Branched-Chain Amino Acids As Critical Switches in Health and Disease
Zhen-Yu Zhang,Daniel Monleon,Peter Verhamme,Jan A. Staessen
DOI: https://doi.org/10.1161/hypertensionaha.118.10919
IF: 9.8968
2018-01-01
Hypertension
Abstract:HomeHypertensionVol. 72, No. 5Branched-Chain Amino Acids as Critical Switches in Health and Disease Free AccessReview ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessReview ArticlePDF/EPUBBranched-Chain Amino Acids as Critical Switches in Health and Disease Zhen-Yu Zhang, Daniel Monleon, Peter Verhamme and Jan A. Staessen Zhen-Yu ZhangZhen-Yu Zhang Correspondence to Zhen-Yu Zhang, KU Leuven Department of Cardiovascular Sciences, Studies Coordinating Centre, Research Unit Hypertension and Cardiovascular Epidemiology, University of Leuven, Campus Sint Rafaël, Kapucijnenvoer 35, Box 7001, BE-3000 Leuven, Belgium. Email E-mail Address: [email protected] From the KU Leuven Department of Cardiovascular Sciences, Research Unit Hypertension and Cardiovascular Epidemiology (Z.-Y.Z., J.A.S.), University of Leuven, Belgium Department of Cardiovascular Disease, Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, China (Z.-Y.Z.) Search for more papers by this author , Daniel MonleonDaniel Monleon Metabolomic and Molecular Image Laboratory, Fundación Investigatión Clínico de Valencia, Spain (D.M.) Search for more papers by this author , Peter VerhammePeter Verhamme KU Leuven Department of Cardiovascular Sciences, Center for Molecular and Vascular Biology (P.V.), University of Leuven, Belgium Search for more papers by this author and Jan A. StaessenJan A. Staessen From the KU Leuven Department of Cardiovascular Sciences, Research Unit Hypertension and Cardiovascular Epidemiology (Z.-Y.Z., J.A.S.), University of Leuven, Belgium Cardiovascular Research Institute, Maastricht University, the Netherlands (J.A.S.). Search for more papers by this author Originally published10 Sep 2018https://doi.org/10.1161/HYPERTENSIONAHA.118.10919Hypertension. 2018;72:1012–1022Other version(s) of this articleYou are viewing the most recent version of this article. Previous versions: September 10, 2018: Ahead of Print The aromatic amino acids (AAAs) histidine, phenylalanine, tryptophan, and tyrosine have a side chain with an aromatic ring and are precursors of dopamine, epinephrine, norepinephrine, serotonin, and thyroxine. The branched-chain amino acids (BCAAs), leucine, isoleucine, and valine, share a structurally similar side chain.1,2 Together with lysine, the AAAs and BCAAs are essential amino acids (EAAs), meaning that mammals, including humans, must meet their metabolic needs via a sufficient dietary intake. This brief review highlights how BCAAs play a key role as gatekeepers feeding into upstream and downstream molecular pathways and in this way are major metabolic regulators involved in the pathophysiology of disease. Maple syrup syndrome, characterized by neurotoxicity, motor disturbances, mental retardation, and premature death, rests on dysregulation of the BCAA catabolism. The disease has a Mendelian inheritance2 and falls outside the scope of this review.MetabolismGastrointestinal and Cellular Uptake of BCAAsBCAAs are neutral (nonpolar and hydrophobic) amino acids. System L is the major Na+-independent system in the intestinal basolateral membrane responsible for the uptake of neutral amino acids.3 It is expressed as 2 different isoforms: SLC7A5 also known as LAT1 (L-type amino acid transporter 1) and SLC7A8 also known as the LAT2 (L-type amino acid transporter 2). Both belong to the SLC7 (solute carrier family-7) gene family and require SLC3A2 (a heavy chain subunit), which brings the light subunit (SLC7A5 or SLC7A8)—the actual transporter—to the apical cell membrane.3 LAT1 and LAT2 are obligatory amino acid exchangers with a 1:1 stoichiometry and transport large neutral amino acids.4 LAT2 has a broader substrate selectivity range than LAT1, also accepting smaller amino acids.4 SLC7A5 regulates the simultaneous efflux of L-glutamine out of cells and the influx of L-leucine and other EAAs into cells, thereby activating downstream signaling, for instance, via mTOR (mammalian target of rapamycin) kinase.5,6 The primary BCAA transporter in the gut is LAT2.3 However, both LAT1 and LAT2 are ubiquitously