- Email: [email protected]
Models of myocardial ischemia
Drug Discovery Today: Disease Models
DRUG DISCOVERY
TODAY
DISEASE
MODELS
Vol. 3, No. 3 2006
Editors-in-Chief Jan Tornell – AstraZeneca, Sweden Andrew McCulloch – University of California, San Diego, USA
Cardiovascular diseases
Models of myocardial ischemia Kirsti Ytrehus Department of Medical Physiology, Institute of Medical Biology, Faculty of Medicine, University of Tromsø, Breivika 9037, Norway
The heart is highly dependent on aerobic metabolism. Given the widespread occurrence of coronary atherosclerosis in the human, myocardial ischemia is a huge
Section Editor: Steve Pogwizd – University of Illinois at Chicago, Chicago, USA
clinical problem. Myocardial ischemic is a molecular disease of cardiomyocytes, but also, as a consequence of reduced heart function, becomes a disease of the whole organism. Experimental models range from large animal models designed to mimic the human clinical situation, to gene modified mice, isolated hearts and cell cultures. Introduction Heart disease includes a large number of different diseases such as genetically derived defects in ion channels, congenital disease, heart valve disorders, cardiomyopathies, myocardial infarction and heart failure. In human medicine, it is assumed that the combat of ischemic heart disease and its consequences will be a worldwide challenge for years to come. Ischemic heart disease is listed among the degenerative diseases and diseases of an aging population. Some of the available Internet resources describing the impact of ischemic heart disease are listed in Links. To establish useful models for ischemic heart disease, we not only need models of disease mechanisms to understand how and why injury and malfunction occur (Box 1), but also need models for further development of interventions like surgery, revascularization and tissue engineering as well as good models for large scale drug testing. When investigating molecular mechanisms, mice models have been increasingly popular in the latest years owing to the possibility of studying genemodified animals and this area is rapidly expanding. Genetically engineered mice strains with single gene knockout and E-mail address: K. Ytrehus ([email protected]) 1740-6757/$ ß 2006 Elsevier Ltd. All rights reserved.
DOI: 10.1016/j.ddmod.2006.10.013
the development of heart-specific conditional knockout techniques are continually expanding our understanding of the gene products regulating susceptibility towards ischemia. Furthermore, the use of cell imaging and cell culture technique has given a new momentum to ischemia research. Techniques previously only used within cell proliferation and cancer research have proven useful. The ischemic heart and ischemic heart disease involves an organ with severe dysfunction and injury at the molecular level, but with great consequences for the integrated organism. This is the background for the diversity of experimental methods needed in this research field.
In vivo models Experimental models aimed at studying myocardial ischemia in the integrated organism are important in human medicine given the limited possibilities for well-controlled human studies. Whether the investigator wants to be close to the situation in human medicine or studies the influence of extra-cardiac factors in ischemic heart disease, the model of choice will be an in vivo model. This also includes development of new clinical methods, instruments and pharmacological agents. Most in vivo experimental models of myocardial ischemia have been established in the dog, pig, rabbit, rat or mouse. Although other species have also been used, the accumulated knowledge of the most often used laboratory animals in models of myocardial ischemia cannot be underscored. In vivo models are usually divided into chronic or acute models. In addition, myocardial ischemia can be studied in 263
Drug Discovery Today: Disease Models | Cardiovascular diseases
Glossary Hibernation: is defined as reversible reduction in contractile function due to reduced coronary perfusion. The original definition relating to chronic ischemia is extended to also include acute hibernation. Hibernation may represent an endogenous regulation resulting in reduction in function to reduce oxygen demand and preserve cellular integrity. With chronic hibernation normalization of ultrastructure and recovery of function may be delayed after normalization of blood flow. Because collateral flow development is sparse in the pig heart, this species is often used for hibernation studies. Ischemic postconditioning: is protection against ischemiareperfusion injury caused by one or multiple very brief, transient