The Emerging Role of Metabolomics in the Diagnosis and ... · contributing to the worsening of...

Listen to this manuscript’s

audio summary by

JACC Editor-in-Chief

Dr. Valentin Fuster.

J O U R N A L O F T H E A M E R I C A N C O L L E G E O F C A R D I O L O G Y V O L . 6 8 , N O . 2 5 , 2 0 1 6

ª 2 0 1 6 B Y T H E A M E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N

P U B L I S H E D B Y E L S E V I E R

I S S N 0 7 3 5 - 1 0 9 7 / $ 3 6 . 0 0

h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j a c c . 2 0 1 6 . 0 9 . 9 7 2

FOCUS SEMINAR: GENETICS

STATE-OF-THE-ART REVIEW

The Emerging Role of Metabolomicsin the Diagnosis and Prognosis ofCardiovascular Disease
John R. Ussher, PHD,a,b Sammy Elmariah, MD, MPH,c Robert E. Gerszten, MD,d Jason R.B. Dyck, PHDa
ABSTRACT

Fro

an

of

va

De

Str

su

sen

Me

Re

Ma

Perturbations in cardiac energy metabolism are major contributors to a number of cardiovascular pathologies. In addition,

comorbidities associated with cardiovascular disease (CVD) can alter systemic and myocardial metabolism, often

contributing to the worsening of cardiac function and health outcomes. State-of-the-art metabolomic technologies

give us the ability to measure thousands of metabolites in biological fluids or biopsies, providing us with a

metabolic fingerprint of individual patients. These metabolic profiles may serve as diagnostic and/or prognostic tools

that have the potential to significantly alter the management of CVD. Herein, the authors review how metabolomics

can assist in the interpretation of perturbed metabolic processes, and how this has improved our ability to

understand the pathology of ischemic heart disease, atherosclerosis, and heart failure. Taken together,

the integration of metabolomics with other “omics” platforms will allow us to gain insight into pathophysiological

interactions of metabolites, proteins, genes, and disease states, while advancing personalized medicine.

(J Am Coll Cardiol 2016;68:2850–70) © 2016 by the American College of Cardiology Foundation.

O ver the last few decades, there has been agrowing appreciation for the importantcontribution that myocardial energy meta-

bolism plays in the regulation of cardiac function.As the heart is the most metabolically demanding or-gan in the body, it is not surprising that perturbationsin cardiac energy metabolism are major contributorsto a number of cardiovascular pathologies. In addi-tion, comorbidities associated with cardiovasculardisease (CVD) pathogenesis can alter systemic andmyocardial metabolism, which often aids in the wors-ening of cardiac function and health outcomes. In nosituation is this more relevant than in obesity and

m the aCardiovascular Research Centre, Department of Pediatrics, Mazan

d Dentistry, University of Alberta, Edmonton, Alberta, Canada; bFaculty o

Alberta, Edmonton, Alberta, Canada; cDivision of Cardiology, Departmen

rd Medical School, Boston, Massachusetts; and the dDivision of Cardiovas

aconess Medical Center, Harvard Medical School, Boston, Massachusetts

oke Foundation of Alberta, NWT & Nunavut, and is a Scholar of the Ca

pported by the American Heart Association (4FTF20440012) and by the

feld Scholar Award. Dr. Gerszten is supported by NIH grant R01-DK08157

dicine, and is supported by the Heart and Stroke Foundation of Canad

search. The authors have reported that they have no relationships releva

nuscript received August 24, 2016; accepted September 9, 2016.

diabetes, where these conditions can cause major sys-temic metabolic disturbances that have a negativeimpact on organs such as the liver, skeletal muscle,adipose tissue, the vasculature, as well as themyocardium. However, even in the absence ofobesity and diabetes, alterations in substrate meta-bolism of numerous organs resulting from the onsetof CVD can contribute to changes in the metabolicprofile of a patient. With numerous advancementsin “omics” technology platforms, including geno-mics, transcriptomics, and proteomics, we now havea much broader understanding of the molecular/cellular/functional changes that take place in CVD,

kowski Alberta Heart Institute, Faculty of Medicine

f Pharmacy and Pharmaceutical Sciences, University

t of Medicine, Massachusetts General Hospital, Har-

cular Medicine, Department of Medicine, Beth Israel

. Dr. Ussher is a New Investigator of the Heart and

nadian Diabetes Association (CDA). Dr. Elmariah is

Massachusetts General Hospital Heart Center Has-

2. Dr. Dyck is a Canada Research Chair in Molecular

a, the CDA, and the Canadian Institutes for Health

nt to the contents of this paper to disclose.

https://s3.amazonaws.com/ADFJACC/JACC6825/JACC6825_fustersummary_09




http://crossmark.crossref.org/dialog/?doi=10.1016/j.jacc.2016.09.972&domain=pdf

http://dx.doi.org/10.1016/j.jacc.2016.09.972

AB BR E V I A T I O N S

AND ACRONYM S

ATP = adenosine triphosphate

BCAA = branched-chain amino

acid

BCKA = branched-chain

a-keto-acid

BNP = B-type natriuretic

peptide

CAD = coronary artery disease

CoA = coenzyme A

CVD = cardiovascular disease

HF = heart failure

HFpEF = heart failure with

preserved ejection fraction

HFrEF = heart failure with

reduced ejection fraction

IHD = ischemic heart disease

L-C = long-chain

LC = liquid chromatography

LV = left ventricular

M-C = medium-chain

MI = myocardial infarction

MS = mass spectrometry

NMR = nuclear magnetic

resonance

PDH = pyruvate

dehydrogenase

PET = positron emission

tomography

S-C = short-chain

T2D = type 2 diabetes

TCA = tricarboxylic acid

TMA = trimethylamine

TMAO = trimethylamine-N-

oxide

J A C C V O L . 6 8 , N O . 2 5 , 2 0 1 6 Ussher et al.D E C E M B E R 2 7 , 2 0 1 6 : 2 8 5 0 – 7 0 Metabolomics in Cardiovascular Disease

2851

as well as predictions of how these changes may influ-ence intermediary metabolism (Central Illustration).Furthermore, state-of-the-art metabolomic technolo-gies that are now available give us the ability to mea-sure thousands of metabolites in biological fluids orbiopsies, providing us with a “snapshot” of the meta-bolic fingerprint of individual patients (Table 1).These snapshots can potentially serve as diagnosticand/or prognostic tools that can be used to identifyimpairments in systemic or myocardial metabolismoccurring during the development and worsening ofCVD, as well as help guide the types and timing ofspecific interventions/therapies. Thus, metabolomicsis emerging as an important tool that can aid clini-cians in better understanding the pathogenesis ofCVD, and has the potential to significantly alter themanagement of CVD.

OVERVIEW OF CELLULAR METABOLISM

Substrate metabolism is an essential component ofcellular heath and survival. Both anabolic and cata-bolic processes are necessary to support the numerouscellular events that contribute to cell, organ, andorganism survival. From a simplistic perspective,when cells in the body generate energy in the formof adenosine triphosphate (ATP), they catabolizethe various energy sources available, either fromendogenous energy stores (primarily glycogen or tri-acylglycerols), or from exogenous substrates circu-lating in the blood (e.g., carbohydrates, fatty acids,amino acids, and ketone bodies, among others). Bycontrast, when cells need to build materials (e.g.,phospholipids for membranes, proteins for growth,fatty acids for de novo lipogenesis, and so on) orperform many cellular processes (e.g., growth, ionichomeostasis, signal transduction, contraction, amongothers), they consume ATP, releasing the energyneeded to support these functions. For the vast ma-jority of cells in our body, carbohydrates (primarily inthe form of glucose and lactate) and fatty acidsrepresent the most common energy substratesmetabolized by our cells to produce ATP.

Of the many substrates that can be used togenerate energy, glucose and fatty acids are themajor contributors to overall ATP production. Forthe former, following transport from the circulationinto the cell, glucose can undergo anaerobic glycol-ysis to produce pyruvate and small amounts of ATP(2 ATP per glucose molecule) (Figure 1). In some celltypes, if the energy need is low, the majority of thispyruvate is converted into lactate, which can exitthe cell into the circulation (Figure 1). However, if

the cellular energy demands are high andoxygen is present, pyruvate derived fromglucose and/or lactate can enter the mito-chondria, where it is converted to acetylcoenzyme A (acetyl-CoA) by pyruvate dehy-drogenase (PDH). Acetyl-CoA is the commonintermediate that links oxidative metabolismof all nutrient energy sources, and thisacetyl-CoA is utilized by the tricarboxylicacid (TCA) cycle (also known as the Krebscycle) to produce reducing equivalents (e.g.,nicotinamide adenine dinucleotide andflavin adenine dinucleotide), which act aselectron donors to drive the proton motiveforce that fuels ATP synthesis (Figure 1). Thislatter process of glucose breakdown frompyruvate to eventual ATP production istermed glucose oxidation, and ultimatelygenerates a significantly greater amount ofATP than glycolysis (31 ATP from glucoseoxidation vs. 2 ATP from glycolysis). Duringcellular fatty acid catabolism, fatty acids areeither transported into and/or passivelyenter the cell, converted into fatty acyl-CoAesters, and then converted into a fatty acyl-carnitine, which permits the fatty acid totraverse the mitochondrial membrane.There, it is reconverted back into a fattyacyl-CoA ester for subsequent mitochondrialb-oxidation (Figures 1 and 2) (1). The mito-chondrial b-oxidation enzymatic machineryproceeds to repeatedly remove acetyl-CoAfrom the fatty acyl-CoA ester until it has beencompletely oxidized. Hence, complete oxida-tion of a fatty acid molecule, such as oleate orpalmitate, produces more acetyl-CoA andreducing equivalents, and thereby a much
larger amount of ATP than the complete oxidation of aglucose molecule (104 ATP from palmitate vs. 31 ATPfrom glucose oxidation).
Although glucose and fatty acids represent themost common fuels used by our cells to produce en-ergy, the majority of cells in our bodies are alsocapable of metabolizing amino acids (e.g., hepato-cytes, skeletal muscle myocytes) and ketone bodies(e.g., neurons) to produce acetyl-CoA for oxidativeenergy metabolism (Figure 1). In particular, cellularmetabolism of the branched-chain amino acids(BCAAs) leucine, isoleucine, and valine, has beenextensively studied, as BCAAs are potent regulatorsof systemic metabolism, energy expenditure, andmuscle protein synthesis (2). In order for BCAAs toenter the mitochondria for oxidative metabolism,

CENTRAL ILLUSTRATION Integration of Various Omics Platforms for Patient Phenotyping

Ussher, J.R. et al. J Am Coll Cardiol. 2016;68(25):2850–70.

