Reach Us +44-1467840001
Assessing the Efficiency of the Diabetic Heart at Subcellular, Tissue and Organ Level | OMICS International
ISSN: 2329-9126
Journal of General Practice
Make the best use of Scientific Research and information from our 700+ peer reviewed, Open Access Journals that operates with the help of 50,000+ Editorial Board Members and esteemed reviewers and 1000+ Scientific associations in Medical, Clinical, Pharmaceutical, Engineering, Technology and Management Fields.
Meet Inspiring Speakers and Experts at our 3000+ Global Conferenceseries Events with over 600+ Conferences, 1200+ Symposiums and 1200+ Workshops on Medical, Pharma, Engineering, Science, Technology and Business

Assessing the Efficiency of the Diabetic Heart at Subcellular, Tissue and Organ Level

Denis S Loiselle1,3*, June-Chiew Han3, Kimberley M Mellor1,3, Toan Pham1-4, Kenneth Tran3, Soyeon (Eden) Goo1,3,
Andrew J Taberner1,2 and Anthony JR Hickey4
1 Departments of Physiology, The University of Auckland, Auckland, New Zealand
2 Departments of Engineering Science, The University of Auckland, Auckland, New Zealand
3 Auckland Bioengineering Institute, The University of Auckland, Auckland, New Zealand
4 School of Biological Sciences, The University of Auckland, Auckland, New Zealand
Corresponding Author : Denis Loiselle
Department of Physiology
School of Biological Sciences
The University of Auckland, Auckland, New Zealand
Tel: 64-9-373-7599
E-mail: [email protected]
Received May 01, 2014; Accepted June 26, 2014; Published June 30, 2014
Citation: Loiselle DS, Han J, Mellor KM, Pham T, Tran K, et al. (2014) Assessing the Efficiency of the Diabetic Heart at Subcellular, Tissue and Organ Level. J Gen Pract 2:168. doi: 10.4172/2329-9126.1000168
Copyright: © 2014 Loiselle D, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Related article at
DownloadPubmed DownloadScholar Google

Visit for more related articles at Journal of General Practice


In this review, we focus on the diabetic heart rather than the vascular complications of diabetes. Focus is further narrowed to a specific, but widely used, animal model: the diabetic rat heart in which diabetes has been induced by a single injection of streptozotocin. Our experimental approach is primarily biophysical and ranges from measurements made in isolated working whole-hearts, to those made from isolated left-ventricular tissues and mitochondria. Our interest is on the effect of severe diabetes on cardiac energetics, in terms of efficiency of cardiac work performance, ATP synthesis and oxygen consumption. By designing experiments to test the energetic performance of the heart and its trabeculae across a wide range of protocols, we have revealed the dependence of efficiency on afterload. This has allowed us to clarify a long-standing uncertainty in the literature; whereas the diabetic heart is unable to work against high afterloads, it nevertheless retains normal peak efficiency. But a further anomaly has been revealed. Whereas there is no evidence that the diabetic myocardium loses peak mechanical efficiency, its mitochondria demonstrate a decreased P:O ratio - i.e., a decreased bioenergetic efficiency. This decrease is consistent with an increase in the rate of production of reactive oxygen species, together with elevated proton leakage across the inner mitochondrial membrane at near maximal phosphorylating respiration states.

In this review, we focus on the diabetic heart rather than the vascular complications of diabetes. Focus is further narrowed to a specific, but widely used, animal model: the diabetic rat heart in which diabetes has been induced by a single injection of streptozotocin. Our experimental approach is primarily biophysical and ranges from measurements made in isolated working whole-hearts, to those made from isolated left-ventricular tissues and mitochondria. Our interest is on the effect of severe diabetes on cardiac energetics, in terms of efficiency of cardiac work performance, ATP synthesis and oxygen consumption. By designing experiments to test the energetic performance of the heart and its trabeculae across a wide range of protocols, we have revealed the dependence of efficiency on afterload. This has allowed us to clarify a long-standing uncertainty in the literature; whereas the diabetic heart is unable to work against high afterloads, it nevertheless retains normal peak efficiency. But a further anomaly has been revealed. Whereas there is no evidence that the diabetic myocardium loses peak mechanical efficiency, its mitochondria demonstrate a decreased P:O ratio - i.e., a decreased bioenergetic efficiency. This decrease is consistent with an increase in the rate of production of reactive oxygen species, together with elevated proton leakage across the inner mitochondrial membrane at near maximal phosphorylating respiration states.
