) and left ventricular outflow tract (LVOT) obstruction, defined as a gradient of ≥ 30 mmHg measured at rest, were recorded. Left ventricular mass (LV mass) was calculated by the formula: 0.80 (1.04 x [PWT + SWT+LVEDD]3 –LVEDD3) + 0.6), where PWT and SWT is posterior and septal wall thickness, respectively, and LVEDD is left ventricular internal diameter at diastole [20].

Tissue Doppler imaging (TDI) velocities of systolic (Sm) and early diastolic motion (Em) were measured from the septal and lateral mitral valve annulus.

Metabolomic analysis

Two analytical platforms for metabolic profiling were applied to all samples: (a) platform for global profiling of small, polar metabolites, based on comprehensive two-dimensional gas chromatography coupled with time-of-flight mass spectrometry (GC×GC-TOFMS), covering molecules such as sugars, sterols, amino acids and various organic acids, including free fatty acids and ketoacids; (b) global lipidomics platform based on ultra performance liquid chromatography coupled to quadrupole time-of-flight mass spectrometry (UPLC-QTOFMS), covering various molecular lipids such as phospholipids, triglycerides, sphingolipids and neutral lipids. Both platforms have been extensively described [21,22], and are also described in the S1 Methods. The raw data from UPLC-QTOFMS platform were processed with MZmine 2 [23], while the GC×GC-TOFMS data were processed with the Guineu software [21].

A total of 699 molecular lipids and 1603 small polar metabolites were detected, of which 238 lipids and 215 metabolites were identified. Only lipids and metabolites identified and found in at least 70% of the samples were used in the statistical analysis.

Cluster analysis

The lipidomics and metabolomics data were scaled with zero mean and unit variance to make the profiles comparable with each other, and Bayesian model-based clustering was applied on the scaled data to group metabolites with similar profiles across all samples. The clustering was performed using the mclust [24] method, implemented in R [25] in the package “metadar” [26]. In mclust, the observed data are viewed as a mixture of several clusters and each cluster comes from a unique probability density function. A number of clusters in the mixture, together with the cluster-specific parameters that constrain the probability distributions, will define a model, which can then be compared with others. The clustering process selects the optimal model and determines the data partition accordingly.

Statistical analyses and linear mixed models

Statistical analyses were performed with R version 2.15.0 for Windows (R Development Core Team (2011), R Foundation for Statistical Computing, Vienna, Austria) and IBM SPSS Version 19.0 (IBM Corp. Released 2010. IBM SPSS Statistics for Windows, Version 19.0. Armonk, NY: IBM Corp.). Most analyses and visualization in R were performed using the “metadar” package for metabolomics data analysis [26]. Clinical, echocardiographic and individual metabolite level data between the study groups were compared with the independent samples T-test or Fisher’s exact test for 2 groups and for 3-groups we employed one-way analysis of variance (ANOVA) with pairwise comparison adjusted with the Bonferroni correction or Tukey’s range test. The average within-cluster metabolite profiles between the three groups were analyzed with one-way ANOVA. Analysis of covariance (ANVOCA) was performed by adjusting the data for age and gender, and then performing one-way ANOVA on the adjusted data. Individual metabolite levels were visualized using beanplots [27], implemented in the ‘beanplot’ R package. Beanplots provide information on the mean metabolite level within each group, the density of the data-point distribution, as well as show individual data points. To assess the association between metabolic profile and echocardiographic parameters in mutation carriers with and without hypertrophy, correlation analyses using Spearman product-moment correlation for pairwise complete data were performed in R, as implemented in “metadar” [26]. Scatterplots and linear regression between metabolite levels and echocardiographic parameters were performed in SPSS. All tests were assessed with a significance level of p≤ 0.05.