The metabolome represents one of the downstream end products of the environment’s interaction with the genome-transcriptome-proteome.

TABLE 1 Perturbations in Cellular Metabolism Inferred via

Metabolomic Profiling

Metabolite Metabolic Perturbation

Lactate Glucose metabolism(anaerobic glycolysis)

Acylcarnitines Fatty acid oxidation

Leucine,Isoleucine,Valine

BCAA metabolism

b-Hydroxybutyrateb-Hydroxybutyrylcarnitine

Ketone body oxidation

Indicates key metabolites measured in the circulation during blood-basedmetabolomic profiling and the changes in the metabolic processes they predict.

BCAA ¼ branched-chain amino acid.

Ussher et al. J A C C V O L . 6 8 , N O . 2 5 , 2 0 1 6

Metabolomics in Cardiovascular Disease D E C E M B E R 2 7 , 2 0 1 6 : 2 8 5 0 – 7 0

2852

they must first be converted into their respectivebranched-chain a-keto-acids (BCKAs), which aresubsequently oxidized via BCKA dehydrogenase(Figure 3). BCAA/BCKA metabolism is of particularrelevance during fasting/starvation, where skeletalmuscle catabolism of BCAAs produces alanine tosupport hepatic gluconeogenesis, and hepaticcatabolism of BCAAs leads to the formation ofprecursors for hepatic ketogenesis (Figure 3). Ofinterest, the increase in cellular metabolism of BCAAsin muscle has been strongly correlated with reducedinsulin sensitivity and impaired glucose homeostasis(see the section Additional Considerations for the Useof Metabolomics in the Diagnosis and Prognosis ofCVD) (2).

FIGURE 1 Cellular Metabolism and Implications of Metabolomic Profiling

The primary substrates used by our cells to produce energy (ATP) include carbohydrates, fatty acids, ketone bodies, and amino acids. In the absence of oxygen, ATP can

be generated from glycolysis (step 1) as glucose is converted into pyruvate and then lactate. In the presence of oxygen, pyruvate can enter the mitochondria and be

oxidized via PDH to produce acetyl-CoA for the TCA cycle. This process is referred to as glucose oxidation (step 2). Similarly, fatty acids are transported into the cell,

converted to fatty acyl-CoA, and then enter the mitochondria via the CPT1/CPT2 system. Following this, the fatty acyl-CoA ester enters the b-oxidation pathway for

further breakdown to produce acetyl-CoA (step 3). BCAAs can also be oxidized to produce acetyl-CoA, but must first be converted into their respective BCKA

derivatives, which are subsequently oxidized by BCKD as they enter the mitochondria, undergoing a series of reactions that lead to the formation of acetyl-CoA

(step 4). Ketone bodies such as b-hydroxybutyrate can also be oxidized (via BDH) in the mitochondria and result in the formation of acetyl-CoA (step 5). Thus, acetyl-

CoA represents the common metabolite linking oxidative energy metabolism of the main substrates identified herein, and this acetyl-CoA feeds into the TCA cycle and

results in the formation of reducing equivalents NADH and FADH2, which fuel ATP production. During cardiometabolic disease progression, metabolic pathways are

often perturbed and lead to the accumulation or loss of various metabolites that can be detected in the circulation via metabolomic profiling. These key metabolites

capable of spilling over into the circulation, and which can reflect a “snapshot” of cellular metabolism, are indicated in blue text. ATP ¼ adenosine triphosphate;

BCAA ¼ branched-chain amino acid; BCKA ¼ branched-chain a-keto-acid; BDH ¼ b-hydroxybutyrate dehydrogenase; CoA ¼ coenzyme A; CPT ¼ carnitine

palmitoyltransferase; FADH2 ¼ flavin adenine dinucleotide; L-C ¼ long-chain; NADH ¼ nicotinamide adenine dinucleotide; PDH ¼ pyruvate dehydrogenase;

TCA ¼ tricarboxylic acid.


2853



2854

MYOCARDIAL METABOLISM

In the healthy adult heart, virtually all ATP produc-tion is derived from mitochondrial oxidative meta-bolism, with the remainder primarily arising fromglycolysis (3). Because myocardial ATP stores arerelatively low (w300 mg of total ATP for an averageheart) compared with the high rate of ATP utilization,there is a nearly complete turnover of the myocardialATP pool every 10 s (3). In order to meet this enor-mous energy demand, the healthy heart metabolizesmore than 30 g of fat and 20 g of carbohydrate perday, and uses the equivalent of nearly 35 liters ofoxygen while doing so (3). Furthermore, the meta-bolic flexibility of the heart is extremely dynamic, asdemonstrated by its ability to rapidly alter its patternof fuel utilization in order to adapt to its substrateand hormonal environment, whereby virtually allenergy substrates, including fatty acids, glucose,lactate, ketone bodies, and amino acids, can beconsumed by the heart to generate ATP.

Interestingly, the normal patterns of cardiac sub-strate metabolism are often perturbed in a variety ofCVDs, and it is widely accepted that these metabolicchanges directly contribute to disease pathophysi-ology. Indeed, alterations in cardiac metabolismassessed via coronary sinus catheterization combinedwith infusion of radiolabeled tracers, positron emis-sion tomography (PET) imaging, or cardiac magneticresonance imaging have been observed in patientswith obesity/diabetes, ischemic heart disease (IHD),and/or heart failure (HF) (3). However, these methodscan be invasive and costly, present potential healthhazards due to use of radioisotopes with long half-lives (e.g., carbon-14 [14C]), and are often deployedin patients who already have established or advancedCVD, thus limiting the diagnostic potential of theseassessments. Perturbations in myocardial metabolismfrequently lead to the accumulation or loss of specificmetabolites from the various metabolic pathwaysdescribed previously, and changes in these metabo-lites are often reflected within the circulation(Figure 1). These metabolites represent the terminalproduct of the environment’s interaction with thegenome-transcriptome-proteome (Central Illustration)in response to a disease process (4,5). Hence, theability to accurately identify and quantify changes inthese metabolites, and to use this information todetect perturbations in a specific metabolic process,adds to the arsenal of diagnostic information avail-able to clinicians. Fortunately, significant advance-ments in metabolomics-based technology platformsnow enable us to detect these various metabolites

with high sensitivity and selectivity. Furthermore, anumber of these metabolites may have clinical utilitywith regard to predicting (screening biomarkers) anddetecting (diagnostic biomarkers) disease, evaluatingdisease progression/remission (prognostic bio-markers), and assessing therapeutic effectiveness.

METABOLOMICS

THE EVOLUTION OF METABOLOMIC PROFILING. In1972, the concept of using quantitative and qualita-tive patterns of metabolites within biological fluids tounderstand the physiological status of a biologicalsystem was introduced by Pauling et al. (6). In thesame year, Horning and Horning (7) coined the termmetabolic profiling to describe data obtained from gaschromatography of a patient sample. Since theseinitial reports, the concept that metabolites cancomprehensively characterize a phenotype, diseasestate, or physiological response to an applied stim-ulus or perturbation, has exponentially evolved overthe past 4 decades into the field of metabolomics, thesystematic assessment of small molecules in biolog-ical fluid or tissue (4,8).

The metabolome encompasses an immense varietyof endogenous small molecules, including aminoacids, sugars, lipids, nucleic acids, amines, organicacids, fatty acids, and urea-cycle and methyl-transfermetabolites, as well as a myriad of exogenous chem-icals, such as pharmacological agents, toxins, andxenobiotics. Chemical characteristics, including hy-drophilicity, lipophilicity, polarity, mass, and charge,vary tremendously across these compound classes, asdoes the dynamic range of metabolite concentrations,varying from picomolar to millimolar amounts (5,8).The broad range of concentrations and the biochem-ical diversity of metabolites preclude the measure-ment of all metabolites using a single analyticaltechnique or platform. Instead, combined analyticalapproaches using several metabolomic-profilingplatforms are utilized to span the diversity of themetabolome (5,8). However, metabolic pathways arehighly conserved across species, readily allowing forthe mechanistic and functional evaluation of humanmetabolomic findings in model organisms (4).

METABOLOMIC ANALYTIC TECHNIQUES. Nuclearmagnetic resonance (NMR) spectroscopy and massspectrometry (MS), often coupled to a chromato-graphic technique, are the primary methods by whichthe metabolome is assessed. Both techniques enablehigh-throughput profiling of large numbers of me-tabolites simultaneously within biological fluids, but

FIGURE 2 Use of Acylcarnitine Profiling to Infer Changes in Fatty Acid Metabolism

During fatty acid oxidation, fatty acyl-CoA esters are converted into fatty acylcarnitine derivatives (via CPT-1), which allows fatty acids to enter the mitochondria

where they are reconverted back into fatty acyl-CoA (via CPT-2) for subsequent mitochondrial b-oxidation. General defects in the enzymes of the b-oxidative

machinery that inhibit the oxidation of L-C fatty acids will increase the accumulation of mitochondrial L-C acyl-CoAs and in turn lead to increases in L-C acylcarnitines via

CPT-2. However, impaired L-C fatty acid b-oxidation will produce decreases in S-C and M-C acyl-CoAs, as L-C fatty acids are not broken down via repeated cycles

of b-oxidation, leading to corresponding reductions in the respective S-C and M-C acylcarnitine derivatives. Therefore, reductions in S-C/L-C or M-C/L-C acylcarnitine

ratios in the circulation often reflect impairments in mitochondrial L-C fatty acid oxidation rates. In contrast, general increases in numerous S-C, M-C, and L-C

acylcarnitines often infer elevations in fatty acid oxidation. If fatty acid oxidation is impaired due to inhibition of mitochondrial fatty acid uptake, decreases in S-C,

M-C, and L-C acylcarnitines are observed. CrAT ¼ carnitine acetyltransferase; M-C ¼ medium-chain; S-C ¼ short-chain; other abbreviations as in Figure 1.


2855



2856

each is associated with different analytical strengthsand weaknesses.