Cardiac oxygen consumption; Pressure-volume work; Myocardial heat production, Force-length work, Mitochondrial efficiency
Readers of the Journal of General Practice will scarcely need be reminded that diabetes is a leading cause of death in both the ‘developed’ and ‘developing’ world [1]. New Zealand is no exception. We do have something of a unique situation, however, since the burden falls disproportionately on the Maori & Polynesian populations, with both the prevalence and the death rates approaching double those of Pakeha ( Given the strong link between obesity and diabetes, the high rates of obesity in New Zealand (31% of adults in 2013) is of increasing concern (
There have been a number of excellent reviews on the subject of diabetes [2-5], surely the most unique of which is that authored by Gottlieb [6], dealing with clinical and epidemiological issues. Since we are neither clinicians nor epidemiologists but rather, physiologists, biophysicists and bioengineers, we offer a different point-of-view. Our motivation commences from the knowledge that upwards of 80% of overall morbidity and mortality associated with diabetes is attributable to cardiovascular disease, involving both the arterial/arteriolar system and the myocardium per se. Specifically, we focus on the ‘diabetic heart’ first described by Rubler et al. [7] from four diabetic patients who had died of diabetic glomerulosclerosis and whose hearts were hypertrophied but showed no signs of arterial disease at autopsy. It is now generally accepted that a specific diabetic cardiomyopathy exists, independent of vascular complications [8]. In our case, the phrase ‘diabetic heart’ implies the heart of an animal rendered diabetic by some means.
Animal Models of Diabetes
Whereas ultimate interest must remain on the diabetic human heart, it is experimentally fortuitous that animal models exist. In fact, there are a plethora of animal models (for a comprehensive review, see Gomes et al., 2013). Some have been induced by genetic manipulation (db/db mouse, ob/ob mouse, OVE26 mouse, GLUT4-KO mouse), while others have arisen from spontaneous genetic mutation (Zucker Diabetic Fatty (ZDF) rat, Zucker Fatty (ZF) rat, Goto-Kakizaki rat), dietary intervention (high fructose diet, high fat diet, high sucrose diet), or pharmacological intervention (streptozotocin, alloxan). Our particular choice has been the heart of the streptozoticin (STZ)-treated rat. It is an excellent model of Type I diabetes (since the pancreatic Islets of Langerhans have been destroyed), but it has the added advantage of also mimicking a number of features of Type II diabetes in human patients [9].
A feature of STZ-induced diabetes that renders it particularly attractive to experimentalists is its simplicity. Some six to eight weeks following a single injection, rats have full-blown diabetic symptoms including blood glucose concentrations in excess of 25 mM. The cardiac complications evident in the STZ rat mimic those observed clinically: hypertrophy, fibrosis and (of particular relevance to this review) prolonged relaxation. The model thus provides an important experimental opportunity for new investigations of the mechanisms of cardiac pathology in diabetes.
One such mechanism, which has not been without controversy, is the proposal that cardiac dysfunction in diabetes is explained by a reduction in efficiency of the heart pump. We consider that our recent investigations have advanced understanding of this issue. Indeed, our findings form the focus of this Review. A secondary objective is to demonstrate the applicability of animal models in this endeavour. But, first, a brief overview of the characteristics of the diabetic heart.
Characteristics of the Diabetic Heart
Diabetic cardiomyopathy is a distinct cardiomyopathy [7-11]. Indeed, in a substantial proportion of otherwise healthy type 2 diabetics, subclinical left ventricular dysfunction presents prior to development of overt signs of coronary vascular disease [10,12]. Abnormal filling of the left ventricle, consistent with a decrease of left-ventricular compliance [13-15] is an early sign of cardiac pathology in the diabetic heart [16,17]. The prevalence of diastolic dysfunction in type 1 and type 2 diabetes may be as high as 40-75% without overt coronary artery disease [18]. Diastolic dysfunction has been reported in type 1 diabetic adolescents [19-22]. These patients have neither vascular nor haemodynamic complications. Nevertheless, they exhibit a primary myocardial defect. Progression from diastolic to systolic dysfunction in late-stage diabetes is observed clinically, and overlap of both forms of failure is common [23]. In type 2 diabetic patients, myocardial systolic dysfunction is independently associated with glycated hemoglobin level [10].
The Meaning of Cardiac Efficiency
‘Efficiency’, in any field, implies the notion of ‘output’ for a given amount of ‘input’. The heart, like any other muscle, is a thermodynamic machine; it is capable of directly converting biochemical energy into mechanical energy. A reduction of its efficiency implies either/both diminished ability to perform work, for a given amount of metabolic input, or increased metabolic energy expenditure for a given work output. In either case, decreased efficiency of the heart signifies that something is amiss with its internal machinery. Malfunction can appear on either the ‘input’ limb (i.e., the production of ATP) or the ‘output’ limb (i.e., the consumption of ATP). It is possible to quantify the efficiency of both limbs of this ‘production-consumption’ cycle.
Mitochondrial efficiency
Mitochondria are plentiful in the heart, occupying some 30%-40% of myocyte volume [24] Input to a mitochondrion is in the form of O2 and substrates derived from glucose, lipids and amino acids to form acetyl-CoA, ADP and inorganic phosphate (Pi). The fundamental output is ATP, which is either utilised directly or acts to rephosphorylate creatine to creatine phosphate, which functions both as a ‘high energy store’ and an ‘energy shuttle’ [25,26]. The citric acid cycle and the electron transport chain establish a chemi-osmotic gradient of protons across the inner mitochondrial membrane [27]. Harnessing of the free energy of the proton gradient (Figure 1) dictates the efficiency of ATP generation from ADP and Pi by the membrane-bound ATP-ADP synthetase [28-31].