ischemic episode (30–60 s) followed by reperfusion (30–60 s) at early reperfusion after a prolonged ischemic insult. The protective mechanism that limits injury is rapidly activated and maintained for several hours, but it can only be activated close in time to the start of reperfusion. Postconditioning can be instituted in vivo as well as ex vivo in isolated perfused hearts. Ischemic preconditioning: is protection against ischemic injury (mainly infarction) caused by one or multiple, transient ischemic episode before a prolonged ischemic insult. Five to ten minutes transient ischemia is usually needed for preconditioning to be triggered. This results in acute (1–2 h) as well as delayed (24–72 h) protection. Preconditioning as phenomenon has proven to be very robust with respect to experimental models and species. Acute preconditioning can be instituted in the in vivo situation as well as in ex vivo isolated perfused hearts. Stunning: refers to reversible post-ischemic dysfunction in the setting of normalised coronary perfusion. The term came into use at the beginning of the 1980s and gained increasing interest after reperfusion became clinically possible. This also led to the understanding that stunning was not only a laboratory phenomenon but occurred also in humans subjected to either therapeutic or spontaneous reperfusion. The stunning phenomenon was extensively reviewed by Mareban and Bolli in 1999 [38] who also stressed the importance of the following statement: stunning is sometimes inappropriately applied to situations in which the persistence of contractile abnormalities in postischemic tissue is due to other causes (such as myocellular death, persistent ischemia or nonischemic injury). It is important to note that the stunning phenomenon is observed and can be investigated both in vivo and in vitro but that the extent and duration of the stunning are highly modeldependent. It is assumed that dysfunction is due to ROS production and decreased myofilamental calcium responsiveness. The ROS-calcium relationship at ischemia-reperfusion seems model dependent and it follows that it will be difficult to design universal treatment regimes.
conscious or anaesthetized state. The majority of studies are acute ischemia in the anaesthetized animal. This requires understanding of the impact of the anaesthetic agent in use and inclusion of sham operated animals and timed controls. In the clinical situation spontaneous or intentional ischemia is regional (coronary occlusion, angioplasty) or global (lethal arrhythmias or coronary bypass surgery). Correspondingly, experimental models are aimed at simulating either global ischemia or regional ischemia.
Dog models Historically, experiments with myocardial ischemia in the anaesthetized dog have led to basic understanding of heart function and overall hemodynamic changes during regional ischemia. Techniques for measurements of oxygen consump264
www.drugdiscoverytoday.com
Vol. 3, No. 3 2006
Box 1. Ischemic injury – a cell perspective Contractile failure – ATP and CP loss – Diastolic calcium overload – Cytosolic proton accumulation – Increase in cellular free phosphate Arrhythmias – Partial depolarization of membrane potential – Conduction delay – Shortening of the action potential Cardiomyocyte cell death (necrosis, apoptosis, oncosis) – Mitochondrial permeability transition – Cytosolic calcium overload – Caspases – Hypercontraction, sarcolemma rupture
tion and substrate metabolism under ischemia combined with estimates of heart work were originally established in the in situ canine heart. Regional ischemia and infarct models were developed for testing potential cardioprotective compounds. An example demonstrating the use of established knowledge about infarction and tissue lipid metabolism in the dog heart combined with newly raised questions is the use of specific cytochrome P-450 (CYP) antagonists for cardioprotection [1]. The role of arachidonic acid metabolites in myocardial ischemia has been questioned throughout the years, more recently in conjunction with the clinical use of cyclooxygenase COX II blockers and the increasing understanding of the diversity of the various CYP products. Nithipatikom et al. 2006 [1] used regional ischemia in anaesthetized dogs to investigate the role of cytochrome P450-hydroxylases and 20-HETE in an infarct model. A reduced level of 20-HETE was associated with reduction in infarct size. COX and lipoxygenase enzymes metabolize arachidonic acid (AA) to 5-, 12- and 15-hydroxyeicosatetraenoic acid (HETE), prostaglandins, prostacyclin, thromboxane and leukotrienes. The third enzymatic pathway for the metabolism of AA is by cytochrome P-450 (CYP) to epoxyeicosatrienoic acids (EETs), dihydroxyeicosatetraenoic acids (DiHETEs) and 19- and 20-HETE, and also other HETEs and reactive oxygen species (ROS). Gottlieb and co-workers indicated in their studies that EETs are beneficial and 20-HETEs detrimental in ischemia. It is known that HETEs induce vasoconstriction [2]. Endothelium produced EET are also described as endothelial derived hyperpolarization factors (EDHF). A second model is a chronic model with a surgically implanted ameroid constrictor placed around the left coronary artery. This results in gradual narrowing of the artery lumen and corresponding gradual reduction in ejection fraction and other contractile parameters of the affected myocardium over a few weeks. The dog heart has a great potential for coronary collateral vessel growth and therefore this model is convenient for studying the regulation of clinically
Vol. 3, No. 3 2006
important physiological adaptive angiogenesis. Therapeutic intervention like supply of endothelial stem cell [3] or angiogenic factors like the cytokine vascular endothelial growth factor (VEGF) and the basic fibroblast growth factor have also been successfully tested in this situation [4]. Thirdly, Lyseggen et al. [5] used an open chest dog heart model to document echocardiographic indices of potential use during evaluation of reperfused myocardium. The study aim was to find means to evaluate post-ischemic viability and therefore whether reperfusion was successful. Measurements based on echocardiography, which can be used in a clinical setting were compared with sonomicrometry, intraventricular pressure measurements, tissue staining for necrosis and coronary perfusion measurements using radionucleotid microspheres for microvascular perfusion. They tested if area of detectable contractile activity as well as dyskinetic or akinetic areas with surviving tissue could be classified as viable with their specific echo measurement. The study postulated that the passive dynamic properties of non-necrotic but akinetic tissue could be distinguished from necrotic tissue owing to the necrotic oedema by echo technique, and in the experimental model they confirmed this was the case.
Porcine models Mainly there are two categories of models of myocardial ischemia in the pig heart. The first one is related to clinical cardiac surgery, advanced instrumentation and a need for mimicking the situation in human surgery. The second one includes models of acute or chronic regional ischemia like models of acute ISCHEMIC PRECONDITIONING or chronic hibernating myocardium (Glossary). Acute ischemic preconditioning is cardioprotection induced by short lasting ischemic episodes before more severe ischemic injury, for example 5 min ischemia followed by 5 min reperfusion before ischemic injury. Relation between heart work and oxygen consumption in the ischemic and postischemic state is important when examining new therapeutic interventions. One aim of new therapeutics would be to increase heart work without increasing oxygen consumption out of proportions. In the setting of hypothermic cardioplegia, it was shown that addition of nicorandil, a potassium K-ATP channel opener and NO-donor, tended to improve mechanoenergetic efficiency when compared to a traditional cardioplegic potassium chloride solution [6]. Recent in vivo studies by Vinten-Johansen and co-workers in the dog heart have put the focus at the early phase of reperfusion [7,8] (Glossary). Transient and repetitive ischemia of 30-s duration applied during early reperfusion (so called postconditioning) has been reported to result in a significant reduction of reperfusion injury both in vivo and in vitro. Postconditioning can be applied clinically in conjunction with therapeutic reperfusion in contrast to clinical use of ischemic preconditioning which would require treatment before the disease is evident. Although promising due
Drug Discovery Today: Disease Models | Cardiovascular diseases
to potential clinical use, more research is needed to clarify under which circumstances postconditioning occurs and what the mechanisms behind the phenomenon are. In an open chest pig heart study of myocardial infarction by 30 min regional ischemia, classical ischemic preconditioning was demonstrated, but the postconditioning protocol failed to protect [9]. This study combined tissue sampling for western blot detection of cell signalling activation of key kinases in the cell survival pathways (Akt, p70s kinase, ERK), these signalling proteins were comparably affected in both the preconditioned and the post-conditioned group whereas only the preconditioned group had significant reduction in infarct size.