Results

Clinical characteristics and echocardiography

The baseline clinical characteristics are presented in Table 1. There were more male subjects in the pooled G+ mutation carrier group compared to the control group. The subjects in the G+ group were only asymptomatic or mildly symptomatic (New York Heart Association Functional Classification class I-II). Some of the subjects in both study groups had hypertension, and a few of the subjects in the G+ group had atrial fibrillation (AF). There was no difference in the body mass index, heart rate, blood pressure or lipid levels between the control and G+ groups. There was no significant difference in age between the control and G+ subjects, but in the subgroups analysis, G+/LVH- subjects were younger than subjects in other two groups (data not shown). Baseline data is listed in S1 Dataset.

thumbnail
Table 1. Clinical characteristics of the MYBPC3-Q1061X mutation carriers (G+) and control group.

https://doi.org/10.1371/journal.pone.0134184.t001

Echocardiographic data in control, G+/LVH- and G+/LVH+ groups are shown in Table 2. Left ventricular dimensions and ejection fraction (LVEF) were within normal range in all three groups. By definition, MWT and LV mass were higher in hypertrophic subjects compared to non-hypertrophic subjects. Left atrial size was larger in the G+/LVH+ group compared to other two groups. There was no difference in LVOT gradient between the three groups, but 7 patients of the G+/LVH+ groups had systolic anterior motion of the mitral valve apparatus (SAM) and two had LVOT obstruction (LVOT gradient > 30 mmHg at rest). Mitral valve E velocity did not differ between the study groups. TDI septal and lateral annular Sm was diminished in the G+/LVH+ group indicating systolic dysfunction despite preserved LVEF. In the G+/LVH+ group, early diastolic velocities at the lateral and septal mitral annulus were significantly lower compared to other two groups, consistent with diastolic dysfunction typical of HCM. The G+/LVH- group had higher TDI lateral Em compared to the control group, presumably because the G+/LVH- group was younger than the control group. Otherwise there were no significant differences in baseline echocardiographic variables between the G+/LVH- and control groups (Table 2). Echocardiographic data is listed in S2 Dataset.

Association of the global metabolome with the MYBPC3-Q1061X mutation and LVH

Following the data processing and filtering as described in Methods, a total of 86 polar metabolites and 238 molecular lipids were included in the data analysis. Metabolite-data are listed in S3 Dataset, as is lipid-data (S4 Dataset). Since many metabolites are co-regulated, it cannot be assumed that the profiles of all measured metabolites are independent [28]. Therefore, all lipids and metabolites were first surveyed by clustering of the data into subsets using Bayesian model-based clustering [24]. The identified lipids and polar metabolites were decomposed into seven and four clusters, respectively (LC1-LC7, MC1-MC4; Table 3). The division of lipid clusters reflects different lipid functional and structural groups, while the metabolite profiles are decomposed into three small clusters and one large cluster comprising different metabolite functional groups. When comparing the three groups (one-way ANOVA), one of the lipid clusters reached significance level (LC3) while one lipid (LC5) and one metabolite (MC4) cluster were marginally significant (Fig 1a).

thumbnail
Table 3. Description of metabolite clusters from the lipidomics (LC) and metabolomics (MC) platforms.

https://doi.org/10.1371/journal.pone.0134184.t003

thumbnail
Fig 1. Metabolomic profiles across the three study groups.

(a) Mean metabolite levels within each cluster for the three groups. The error bars show standard error of the mean (SEM), and cluster LC3 is the only significant cluster. Nominal p-values are shown (one-way ANOVA). (b) Profiles of selected representative metabolites from different clusters in all three groups. The metabolite levels are shown as beanplots [27], which provide information on the mean level (solid line), individual data point (short black lines), and the density of the distribution.

https://doi.org/10.1371/journal.pone.0134184.g001

The lipid cluster LC3 which mainly includes long-chain polyunsaturated fatty acid (PUFA) containing triglycerides (TGs), and phospholipids (plasmalogens) was significantly elevated in HCM patients (p = 0.037), especially when compared to the control group (p = 0.047, Tukey’s range test). This indicates that increased levels of TGs and phospholipids with a high degree of double bonds are associated with HCM caused by MYBPC3-Q1061X. The remaining lipid clusters also tended to be higher in HCM patients, although none of these differences were statistically significant.

One of the smaller clusters comprising polar metabolites, MC4, contains two branched chain amino acids (BCAAs) leucine and isoleucine, while two other small clusters contain carboxylic acids. The large cluster contains various metabolites, among others 4-methyl-2-oxovaleric acid, a catabolic product of leucine, and several organic acids involved in the citric acid cycle, including α-ketoglutaric acid and malic acid.