Since the output of this system is ATP and the input is oxygen, it is natural to define the efficiency of the ‘metabolic chain’ as the ratio of the number of moles of ATP produced per mole of oxygen consumed. Biochemists typically measure the O2 flux and infer the ATP by monitoring transitions in respiration rates (or states 2 to 3 to 4) following the addition and depletion of limiting amounts of ADP. A simple P:O ratio is derived, where P signifies ‘ATP’ and O signifies molecular oxygen. Biophysicists, on the other hand, typically measure only the consumption of oxygen (VO2), converting it to units of energy using a conversion factor known as the ‘energetic equivalent of oxygen’.
The energetic equivalent of oxygen: This quantity, classically evaluated as 20 kJ per litre of oxygen (marginally less for fatty acids), arises from an experiment that is conceptually and technically very simple–namely, the complete oxidation of glucose in a bomb-calorimeter. Readers may recall that this yields 36-38 molecules of ATP, giving the classical P:O ratio of 3.03.2. However, re-evaluation of these quantities over the past decade has led to their reduction in magnitude. The revised values are 29.4-29.9 ATP/glucose producing an updated P:O ratio of 2.4-2.5 [32,33]. This reduction of efficiency of approximately 18% is primarily attributable to more recent experiments being conducted at physiological temperature (37°C) rather than at ‘standard’ (20°C) or ‘room’ (24°C) temperatures. At body temperature, the inner mitochondrial membrane becomes more permeable to protons and this diminishes the apparent efficiency of oxidative phosphorylation (OXPHOS).
The measurement of mitochondrial efficiency
Recent technological advances have reinvigorated the measurement of mitochondrial energetics, largely through the application of fluorescence-based measures that permit analysis of reactive oxygen species (ROS), mitochondrial membrane potential and ATP synthesis [34-36]. ‘Substrate-inhibitor-titration’ protocols have been developed [37-41]. These protocols have provided approaches that permit dissection of the electron transport system (ETS) and OXPHOS. Essentially different protocols measure O2 uptake in a high-resolution respirometer while metabolic substrates, or specific poisons, are added in an order generally specified by the ‘linear’ electron-stripping arrangement of Complexes I to IV. One of our general experimental protocols, which is coupled to the measurement of ATP, is as follows.
Measurement of the rate of leak respiration: Complex I-mediated “leak” respiration can be determined using saturating levels of Pi and substrates that generate NADH (eg. malate, glutamate and pyruvate). Since ADP is absent, any resulting respiration is due to proton conductance across the inner membrane, or electron slippage in the ETS. If saturating ADP is added, OXPHOS is activated and a ratio of Leak/OXPHOS can provide a measure of the potential turnover of oxygen that does not contribute to ATP synthesis. This can also be performed following the addition of succinate (Figure 2). Following the addition of succinate, FADH2 is formed (bound within Complex II) and is re-oxidised to donate electrons to the ETS. Following the addition of saturating ADP, OXPHOS is activated. The ratio of Leak/OXPHOS can be derived, and this provides a measure of the potential turnover of oxygen that does not contribute to ATP synthesis, i.e. efficiency. While useful, this ratio assumes that the proportion of leak remains constant, and does not decrease as phosphorylation commences, when the membrane potential declines. Leak ratios are also lower with Complex II substrates as the redox potential of FADH2 is too low to mediate proton pumping (unlike Complex I, see Figure 1), and therefore Complex II is less “coupled” to OXPHOS. The combination of Complex I and II substrates results in an intermediary Leak/OXPHOS ratio, or coupling in terms of the P:O ratio, but this likely reflects mitochondrial function and efficiencies in vivo, as both sets of substrates are present in vivo [42]. Electron transfer flavoproteins (ETF) involved in fatty acid ß-oxidation, are also FAD containing enzymes and, like Complex II, are also less coupled. Therefore the Leak respiration rate and efficiency vary with substrate, and substrate contributions likely differ given that the diabetic heart has a greater dependence on lipid oxidation [43-45].
Leak and uncoupled respiration: Leak respiration results from different known and hypothetical sources. Those that are known result from proton conductance directly through the inner membrane, through Uncoupling Proteins (UCPs) [46,47], and the adenine nucleotide transporter (ANT, Figure 1). In addition, some lipids can act as uncoupling agents. Less understood sources of leak may result from electron slippage, and some of this may be attributed to reactive oxygen species (ROS) production from the ETS, yet this likely contributes a minimal loss of efficiency [34]. UCPs are positively regulated by the reactive oxygen species: O2.-, and their activation will decrease coupling efficiency [46,47]. The source of leak respiration can be further explored by addition of oligomycin, a blocker of the F1F0-ATPase, and/or atracyloside or carboxy atractyloside. While both inhibit OXPHOS, the atractyloside compounds also inhibit proton conductance through ANT. Guanine Diphosphate (GDP) is also a potent inhibitor of UCPs [48]. Therefore the OXPHOS components of leak can be dissected.