Metabolite profiling of biological samples usingNMR spectroscopy was pioneered in 1983. This tech-nique identifies metabolites by chemical shifts inresonance frequency when they are subjected to amagnetic field. Within a strong magnetic field, nucleiabsorb electromagnetic radiation in a characteristicfrequency that can be used to assign local molecularstructure. Measuring all frequencies can identifymetabolites within a sample. Advantages of NMRinclude that it is robust, reproducible, requires mini-mal sample preparation, is low in cost, and nonde-structive. Although NMR spectroscopy is alsoquantitative, without the need for standards, itssensitivity, which relates to the strength of the mag-net, is limited, especially for low abundance metab-olites (4). Using standard equipment, NMR is able toresolve metabolite concentrations down to the lowmicromolar range (4,8). In addition, within a biolog-ical sample, current technology enables the accuratequantification of approximately 100 of the mostabundant metabolites (8).

MS identifiesmetabolite species on the basis of theirmass/charge ratio (m/z). Although samples can beinfused directly into the mass spectrometer, the sep-aration of metabolite components using either liquidchromatography (LC) or gas chromatography is oftenfirst performed to facilitate analyte identification andquantification. Chromatographic separation involvesdissolving the sample into a solvent and passing thismobile phase through a stationary phase, such as acolumn containing surfaces with specific interactionchemistries that allow the dissolved metabolites toremain in the column or to pass through at varyingspeeds. Analytes are subsequently ionized for intro-duction into themass separation unit. Mass separationcan be performed by several techniques, includingtime-of-flight, quadrupole, and ion trap mass ana-lyzers, each of which has varying dynamic ranges,resolution, and accuracy. Tandem MS (MS/MS) allowsfor the use of 2 or more stages of mass analysis to focuson the fragmentation of an ion within a mixture in or-der to enhance resolution and accuracy (5). Integrationof these serial analyte manipulations provides thehighly sensitive capability of identifying and analyzingthousands of metabolites with concentrations down tothe low femtomolar range (5,8).

Metabolite profiling using NMR or MS can beperformed in a targeted or untargeted manner (5).Targeted metabolomics measures a distinct, well-characterized set of metabolites within a biologicalsample. Absolute quantification of analytes ispossible by spiking in internal standards, usually

isotope-labeled versions of endogenous metabolites,across a range of concentrations. Using the targetedapproach, key metabolites within metabolic pathwaysact as sentinels to reflect changes within a metabolicpathway (Table 1). Although instruments functioningwithin a targeted mode are more sensitive, targetedmetabolomics only permits the assessment of severalhundred known metabolites at a time. By contrast,untargeted metabolomics attempts to analyze allsmall molecules within a sample in an unbiasedmanner and is often not quantitative. In untargetedmetabolite profiling, closer to 8,000 metabolites areanalyzed, the identities of most of which are un-known (9). This approach relies on patterns or sig-natures reflective of a metabolic state or,alternatively, the investment of tremendous time andresources in order to identify unknown metabolites.

LIMITATIONS WITH THE USE OF METABOLOMICS TO

ASSESS INTERMEDIARY METABOLISM. Metabolomicprofiling performed on blood or urine samples canprovide significant insight into metabolic pathwaysthat may be altered during the progression ofnumerous pathologies (Figures 4 to 6, Table 1). Manyof these diseases, such as obesity and/or type 2 dia-betes (T2D) are complex syndromes that involvepathological changes in multiple organ systems, eachof which may significantly contribute to the blood-and urine-based metabolomic profile. As such, usingblood- and urine-based metabolomics as a diagnosticor prognostic tool is valuable, but cannot be consid-ered definitive in informing us about new molecularpathophysiological processes that are occurringwithin the heart or any specific organ. Indeed,assessment of metabolism at the organ level neces-sitates the measurement of changes in metabolitesacross the organ. For example, the delta in metabolitelevels between arterial blood and the coronary sinusfor the heart or the renal artery and vein for the kid-ney provides more precise information as to themetabolic changes that are occurring in those organs.In addition, because metabolomics only provides asnapshot of metabolic flux through a particularpathway, the reason for increased or decreasedmetabolite levels may not be readily apparent. Forinstance, an aberrant metabolite concentration maybe due to a defective metabolic pathway, changes insubstrate supply, or abnormal metabolic rate of theprimary product. Thus, predictions of defectivemetabolic processes on the basis of blood- or urine-based samples alone must be interpreted withcaution. Regardless of these limitations, identifica-tion of numerous altered metabolites related to acommon metabolic pathway in the blood or urine can

FIGURE 3 Cellular Fates of BCAA Catabolism

Although Figure 1 depicts that BCAA metabolism can be used to generate acetyl-CoA for oxidative energy metabolism, BCAA catabolism serves a number of other

purposes. In particular, during fasting/starvation, the first step in BCAA catabolism involves transfer of amino groups to a-ketoglutarate, followed by subsequent

amino group transfer to pyruvate, producing alanine. This is of particular importance in skeletal muscle, where this alanine is exported to the liver to support hepatic

gluconeogenesis (step 1). Conversely, enhanced BCAA (leucine and isoleucine) catabolism in the liver during fasting/starvation can lead to the formation of precursors

for the biosynthesis of ketone bodies (step 2). Finally, BCAA catabolism can be used to refuel TCA cycle intermediates through anaplerosis, either at the level of

a-ketoglutarate, or through formation of succinyl-CoA produced via subsequent breakdown of valine or isoleucine (step 3). Abbreviations as in Figure 1.


2857



2858

yield significant insight into potential perturbationsin metabolism that take place during disease pro-gression. As such, metabolomic signatures in bloodor urine can often yield hypothesis-generating find-ings. These findings ultimately can lead to intricatemolecular explorations in a diseased organ,cell, or animal model, which in turn will enhanceclinical utility and our understanding of diseasepathobiology.

METABOLOMICS AND IHD

IHD is caused by inadequate perfusion of themyocardium and encompasses both stable and un-stable angina, as well as myocardial infarction (MI).IHD represents the most common form of CVD, andalthough survival rates following an acute MI havegreatly improved in recent years, mortality due toIHD still remains the leading cause of cardiovasculardeath globally (10). Because oxygen and nutrientsupply/delivery to the myocardium are markedlyreduced during ischemic periods, there are a numberof key changes that manifest with regard to myocar-dial intermediary energy metabolism (Figure 4), oftenmarked by a reduction in overall oxidative meta-bolism. In response to this decline in oxidativemetabolism, glycogen breakdown and glycolysis ratesare increased, as glycolysis can produce ATP anaero-bically at a rapid rate (3). The extent of the increase inmyocardial glycolysis is dependent on the severity ofischemia and its duration. Whether these myocardialmetabolic alterations during myocardial ischemia canbe detected via use of blood-based metabolomics is aquestion of great interest due to its diagnosticpotential.

TCA CYCLE ACTIVITY. Application of LC/MS blood-based metabolomic profiling in 36 individuals sub-jected to exercise stress testing demonstrated thatindividuals exhibiting myocardial ischemia (the ma-jority confirmed via coronary angiography) had re-ductions in a number of TCA cycle metabolites (e.g.,oxaloacetate), consistent with myocardial ischemia-induced reductions in oxidative metabolism (11). Inpatients undergoing alcohol septal ablation for hy-pertrophic obstructive cardiomyopathy, a model ofplanned MI, serial blood samples collected as early as10 min and up to 24 h post-ablation from the coronarysinus and periphery revealed alterations in numerousmetabolites (12). This included changes in a numberof organic acids, including increases in succinic andmalic acid, reflecting dysregulated TCA cycle activity.Hence, blood-based metabolomics frequently iden-tifies defects in TCA cycle activity and oxidativemetabolism in IHD patients.

FATTY ACID UTILIZATION. Fatty acid oxidationrates are decreased in the ischemic myocardium asa direct result of reduced blood flow and oxygendelivery. As such, paired collection of arterial andcoronary sinus blood before cardiac surgerydemonstrated significant reductions in the myocar-dial extraction of free fatty acids in patients withcoronary artery disease (CAD) versus those withoutCAD (13). However, one must be cautious wheninterpreting circulating metabolomic profiles in pa-tients with IHD, because metabolism within theischemic myocardium differs from that of the viablenonischemic myocardium. Consistent with this,fatty acid oxidation rates in the viable nonischemicmyocardium appear normal, and thus the circu-lating metabolomic profile from this region maymask the metabolic changes present withinischemic tissue. Likewise, direct coronary sinuscatheterization combined with infusion of [1-14C]-oleate or [1-14C]-palmitate demonstrated similarfatty acid oxidation rates in patients who had justunderwent coronary angiography for symptoms ofIHD versus healthy volunteers (14). Paired collectionof arterial and coronary sinus blood before cardiacsurgery has also demonstrated increases in baselineshort-chain (S-C) dicarboxylacylcarnitines, whichwere shown to predict risk for cardiovascular eventsin patients with CAD (15). In addition, LC/MS/MS-based serum metabolomic profiling in elderlypatients with established CAD, but without HF,demonstrates that medium-chain (M-C) and long-chain (L-C) acylcarnitines are increased, and pre-dict incident risk for subsequent cardiovascularevents independent of standard predictors (16). Thesource of circulating S-C dicarboxylacylcarnitinesand how they may contribute to the pathology ofCAD are unknown. Because peroxisomes areanother cellular organelle capable of oxidizing fattyacids like mitochondria, such changes may reflectalterations in peroxisomal fatty acid metabolismbecause L-C dicarboxylic acids are often shortenedvia peroxisomal oxidation (17).

GLUCOSE UTILIZATION. Glucose oxidation rates arealso markedly depressed in the ischemic myocar-dium, whereas glycolytic rates are robustly increaseddue to the stimulation of glycogenolysis (14,18). Themore serious the ischemic episode, the fastermyocardial lactate accumulates, because there islimited flow to wash out lactate and other glycolyticintermediates (e.g., protons), which also contributesto the eventual inhibition of glycolysis during pro-longed ischemia (18). Blood-based metabolomicprofiling in patients undergoing global ischemia/

FIGURE 4 Metabolomics in IHD and Atherosclerosis

Duringmyocardial ischemia, a reduction in bloodflowandoxidativemetabolismdecreases substrate extractionby theheart,which contributes to theelevation

in circulating FFAs and BCAAs, whereas anaerobic glycolysis increases to produce ATP in the absence of oxygen. Hence, increases in circulating lactate levels

are frequently observed during ischemia. CAD resulting from atherosclerosis is often a major contributor to ischemic heart disease or myocardial infarction. It

has been demonstrated that gut microbiota are capable of converting phosphatidylcholine/choline and L-carnitine into TMA, which the liver converts into

TMAO. TMAO increases the risk for atherosclerosis potentially via interfering with reverse cholesterol transport and promoting plaque formation, which

promotes IHD. Together, all of these changes in themetabolomic profile can provide significant diagnostic insight into the risk for atherosclerosis or presence

of CAD. CAD ¼ coronary artery disease; FFA ¼ free fatty acid; IHD ¼ ischemic heart disease; TMA ¼ trimethylamine; TMAO ¼ trimethylamine-N-oxide;

other abbreviations as in Figure 1.