Association of molecular lipids and polar metabolites with the MYBPC3-Q1061X mutation and LVH

When analyzing the lipids at the molecular level (one-way ANOVA), several molecular lipids were elevated in the G+/LVH+ group (Table 4). Out of the 238 lipids, 28 were different between the groups. Several TGs were increased in the G+/LVH+ group compared to the control group, and some also to the G+/LVH- group. These TGs all had long-chain, even-numbered (54–58 carbons) fatty acids and were mostly polyunsaturated (4–7 double bonds). The only TG which was decreased in the G+/LVH+ group as compared to both the control group and G+/LVH- group was a triglyceride with an odd number of acyl carbons; TG(53:7) (p = 0.024 and p = 0.026, respectively). Most of the TGs were marginally increased in the G+/LVH- group compared to the control group, although none of the differences was statistically significant. The two ether phospholipids PE(38:6e) and PE(38:7e) were increased in the G+/LVH+ group compared to the control group.

thumbnail
Table 4. Mean metabolite-levels for significant molecular lipids and polar metabolites.

https://doi.org/10.1371/journal.pone.0134184.t004

When analyzing individual polar metabolites, six were different between the groups (Table 4). All three BCAAs were increased in the G+/LVH+ group. Valine was 1.12-fold higher compared to the control group and 1.05-fold compared to the mutation carriers in the G+/LVH- group (p = 0.021). Leucine and its metabolic product, 4-methyl-2-oxovaleric acid (ketoleucine), had even higher concentrations in the G+/LVH+ group compared to the other groups (p = 0.05 and p = 0.007, respectively). Leucine was increased 1.18-fold compared to the control group (p = 0.041), and 4-methyl-2-oxovaleric acid was increased 1.24-fold (p = 0.013) and 1.2-fold (p = 0.041) compared to the control group and mutation carriers without LVH, respectively. Because of the difference in age and gender between the groups, we also performed an ANCOVA adjusted for these two variables. Valine and leucine were then only marginally different (p = 0.060 and p = 0.059), while 4-methyl-2-oxovaleric acid was not significant (S1 Table). Isoleucine was also increased in the G+/LVH+ group, by 1.19-fold compared to the control group and 1.04-fold compared to mutation carriers without LVH, although the differences were not statistically significant.

Selected lipids and polar metabolites are shown in Fig 1b.

Correlation analysis of metabolomic variables to echocardiographic measurements

Spearman correlation coefficients for the relationship between the metabolites and the echocardiographic parameters were obtained for the pooled G+/LVH- and G+/LVH+ groups (Fig 2). In subjects with the MYBPC3-Q1061X mutation, 4-methyl-2-oxovaleric acid, alpha-ketoglutaric acid, benzenepropanoic acid and glutamic acid were positively correlated with several of the echocardiographic variables, such as LVEDD, MWT, LV mass and LAD. The lysophospholipids, phospholipids and triglycerides were positively correlated with MWT and LV mass. Several of the lipids were negatively correlated to TDI septal Sm, TDI lateral Sm, TDI septal Em and TDI lateral Em, with the PE plasmalogens being significantly correlated. The odd-numbered triglyceride TG(53:7) showed an opposite correlation pattern as compared to the other TGs.

thumbnail
Fig 2. Correlation heatmap and scatterplots.

(a) Heatmap of Spearman product-moment correlation coefficients between metabolites and echocardiographic parameters for pooled G+/LVH- and G+/LVH+ groups. The colors represent the correlation, with red being more positive and blue more negative. Significance is given as *(p<0.05), **(p<0.01) and ***(p<0.001). (b) Scatterplots showing the relationship between lysoPE(18:2) and MWT (left), and PE(40:8e) and MWT (right).

https://doi.org/10.1371/journal.pone.0134184.g002

Discussion

Principal findings

We have shown that the patients with HCM attributable to the Q1061X mutation of MYBPC3 display elevated levels of several triglycerides and amino acids including BCAAs, as well as some phospholipids, compared to control subjects without LVH. Although we did not detect any significant metabolite-differences between mutation carriers without hypertrophy and the control group, there is a trend of several of the metabolites in the non-hypertrophic mutation carrier group in the same direction as the metabolites in the hypertrophic group. In the pooled group analysis including all subjects with the MYBPC3-Q1061X mutation, several metabolites, including several TGs, ether PEs and lysophospholipids were significantly correlated with indices of LV remodeling such as hypertrophy, LV mass, and systolic and diastolic dysfunction, indicating a shift in the metabolic profile with manifest hypertrophic cardiomyopathy.