Testing limits on the OXPHOS system: Repeated titrations of uncoupling agents such as dintrophenol, or more commonly carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (FCCP), are used to uncouple and depolarise mitochondria [49]. This then provides a measure of the ETS capacity relative to the OXHPHOS, and ratios ETS/OXPHOS. Typically there is an excess capacity of the ETS/OXPHOS. A decrease in this ratio can indicate decreased ETS capacities [39]. Preliminary data from our group indicates that there is less ETS capacity in STZ-diabetic rat heart mitochondria and similar findings have been reported by others [50]. In this regard, it is germane to note that cardiac mitochondrial respiration is regulated by nitric oxide [51], thereby likely limiting ATP synthesis in vivo, and that nitric oxide production is elevated in chronic states of cardiac disease [52].
Background respiration and Complex IV (Cytochrome c oxidase): Following the addition of antimycin A, respiration is inhibited at Complex III and this provides a measure of non-mitochondrial respiration. The addition of ascorbate and N,N,N',N'-Tetramethyl-p-Phenylenediamine (TMPD) supplies electrons to Complex IV (Figure 2).
Real-time measurement of ATP synthesis: Many studies of the diabetic heart have shown metabolic derangements at the level of the mitochondria (see the Review by Bugger and Abel [53]). One study, testing permeabilised fibres of type 2 diabetic human atrial appendage, showed altered (Complex I-supporting) lipid and glutamate flux [54]. Other studies, using animal models of type II diabetes mellitus, showed decreased OXPHOS capacities and P:O ratios [48,55,56], but see [57]. But the reported effects of diabetes on coupling efficiency have ranged from negligible to reduced [53]. Some of this variability may be attributable to the fact that, until recently, P:O ratios have been determined using sub-optimal amounts of ADP. This, in turn, limits the maximal flux of respiration in the OXPHOS state, thereby restricting interpretation in the context of high work rates.
Methods have now been developed that permit fluorescence-based measurements of ATP/ADP exchange [58] under non-limiting conditions and we have integrated this system to examine ATP synthesis simultaneously with mitochondrial respiration. The magnitude of ATP/ADP exchange can be measured in real time by fluorometric quantitation of Mg2+ (Figure 3) which preferentially chelates ATP. We follow the fluorescence of the fluorophore Magnesium Green (MgG, Molecular Probes™) since Mg2+-MgG fluorescence decreases as ATP is formed [58]. The inhibitors ouabain and blebbistatin are used to inhibit Na+-K+ ATPase and the myosin heavy-chain [59], respectively. A sample (2 mg of tissue homogenate, or 0.125 mg/ml of mitochondrial protein) is added to Mg2+-containing media within the oxygraph chambers and allowed to equilibrate. Substrates (pyruvate, glutamate, malate and succinate) are then added followed by excess Mg2+-free ADP. The added ADP also serves as a reference/standard since, whereas ADP also binds Mg2+, it does so with a 9-fold lower affinity and any ATP produced decreases the MgG fluorescence further. Separate experiments, titrating either ADP or ATP, are conducted to determine correction factors so that the ADP signal can be used as a calibration point. OXPHOS now proceeds and ATP production/turnover can be measured in real time with, and even without, oxygen [34]. An advantage of this approach is that ATP dynamics can be measured at saturating ADP concentrations, in pseudo-steady states and into anoxia. Oligomycin can then be added to determine the rate of mitochondrial ATP hydrolysis, as anoxic mitochondria become net ATP consumers. This provides insight into issues likely occurring in settings of cardiac infarct. Recent data indicate that, whereas type I diabetic rat heart have depressed OXPHOS capacity, anoxic mitochondrial ATP hydrolysis capacity matches that of normal heart mitochondria. That is, upon reperfusion, diabetic hearts will be at a greater ATP deficit than healthy hearts.
Measurement of rate of production of reactive oxygen species (ROS) simultaneously with respiration. An additional, although minor, loss of mitochondrial efficiency occurs through the loss of electrons from the ETS. The superoxide radical (O2.) is formed from molecular oxygen as a by-product of oxidation at mitochondrial Complex I and Complex III. In vivo, O2. is normally degraded by superoxide dismutases (SOD) to hydrogen peroxide (H2O2), or consumed by antioxidants and antioxidant enzyme systems (which generally reduce H2O2 to H2O and less harmful products) and by peroxidases and catalase to O2 and H2O [60].
Net ROS production can be measured simultaneously with respirational flux using the fluorophore Amplex Ultrared (Molecualr Probes). Addition of exogenous SOD can also enhance ROS detection. The combined mitochondrial H2O2 and exogenous SOD-derived H2O2 is then linked to horseradish peroxidase which, in turn, reacts with Amplex Ultrared to form a fluorescent product [35,61]. The production of H2O2 from various sources is shown in Figure 4a for samples of heart tissue from STZ-treated rats and their SHAM-injected controls. The relatively minor production of reactive oxygen species, is made apparent in Figure 4b, where the least activity is seen to occur during oxidative phosphorylation.