expressed throughout the human body, in particular, in organs that play an important role in the homeostasis or usage of BCAAs, including the kidney, liver, brain, heart, lung, and muscle.Recent studies integrating the serum metabolome, the gut microbiome, and clinical phenotypes show that gut microbiota play an important role in regulating the bioavailability of BCAAs in the intestine and their transport through the gut wall.7 Several observations underpin this concept. First, in humans, the serum concentrations of BCAAs, bacterial BCAA biosynthesis, and BCAA transport through the intestinal wall are functionally interrelated.7 Second, insulin resistance—a phenotype associated with the BCAA level5—is also correlated with circulating bacterial cometabolites.7 Third, specific microbial species, including Prevotella copri and Bacteroides vulgatus are the main drivers of the association between the intraintestinal biosynthesis of BCAAs and insulin resistance.7 Finally, in mice, P. copri can augment circulating levels of BCAAs, induce insulin resistance, and aggravate glucose intolerance.7BCAA CatabolismA large proportion of BCAAs from dietary sources is absorbed from the intestine, but bypasses the liver, and is directly delivered to the peripheral tissues.8 BCAAs account for >50% of the splanchnic output of amino acids, even if they only constitute 20% of the ingested protein source.9 Although the enzymatic machinery for BCAA metabolism is active in the liver, BCAA catabolism occurs mainly in other tissues, including skeletal muscle (50%), heart, adipose tissue, renal tubular cells, and neurons.1The initial step in BCAA catabolism (Figure 1) consists of a transamination catalyzed by a BCAT (branched-chain amino acid transferase)—a readily reversible reaction that yields the corresponding branched-chain α-keto acid (BCKA).1 The human BCAT exists as 2 isoenzymes, encoded by separate genes and, respectively, expressed in the cytosol (BCAT1) and in mitochondria (BCAT2).10–12 The mitochondrial isoenzyme also functions as the branched-chain α-keto transport protein, which determines the rate of efflux of BCKAs from muscle tissue, when perfused with BCAAs.11,13 BCKAs are irreversibly catabolized by the BCKA dehydrogenase complex (BCKDC)—a multiprotein enzyme composed of 3 subunits: E1, E2, and E3.1,12,14 The E1 subunit of the BCKD complex (BCKDC) is a tetramer composed of 2 α-components and 2 β-components (α2β2).12 BCKDC has a molecular mass close to 4×106 Da and an overall diameter of ≈400 Å.14 It belongs to the mitochondrial α-keto acid dehydrogenase family, including also pyruvate dehydrogenase and α-ketoglutarate dehydrogenase, which are key enzymes that function in the tricarboxylic acid cycle (TCA), also known as the Krebs cycle.14 Two proteins are attached to the E1 subunit: a PP2Cm (mitochondrial protein phosphatase) and a BCKDK (BCKA dehydrogenase kinase). They function as key regulators of BCKDC activity, which is also allosterically modulated by the availability of substrates (Figure 1). PP2Cm is a required cofactor for the catabolism of BCKAs and activates the BCKDC complex.12,15 The BCKDK-mediated reaction is the rate-limiting step in the BCAA catabolic pathway. BCKDC expression and activity depend on phosphorylation status of its regulatory subunit E1α (Figure 1), catalyzed by BCKDK.12,14 When the BCAA level is low, E1α is hyperphosphorylated, leading to inhibition of BCKDC activity and preservation of free BCAAs, whereas when the BCAA level is high, E1α is dephosphorylated, leading to BCKDC activation and reduction of the total BCAA levels.16 The terminal metabolites of isoleucine catabolism are propionyl-CoA and acetyl-CoA (Figure 1). The isoleucine degradation products are, therefore, both glycogenic and ketogenic.1 Leucine yields acetoacetate and acetyl-CoA and is ketogenic.1 Valine yields succinyl-CoA and is glycogenic.1 Acetyl-CoA can enter the TCA15 and acetoacetate and succinyl-CoA are tricarboxylic acid intermediates, so that the terminal BCAA catabolites contribute to energy generation (Figure 1).Download figureDownload PowerPointFigure 1. Catabolic pathways of branched-chain amino acids (BCAAs) and downstream signaling initiated by catabolites (for explanation, see BCAA Catabolism section). 