In the context of current literature

To our knowledge, no metabolomics studies of HCM in humans have so far been reported. In earlier metabolomics studies using a similar design on dilated cardiomyopathy (DCM), potential biomarkers such as steroid metabolites, glutamine, threonine and histidine were found to be diminished and TCA cycle intermediates and lipid β-oxidation products were found to be increased in DCM patients compared to control individuals [29]. We have previously reported that odd-chain triglycerides are diminished in patients with DCM attributable to the mutations in the lamin A/C gene (LMNA) [30]. In hamster models with a pathophysiology similar to human DCM, several metabolites involved in glycolysis, pentose phosphate pathway and TCA cycle, together with triglycerides, were decreased as compared to the control animals [18]. These studies suggest altered energy production and tissue remodeling in DCM. Comparing the metabolomic changes in DCM to the results of this study on HCM, both exhibit a decrease in glutamate and increase in threonine levels, but in contrast the levels of triglycerides are decreased in DCM and increased in HCM.

Possible mechanisms

In our study, levels of several TGs with even-numbered chains were increased in hypertrophic mutation carriers. The levels of these TGs correlated with several echocardiographic variables such as increased wall thickness and reduced systolic and diastolic function. Substrates for myocardial energy production in humans are variable throughout life. While the healthy adult heart primarily oxidize long-chain fatty acids as energy supply [31], the hypertrophic and failing heart shifts towards glucose and lactate metabolism, which is also seen in conditions such as ischemia, hypothyroidism, hypoxia, diabetes and athrophy [32]. This transition from oxidative fatty acid metabolism to glucose metabolism in cardiac hypertrophy is thought to be associated with the pathological remodeling of the heart [33]. A Finnish study of patients with HCM attributable to the Asp175Asn mutation in the α-tropomyosin gene revealed increased myocardial oxidative metabolism and free fatty acid metabolism in HCM patients with mild LVH, which decreased with advanced LV hypertrophy [34]. The increased levels of circulating TGs in HCM patients in this study might represent a shift in myocardial metabolism substrates.

Phospholipids are major components in heart tissue, and patients with HCM exhibited elevated levels of several ether phospholipids compared to mutation carriers without LVH. Earlier animal studies on hamsters with pathophysiology similar to human DCM [18] and rats with aortic constriction-induced LVH from birth [35] have shown contradictory results on phospholipid composition. Although these remodeling processes probably are different from that of HCM, they indicate a dysregulation of membrane phospholipid homeostasis which may contribute to the development of several cardiomyopathies [36].

In addition to increased lipid levels, HCM patients displayed increased levels of the BCAAs valine and leucine, which are catabolized in non-hepatic tissue, such as cardiac muscle [37]. We found that leucine clusters with isoleucine, meaning they show a similar concentration-pattern among the subjects, while 4-methyl-2-oxovaleric acid clusters together with valine in the larger MC1 cluster. BCAAs are converted into branched-chain α-keto-acids (BCKA) and further into propionyl-CoA by the BCKA dehydrogenase complex (BCKD), before it is degraded into acetyl-CoA or succinyl-CoA and enters the citric acid cycle. The BCKD complex must be dephosphorylated by a protein phosphatase, PP2C, in order to be active [38]. Studies show a high expression of mitochondrial PP2C (PP2Cm) in cardiac muscles in zebra-fish embryos and adult mice, with stress being an important regulator, as demonstrated by significantly reduced expression of both PP2C mRNA and protein in hypertrophic and failing hearts [39]. This reduction leads to elevated plasma levels of BCAA and BCKA in PP2Cm-deficient mice [38].