Does Diabetes Diminish Mitochondrial Efficiency?
This question has been addressed in a variety of ways for the past half century, with a remarkable degree of consensus.
Evidence from the literature
In an early paper, Parks et al. [62] showed a decrease in the P:O ratio of alloxan-induced type 1 diabetic rats, albeit from liver mitochondria. Comparable behaviour was reported for mitochondria isolated from the hearts of STZ-induced diabetic rats by Pierce and Dhalla [63]. Since those early studies, functional impairment of diabetic cardiac mitochondria has been reported in a variety of animal models: genetically diabetic mice [64], STZ-induced type 1 diabetic rats [65], insulin-resistant and obese mice [43,44] and Zucker Diabetic Fatty rats at submaximal rates of ADP-stimulated respiration [66], even prior to the development of symptoms of diabetic cardiomyopathy [67]. Our recent (unpublished) results (Figure 3) are in general accord.
Possible modi operandi
It has been recognised for half a century that diabetes is characterised by a reduction of glucose metabolism in favour of fatty acids [68]. That original study was done in the rat heart but the result has been shown to be equally valid in human patients suffering Type I diabetes [69]. It is this switch from carbohydrate to fatty acid metabolism that is presumed to diminish cardiac mitochondrial efficiency. Yet a number of causes have been touted and tested. Boudina and Abel [70] discuss various deleterious effects of increased fatty acid oxidation that characterise both Type I and Type II diabetes, among which is increased mitochondrial production of reactive oxygen species (ROS) including superoxide radicals known to be capable of activating uncoupling proteins [55]. The result of such a scenario would be a decrease of mitochondrial efficiency, since the proton gradient would be ‘squandered’ (Figure 1) and the ETS would demand more O2. Boudina and Abel [71] further contend that the ‘activating’ action of reactive oxygen species “…amplify the hyperglycemia-induced activation of protein kinase C isoforms [and] increase formation of glucose-derived advanced glycation end-products”. The hyperglycaemic setting of diabetes, combined with elevated levels of ROS, provides the ideal environment for glycation to occur–irreversible and obstructive attachment of an advanced glycation end product to a protein, adversely affecting protein structure and function.
Oxygen radicals, in particular, are expected to reduce mitochondrial efficiency by damaging proteins and lipid membranes, cardiolipin in particular [72], and elevating proton leak through uncoupling proteins [70]. But whatever the modi operandi of ROS in diminishing mitochondrial efficiency, it is sobering to learn that a high-fructose diet per se is capable of generating superoxide in the murine heart, despite only mild hyperglycemia and in the absence of overt cardiac dysfunction [73,74], thus supporting the contention that ROS-induced intracellular damage is an early manifestation of diabetic cardiac pathology. Over-consumption of fructose, particularly as a consequence of its insidious use as a sweetener in foods and soft-drinks, is implicated in the ‘insulin resistance syndrome’ and the ‘obesity epidemic’, which runs in parallel with the ‘diabetes epidemic’ [75].
The Output Efficiency of the Heart
Having dealt with the input (mitochondrial) efficiency of the heart, we now turn our attention to its output (work) efficiency. In order to do so, we first remind readers of the commonly-adopted partitioning of the total energy expenditure.
Partitioning of cardiac energy expenditure
The major division is between basal and suprabasal metabolic rate. The basal component requires that the heart be at rest. Since this does not occur during life, some experimental intervention (such as high K+ arrest) is required for this measurement. The suprabasal moiety reflects the metabolic cost of the actively beating heart. This component of metabolism is again divided between the energy to activate contraction and the energy to effect contraction. The latter reflects the hydrolysis of ATP by crossbridge ATPase activity. The former includes the hydrolysis of ATP by the sarcolemmal Na+-K+ ATPase to restore transsarcolemmal ion gradients following the action potential, plus the cost associated with the uptake of Ca2+ by the sarcoplasmic reticular (SR) Ca2+-ATPase. The SR is the internal store of Ca2+ in the cell. This Ca2+ is released from the SR store and binds to contractile proteins, thereby inducing contraction. The SR Ca2+ ATPase transports the Ca2+ back into the SR thus allowing relaxation to occur.
Early studies [76,77] (for a review, see Cesario et al. [78]) attributed the diminished contractility of the diabetic heart to reduced Ca2+ release from the sarcoplasmic reticulum (SR). However, more recent studies of streptozotocin-induced diabetic hypertrophy [79,80] have shown no detectable diminution in magnitude of the ‘Ca2+ transient’.