2859



2860

reperfusion as a result of planned cardiac surgerywith cardioplegic arrest or in patients undergoingalcohol septal ablation for hypertrophic obstructivecardiomyopathy leads to marked increases in circu-lating lactate levels, reflective of enhanced myocar-dial anaerobic glycolytic metabolism (12,13). Thesefindings have been validated in a cohort of patientswho underwent spontaneous MI (12), and have alsobeen reported in stable angina patients undergoingcoronary angioplasty, where ischemia for at least1 min following balloon inflation produced markedincreases in circulating lactate levels collected10 min post-ischemia (19).

AMINO ACID UTILIZATION. Current informationsuggests that the normal heart’s reliance on aminoacids as a source of ATP is minimal, as infusion of L-[ring-2,6-3H]phenylalanine demonstrates that aminoacids are used primarily for anabolic purposes in themyocardium (20). This is supported by measurementsof leucine oxidation in the isolated rat heart, whichcontributes between 3% to 5% of overall cardiac oxy-gen consumption, depending on experimental condi-tions (21). Regardless, it has been suggested that aminoacid metabolism in the heart may be particularlyimportant during ischemia. Amino acids such asglutamate and glutamine, as well as the BCAAs, canserve as anaplerotic substrates that refuel the TCAcycle via intraconversion into a-ketoglutarate orsuccinyl-CoA, respectively (Figure 3) (22). Moreover,guanosine triphosphate (which can be converted toATP) can be generated in the absence of oxygen duringthe metabolism of glutamate and glutamine viasubstrate-level phosphorylation (22). Direct compari-sons of arteriovenous differences in 8 healthy and 11CAD patients demonstrates that IHD is associated witha net myocardial release of alanine and net myocardialuptake of glutamate (23). Interestingly, circulatingBCAA signatures have also been shown to predict bothprevalent and subsequent cardiovascular risk in pa-tients with CAD recruited through the CATHGEN(CATHeterization GENetics) biorepository at the DukeUniversity Medical Center (15,24). However, a limita-tion of these studies is that all patients underwentcardiac catheterization for suspected CAD and werenot healthy subjects. However, recent studies directlycomparing blood-based metabolite profiles alsoobserved that serum BCAA levels assessed via LC/MS/MS-based metabolomics were elevated in age- andsex-matched CAD patients versus healthy individuals,and this was found to be independent of other tradi-tional risk factors, including diabetes, hypertension,and dyslipidemia (25). Although studies to date arelimited in scope, BCAAuptake appears negligible in the

ischemic heart (22) and, in general, our mechanisticunderstanding of the relevance of amino acid andprotein metabolism in the progression of IHD remainspoorly understood.

METABOLOMICS AND ATHEROSCLEROSIS

Because the vast majority of IHD is due to the pres-ence of atherosclerotic plaques in the coronary ves-sels, there has been great interest in determiningwhether metabolomic profiling can identify in-dividuals at increased risk for IHD due to underlyingatherosclerosis. This is especially important consid-ering a large number of patients with clinically diag-nosed CAD exhibit derangements in myocardialmetabolism, and the use of techniques such as PETimaging to confirm these derangements is costly.Hence, the use of blood-based metabolomic profilingto identify subclinical atherosclerosis and early CADmay be of clinical value. Indeed, a number ofmetabolomic studies in patient cohorts have identi-fied a variety of potential biomarkers that can predictrisk for CAD and subsequent cardiovascular events,including MI and death. Most notably, sophisticatedstudies have demonstrated that circulating trime-thylamine-N-oxide (TMAO) is a significant predictorfor atherosclerosis and incident risk for MI and stroke(26,27), with the gut microbiome being implicated as acritical factor regulating this process (26). Specif-ically, dietary phosphatidylcholine, choline, andcarnitine are substrates for the generation of trime-thylamine (TMA) via the host gut microbiome. ThisTMA is subsequently released into the circulation andis then oxidized by the liver to TMAO. TMAO, in turn,promotes the progression of atherosclerosis, poten-tially via interfering with reverse cholesterol trans-port, which subsequently increases the risk ofcardiovascular events (Figure 4) (26). Recent studiesindicate that TMAO may also increase platelet hyper-reactivity, as application of exogenous TMAO atphysiological levels to human platelet-rich plasmaenhanced platelet aggregation, while increasing theadherence of fluorescently labeled platelets withinhuman whole blood to collagen (27). Furthermore,studies have shown that interfering with the gutmicrobiome’s ability to convert dietary choline orcarnitine into TMA, via inhibiting bacterial TMA lyasewith 3,3-dimethyl-1-butanol in a nonlethal manner,reduces circulating TMAO levels and the subsequentprogression of atherosclerotic lesions in a mousemodel of atherosclerosis (28). Interestingly, thesestudies support that increased circulating levels ofphosphotidylcholine/choline and/or carnitine areassociated with increased cardiovascular risk, yet

FIGURE 5 Metabolomics in HF

The failing heart undergoes a well-characterized metabolic remodeling that consists of general reductions in mitochondrial function and oxidative energy

metabolism, which is compensated via an increase in anaerobic glucose metabolism (glycolysis). Hence, circulating metabolomic profiles often yield

increases in L-C acylcarnitines and lactate in HF patients. Circulating BCAAs are also often elevated in HF patients, although reasons for this increase are

unknown. As recent studies in HF patients demonstrate impairments in the heart’s ability to break down BCKAs (Figure 3), this may lead to a buildup of

myocardial BCAAs that spill over into the circulation. Furthermore, recent studies have identified that the failing heart may increase its reliance on

ketone bodies as an oxidative energy source to compensate for the reduction in mitochondrial fatty acid oxidation, and circulating ketone bodies

(e.g., b-hydroxybutyrate and acetoacetate) are often decreased in HF patients. Of interest, decreased circulating ketone bodies and increased circulating

L-C acylcarnitines may also have diagnostic capability in differentiating HFrEF from HFpEF patients, in that serum concentrations of ketone bodies and L-C

acylcarnitines are decreased and increased, respectively, in HFrEF patients compared with HFpEF patients. HF ¼ heart failure; HFpEF ¼ heart failure with

preserved ejection fraction; HFrEF ¼ heart failure with reduced ejection fraction; other abbreviations as in Figures 1 and 2.


2861



2862

only appear to provide true prognostic value in thoseindividuals in which TMAO is also elevated (26).Together, these findings support TMAO as a novelbiomarker for the diagnosis of atherosclerosis/CAD,and exemplify the power of metabolomics to producehypothesis-generating findings that can lead tointricate molecular explorations, which ultimatelyimprove our understanding of disease pathology.

Other identified biomarkers from blood-basedmetabolomic profiling include observations thatincreased levels of 18:2 monoglyceride or decreasedlevels of 18:2 lysophosphatidylcholine and 28:1sphingomyelin are associated with increased risk forincident CAD events in pooled data from 3 case-cohort studies (29). In patients with angina or MIversus healthy subjects, an increased prevalence forCAD was associated with increased circulating levelsof lysophosphatidylcholines containing unsaturatedfatty acids, and decreased circulating levels oflysophosphatidylcholines containing saturated fattyacids (30). In addition, this study also identified thatcirculating phosphatidylcholines containing cer-amide, sphingomyelin, diacylglycerol, or palmiticacid were associated with increased prevalence forMI. The diagnostic information these circulatingmetabolites harbor is currently unknown, but onceagain illustrates the ability of blood-based metab-olomics to potentially identify novel metabolic pro-cesses that may contribute to disease pathology. Inobese and T2D patients, numerous tissues oftendisplay marked accumulation of lipid intermediates,including sphingomyelins, ceramides, and DAGs (31).These lipids often contribute to a phenomenonreferred to as lipotoxicity to reflect lipid-inducedcellular toxicity and dysfunction. If increases inthese circulating metabolites also reflect accumula-tion of these metabolites in organs such as theliver and skeletal muscle, perhaps these metabolitesmay have clinical utility in the diagnosis and prog-nosis of insulin resistance and T2D (see the sectionAdditional Considerations for the Use ofMetabolomics in the Diagnosis and Prognosis ofCVD). As alluded to previously, circulating BCAAspredict risk for CAD, often even after correcting forT2D (15,24,25), and LC/MS-based metabolic profilingin a cross-sectional study of 472 Chinese subjectsdemonstrated that serum BCAAs are increased andpositively correlated with carotid intima-mediathickness, a well-established index of subclinicalatherosclerosis (32). Importantly, if changes in thesecirculating metabolites are truly indicative ofatherosclerotic risk, the detection of these bio-markers may allow for early intervention with theappropriate therapies.

METABOLOMICS AND HF

HF affects more than 5 million people in NorthAmerica, withmore than one-half-million new cases inthe United States every year (33,34). Despite theadvent of several newmedications and devices used tomanage HF, there remains a high mortality rate of ap-proximately 30% to 50% within 5 years (35). Althougha vast array of pathophysiological and molecularevents contributes to the development and progres-sion of HF (36–39), a long-standing concept in HF isthat the failing heart has defects inmetabolic processesthat normally permit proper ATP production necessaryto maintain contractile function (39) (Figure 5). Withinthe failing heart, oxidative phosphorylation isimpaired, oxygen consumption is depressed, and ATPproduction is compromised (39), with failing humanhearts demonstrating approximately 30% lower ATPlevels than nonfailing hearts (40).