Morpholinos employed to suppress expression of PP2Cm in developing zebra-fish embryos resulted in dose-dependent development defects, contractile dysfunction and heart failure [39]. BCAAs are also efficient activators of the mTOR (mammalian target of rapamycin) pathway, which regulates protein synthesis. A local elevation of BCAA concentrations may lead to chronic induction of mTOR, promoting cardiac hypertrophy by altering insulin sensitivity [37]. The early development of HCM is characterized by reduced energy efficiency as shown previously by diminished PCr/ATP ratio and myocardial external efficiency even in carriers of HCM mutations without hypertrophy [40,41]. The MYBPC3-Q1061X mutation most likely results in haploinsufficiency leading to contractile impairment and impaired energy utilization [42,43]. The LV septum requires most energy in contraction and Ashrafian et al. have hypothesized that this chronic energy depletion leads through myocyte dysfunction, AMP-activated protein kinase activation and increase in cytosolic Ca2+ into the characteristic asymmetric septal hypertrophy most commonly seen in HCM [44].

This is the first study to our knowledge to incorporate comprehensive metabolomics analysis and modern echocardiographic data in carriers of a single HCM-causing mutation. The metabolomic profile of increased TGs and BCAAs in HCM patients might indicate a change in energy metabolism similar to that observed in heart failure and hypertrophy due to other causes. The observed trend of these metabolites in the G+/LVH- group, in the same direction as in the HCM patients, might indicate that some changes in metabolism occur already before clinical hypertrophy, despite the lack of significance. Metabolomics might aid in finding non-hypertrophic mutation carriers who are at risk of developing HCM.

Limitations of the study

Even though the study population was relatively small and heterogeneous, and some HCM patients had hypertension, all participants with HCM had the MYBPC3-Q1061X mutation primarily responsible for hypertrophic changes. Also, because the metabolome is sensitive to various changes unrelated to the disease process in question, we tried to minimize these effects with cluster analysis, use of control subjects from the same families to minimize genetic differences, and age-adjustment of comparisons. The study subjects had normal Finnish diet. Finally, although the MYBPC3-Q1061X mutation is rare outside Finland, current literature suggest that most of the known MYBPC3 mutations lead to similar haploinsufficiency through truncated proteins, which are not detectable in the myocardium [42,43]. Therefore, the presented metabolic changes may not just be specific to the MYBPC3-Q1061X mutation, but may represent changes common to other MYBPC3 mutations resulting in truncated proteins.

Conclusions

In conclusion, we have identified several metabolic changes that appear to be associated with HCM attributable to the MYBPC3-Q1061X mutation. Especially the branched chain amino acids valine and leucine, 4-methyl-oxovaleric acid and PUFA-containing long chain TGs tend to show an increasing trend between the three groups, with significant differences between the G+/LVH+ and the control groups. Possible clinical implications for the findings in this study may include the use of the elevated levels of metabolites as biomarkers of the LV remodeling process in HCM, but more studies are needed to investigate the temporal relationship between the levels of metabolites and echocardiographic findings in LV remodeling.

Supporting Information

S1 Dataset. Baseline data.

Subject-index is listed in the first row, group in the second row, and baseline-data listed in separate rows.

https://doi.org/10.1371/journal.pone.0134184.s001

(XLSX)

S2 Dataset. Echocardiographic parameters.

Subject-index is listed in the first row, group in the second row, and echocardiographic parameters listed in separate rows.

https://doi.org/10.1371/journal.pone.0134184.s002

(XLSX)

S3 Dataset. Metabolite data.

Subject-index is listed in the first row, group in the second row, and individual metabolite-levels listed in separate rows.

https://doi.org/10.1371/journal.pone.0134184.s003

(XLSX)

S4 Dataset. Lipid data.

Subject-index is listed in the first row, group in the second row, and individual lipid-levels listed in separate rows.

https://doi.org/10.1371/journal.pone.0134184.s004

(XLSX)

S1 Fig. Overview of MYBPC.