Because cardiac dysfunction in diabetes is not always linked to changes in the Ca2+ transient, we have focussed our attention downstream, on the mechanical behaviour, and energetic cost, of the Ca2+-triggered, actin-activated, myosin ATPase of the contractile proteins. The acto-myosin complex is the principal consumer of metabolic energy in the cardiac myocyte. It directly converts the Free Energy of ATP hydrolysis into cycling of actin-myosin cross-bridges, thereby allowing shortening and force production by the sarcomeres. Our focus is on the efficiency of this energy transduction process [34]. We measure cardiac efficiency at two distinct levels: the isolated whole-heart and isolated ventricular trabeculae. In the former case, efficiency is given by the ratio of pressure-volume work to the energetic equivalent of oxygen [81]. In the latter case, efficiency is given by the ratio of force-length work to the sum of work and heat production [82-84]. We hope to demonstrate to readers the advantage of studying the diabetic myocardium at these two extreme levels of structural organisation of the heart. But, prior to embarking on that undertaking, we wish to dispel that notion that there are other, arbitrary, definitions of ‘efficiency’.
“PVA ‘efficiency’”: Many readers will have encountered the notion of PVA (pressure-volume-area) efficiency as introduced and extensively promulgated by Suga and colleagues [85-89]. The notion is based on the assumption that the area between the abscissa and the (linear) end-systolic pressure-volume relation quantifies the external work performed by the heart. The ratio of this ‘work’ term to the volume of oxygen consumed in performing it comprises ‘PVA efficiency’. The notion is appealing but, in our view, fundamentally flawed for the following reasons. Firstly, efficiency should not be invariant with afterload. In particular it should be zero when no external work is performed–i.e., under both zero-load contractions and isovolumic contractions. But ‘PVA efficiency’ of the healthy heart is reported to be some 40%, independent of loading conditions [89]. Secondly, the numerator should not contain a component that correlates with twitch heat production, as recent experimental and theoretical analyses have shown to be the case [90,91]. Thus, whereas the ratio of PVA to VO2 may be of clinical or other interest, we advocate that it no longer be referenced as a measure of cardiac efficiency.
In order to study the efficiency of the heart, we use isolated hearts and isolated trabecula preparations. These experimental models allow cardiac investigations without the complication of either neural or humoral input. Details of these two experimental approaches follow.
The isolated rat whole-heart model
In our experiments, an STZ-induced diabetic rat heart is excised under deep anaesthesia and mounted in a working-heart rig [34,81,84] in which preload and afterload can be varied independently. In order to avoid the vexing problem of variation of heart rate between specimens, the hearts are electrically paced at a rate in excess of their spontaneous beat frequencies. Aortic and coronary flows of a glucose-supplemented saline perfusate, together with their oxygen partial pressures, are measured continuously, allowing calculation of oxygen consumption (VO2) at any desired combination of preload and afterload (Figure 5).
Examples of (continuous) output from selected transducers located in the whole-heart apparatus are presented in Figure 6.
Whereas isovolumic contractions cannot be achieved (since the ‘systemic’ (aortic) and coronary (pulmonary arterial) flows are in parallel), they can be approximated by increasing the afterload to the point where the left ventricle is no longer capable of ejecting a stroke volume. Thus the metabolic energy demand can be examined over a very wide range of afterloads. This fact is crucial to many of our interpretations, as will be elaborated below.
The measured variables shown in Figure 6 form the basis of derived quantities of interest. For example, stroke volume is given by the quotient of total flow (aortic+coronary) and pacing frequency. External work is then calculated as the product of stroke volume and mean aortic pressure. Oxygen consumption (VO2) is given by the product of coronary flow, the solubility of oxygen and the arterio-venous difference in partial pressure across the coronary circulation. Cardiac efficiency is then given by the ratio of pressure-volume work to VO2. Calculated (and curve-fitted) values of work, oxygen consumption and their ratio, cardiac efficiency, are shown in Figure 7 for Control and STZ-treated hearts.
As is evident in Figure 7, the ability of the heart to perform external, macroscopic, pressure-volume work (given by the product of stroke volume and mean aortic pressure, Panel A) is a complex function of after-load, being zero at both the origin and at the extrapolated isovolumic point, and peaking at some intermediate afterload. By contrast, the metabolic energy expenditure (indexed by VO2, Panel A) has a roughly monotonic dependence on afterload. In consequence, the ratio of mechanical energy output to metabolic energy input (efficiency, εTotal, Panel B) mimics the afterload-dependence of work. The subscript ‘Total’ on ‘ε’ signifies that we are including the total oxygen consumption of the heart and not merely its supra-basal value. If required, the basal rate of cardiac metabolism can be estimated by arresting the heart, commonly achieved by using either a high K+ [93, 94] or a low Ca2+ [93,95] perfusate. Subtracting the basal oxygen consumption from the total clearly identifies the metabolic cost of activating contraction plus that of the work performance per se. Separation of the ‘activation’ component of work from the contractile component is not readily achieved in the whole heart, although Suga and colleagues developed a pharmacologically-based protocol [96,97] for its estimation. We take a different (non-pharmacological) approach as will be shown in the section on energetics of trabeculae.