4E-BP1 indicates eukaryotic initiation factor 4E-binding protein 1; BCAT, branched-chain amino acid aminotransferase; BCKA, branched-chain keto acid; BCKDC, branched-chain α-keto acid dehydrogenase complex; BCKDK, branched-chain keto acid dehydrogenase kinase; INS, insulin; INSR, insulin receptor; mTOR, mammalian target of rapamycin; p+/p−, phosphorylation/dephosphorylation; PP2Cm, mitochondrial protein phosphatase; SK6β1, ribosomal protein S6 kinase β1; SLC3A2, solute carrier family 3 member 2; SLC7A5, L-type amino acid transporter 1; SLC7A8, L-type amino acid transporter 2; and TCA, tricarboxylic acid cycle.The activity of BCKDC across organs varies directly with the mitochondrial tissue content and with the subcellular distribution of BCAT between the mitochondrial and cytosolic compartments.10 The tissue distribution of BCKDK in mice by Western blot17 and in rats by mRNA detection18 shows similar patterns. Skeletal muscles contained the highest concentration of BCKDK.17,18 BCKDK in the liver was below the detection limit.17,18 Heart,17,18 kidney,17,18 lung,18 and small intestine18 had intermediate levels with lower levels also being detectable in the brain17,18 and eye.18 In humans,19 kidneys have the highest BCKDC activity, followed by liver, brain, heart, colon, skeletal muscle, and stomach. Subcutaneous adipose tissue and small intestine showed the lowest BCKDC activity.19 In line with the metabolic role of BCAAs, obese women, compared with nonobese controls, had lower expression and protein levels of BCAA-catabolizing enzymes in visceral adipose tissue.20FunctionBCAAs fulfill 3 distinct roles. First, they provide building blocks for protein synthesis and promote protein synthesis. Second, when catabolized via the TCA (described above), they act as a fuel source. Finally, they stimulate cell signaling via activation of mTOR, which uncouples insulin signaling5 and regulates protein translation, thereby balancing cell growth and autophagy (Figure 1).As protein building blocks, BCAAs account for ≈35% of the indispensable amino acids in muscle proteins and 40% of amino acids required for protein synthesis in mammals. BCAAs feed into the aminoacyl-tRNA biosynthesis.21 The enzyme aminoacyl-tRNA synthetase activates binding of an amino acid to tRNA to form an aminoacyl-tRNA ester. This is the crucial step in protein synthesis by determining how the genetic code is interpreted as an amino acid sequence.21 BCAAs are among the most hydrophobic of amino acids. Generally, the interior of water-soluble globular proteins consists, largely, of hydrophobic amino acids, principally leucine, isoleucine, valine, phenylalanine, and methionine.8,22 This configuration is important for the stability of the folded proteins and their function.8,22,23Because of their poor hepatic catabolism, circulating BCAAs, in particular, leucine, act as signaling molecules and nutritional sensors. They promote protein synthesis, cellular metabolism, and cell growth in an mTOR-dependent manner,24–26 thereby uncoupling insulin signaling (Figure 1). In this way, BCAAs are essential for normal growth and function at cellular and organism level.BCAAs in Health and DiseaseOur literature review highlighted the multifaceted role of BCAAs in various organ systems.Insulin ResistanceThe common feature of metabolic syndrome and type-2 diabetes mellitus (T2DM) is an impaired glucose-insulin homeostasis, commonly on a background of obesity.20,27 As evidenced by experiments in healthy volunteers,27,28 EAAs stimulate the insulin secretion without effect on insulin clearance from the blood. Glucose ingestion after an overnight fast triggers an insulin-dependent homeostatic response, which involves the 4 key axes of insulin action: proteolysis, lipolysis, ketogenesis, and glycolysis.29 Among prediabetic patients, the responses along all 4 axes are blunted,29 with increases in circulating lactate (marker of glycolysis) and declines in glycerol (lipolysis), leucine/isoleucine (proteolysis) and β-hydroxybutyrate (ketogenesis).29 Several cross-sectional,30–39 longitudinal,40–47 and genetic association48 studies supported the experimental studies27,28 showing interference of BCAAs with insulinotropic actions.Cross-Sectional StudiesIn the cross-sectional INTERMAP project (International Study of Macro/Micronutrients and Blood Pressure), the risk of stroke and cardiovascular complications was higher in Northern (Beijing and Shanxi) compared with Southern (Guangxi) China.49 A further metabolome-wide association study also revealed a higher urinary excretion of BCAAs in Northern China.30 This led the INTERMAP investigators to speculate that dietary factors and endogenous metabolic influences, partly emanating from the gut flora,7 might underlie the higher cardiovascular risk in the north.30Several other cross-sectional studies reported positive associations31–39 of circulating BCAAs with a high cardiovascular risk profile,34 hypertension,31,38 or with metabolic dysregulation, including an elevated body mass index,20,35 central obesity,20,35,38 impaired fasting glucose,31,33,39 insulin resistance estimated from the Homeostatic Model Assessment (HOMA-IR) index,31,32,35 the metabolic syndrome,33,34,36–38 diabetes mellitus,38 or dyslipidemia.35,38 These cross-sectional associations30–39 were replicated across diverse ethnic groups, women and men, a wide age range, obese and nonobese people, and in population and patients.Longitudinal StudiesMoving from cross-sectional30–39 to longitudinal40–47 studies, in the Framingham Heart Study,40 2422 nondiabetic subjects underwent a routine examination between 1991 and 1995, of whom 201 developed new-onset diabetes mellitus during a 12-year follow-up period and of whom 189 could be propensity matched with controls from the same baseline examination who did not develop diabetes mellitus. Cases and controls had a similar sex distribution (42% women), age (56.5 years), body mass index (30.2 kg/m2), and fasting glucose (105 mg/dL). The risk of developing diabetes mellitus was greater with higher baseline plasma levels of the 3 BCAAs (isoleucine, leucine, and valine) and 2 AAAs (tyrosine and phenylalanine). When compared with single amino acids, combinations of 3 amino acids further improved the prediction of diabetes mellitus with increments in the −2 log likelihood ratio from 6 to 9 (P≤0.035).40 In an independent replication sample, from the Malmö Diet and Cancer study consisting of 163 cases and 163 controls (55% women; mean age, 58 years), 4 of the 5 individual amino acids (leucine, valine, tyrosine, and phenylalanine) were significantly associated with incident diabetes mellitus (adjusted odds ratios per 1-SD increment were similar to Framingham, 1.37–2.01; P≤0.04). The remaining amino acid, isoleucine, had a nonsignificant association in the replication study (P=0.09).40 One potential limitation of these Framingham findings is that they might not be generalizable to a low-risk population sample.50After the 2011 Framingham report,40 a systemic review51 identified 7 other prospective studies,41–47 published from 201241,43 until 2015,42,46,47 which estimated the risk of developing prediabetes mellitus42,44,46 or T2DM41–47 from baseline levels of circulating metabolites. The sample size of these 7 additional studies, including the replication sample, if applicable,41,44,45 ranged from 14746 to 467845 and follow-up from 344 to 1947 years. Study participants were recruited in Germany,42,45 Finland,43 Framingham, Massachusetts, and Malmö, Sweden,41 multiple European countries,44 the United States,46 and the United Kingdom.47 Two studies were multiethnic,46,47 and 2 enrolled only men.43,47 Combining the initial Framingham study40 with the 7 additional prospective reports41–47 provided summary statistics encompassing 8000 individuals, of whom 1940 had T2DM. Per study-specific SD difference, the pooled relative risk of T2DM amounted to 1.36 (95% CI, 1.24–1.48; I2=9.5%) for isoleucine, 1.36 (CI, 1.17–1.58; I2=37.4%) for leucine, 1.35 (CI, 1.19–1.53; I2=45.8%) for valine, 1.36 (CI, 1.19–1.55; I2=51.6%) for tyrosine, and 1.26 (CI, 1.10–1.44; I2=56%) for phenylalanine.51 Glycine and glutamine were inversely associated with T2DM with relative risks amounting to 0.89 (CI, 0.81–0.96; I2=0%) and 0.85 (CI, 0.82–0.89; I2=0%), respectively.51Genetic Association StudiesA polymorphism in the PPM1K gene (rs1440581; C>T), encoding the mitochondrial protein phosphatase PP2Cm (Figure 1), affects the ratio of circulating BCAAs to AAAs, which both have been related to insulin