Metabolomic profiling, when used alone or in com-bination with standard HF biomarkers (e.g., B-typenatriuretic peptide [BNP]), has significant diagnosticand prognostic value (40). Although the majority ofreports utilizing metabolomic analyses in HF arecomparing non-HF controls with HF patients (41–43),other studies have shown that blood-based metab-olomics can also differentiate between the degree ofHF severity (44), or between HF with preserved ejec-tion fraction (HFpEF) and HF with reduced ejectionfraction (HFrEF) (42,45). As a result of the profoundchanges that occur in energy metabolism during HF,the identification of blood- or urine-based metabolicsignatures for these patients may uniquely assist cli-nicians with the clinical management of HF throughdetection, classification, and/or markers of effectivetreatment (46,47). Even though changes in the levels ofnumerous metabolites have been shown to occur inHF, there appear to be some keymetabolite classes thatconsistently change in metabolomic profiles of HFpatients (Table 2), which will be discussed herein.

FATTY ACID METABOLISM. It has been well estab-lished that impaired fatty acid flux in the mitochon-dria is associated with HF (48). Although multiplefatty acid metabolites are altered in the blood and/orurine of HF patients, attention has primarily focusedon changes in acylcarnitine profiles, as acylcarnitinesare derivatives of fatty acyl-CoAs that can reflectchanges in fatty acid oxidation rates (Figure 2). Indeed,changes in numerous S-C, M-C, and L-C acylcarnitines,as well as the ratios of S-C/M-C, S-C/L-C, or M-C/L-Coften indicate perturbations in fatty acid oxidation.Decreases in the ratios of S-C/M-C or M-C/L-C acyl-carnitines often reflect specific defects in the

FIGURE 6 Metabolomics in Obesity and Diabetes

Obesity and type 2 diabetes contribute to a number of deleterious changes in multiple organs in the body. These changes include adipose tissue

inflammation, skeletal muscle insulin resistance, nonalcoholic fatty liver disease, b-cell dysfunction, vascular dysfunction, and cardiomyopathy. Associated

with these changes are metabolic derangements in these organs that can contribute to an increased risk for CVD in these individuals. Blood-based

metabolomic profiles of individuals with obesity and type 2 diabetes often yield increased acylcarnitines, lactate, and BCAAs. CVD ¼ cardiovascular

disease; other abbreviations as in Figure 1.


2863

TABLE 2 Seminal HF Studies of Metabolomics in CVD and their Principal Findings

Patient Population Principal Metabolic FindingMethod of

Metabolite Profiling Ref. #

HFpEF patients(LVEF >45%)

HFrEF patients(LVEF <45%)

Circulating acylcarnitines areincreased in HFpEF patientsand further increased inHFrEF patients

MS/MS (45)

Chronic HF patients(LVEF 20%–30%)

End-stage HF patients(LVEF 15%–20%)

Circulating C16 and C18:1acylcarnitines are increased inend-stage HF vs. chronic HF,and are corrected viaLVAD placement

MS (46)

End-stage HF patientswith nonischemiccardiomyopathy(mean LVEF 17.5%)

S-C, M-C, and L-C acylcarnitinesare decreased in transmural LVmyocardial biopsies fromend-stage HF patients

b-Hydroxybutyryl-CoA isincreased in transmural LVmyocardial biopsies fromend-stage HF patients

LC/MS (51)

HFpEF patients(LVEF >45%)

HFrEF patients(LVEF <45%)

Circulating ketone bodies aredecreased in HFpEF patientsand further decreased inHFrEF patients

LC-MS/MS and 1H-NMR (42)

HF patients(LVEF #30%)with dilatedcardiomyopathy

BCKAs are increased in LVbiopsies of HF patients withdilated cardiomyopathy

LC/MS (63)

The table highlights some of the most important studies using metabolomics to assess metabolic perturbations inheart failure, with brief summaries of the key metabolite change observed via metabolomic profiling and themethod of metabolite quantification.

1H-NMR ¼ proton nuclear magnetic resonance; BCKA ¼ branched-chain a-keto-acid; CoA ¼ coenzyme A;CVD ¼ cardiovascular disease; HF ¼ heart failure; HFpEF ¼ heart failure with preserved ejection fraction; HFrEF¼heart failure with reduced ejection fraction; L-C ¼ long-chain; LC/MS ¼ liquid chromatography/mass spec-trometry; LC-MS/MS ¼ liquid chromatography/tandem mass spectrometry; LV ¼ left ventricular; LVAD ¼ leftventricular assist device; LVEF ¼ left ventricular ejection fraction; M-C ¼ medium-chain; MS ¼ mass spec-trometry; MS/MS ¼ tandem mass spectrometry; S-C ¼ short-chain.



2864

mitochondrial b-oxidation machinery, whereas in-creases in these ratios often reflect increases in fatty-acid oxidation (49). In contrast, decreases innumerous S-C, M-C, and L-C acylcarnitines oftenimply a defect in mitochondrial fatty acid uptake andsubsequent oxidation, whereas general increases overa broad range of S-C, M-C, and L-C acylcarnitinesindicate increases in fatty acid oxidation (Figure 2)(49,50).

Metabolomic profiling in a subset of patientsfrom the HF-ACTION (Exercise Training Program toImprove Clinical Outcomes in Individuals WithCongestive Heart Failure) trial demonstrated eleva-tions in circulating C16 and C18:1 acylcarnitines inpatients with end-stage HF versus those withchronic systolic HF, which were associated withincreased risk for mortality and hospitalization forHF (46). Intriguingly, metabolomic profilingfollowing left ventricular (LV) assist device im-plantation in these end-stage HF patients resultedin a decrease in these circulating L-C acylcarnitines.Consistent with these results, in a cohort of patientstaken from the CATHGEN biorepository, Hunteret al. (45) identified that circulating L-C

acylcarnitines are increased in HFpEF patientscompared with non-HF controls, and are furtherincreased in HFrEF patients. Conversely, Bedi et al.(51) observed a reduction in a broad range of S-C,M-C, and L-C acylcarnitines in myocardial tissuefrom end-stage HF patients at the time of transplantcompared with levels in heart tissue from subjectswith no history of HF. Reasons for the discrepancyin circulating versus myocardial acylcarnitine pro-files in these studies may be attributed to a largefraction of the patients from the HF-ACTION trialand CATHGEN biorepository being diabetic, whereasno end-stage HF group patients were diabetic in thestudy by Bedi et al. Circulating acylcarnitines areoften elevated in obese and diabetic subjects (52),and may actually reflect general elevations inskeletal muscle fatty acid oxidation (see thesection Additional Considerations for the Use ofMetabolomics in the Diagnosis and Prognosis ofCVD), although proper interpretation of thisdataset also requires assessment of S-C and M-Cacylcarnitine profiles (49). Conversely, the generalreduction in myocardial S-C, M-C, and L-C acylcar-nitine content observed by Bedi et al. (51) isreflective of impaired mitochondrial function andsubsequent fatty acid oxidation, which is entirelyconsistent with the reduction in fatty acid oxidationobserved in the more severe stages of HF (1,3,38).These findings observed from metabolomic analysesalign with molecular/metabolic studies in mice,whereby both protein expression and activity offatty acid oxidation enzymes only exhibit a milddecrease in compensated hypertrophy, but aremarkedly decreased in mice with decompensatedHF (1). As there is controversy surroundingmyocardial fatty acid oxidation rates in the failingheart, with some studies reporting an increase, nochange, or a decrease in fatty acid oxidation (1), it isplausible that many of these discrepancies can beexplained by the severity of HF (e.g., compensatedhypertrophy vs. decompensated HF), the overalldecline in LV function (HFpEF vs. HFrEF), and thepresence of underlying obesity/diabetes. Hence,future metabolomic studies in HF populations willneed to make distinct comparisons between thesesubgroups of HF patients, while taking into accounttheir associated comorbidities (e.g., obesity and/orT2D) and the type of sample collected (e.g., serum/plasma vs. tissue biopsy).

GLUCOSE METABOLISM. A number of studies havedemonstrated that mitochondrial glucose oxidationis defective within the failing myocardium (38,53).However, there is also reported controversy


2865

regarding glucose oxidation alterations in the failingheart, where it has also been suggested thatmyocardial glucose oxidation rates are increased inHF (38). This discrepancy may be explained by themagnitudes of difference between the higher abso-lute rates of glycolysis versus glucose oxidation inthe heart (w10-fold higher). Those studies reportingan increase in glucose oxidation in the failingmyocardium may simply be observing the result ofthe mass increase in glycolysis-derived pyruvate forPDH, despite PDH activity being impaired. More-over, increased circulating pyruvate and lactatelevels are often observed in metabolomic panelsfrom HF patients, consistent with the HF-mediatedincrease in myocardial glucose uptake and glycol-ysis (43,54). The elevation in circulating lactatelevels is likely due to a combination of the increasein glycolysis and the inability of the failing heart tosufficiently oxidize the increased pyruvate gener-ated from glycolysis (Figure 5) (38,53). Thus, blood-based metabolomic profiles from failing hearts areconsistent with the increased glycolysis rates andthe general defect in absolute glucose oxidationrates observed in animal and human studies directlyassessing myocardial metabolism (1,38).

KETONE BODY METABOLISM. Early metabolomicprofiling of blood from HF patients showed higherlevels of ketone bodies in HF patients than in healthycontrols (55,56). More recent work also showed that b-hydroxybutyrate and acetone (a breakdown productof acetoacetate and b-hydroxybutyrate) levels weresignificantly increased in HF patients (57). As such, ithas been proposed that the failing heart cannotadequately extract ketone bodies from the blood touse for energy production, thus leading to theelevation in circulating ketone bodies (58). On thecontrary, we have shown that the serum concen-trations of the ketone bodies, acetoacetate,a-hydroxybutyrate, and b-hydroxybutyrate, werelower in HFrEF patients than in non-HF controls (42).Interestingly, we also showed that serum ketonebodies were lower in HFrEF patients compared withHFpEF patients (42), suggesting potential changes inketone body metabolism on the basis of the “type” ofHF, similar to what is observed with L-C acylcarni-tines (Figure 5). In agreement with our studies, lowerconcentrations of a-hydroxybutyrate were reportedin HF patients with severe LV dysfunction(LVEF <35%) compared with HF patients with lesssevere declines in LVEF and healthy controls (44). Ithas been proposed that the extent to which the heartextracts ketone bodies from the blood is dependenton their circulating concentration (44). Thus, it is

apparent that the absolute levels of ketone bodies inthe blood represent more than just ketone body uti-lization by the heart, and other factors, such as theseverity and type of HF (i.e., compensated hypertro-phy vs. decompensated HF and/or HFpEF vs. HFrEF),other comorbidities, and/or the influence thatother organs have in regulating alterations incirculating ketone bodies are also involved. Likewise,recent studies also support that the failing heartexhibits elevations in ketone body oxidation,because metabolomic profiling in myocardialextracts from end-stage HF patients revealed eleva-tions in b-hydroxybutyryl-CoA, which were coupledto decreases in b-hydroxybutyrate and increasedexpression of enzymes involved in ketone bodyoxidation (51). Metabolomic profiling from myocar-dial extracts of mice subjected to HF via combinedtransverse aortic constriction and apical MI haveproduced similar findings (59). Although we cannotyet determine the physiological importance of thesemetabolomic findings in HF, the potential HF-mediated increase in myocardial ketone body oxida-tion rates may represent an adaptive mechanism tocompensate for the decline in myocardial fatty acidoxidation.