The rat isolated LV trabecula model
Trabeculae are minute, naturally arising, thread-like collections of linearly arranged myocytes (commonly 2-3 mm in length and 250 µm in diameter). This ‘linear’ microstructure largely eliminates any ambiguity in the measurement of extent of shortening and the development of force. For this reason, they have become the preparation-of-choice in the study of cardiac mechanics, since their introduction by ter Keurs et al. some 35 years ago [98]. In Figure 8, we present photographs of the steps involved in dissecting and mounting a suitable rat LV trabecula.
Work-loop microcalorimetry: Our laboratory has developed a unique device with which to drive trabeculae through force-length work-loops designed to measure the pressure-volume work-loops of the heart, while simultaneously measuring their heat output. This device [82], which we call a work-loop calorimeter, measures displacement in the order of micrometres, forces in the order of micrograms and heat output in the order of nanowatts. Displacement is achieved by a linear motor and is measured by one arm of a heterodyne laser. Another arm of the laser, using differential laser-Doppler techniques, measures the bending of a very stiff stainless-steel lever to provide a measure of force production (Figure 9). The product of force and displacement provides an estimate of external work performance.
The principal of heat measurement by the device is conceptually simple. The trabecula is tethered to ‘upstream’ and ‘downstream’ platinum hooks and located mid-stream. Since it is respiring and producing heat, the temperature of the downstream superfusate is higher than that upstream. This difference is sensed by arrays of thermopiles, the voltage outputs of which are proportional to the difference of temperature of the upstream and downstream superfusate (see the inset of Figure 9).
Knowing the rate of flow of the bathing solution and the temperature sensitivity of the thermopiles, it is straightforward to calculate the rate of heat production of the trabecula. With this facility, coupled with the ability to measure and control both force and shortening during a single muscle twitch, we can subject trabeculae to work-loops such as those exemplified in Figure 10. Note that these are force-length loops, designed to mimic the pressure-volume loops performed by the heart with each beat.
Work, heat and contractile efficiency of trabeculae: Typical experimental records from a ‘work-loop experiment’ are presented in Figure10.
From records such as those presented in Figure 10, arising from an individual experiment, it is straightforward to calculate values of work, heat and efficiency, as functions of afterload.
Repeating the procedure for a number of trabeculae allows us to calculate average values for each variable and, thus, to make statistical comparisons between diabetic and non-diabetic hearts. Examples of both individual (insets) and group results are displayed in Figure 11 where the thick lines denote STZ-treated hearts and the thin lines their controls.
It can be seen in Figure 11A that twitch heat production is linearly related to afterload. Hence, extrapolation of the line to the ordinate, where the magnitude of the force is zero, yields an estimate of the force-independent or ‘activation’ heat. As described, above, ‘activation’ is the name given to the processes that take place between the arrival of an action potential (i.e., depolarisation of the sarcolemmal membrane) and release of Ca2+ from its intracellular stores (the sarcoplasmic reticulum). Activation heat is the thermal accompaniment of the work of ion pumps that restore the cell to its diastolic ionic equilibrium. We estimate this value by quantifying the heat production in the absence of force production, i.e., the intercept of the heat-afterload relation.
It can also be seen in Figure 11A that the Work component is a modest fraction of the total Energy expenditure (note the double scale on the ordinate), implying equally modest values of efficiency. This implication is borne out by the data shown in Figure 11B, where it can be seen that the maximum value of total efficiency (which occurs at a relative afterload of about 0.4) is scarcely 17%. Since this index of ‘total’ efficiency includes the activation heat in the denominator of the ratio, the contractile efficiency of the crossbridges per se can be extracted by subtracting the activation heat from the total heat in the denominator of the efficiency ratio. When this is done, the result (Figure 11C) shows a peak value in the vicinity of 40%. As far as the crossbridges are concerned, the Activation Heat is pure ‘overhead’.
An apparent paradox
Given the amount of information already gleaned from Figure 11, it is easy to overlook an apparent paradox. There is no difference of performance between trabeculae originating from diabetic hearts and controls. The dependencies on afterload of work output, heat evolution, total energy expenditure (work+heat), contractile efficiency and cross-bridge efficiency are indistinguishable between the two cohorts. At face value, these results contradict those observed in working whole-hearts. How can the peak twitch force of STZ trabeculae and their controls not differ in the ‘one-dimensional’ case of isolated trabeculae (Figure 11), when the peak afterload achievable by the ‘three dimensional’ whole-heart (Figure 7) is so severely compromised by diabetes? How is this contradiction to be reconciled?