Schematic drawing of MYBPC, indicating domains, myosin-binding region, titin-binding region, amino acid positions, and location of the MYBPC3-Q1061X mutation.

https://doi.org/10.1371/journal.pone.0134184.s005

(TIF)

S1 Methods. Lipidomic and metabolomic methods.

https://doi.org/10.1371/journal.pone.0134184.s006

(DOC)

S1 Table. Metabolite-levels for significant molecular lipids and polar metabolites after adjusting for age and gender.

https://doi.org/10.1371/journal.pone.0134184.s007

(DOC)

Acknowledgments

The authors would like to thank Ulla Lahtinen, Leena Öhrnberg, Anna-Liisa Ruuskepää, Päivi Pöhö and Ismo Mattila for sample extraction and instrument processing, Gopalacharyulu Peddinti and Marko Sysi-Aho for assistance with bioinformatics analysis (VTT Technical Research Centre of Finland), and Sini Weckström, and Satu Nenonen for their assistance in data collection.

Author Contributions

Conceived and designed the experiments: T. Heliö MO JK. Performed the experiments: BJ MJ. Analyzed the data: BJ MJ MH MO JK. Contributed reagents/materials/analysis tools: T. Heliö PJ M. Laine MSN M. Laakso T. Hyötyläinen MO JK. Wrote the paper: BJ MJ MO JK.