AMINO ACID METABOLISM. The contribution of car-diac amino acid metabolism to changes in themetabolome of HF patients is currently unknown.As mentioned in the section Metabolomics and IHD,amino acids appear to make a minimal contribution tooverall cardiac energy metabolism (20,21). Neverthe-less, renewed interest in studying amino acidmetabolism in the myocardium stems from demon-strations that decreases in the circulating methyl-arginine/arginine ratio and essential amino acidlevels from metabolomic panels in HF patients yield asimilar diagnostic value to BNP, while having aprognostic value that exceeds BNP for distinguishingbetween control and HF patients (41). Consistent withthis, circulating levels of essential and nonessentialamino acids were found to be lower in patients withchronic HF compared with non-HF controls (43), andtreatment that results in improvement of HF (LVfunction, quality of life, and survival) also partiallynormalizes circulating levels of these amino acids(47). Furthermore, amino acids, such as leucine andisoleucine, can also be catabolized and used togenerate ketone bodies (Figure 3), suggesting theremay be a link between the levels of ketogenic aminoacids and ketone bodies in HF (60).

In addition to the amino acids mentioned in thepreceding text, BCAAs may also play a role in HFpathogenesis. Interestingly, it has been shown that



2866

circulating leucine and isoleucine levels are elevatedin chronic HF patients compared with healthy con-trols (43). These results appear independent ofobesity (mean body mass index of 25) or dyslipide-mia (14% of HF patients demonstrated dyslipidemia),although it was not considered whether diabetes wasa potential confounder, because w70% of these HFpatients had T2D. Furthermore, animal models sug-gest that BCAA metabolism may impair glucose uti-lization in the heart (61), and that BCAAs mayaccumulate in the heart and interfere with insulinreceptor-mediated signal transduction (2,62).Studies of myocardial extracts revealed increasedlevels of the BCKAs a-ketoisovalerate (valine-derivedBCKA), a-ketoisocaproate (leucine-derived BCKA),and a-keto-b-methylvalerate (isoleucine-derivedBCKA) in mice and humans with HF, whereas phar-macological manipulations that enhanced systemicBCKA metabolism attenuated pressure overload-induced HF (63). Despite a number of studies nowsupporting aberrant BCAA metabolism in HF,whether abnormal circulating BCAA levels in HFpatients is a component of the metabolomic signa-ture of HF itself, or is due to associated comorbid-ities, such as insulin resistance and/or T2D, iscurrently unknown (see the section AdditionalConsiderations for the Use of Metabolomics in theDiagnosis and Prognosis of CVD).

METABOLOMICS AND CVD IN

OBESITY AND DIABETES

The prevalence of T2D is increasing at an alarmingrate, due in large part to the obesity epidemic and ouraging population (64–66). The disease usually de-velops later in life, and is often detected incidentallyor after sequelae of the associated vascular disease orCVD has occurred (65,67). The application of high-throughput metabolomics to populations at risk ofT2D has generated considerable excitement, giventhe potential to identify biomarkers predictive ofincident T2D and its associated sequelae, and thepossibility of gaining insight into novel modifiabledisease pathways and therapeutic strategies forreversing the disease and its associated comorbidities(Figure 6) (68). Furthermore, identification of circu-lating metabolites that correlate with risk for insulinresistance (e.g., BCAAs) via metabolomics may aid inthe treatment and overall prognosis of CVD. Hence,the use of metabolomics to screen for numerousmetabolites linked to insulin resistance could theo-retically aid in the diagnosis, prognosis, and man-agement of CVD in obese and/or T2D patients. In thelast decade, metabolomic profiling in obese and/or

T2D subjects has identified a number of potentialmetabolites that may be mechanistically involved inthe pathogenesis of these conditions and their asso-ciated comorbidities. We will herein discuss some ofthese perturbed metabolic pathways and their asso-ciated metabolites that may reflect perturbedmyocardial metabolism, as well as their potentialdiagnostic implications.

PET imaging with 1-11C-palmitate demonstratesincreases in both myocardial fatty acid uptake andoxidation in obese and insulin-resistant women, aswell as in men with uncomplicated T2D (1,69).Furthermore, numerous studies directly assessingfatty acid oxidation rates in hearts from obese and/orT2D animals have observed similar results (1,69).Though no assessment of the acylcarnitine profile hasbeen made in myocardial biopsies in obese and/orT2D subjects, myocardial extracts in obese versuslean mice demonstrate elevations in S-C, M-C, and L-C acylcarnitines (70), reflecting observations seen incirculating and skeletal muscle acylcarnitine profilesof obese and/or T2D humans and animals (see thesection Additional Considerations for the Use ofMetabolomics in the Diagnosis and Prognosis ofCVD) (50,52). Although the heart’s contribution tothe circulating acylcarnitine pool remains unknown,the expression of carnitine acetyltransferase, a keyenzyme involved in mediating carnitine transfer toacetyl-CoA for mitochondrial export (Figure 2), isgreater in the heart than the skeletal muscle (71). Thisobservation supports the possibility that the heartmay contribute to the circulating acylcarnitineprofile, and that increases in S-C, M-C, and L-Cacylcarnitines in the circulation of obese and/or T2Dpatients do reflect increases in myocardial fatty acidoxidation rates. With regard to myocardial glucosemetabolism in obese and/or T2D patients, the mostfrequently observed defect appears to take place atthe level of glucose oxidation (1,70). Nevertheless,the vast majority of studies supporting this concepthave been performed in animals, due to the lack ofavailable methods to accurately assess myocardialglucose oxidation rates in humans. Currently, studiesof myocardial ketone body and amino acid meta-bolism in humans are limited, and even more so inobese and/or T2D patients. Nevertheless, with recentmetabolomic findings illustrating that the failingheart relies on ketone bodies as an adaptive mecha-nism to meet its oxidative energy needs, whereasmyocardial BCAA/BCKA metabolism is impaired (seethe section Metabolomics and HF) (59,63), it is likelythat investigations on myocardial ketone body andBCAA metabolism in obesity and/or T2D will increasein order to elucidate their link with HF (72,73).


2867

ADDITIONAL CONSIDERATIONS FOR THE USE

OF METABOLOMICS IN THE DIAGNOSIS AND

PROGNOSIS OF CVD

Although the majority of this review has focused onhow blood-based metabolomics can aid in the diag-nosis and prognosis of various CVDs, patients withCVD often also have associated comorbidities, such askidney disease and hepatic steatosis, as well asobesity and/or T2D (8,17). In fact, obesity and/or T2Dmay add extreme complexity to the understanding ofthe metabolomic profile, because these conditionsoften result in significant pathology and perturbedmetabolism in other peripheral tissues (e.g., skeletalmuscle, liver, adipose tissue, among others), andthese peripheral tissues also contribute to the circu-lating metabolomic profile measured in patients withCVD. Indeed, systemic insulin resistance is a hallmarkfeature of obesity and T2D, and metabolomic profilingin obese and/or diabetic subjects often reveals in-creases in circulating lactate levels (74,75), which onthe surface suggest increases in anaerobic glycolysis(Figure 1). However, increases in anaerobic glycolysisare not consistent with systemic insulin resistance,which impairs glucose metabolism in numeroustissues, including skeletal muscle. Conversely, pro-teomic studies have observed increases in theexpression of glycolytic enzymes in vastus lateralisbiopsies from obese and T2D patients (76). Further-more, the source of elevated lactate in the circulationis difficult to ascertain, and can arise from othersources in addition to skeletal muscle. Because adi-pose tissue dysfunction in obesity is often associatedwith tissue hypoxia (77), increased anaerobic glycol-ysis in the expanded adipose depots of obese in-dividuals may also contribute to this elevation incirculating lactate.

As circulating free fatty acids and triacylglycerolsare markedly increased during obesity/T2D, fatty acidsupply, uptake, and metabolism are frequentlyincreased in peripheral tissues, such as the skeletalmuscle and heart. Metabolomic profiling to assessalterations in fatty acid metabolism in obese and/ordiabetic subjects has revealed increases in numerouscirculating acylcarnitine species (52), whereas otherstudies have reported no differences (78). It has beensuggested that a very large portion of the circulatingacylcarnitine pool arises from skeletal muscle-mediated export, and muscle biopsies in obesehumans also demonstrate increases in M-C and L-Cacylcarnitines compared with lean controls (79).Thus, assessment of acylcarnitine profiles via blood-based metabolomics supports increases in both

skeletal muscle and myocardial fatty acid oxidationrates (see the section Metabolomics and CVD inObesity and Diabetes), although additional investi-gation is required to determine the contributions in-dividual organs make to the circulating acylcarnitineprofile.

Circulating ketone bodies (e.g., b-hydroxybutyrateand acetoacetate) are frequently elevated in obeseand/or T2D patients (80), and elevations in circu-lating ketone bodies predict worsening of hypergly-cemia in men with or without newly diagnosed T2D(81). Changes in circulating b-hydroxybutyr-ylcarnitine may reflect changes in peripheral tissueketone body oxidation because b-hydroxybutyr-ylcarnitine is a mitochondrial intermediate deriveddirectly from b-hydroxybutyryl-CoA (30). Of interest,recent findings from the EMPA-REG OUTCOME trialhave observed that treatment with the sodium-glucose cotransporter 2 inhibitor, empagliflozin,markedly lowers cardiovascular risk in T2D patients,including a 35% reduction in hospitalization for HF(82). As these exciting findings were associated with asignificant increase in circulating b-hydroxybutyratelevels, it has been suggested that empagliflozin re-duces cardiovascular risk by increasing ketone bodyuptake and oxidation in the heart (83). This isconsistent with increased ketone body oxidation be-ing a potential beneficial and adaptive mechanism inthe failing heart (see the section Metabolomics andHF), and highlights the need to understand the sig-nificance of blood-based metabolomic profilesrevealing perturbations in ketone body metabolism.