Rabbit models In vivo models of myocardial ischemia in the rabbit can be acute models as well as chronic models. The reader is referred to papers by Downey and co-workers describing the use of the rabbit heart for in vivo studies of regional myocardial ischemia, ischemic preconditioning, post-conditioning and pharmacological cardio-protection [10–13] (Glossary). One advantage of in vivo rabbit models is the possibility to transfer results from the in vitro isolated perfused heart into the in vivo setting without change of animal species. This advantage is also found in rat and mice models. Another advantage is that through special feeding regimes atherosclerotic conditions can be induced in the rabbit.
Rat The chronic in vivo rat model of regional ischemia and infarction has been used for more than three decades. When used without reperfusion, this model is mostly suited for postinfarction remodelling and failure, and for the investigation of scar tissue and inflammation related to tissue repair [14]. For the study of infarct size limitation, reperfusion is obligate because the rat heart (like the rabbit and mice heart) has no collaterals; occlusion of blood flow to an area eventually leads to loss of myocardial cells in that area. With standardized reperfusion and quantification of infarct size, cell death delay protocols for cardio-protection are easy to establish in vivo in a rat model [12]. Possibilities for hemodynamic measurements are limited in the rat heart compared to the complex instrumentation that is possible in larger animals and also during human heart-surgery. However, a major improvement has come with the newest noninvasive imaging modalities, namely echo Doppler technique, NMR imaging and spectroscopy, and PET, which make possible the continuous in vivo monitoring of metabolic activity and also allows the study of the distribution of specific intra and/or extracellular tracers.
Mice Several experimental models developed for rats have now been adjusted for use in mice. This involves miniaturization www.drugdiscoverytoday.com
265
Drug Discovery Today: Disease Models | Cardiovascular diseases
of equipment and, in most cases, optical aid to identify structures. The anatomical location of the left coronary artery is constant between individuals in mice (as well as in rats). Established technique for intubation and artificial ventilation makes chronic experiments with regional ischemia – reperfusion possible in mice because the animal can recover after surgery. Within the field of transgenic mice research, Internet resources have also been developed by commercial animal supply companies, universities and other research institutions – one example being the one provided by The University of Michigan: Transgenic animal web (see Links). An interesting finding with respect to myocardial ischemia has been the role of erythropoietin in non-hematopoetic tissue. Tada et al. [15] were able to show that endogenous nonhematopoetic erythropoietin receptors modulate ischemic injury. Selective knock out of these receptors in mice resulted in worsening of ischemia outcome. Correspondingly, studies have shown that the addition of erythropoietin protects [16] and combined with the protection afforded by ischemic preconditioning has a cumulative protective effect through inhibition of GSK3beta activity. Across species and experimental models, adenosine released from ischemic tissue plays a very central role in modulating the ischemic process in the heart. Adenosine binds to receptors on cells in the vessel wall, cardiomyocytes, as well as on other cell types in the heart and blood stream and is also a precursor in adenine nucleotide synthesis. Adenosine given in short episodes before myocardial ischemia mimics the effect of ischemic preconditioning (pharmacological preconditioning). This effect is based on stimulation of the A1 or A3 adenosine receptor in the myocardium. However, continuous use of adenosine receptor stimulation by adenosine or receptor subtype specific analogues for pharmacological preconditioning fail to maintain the heart in a protected state thus limiting potential clinical use [13]. The role of adenosine receptor agonism as a cardioprotective principle was confirmed in a study using a transgenic mice model of moderate adenosine A1 receptor overexpression [17]. When subjected to a standardized 45-min regional ischemia and 24-h reperfusion, infarcts were significantly smaller in hearts from transgenic mice compared to wild type. Other genes have also been reported to induce a cardioprotective phenotype when over-expressed in rat or mice, for example heat shock protein 70, and the antioxidant enzymes manganese superoxide dismutase and glutathione peroxidase [18–20]. It is important to distinguish changes directly induced by selective gene modification from those secondary adaptive changes. When results on modulation of ischemia by use of specific inhibitors or activators in wild type animals are confirmed in dedicated transgenic animal models this provide a strong case for a mechanistic connection with the gene product in question. 266
www.drugdiscoverytoday.com
Vol. 3, No. 3 2006
Ex vivo and in vitro models In the present text we will include ex vivo models under the in vitro heading, referring to organ (heart), tissue (muscle) and primary cells isolated from animals.