resistance and diabetes mellitus in prospective cohort studies.52 In a 2-year intervention trial of 734 overweight or obese patients randomly assigned to 4 diets with varying content of fat (high versus low) and protein (high versus average), dietary fat significantly modified the genetic effects on changes in weight, fasting insulin, and HOMA-IR at 6 months.48 In the high-fat group, the C allele was related to less weight loss and smaller decreases in serum insulin and HOMA-IR (P≤0.02), whereas an opposite genotypic effect on changes in insulin and HOMA-IR was observed in the low-fat diet group (P≤0.04). At 2 years, the gene-diet interactions remained significant for weight loss (P=0.008) but became null for changes in serum insulin and HOMA-IR as a result from weight regain.48 These observations indirectly demonstrated how nutritional and genetic determinants interactively regulate the circulating levels of BCAAs and AAAs and their impact on metabolic traits.27,28,48Synergism Between Lipids and BCAAs in Relation to Insulin ResistanceDecades of research established a strong relation between glucose and lipid metabolism.53,54 However, accruing evidence suggests that BCAAs synergize with fatty acids and their metabolites in promoting insulin resistance.25,26,55,56 A plasma metabolomic study of >100 circulating metabolites were grouped by principal component analysis into 18 factors. It was demonstrated that the best factor discriminated between 74 obese insulin resistant patients and 67 lean insulin-sensitive controls (HOMAscore), was not lipid related but comprised the BCAAs valine, leucine/isoleucine, the AAAs phenylalanine and tyrosine, C3 and C5 acylcarnitines and glutamate/glutamine, and alanine.25 The preferential association of this BCAA-related metabolite cluster with insulin resistance was subsequently confirmed in a cross-sectional study of sedentary patients with a disturbed glucose tolerance test55 and in cohorts of Chinese and Asian-Indian men in Singapore, matched for body mass index.56Recent publications contributed to a better understanding of the mechanistic link between insulin resistance and BCAA catabolism.57,58 Insulin resistance in skeletal muscle stems from the excess accumulation of lipids—a process that requires blood-borne lipids to traverse the blood vessel wall.53,54,57 In mice, 3-hydroxyisobutyrate—a catabolic intermediate of valine—is secreted from muscle cells, activates endothelial fatty acid transport, stimulates muscle fatty acid uptake in vivo, and promotes lipid accumulation in muscle, leading to insulin resistance.58 Conversely, in a cell model consisting of mouse myoblasts and human umbilical vein endothelial cells, inhibition of the synthesis of 3-hydroxyisobutyrate reduced the endothelial fatty acid uptake.58 Along similar lines, 3-hydroxyisobutyrate levels were elevated in muscle from db/db mice58,59 with diabetes mellitus and from human patients with diabetes mellitus,60 as compared with those without diabetes mellitus. All catabolic products of the BCKAs are trapped inside the cell by covalent linkage to coenzyme A, with the single exception of 3-hydroxyisobutyrate, which is thus ideally suited to act as a secreted reporter of the BCAA catabolic flux in muscle.58 These findings mechanistically link the catabolic flux of BCAAs to insulin resistance.58HypertensionThe scarce studies that addressed the association of hypertension with dietary,61,62 urinary,30 or circulating63 BCAAs produced inconsistent results. Among 1898 female twins,61 aged 18 to 75 years, dietary intake of 7 amino acids was assessed from food frequency questionnaires and food composition tables. In multivariable-adjusted cross-sectional analyses comparing the bottom versus the top fifth of the distribution of dietary leucine consumption (1.0% versus 1.6% of energy intake), peripheral (−6/−3 mm Hg) and central (−5/−3 mm Hg) systolic/diastolic blood pressure and aortic pulse wave velocity (−0.5 m/s) were lower (P for trend, ≤0.02) with higher intake.61 Total protein intake from vegetable and animal sources was not associated with any of the outcomes, but the inverse association between pulse wave velocity and leucine intake was only observed for leucine from animal protein sources.61 In the Tehran Lipid and