References

  1. 1. Maron BJ. Hypertrophic cardiomyopathy: A systematic review. JAMA. 2002;287(10):1308–1320. pmid:11886323
  2. 2. Maron BJ, Maron MS, Semsarian C. Genetics of hypertrophic cardiomyopathy after 20 years: Clinical perspectives. J Am Coll Cardiol. 2012;60(8):705–715. pmid:22796258
  3. 3. Jääskeläinen P, Kuusisto J, Miettinen R, Kärkkäinen P, Kärkkäinen S, Heikkinen S et al. Mutations in the cardiac myosin-binding protein C gene are the predominant cause of familial hypertrophic cardiomyopathy in eastern finland. J Mol Med (Berl). 2002;80(7):412–422.
  4. 4. Jääskeläinen P, Miettinen R, Kärkkäinen P, Toivonen L, Laakso M, Kuusisto J. Genetics of hypertrophic cardiomyopathy in eastern finland: Few founder mutations with benign or intermediary phenotypes. Ann Med. 2004;36(1):23–32. pmid:15000344
  5. 5. Jääskeläinen P, Heliö T, Aalto-Setälä K, Kaartinen M, Ilveskoski E, Hämäläinen L et al. Two founder mutations in the alpha-tropomyosin and the cardiac myosin-binding protein C genes are common causes of hypertrophic cardiomyopathy in the finnish population. Ann Med. 2013;45(1):85–90. pmid:22462493
  6. 6. Charron P, Dubourg O, Desnos M, Bennaceur M, Carrier L, Camproux AC et al. Clinical features and prognostic implications of familial hypertrophic cardiomyopathy related to the cardiac myosin-binding protein C gene. Circulation. 1998;97(22):2230–2236. pmid:9631872
  7. 7. Niimura H, Bachinski LL, Sangwatanaroj S, Watkins H, Chudley AE, McKenna W et al. Mutations in the gene for cardiac myosin-binding protein C and late-onset familial hypertrophic cardiomyopathy. N Engl J Med. 1998;338(18):1248–1257. pmid:9562578
  8. 8. Maron BJ, Spirito P, Wesley Y, Arce J. Development and progression of left ventricular hypertrophy in children with hypertrophic cardiomyopathy. N Engl J Med. 1986;315(10):610–614. pmid:2942774
  9. 9. Spirito P, Maron BJ. Absence of progression of left ventricular hypertrophy in adult patients with hypertrophic cardiomyopathy. J Am Coll Cardiol. 1987;9(5):1013–1017. pmid:2952700
  10. 10. Nishimura RA, Ommen SR, Tajik AJ. Cardiology patient page. hypertrophic cardiomyopathy: A patient perspective. Circulation. 2003;108(19):e133–5. pmid:14609998
  11. 11. Orešič M, Simell S, Sysi-Aho M, Nanto-Salonen K, Seppänen-Laakso T, Parikka V et al. Dysregulation of lipid and amino acid metabolism precedes islet autoimmunity in children who later progress to type 1 diabetes. J Exp Med. 2008;205(13):2975–2984. pmid:19075291
  12. 12. Illig T, Gieger C, Zhai G, Romisch-Margl W, Wang-Sattler R, Prehn C et al. A genome-wide perspective of genetic variation in human metabolism. Nat Genet. 2010;42(2):137–141. pmid:20037589
  13. 13. Lenz EM, Bright J, Wilson ID, Hughes A, Morrisson J, Lindberg H et al. Metabonomics, dietary influences and cultural differences: A 1H NMR-based study of urine samples obtained from healthy british and swedish subjects. J Pharm Biomed Anal. 2004;36(4):841–849. pmid:15533678
  14. 14. Nikkila J, Sysi-Aho M, Ermolov A, Seppänen-Laakso T, Simell O, Kaski S et al. Gender-dependent progression of systemic metabolic states in early childhood. Mol Syst Biol. 2008;4:197. pmid:18523432
  15. 15. Velagapudi VR, Hezaveh R, Reigstad CS, Gopalacharyulu P, Yetukuri L, Islam S et al. The gut microbiota modulates host energy and lipid metabolism in mice. J Lipid Res. 2010;51(5):1101–1112. pmid:20040631
  16. 16. Orešič M. Metabolomics, a novel tool for studies of nutrition, metabolism and lipid dysfunction. Nutr Metab Cardiovasc Dis. 2009;19(11):816–824. pmid:19692215
  17. 17. Griffin JL, Atherton H, Shockcor J, Atzori L. Metabolomics as a tool for cardiac research. Nat Rev Cardiol. 2011;8(11):630–643. pmid:21931361
  18. 18. Maekawa K, Hirayama A, Iwata Y, Tajima Y, Nishimaki-Mogami T, Sugawara S et al. Global metabolomic analysis of heart tissue in a hamster model for dilated cardiomyopathy. J Mol Cell Cardiol. 2013;59:76–85. pmid:23454301
  19. 19. Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA et al. Recommendations for chamber quantification. Eur J Echocardiogr. 2006;7(2):79–108. pmid:16458610
  20. 20. Devereux RB, Alonso DR, Lutas EM, Gottlieb GJ, Campo E, Sachs I et al. Echocardiographic assessment of left ventricular hypertrophy: Comparison to necropsy findings. Am J Cardiol. 1986;57(6):450–458. pmid:2936235
  21. 21. Castillo S, Mattila I, Miettinen J, Orešič M, Hyötyläinen T. Data analysis tool for comprehensive two-dimensional gas chromatography/time-of-flight mass spectrometry. Anal Chem. 2011;83(8):3058–3067. pmid:21434611
  22. 22. Nygren H, Seppänen-Laakso T, Castillo S, Hyötyläinen T, Orešič M. Liquid chromatography-mass spectrometry (LC-MS)-based lipidomics for studies of body fluids and tissues. Methods Mol Biol. 2011;708:247–257. pmid:21207295