The literature is replete with reports showing a constellation of consequences for the diabetic heart: prolongation of the duration of the action potential [100-104], thereby diminishing the period of electrical diastole, slowing the relaxation phase of the Ca2+ transient [79,80,105,106], thereby prolonging the ‘activated state’, and reduced activity of the actomyosin ATPase [107-112], consistent with a switch from the a (fast twitch) to the ß (slow twitch) isoform of the actomyosin ATPase [107,109,113-115]. Each of these phenomena contributes to prolongation of twitch duration and, effectively, an abbreviation of the period of mechanical diastole. Thus, at even physiological heart rates (4-5 Hz for the rat heart), the diastolic heart has less time to fill. Hence, its diastolic volume is reduced and, since the sarcomeres will now be operating at shorter length, peak systolic pressure development is reduced. By contrast, when an isolated trabecula undergoes a force-length loop, it is forcibly returned to Lo (optimal sarcomere length for developing force) during the diastolic interval. Realisation of this fundamental difference of diastolic behaviour allows reconciliation of the apparent contradiction. Diabetes does NOT reduce the ability of the tissues of the heart to develop force or pressure. The abbreviation of the diastolic period is the culprit. This insight could not have been gleaned had we not had experimental access to both whole-heart and isolated tissue preparations.
The Paradoxical Results between Mitochondrial (Input) Efficiency and Mechanical (Output) Efficiency
Despite resolving a contradiction between the behaviours of the intact heart and its tissues, we are nevertheless left with something of a contradiction. The efficiency of the mitochondria from diabetic hearts is about 15% lower than that of healthy hearts (Figure 2). But the peak efficiency of either intact diabetic hearts (Figure 7) or their trabeculae (Figure 11) is not. We can offer no definitive explanation but, given the resolution of the apparent contradiction outlined in 5. (vi), above, we are tempted to wonder if some equivalent phenomenon awaits for the measurement of mitochondrial efficiency. Is there some combination of concentrations of metabolic substrates that would allow submaximal rates of oxidative phosphorylation to yield equivalent P:O ratios between diabetic and non-diabetic hearts? As is invariably the case, more research is needed.
Summary and Conclusions
Provided with oxygen and metabolic substrates, the mitochondria, produce ATP. The efficiency of this oxidative phosphorylation (input) process can be quantified. The ATP produced can then be used to drive ion pumps in the sarcolemma and sarcoplasmic reticulum, and the relative sliding motion of the actin and myosin filaments. The efficiency of this overall ‘activation-contraction’ (output) process can also be measured–in either the isolated whole-heart or in samples of tissue dissected from it. We (and many others) observe that mitochondrial efficiency is depressed in the diabetic heart.
We find no difference in the peak output efficiency (i.e., the ratio of work to oxygen consumption) between STZ-diabetic and healthy hearts, although the mechanical output of the heart is reduced. This finding underscores the importance of examining cardiac mechano-energetics across a wide range of afterloads.
Measurement of the output efficiency (i.e., the ratio of work to (work plus heat)) of isolated trabeculae likewise reveals no difference between diabetic and healthy tissue. But, in striking contrast with the inability of the isolated whole heart to develop high left-ventricular pressures in vitro, there is no reduction in the ability of isolated trabeculae to produce maximal force. Resolution of this apparent conflict provides new insight into the mechanical behaviour of the diabetic heart in vivo. Dysfunction observed in the diabetic heart cannot be wholly attributed to a reduction of force development by myocardial tissues per se (at least at this time-point of diabetic insult) but, rather, involves complications at the whole organ level-most likely related to diminished chamber compliance. This result underscores the advantage of studying isolated organs and their tissues organs.
These investigations provide important clarification for the field of diabetic cardiac pathology by elucidating the effects of diabetes on cardiac efficiency and reconciling inconsistent findings in the literature. It is now clear that previous reports of a reduction in cardiac efficiency in diabetes can be explained by the fact that investigators have performed measurements at only a single afterload. By evaluating the effect of diabetes on cardiac efficiency over a range of afterloads, we now see that peak efficiency is not affected by diabetes, but the capacity of the diabetic heart to pump at high afterloads is diminished. Our experimental protocol (the use of crystalloid perfusates) may underestimate this effect with respect to blood-perfusion, since diabetes is known to increase the viscoclity of serum [116]. These findings have important implications for understanding the mechanisms of cardiac dysfunction in diabetes and support the contention that targeting treatment interventions aimed at improving the efficiency of the diabetic heart may not be an effective approach but strategies for attenuating the shortened diastolic period should be pursued with vigour.
This study was supported by grants from the Health Research Council of New Zealand, the Royal Society of New Zealand, the National Heart Foundation of New Zealand, the Maurice and Phyllis Paykel Trust, and the FRDF funds of the Faculty of Engineering and Auckland Bioengineering Institute, and the Virtual Physiological Rat Centre funded through NIH (P50-GM094503).

Figures at a glance

image   image   image   image   image   image
Figure 1   Figure 2   Figure 3   Figure 4   Figure 5   Figure 6


image   image   image   image   image
Figure 7   Figure 8   Figure 9   Figure 10   Figure 11
Select your language of interest to view the total content in your interested language
Post your comment

Share This Article

Article Usage

  • Total views: 13249
  • [From(publication date):
    July-2014 - Feb 17, 2020]
  • Breakdown by view type
  • HTML page views : 9377
  • PDF downloads : 3872