Glucose Study,62 investigators administered a food frequency questionnaire to 4288 normotensive adults, who were followed up for 3 years. Principal component analysis of 8 amino acid groups identified 3 factors, of which one had a high loading on BCAA intake (0.95). In adjusted analyses, contrasting the highest versus the lowest fourth of the distribution of the BCAA-related factor was associated with a 83% higher risk of hypertension (P for trend, 0.002), whereas the 2 other factors were not associated with the hypertension risk.62 The BCAA-related factor was positively correlated with the dietary intake of animal proteins and dairy products but negatively with the consumption of plant proteins.62 In the INTERMAP study,30 blood pressure and the urinary excretion of BCAAs were higher in Northern than Southern China. Finally, in the Project Viva study,63 109 girls and 104 boys, aged 6 to 10 years at enrollment, were followed up for 5 years. The factor derived by principal component analysis associated with circulating BCAAs was not associated with the changes in blood pressure in either sex in unadjusted or adjusted analyses.63HeartOne of the well-studied metabolic signatures of heart failure is the shift from a fatty acid dominant bioenergetic state into a more glycolytic state, viewed as a compensatory response with the purpose of enhancing the utilization of energy substrate and oxygen.64 However, a recent paradigm put emphasis on the metabolism of amino acids instead of glucose and fatty acids.15 Both experimental65–67 and clinical68 studies support the implication of BCAAs in modulating left ventricular (LV) function.A seminal article demonstrated that defective BCAA catabolism was the most important metabolic change in the failing heart of mice subjected to transaortic constriction.67 Accumulation of BCKAs promoted pressure overload-induced heart failure and was associated with oxidative injury. In mice, pharmacological stimulation of BCAA catabolism by compound BT2 (3,6-dichlorobenzo[b]thiophene-2-carboxylic acid)69 preserved cardiac function after pressure overload.67 In a salt-loaded rat model of heart failure65 and middle-aged mice,66 BCAA dietary supplements prolonged cardiomyocyte survival,65 preserved cardiac function,66 and lengthened life span.65,66 Expression of PP2Cm in the heart is dynamically regulated by stressors.16,69 In hypertrophic and failing hearts, its expression is significantly reduced at both the mRNA and protein level.16KLF15 (Krüppel-like factor 15) is a direct transcriptional activator of BCAT2.70,71 In cultured cardiomyocytes, overexpression of KLF15 induces mRNA expression of BCAT2, the E1α, E1β, and E2 BCKDC subunits, and PP2Cm, with the exception of BCKDK.67 The opposite occurs in KLF15-deficient hearts. Also similar to what was observed in failing human and mouse heart samples, BCKA levels were elevated in the myocardium of KLF15-null hearts.67 Thus, suppression of BCAA catabolism disrupts glucose metabolism and sensitizes the heart to injury.72 Two observations emphasize the relevance of these experimental data with respect to the pathophysiology of heart failure in humans. First, the findings in the failing mouse heart were replicated in mRNA expression studies of human hearts with terminal dilated cardiomyopathy.67 Second, in patients with heart failure treated with cardiac resynchronization therapy, responders had higher pretreatment circulating BCAA levels than nonresponders.68Mitochondrial proteins are almost all downregulated in end-stage diseased hearts, irrespective of the cause of heart failure.73 As mentioned above, this also applies to the BCAT2,67,70,71 the E1α, E1β, and E2 components of the BCKDC,67 and PP2Cm,67 which seem to share a transcriptional or posttranscriptional regulatory circuit.67 BCKA oxidation is impaired in heart failure with reduced ejection fraction.74 However, our literature search did not reveal any data showing association in experimental or human studies between heart failure with preserved ejection fraction and the metabolism of BCAAs.Experimental findings and observations in symptomatic patients cannot be extrapolated to subclinical disease, in which biomarkers would have the greatest diagnostic or prognostic relevance.75 Asymptomatic diastolic