  23. 23. Pluskal T, Castillo S, Villar-Briones A, Orešič M. MZmine 2: Modular framework for processing, visualizing, and analyzing mass spectrometry-based molecular profile data. BMC Bioinformatics. 2010;11:395-2105-11-395. pmid:20650010
  24. 24. Fraley C, Raftery AE. Model-based methods of classification: Using the mclust software in chemometrics. J Stat Softw. 2007;18(6):1–13.
  25. 25. Dalgaard P. Introductory statistics with R. 2nd ed. New York: Springer Verlag; 2008.
  26. 26. Peddinti G. Metadar; R package for metabolomics data analysis. http://code.google.com/p/metadar. Updated 2013.
  27. 27. Kampstra P. Beanplot: A boxplot alternative for visual comparison of distributions. J Stat Soft. 2008;28:1–9.
  28. 28. Steuer R, Kurths J, Fiehn O, Weckwerth W. Observing and interpreting correlations in metabolomic networks. Bioinformatics. 2003;19(8):1019–1026. pmid:12761066
  29. 29. Alexander D, Lombardi R, Rodriguez G, Mitchell MM, Marian AJ. Metabolomic distinction and insights into the pathogenesis of human primary dilated cardiomyopathy. Eur J Clin Invest. 2011;41(5):527–538. pmid:21155767
  30. 30. Sysi-Aho M, Koikkalainen J, Seppänen-Laakso T, Kaartinen M, Kuusisto J, Peuhkurinen K et al. Serum lipidomics meets cardiac magnetic resonance imaging: Profiling of subjects at risk of dilated cardiomyopathy. PLoS One. 2011;6(1):e15744. pmid:21283746
  31. 31. Lopaschuk GD, Ussher JR, Folmes CD, Jaswal JS, Stanley WC. Myocardial fatty acid metabolism in health and disease. Physiol Rev. 2010;90(1):207–258. pmid:20086077
  32. 32. Taegtmeyer H, Sen S, Vela D. Return to the fetal gene program: A suggested metabolic link to gene expression in the heart. Ann N Y Acad Sci. 2010;1188:191–198. pmid:20201903
  33. 33. Kolwicz SC Jr, Tian R. Glucose metabolism and cardiac hypertrophy. Cardiovasc Res. 2011;90(2):194–201. pmid:21502371
  34. 34. Tuunanen H, Kuusisto J, Toikka J, Jääskeläinen P, Marjamäki P, Peuhkurinen K et al. Myocardial perfusion, oxidative metabolism, and free fatty acid uptake in patients with hypertrophic cardiomyopathy attributable to the Asp175Asn mutation in the alpha-tropomyosin gene: A positron emission tomography study. J Nucl Cardiol. 2007;14(3):354–365. pmid:17556170
  35. 35. Novák F, Kolár F, Hamplová B, Mrnka L, Pelouch V, Ostádal B et al. Myocardial phospholipid remodeling under different types of load imposed during early postnatal development. Physiol Res. 2009;58 Suppl 2:S13–32. pmid:20131931
  36. 36. Wende AR, Abel ED. Lipotoxicity in the heart. Biochim Biophys Acta. 2010;1801(3):311–319. pmid:19818871
  37. 37. Huang Y, Zhou M, Sun H, Wang Y. Branched-chain amino acid metabolism in heart disease: An epiphenomenon or a real culprit? Cardiovasc Res. 2011;90(2):220–223. pmid:21502372
  38. 38. Lu G, Sun H, She P, Youn JY, Warburton S, Ping P et al. Protein phosphatase 2Cm is a critical regulator of branched-chain amino acid catabolism in mice and cultured cells. J Clin Invest. 2009;119(6):1678–1687. pmid:19411760
  39. 39. Lu G, Ren S, Korge P, Choi J, Dong Y, Weiss J et al. A novel mitochondrial matrix serine/threonine protein phosphatase regulates the mitochondria permeability transition pore and is essential for cellular survival and development. Genes Dev. 2007;21(7):784–796. pmid:17374715
  40. 40. Crilley JG, Boehm EA, Blair E, Rajagopalan B, Blamire AM, Styles P et al. Hypertrophic cardiomyopathy due to sarcomeric gene mutations is characterized by impaired energy metabolism irrespective of the degree of hypertrophy. J Am Coll Cardiol. 2003;41(10):1776–1782. pmid:12767664
  41. 41. Timmer SA, Germans T, Brouwer WP, Lubberink M, van der Velden J, Wilde AA et al. Carriers of the hypertrophic cardiomyopathy MYBPC3 mutation are characterized by reduced myocardial efficiency in the absence of hypertrophy and microvascular dysfunction. Eur J Heart Fail. 2011;13(12):1283–1289. pmid:22021246
  42. 42. van Dijk SJ, Dooijes D, dos Remedios C, Michels M, Lamers JM, Winegrad S et al. Cardiac myosin-binding protein C mutations and hypertrophic cardiomyopathy: Haploinsufficiency, deranged phosphorylation, and cardiomyocyte dysfunction. Circulation. 2009;119(11):1473–1483. pmid:19273718
  43. 43. Marston S, Copeland O, Jacques A, Livesey K, Tsang V, McKenna WJ et al. Evidence from human myectomy samples that MYBPC3 mutations cause hypertrophic cardiomyopathy through haploinsufficiency. Circ Res. 2009;105(3):219–222. pmid:19574547
  44. 44. Ashrafian H, Redwood C, Blair E, Watkins H. Hypertrophic cardiomyopathy:A paradigm for myocardial energy depletion. Trends Genet. 2003;19(5):263–268. pmid:12711218