Numerous studies have illustrated a tight correla-tion between circulating BCAA levels and risk forinsulin resistance/T2D (2,75). Since these initial ob-servations, several investigators have utilized novelmetabolomic profiling techniques to implicate BCAAsin obesity and insulin resistance (2,17), with BCAAsand aromatic amino acids emerging as predictors offuture T2D. Intriguingly, generation of an aminoacid score on the basis of 3 of these aminoacids (isoleucine, phenylalanine, and tyrosine) wasfound to significantly predict incident T2D (84). Inlight of findings suggesting impaired myocardialBCAA/BCKA metabolism in HF (63), it will be impor-tant to understand how the circulating BCAA profilein obesity reflects deranged systemic and myocardialmetabolism.

Taken together, although blood-based metab-olomics is proving to be a powerful tool that can assistin the diagnosis and prognosis of CVD, in the contextof obesity and/or T2D, caution must be applied inhow we infer the associated peripheral metabolic



2868

perturbations. In such instances, assessment ofchanges in metabolites across the diseased organwould prove more valuable, such as sampling arterialblood and the coronary sinus for the heart (see limi-tations of metabolomics described in the sectionMetabolomics), or having a tissue biopsy to compareagainst the serum/plasma sample. Nonetheless, dueto the ease of obtaining a serum/plasma sampleversus collecting a tissue biopsy or invasive cathe-terization, we must rely on molecular explorations inanimal models of disease to confirm and/or under-stand novel metabolic alterations identified fromhuman blood-based metabolomic signatures.

FINAL SUMMARY AND IMPLICATIONS

Metabolomic profiling technologies have significantlyevolved to a point where high-resolution and high-throughput assessment of thousands of metaboliteswithin a biological sample can be performed withrelative ease. Due to the expansive biochemical di-versity of metabolites within the human metabolome,several complementary platforms and techniques arenecessary for comprehensive metabolic characteriza-tion. These techniques, MS and NMR coupled with LCor gas chromatography, are exquisitely sensitive,generating the metabolic fingerprint of a biologicalsample that not only reflects the condition of interest,but also diet, drug effects, sex, and other comorbid-ities and exposures. Thus, human metabolomicstudies are consequently prone to clinical confou-nding factors that may lead to spurious diagnoses/conclusions. Yet the application of metabolomicprofiling to several large population-based epidemi-ological cohorts has allowed for robust statisticaladjustment for potential confounders, and has

resulted in externally reproducible findings. In addi-tion, metabolomics has been applied in smallerstudies of unique clinical scenarios that allow for se-rial sampling before and after a controlled biologicalperturbation (e.g., exercise testing, planned MI, drugadministration, and so on). Such paired experimentsallow subjects to serve as their own controls, miti-gating the risk of confounding factors. In addition,emerging investigational techniques are integratingmetabolomics with other “omics” platforms in orderto gain insight into pathophysiological interactions ofmetabolites, proteins, genes, and disease states.Furthermore, the state-of-the-art metabolomic tech-nologies now available provide us with a snapshot ofthe metabolic fingerprints of individual patients,which can serve as diagnostic and/or prognostic toolsthat can be used to identify impairments in systemicor myocardial metabolism occurring during thedevelopment and worsening of CVD, as well as iden-tifying the types and timing of specific interventions/therapies. Thus, metabolomics is a powerful tech-nology that is transforming our ability to predict,detect, and understand a myriad of cardiometabolicdisease states, and to monitor the effectiveness oftherapeutic interventions. In doing so, metabolomicscontinues to advance our societal objective ofpersonalizing the practice of medicine.

REPRINT REQUESTS AND CORRESPONDENCE: Dr.Jason R.B. Dyck, Cardiovascular Research Centre,Department of Pediatrics, Mazankowski Alberta HeartInstitute, Faculty of Medicine and Dentistry, Univer-sity of Alberta, 458 Heritage Medical Research Centre,Edmonton, Alberta T6G 2S2, Canada. E-mail: [email protected].

RE F E RENCE S

1. Lopaschuk GD, Ussher JR, Folmes CD, et al.Myocardial fatty acid metabolism in health anddisease. Physiol Rev 2010;90:207–58.

2. Newgard CB. Interplay between lipids andbranched-chain amino acids in development ofinsulin resistance. Cell Metab 2012;15:606–14.

3. Taegtmeyer H, Young ME, Lopaschuk GD, et al.,American Heart Association Council on Basic Car-diovascular Sciences. Assessing cardiac meta-bolism: a scientific statement from the AmericanHeart Association. Circ Res 2016;118:1659–701.

4. Lewis GD, Asnani A, Gerszten RE. Application ofmetabolomics to cardiovascular biomarker andpathway discovery. J AmColl Cardiol 2008;52:117–23.

5. Roberts LD, Gerszten RE. Toward new bio-markers of cardiometabolic diseases. Cell Metab2013;18:43–50.

6. Pauling L, Robinson AB, Teranishi R, et al.Quantitative analysis of urine vapor and breath bygas-liquid partition chromatography. Proc NatlAcad Sci U S A 1971;68:2374–6.

7. Horning EC, Horning MG. Metabolic profiles:gas-phase methods for analysis of metabolites.Clin Chem 1971;17:802–9.

8. Shah SH, Kraus WE, Newgard CB. Metabolomicprofiling for the identification of novel biomarkersand mechanisms related to common cardiovascu-lar diseases: form and function. Circulation 2012;126:1110–20.

9. Wishart DS, Tzur D, Knox C, et al. HMDB: theHuman Metabolome Database. Nucleic Acids Res2007;35:D521–6.

10. GBD 2013 Mortality and Causes of Death Col-laborators. Global, regional, and national age-sex

specific all-cause and cause-specific mortality for240 causes of death, 1990-2013: a systematicanalysis for the Global Burden of Disease Study2013. Lancet 2015;385:117–71.

11. Sabatine MS, Liu E, Morrow DA, et al. Metab-olomic identification of novel biomarkers ofmyocardial ischemia. Circulation 2005;112:3868–75.

12. Lewis GD, Wei R, Liu E, et al. Metaboliteprofiling of blood from individuals undergoingplanned myocardial infarction reveals earlymarkers of myocardial injury. J Clin Invest 2008;118:3503–12.

13. Turer AT, Stevens RD, Bain JR, et al. Metab-olomic profiling reveals distinct patterns ofmyocardial substrate use in humans with coronaryartery disease or left ventricular dysfunction

mailto:[email protected]

mailto:[email protected]

http://refhub.elsevier.com/S0735-1097(16)36731-6/sref1




















































2869

during surgical ischemia/reperfusion. Circulation2009;119:1736–46.

14. Wisneski JA, Gertz EW, Neese RA, et al.Myocardial metabolism of free fatty acids. Studieswith 14C-labeled substrates in humans. J ClinInvest 1987;79:359–66.

15. Shah SH, Bain JR, Muehlbauer MJ, et al. As-sociation of a peripheral blood metabolic profilewith coronary artery disease and risk of subse-quent cardiovascular events. Circ CardiovascGenet 2010;3:207–14.

16. Rizza S, Copetti M, Rossi C, et al. Metab-olomics signature improves the prediction of car-diovascular events in elderly subjects.Atherosclerosis 2014;232:260–4.

17. Shah SH, Newgard CB. Integrated metab-olomics and genomics: systems approaches tobiomarkers and mechanisms of cardiovasculardisease. Circ Cardiovasc Genet 2015;8:410–9.

18. Vary TC, Reibel DK, Neely JR. Control of en-ergy metabolism of heart muscle. Annu RevPhysiol 1981;43:419–30.

19. Bodi V, Sanchis J, Morales JM, et al. Metab-olomic profile of human myocardial ischemia bynuclear magnetic resonance spectroscopy of pe-ripheral blood serum: a translational study basedon transient coronary occlusion models. J Am CollCardiol 2012;59:1629–41.

20. Young LH, McNulty PH, Morgan C, et al.Myocardial protein turnover in patients with cor-onary artery disease. Effect of branched chainamino acid infusion. J Clin Invest 1991;87:554–60.

21. Ichihara K, Neely JR, Siehl DL, et al. Utilizationof leucine by working rat heart. Am J Physiol1980;239:E430–6.

22. Drake KJ, Sidorov VY, McGuinness OP, et al.Amino acids as metabolic substrates during cardiacischemia. Exp Biol Med (Maywood) 2012;237:1369–78.

23. Mudge GH Jr., Mills RM Jr., Taegtmeyer H,et al. Alterations of myocardial amino acid meta-bolism in chronic ischemic heart disease. J ClinInvest 1976;58:1185–92.

24. Bhattacharya S, Granger CB, Craig D, et al.Validation of the association between a branchedchain amino acid metabolite profile and extremesof coronary artery disease in patients referred forcardiac catheterization. Atherosclerosis 2014;232:191–6.

25. Yang RY, Wang SM, Sun L, et al. Association ofbranched-chain amino acids with coronary arterydisease: a matched-pair case-control study. NutrMetab Cardiovasc Dis 2015;25:937–42.

26. Tang WH, Hazen SL. The contributory role ofgut microbiota in cardiovascular disease. J ClinInvest 2014;124:4204–11.

27. Zhu W, Gregory JC, Org E, et al. Gut microbialmetabolite TMAO enhances platelet hyperreac-tivity and thrombosis risk. Cell 2016;165:111–24.

28. Wang Z, Roberts AB, Buffa JA, et al. Non-lethal inhibition of gut microbial trimethylamineproduction for the treatment of atherosclerosis.Cell 2015;163:1585–95.

29. Ganna A, Salihovic S, Sundström J, et al.Large-scale metabolomic profiling identifies novel

biomarkers for incident coronary heart disease.PLoS Genet 2014;10:e1004801.

30. Park JY, Lee SH, Shin MJ, et al. Alteration inmetabolic signature and lipid metabolism in pa-tients with angina pectoris and myocardial infarc-tion. PLoS One 2015;10:e0135228.

31. Muoio DM, Newgard CB. Molecular and meta-bolic mechanisms of insulin resistance and b-cellfailure in type 2 diabetes. Nat Rev Mol Cell Biol2008;9:193–205.