Isolated perfused hearts With a steady increase over the past 20–30 years, isolated perfused hearts are now extensively used in the study of myocardial ischemia (Fig. 1). The simplicity of the technique and the relatively low cost, usually permits larger number of experiments and more hypotheses to be tested compared with the more complex in vivo models. The isolated perfused rat heart is used for biochemical tissue analyses, various
Figure 1. (a) Isolated perfused hearts have been instrumental in understanding the basic principles of cardio-protection, in understanding cell signalling and last but not least in understanding time dependent metabolic changes in the heart muscle during limited oxygen supply. The Langendorff perfusion technique dates back to more than 100 years, and is still used extensively in basic heart research worldwide. Although many modifications of this technique are possible, basically hearts are mounted by aorta on a perfusion apparatus and buffer delivered retrogradely in the aorta line and into the coronaries above the closed aortic valve by constant flow or constant pressure. Global or graded ischemia is produced by clamping the supply line. Hypoxic perfusion is also used. The working heart model was developed and further modified by the work of Morgan, Neely and others almost 50 years ago. When the heart is perfused in working mode, buffer is delivered into the left atrium, and is pumped trough the aortic valve by the heart itself, corresponding to the in vivo situation. The heart is responsible for its own oxygen and substrate supply in the working mode and spontaneous recovery of contractile function after global ischemia is often lost after short time ischemia. With the Langendorff perfusion technique more prolonged ischemia can be tested without complete loss of measurable contractile function. (b) Regional ischemia in the isolated perfused heart is obtained by ligation of the left coronary artery or one of its main braches depending on species. In the rat and mice heart, the anatomy of the artery is constant. In the rabbit (this figure), the artery divides into several branches close to the aortic root. Normally perfused myocardium, ischemic risk zone and infarct are indicated by difference in colours (normally perfused myocardium: red and pink; ischemic risk zone: violet; infarct: blue).
Vol. 3, No. 3 2006
Drug Discovery Today: Disease Models | Cardiovascular diseases
died. In hearts from male db/db mice reduced ischemic tolerance developed in parallel with metabolic changes (increase in fatty acid oxidation and decrease in glucose oxidation). Interestingly, the use of peroxisome proliferator-activated receptor-alpha (PPARalfa) treatment did not affect sensitivity to ischemia-reperfusion, even though carbohydrate oxidation was increased and palmitate oxidation was decreased [24]. Age and gender dependent changes vary considerably between species, animal strains, experimental models and study endpoint used [25]. The impact of age and gender upon response to 20 min of ischemia has been investigated in mice (C57/B16 strain) using isolated perfused heart. The most significant age dependent change in post-ischemic recovery of heart function was reduced diastolic function in line with the reduction in compliance reported with aging in human medicine [25]. Figure 2. Mitochondria and myocardial ischemia. Cell signalling in acute myocardial ischemia – death versus survival signals. Mitochondria play a central role and promote cell death as well as cell survival: cell death by disrupting electron transport and energy supply, and by releasing and activating proteins that mediate apoptosis; cell survival by ability to change into a stress tolerant state in response to metabolic challenges, extra and intracellular stimuli. Cell survival signalling and death signalling is controlled by several protein kinases like MAPkinases, protein kinase Cs and glycogen synthetase kinase [27,40]. MPT, mitochondrial membrane permeability transition; ROS, reactive oxygen species.