LV dysfunction affects >25% of the general population.76,77 It carries a 10% risk of further deterioration during 5 years78 and is a forerunner of cardiovascular complications.79 In 570 randomly recruited Flemish,80 we investigated the associations of echocardiographic diastolic LV function with 43 circulating metabolites measured by nontargeted nuclear magnetic resonance spectroscopy (Figure 2). Echocardiographic LV function was assessed in 2005 to 2010 (baseline) and 2009 to 2013 (follow-up). Median follow-up was 4.7 years. In early diastole, higher peak velocity of the mitral annulus (e′ [early LV relaxation]) and lower ratio of the transmitral blood velocity to mitral annular velocity (E/e′ [LV filling pressure]) likely reflect a more performant diastolic LV function. This interpretation rests on the observation that the ranges of e′ and E/e′ differ widely between people randomly recruited from populations and patients81 and that in the general population, lower e′ and higher E/e′ predict worse outcome.79,82 In multimarker analyses of the baseline and follow-up data and in analyses predicting follow-up e′ and E/e′, the BCAAs leucine and valine were associated with more performant diastolic LV function (Figure 2).80 Valine, leucine, and isoleucine metabolism (−log10P, 12.8) and aminoacyl-tRNA biosynthesis (10.2) were the top metabolic pathways associated with diastolic LV function.80 These findings among people covering the range from normal diastolic LV function to asymptomatic dysfunction illustrate how BCAAs may act as critical switches in health and disease. Elevated circulating BCAA levels may reflect a healthy nutritional state and may be beneficial in normal or slightly dysfunctional hearts but become harmful when BCAA catabolism is defective and cardiotoxic BCKAs accumulate in the failing heart (Figure 3).Download figureDownload PowerPointFigure 2. Results from the Flemish Study on Environment Genes and Health Outcomes, relating echocardiographic diastolic left ventricular (LV) function to circulating metabolites using partial least squares analysis. The graphics are overlaid V plots for the cross-sectional associations of e′ (A) and E/e′ (B) with the metabolites at baseline (2005–2010) and follow-up (2009–2013) and for the longitudinal associations of e′ (C) and E/e′ (D) at follow-up (2009–2013) with the baseline metabolites (2005–2010). The variable importance in projection (VIP) score indicates the importance of a metabolite in the construction of the partial least squares factors. VIP is plotted against the centered and rescaled correlation coefficients, which reflect the associations of the echocardiographic measurements with the metabolites. For E/e′, the x axis was inverted to be directionally similarly associated with diastolic LV function as for e′. Markers in the top left quadrant (lower e′ and higher E/e′) and top right quadrant (higher e′ and lower E/e′) of the V plots are associated with worse and better diastolic LV function, respectively. Numbers identify metabolites: (1) pentanoate, (2) glucose+glutamine, (3) 4-hydroxybutyrate, (4) 2-aminobutyrate, (5) leucine, and (6) valine. Reprinted from Zhang et al80 with permission. Copyright © 2018, European Journal of Preventive Cardiology.Download figureDownload PowerPointFigure 3. Differential role of the branched-chain amino acid catabolic pathway in normal and failing hearts. The branched-chain amino acids (BCAAs) are essential nutrients. In the myocardium, BCAAs are catabolized to branched-chain keto acids (BCKAs) by BCAT (branched-chain amino acid aminotransferase) and to acetyl-CoA by branched-chain α-keto acid dehydrogenase complex (BCKDC) under control of PP2Cm (mitochondrial protein phosphatase) and BCKDK (branched-chain keto acid dehydrogenase kinase). Acetyl-CoA can enter the tricarboxylic acid (TCA) cycle and contribute to energy generation. Circulating BCAA levels, reflecting the balance between nutritional intake and catabolism, are beneficial in normal hearts. Expression of the mitochondrial machinery to catabolize BCAAs is reduced in failing hearts, so that cardiotoxic BCKAs accumulate.Skeletal MusclesStriated skeletal muscle is the largest organ in the human body representing on average 40% of body weight and 20% of protein