32. Yang R, Dong J, Zhao H, et al. Association ofbranched-chain amino acids with carotid intima-media thickness and coronary artery disease riskfactors. PLoS One 2014;9:e99598.

33. Kim DH, Hunt SA. Heart failure management:caregiver versus care plan. Circulation 2003;108:129–31.

34. Owan TE, Hodge DO, Herges RM, et al. Trendsin prevalence and outcome of heart failure withpreserved ejection fraction. N Engl J Med 2006;355:251–9.

35. Roger VL. The heart failure epidemic. Int JEnviron Res Public Health 2010;7:1807–30.

36. Oka T, Komuro I. Molecular mechanisms un-derlying the transition of cardiac hypertrophy toheart failure. Circ J 2008;72 Suppl A:A13–6.

37. Sharma K, Kass DA. Heart failure with pre-served ejection fraction: mechanisms, clinicalfeatures, and therapies. Circ Res 2014;115:79–96.

38. Stanley WC, Recchia FA, Lopaschuk GD.Myocardial substrate metabolism in the normaland failing heart. Physiol Rev 2005;85:1093–129.

39. Neubauer S. The failing heart—an engine outof fuel. N Engl J Med 2007;356:1140–51.

40. Nascimben L, Ingwall JS, Pauletto P, et al.Creatine kinase system in failing and nonfailinghuman myocardium. Circulation 1996;94:1894–901.

41. Cheng ML, Wang CH, Shiao MS, et al. Meta-bolic disturbances identified in plasma are associ-ated with outcomes in patients with heart failure:diagnostic and prognostic value of metabolomics.J Am Coll Cardiol 2015;65:1509–20.

42. Zordoky BN, Sung MM, Ezekowitz J, et al.Metabolomic fingerprint of heart failure withpreserved ejection fraction. PLoS One 2015;10:e0124844.

43. Wang J, Li Z, Chen J, et al. Metabolomicidentification of diagnostic plasma biomarkers inhumans with chronic heart failure. Mol Biosyst2013;9:2618–26.

44. Deidda M, Piras C, Dessalvi CC, et al. Metab-olomic approach to profile functional and meta-bolic changes in heart failure. J Transl Med 2015;13:297.

45. Hunter WG, Kelly JP, McGarrah RW III, et al.Metabolomic profiling identifies novel circulatingbiomarkers of mitochondrial dysfunction differ-entially elevated in heart failure with preservedversus reduced ejection fraction: evidence forshared metabolic impairments in clinical heartfailure. J Am Heart Assoc 2016;5:e003190.

46. Ahmad T, Kelly JP, McGarrah RW, et al.Prognostic implications of long–chain acylcarni-tines in heart failure and reversibility with

mechanical circulatory support. J Am Coll Cardiol2016;67:291–9.

47. Nemutlu E, Zhang S, Xu YZ, et al. Cardiacresynchronization therapy induces adaptivemetabolic transitions in the metabolomic profile ofheart failure. J Card Fail 2015;21:460–9.

48. Lai L, Leone TC, Keller MP, et al. Energymetabolic reprogramming in the hypertrophiedand early stage failing heart: a multisystemsapproach. Circ Heart Fail 2014;7:1022–31.

49. Bain JR, Stevens RD, Wenner BR, et al.Metabolomics applied to diabetes research: mov-ing from information to knowledge. Diabetes2009;58:2429–43.

50. Koves TR, Ussher JR, Noland RC, et al. Mito-chondrial overload and incomplete fatty acidoxidation contribute to skeletal muscle insulinresistance. Cell Metab 2008;7:45–56.

51. Bedi KC Jr., Snyder NW, Brandimarto J, et al.Evidence for intramyocardial disruption of lipidmetabolism and increased myocardial ketone uti-lization in advanced human heart failure. Circula-tion 2016;133:706–16.

52. Mihalik SJ, Goodpaster BH, Kelley DE, et al.Increased levels of plasma acylcarnitines in obesityand type 2 diabetes and identification of a markerof glucolipotoxicity. Obesity (Silver Spring) 2010;18:1695–700.

53. Gupte AA, Hamilton DJ, Cordero-Reyes AM,et al. Mechanical unloading promotes myocardialenergy recovery in human heart failure. Circ Car-diovasc Genet 2014;7:266–76.

54. Desmoulin F, Galinier M, Trouillet C, et al.Metabonomics analysis of plasma reveals thelactate to cholesterol ratio as an independentprognostic factor of short-term mortality in acuteheart failure. PLoS One 2013;8:e60737.

55. Lommi J, Koskinen P, Näveri H, et al. Heartfailure ketosis. J Intern Med 1997;242:231–8.

56. Lommi J, Kupari M, Koskinen P, et al. Bloodketone bodies in congestive heart failure. J AmColl Cardiol 1996;28:665–72.

57. Du Z, Shen A, Huang Y, et al. 1H-NMR-basedmetabolic analysis of human serum revealsnovel markers of myocardial energy expenditurein heart failure patients. PLoS One 2014;9:e88102.

58. Tenori L, Hu X, Pantaleo P, et al. Metabolomicfingerprint of heart failure in humans: a nuclearmagnetic resonance spectroscopy analysis. Int JCardiol 2013;168:e113–5.

59. Aubert G, Martin OJ, Horton JL, et al. Thefailing heart relies on ketone bodies as a fuel.Circulation 2016;133:698–705.

60. Turer AT. Using metabolomics to assessmyocardial metabolism and energetics in heartfailure. J Mol Cell Cardiol 2013;55:12–8.

61. Sansbury BE, DeMartino AM, Xie Z, et al.Metabolomic analysis of pressure-overloaded andinfarcted mouse hearts. Circ Heart Fail 2014;7:634–42.

62. Lynch CJ, Adams SH. Branched-chain aminoacids in metabolic signalling and insulin resistance.Nat Rev Endocrinol 2014;10:723–36.





































































































































































































2870

63. Sun H, Olson KC, Gao C, et al. Catabolic defectof branched-chain amino acids promotes heartfailure. Circulation 2016;133:2038–49.

64. Hu FB, Satija A, Manson JE. Curbing the dia-betes pandemic: the need for global policy solu-tions. JAMA 2015;313:2319–20.

65. Paneni F, Beckman JA, Creager MA, et al.Diabetes and vascular disease: pathophysiology,clinical consequences, and medical therapy: part I.Eur Heart J 2013;34:2436–43.

66. Wild S, Roglic G, Green A, et al. Global prev-alence of diabetes: estimates for the year 2000and projections for 2030. Diabetes Care 2004;27:1047–53.

67. Battiprolu PK, Lopez-Crisosto C, Wang ZV,et al. Diabetic cardiomyopathy and metabolicremodeling of the heart. Life Sci 2013;92:609–15.

68. Herder C, Karakas M, Koenig W. Biomarkers forthe prediction of type 2 diabetes and cardiovasculardisease. Clin Pharmacol Ther 2011;90:52–66.

69. Zlobine I, Gopal K, Ussher JR. Lipotoxicity inobesity and diabetes-related cardiac dysfunction.Biochim Biophys Acta 2016;1860:1555–68.

70. Ussher JR, Koves TR, Jaswal JS, et al. Insulin-stimulated cardiac glucose oxidation is increased inhigh-fat diet-induced obese mice lacking malonylCoA decarboxylase. Diabetes 2009;58:1766–75.

71. Noland RC, Koves TR, Seiler SE, et al. Carnitineinsufficiency caused by aging and overnutrition

compromises mitochondrial performance andmetabolic control. J Biol Chem 2009;284:22840–52.

72. McMurray JJ, Gerstein HC, Holman RR, et al.Heart failure: a cardiovascular outcome in diabetesthat can no longer be ignored. Lancet DiabetesEndocrinol 2014;2:843–51.

73. Riehle C, Abel ED. Insulin signaling and heartfailure. Circ Res 2016;118:1151–69.

74. Hansen JS, Zhao X, Irmler M, et al. Type 2diabetes alters metabolic and transcriptional sig-natures of glucose and amino acid metabolismduring exercise and recovery. Diabetologia 2015;58:1845–54.

75. Newgard CB, An J, Bain JR, et al. A branched-chain amino acid-related metabolic signature thatdifferentiates obese and lean humans and con-tributes to insulin resistance. Cell Metab 2009;9:311–26.

76. Giebelstein J, Poschmann G, Højlund K, et al.The proteomic signature of insulin-resistant hu-man skeletal muscle reveals increased glycolyticand decreased mitochondrial enzymes. Dia-betologia 2012;55:1114–27.

77. Sun K, Tordjman J, Clément K, et al. Fibrosisand adipose tissue dysfunction. Cell Metab 2013;18:470–7.

78. Baker PR II, Boyle KE, Koves TR, et al.Metabolomic analysis reveals altered skeletal

muscle amino acid and fatty acid handling in obesehumans. Obesity (Silver Spring) 2015;23:981–8.

79. Glynn EL, Piner LW, Huffman KM, et al. Impactof combined resistance and aerobic exercisetraining on branched-chain amino acid turnover,glycine metabolism and insulin sensitivity in over-weight humans. Diabetologia 2015;58:2324–35.

80. Robinson AM, Williamson DH. Physiologicalroles of ketone bodies as substrates and signals inmammalian tissues. Physiol Rev 1980;60:143–87.

81. Mahendran Y, Vangipurapu J, Cederberg H,et al. Association of ketone body levels with hy-perglycemia and type 2 diabetes in 9,398 Finnishmen. Diabetes 2013;62:3618–26.

82. Zinman B, Wanner C, Lachin JM, et al., EMPA-REG OUTCOME Investigators. Empagliflozin, car-diovascular outcomes, and mortality in type 2diabetes. N Engl J Med 2015;373:2117–28.

83. Ferrannini E,MarkM,MayouxE. CVprotection inthe EMPA-REG OUTCOME trial: a “thrifty substrate”hypothesis. Diabetes Care 2016;39:1108–14.

84. Wang TJ, Larson MG, Vasan RS, et al. Metab-olite profiles and the risk of developing diabetes.Nat Med 2011;17:448–53.

KEY WORDS diabetes mellitus,diagnostics, energy metabolism, heartfailure, ischemic heart disease, obesity



















































































The Emerging Role of Metabolomics in the Diagnosis and ... · contributing to the worsening of...

Documents

Transcript of The Emerging Role of Metabolomics in the Diagnosis and ... · contributing to the worsening of...