functional measurements and evaluation of infarction size in parallel with time-limited use of pharmacological probes (non recirculating perfusion solution permits washout) (Fig. 1). We were able to show that transient pre-ischemic treatment with pharmacological concentrations of 17betaestradiol followed by 5–15 min of washout before prolonged regional ischemia induced cardio-protection, and that this was related to increase in ROS levels before regional ischemia [21], consistent with a redox signalling function of ROS resulting in activation of cell survival pathways (Fig. 2). Using groups of isolated rat hearts perfused under constant flow or under constant pressure, Penna et al. [22] showed that ISCHEMIC POSTCONDITIONING was cGMP dependent and that the treatment resulted in more robust protection when hearts are perfused at constant coronary flow. With the use of mice hearts and development of transgenic animals, the use of ex vivo perfusion techniques will remain high. The technique was reviewed by Sutherland et al. 2003 [23], and the reader is referred to this review and the corresponding list of references for information. The leptin receptor deficient db/db mouse is a well-described animal model for age dependent diabetic cardiomyopathy [24]. Using mouse isolated perfused heart, the ischemia tolerance in male hearts from a well-described natural genetic (db/db mouse) model of age dependent diabetic cardiomyopathy was stu-
Multi-cellular preparations Multi-cellular preparations are mostly papillary muscle or atria trabecula muscle harvested from different species and superfused with buffer. Suitable endpoints for the study of ischemia in these preparations are contractile force and enzyme release (troponin T) during and after simulated ischemia (usually hypoxia or anoxia). The technique has been adopted for use in human atria trabecula [26]. With a micropipette technique, single cell action potentials can be recorded from the surface of superfused muscle preparations and electrophysiological response to ischemia can be investigated. The role of ischemia induced opening of ATP dependent sarcolemmal potassium channels and action potential shortening is easily demonstrated (Box 1).
Cardiomyocytes Cardiomyocytes can be isolated acutely from hearts of all species and maintained in suspension for hours. Short time culture is also possible. Manipulation of the incubation buffer, anoxia or hypoxia, or ‘pelleting’ and sealing under paraffin can be used to simulate ischemia. By applying confocal laser scanning microscopy-based cell imaging technique to isolated adult rat cardiomyocytes in suspension mitochondrial function and integrity can be studied in intact cells. In this model, the importance of reactive oxygen species (ROS) related loss of mitochondrial membrane potential at reperfusion was demonstrated [27]. Opening of the mitochondria permeability transition pore (MPT) lead to rapid loss of the potential, and it was demonstrated that protection by preconditioning could protect against such pore opening [27]. Figure. 2 illustrates the principle role of ROS and MPT in cell death and cell survival signal transduction.
Cell culture techniques Cell culture techniques have considerably rationalized cell studies; cultures most often used are either based on cells www.drugdiscoverytoday.com
267
Drug Discovery Today: Disease Models | Cardiovascular diseases
isolated from neonatal or embryonic hearts of rats and mice [28,29] or obtained commercially as immortalized cell lines. Neonatal mice heart cell cultures can be harvested from genetically engineered mice hearts; transfection techniques as well as silencing RNA technique can be used in cell cultures. HL-1 cells are immortalized cells derived from an atrial tumour (myoma) in female mice. The cell line has been developed and extensively characterized by Claycomb and coworkers [29]. Compared to the use of primary isolation of ventricular myocytes, the advantages of having a good immortalized cell culture model for use in ischemia studies involve the following: (1) no animal use and care, (2) no costly and lengthy preparation time before any experiments can take place, (3) no variable and limited yield for highthroughput approaches and (4) no problems with heterogeneous cell population. The HL-1 cells and the following generations of these cells have been used successfully for ischemia related experiments; overall results correlate well with results obtained, for example, isolated hearts. One example of the use of this model [30,31] is the study of pharmacological preconditioning with opioid agonists and the detailed delineation of down stream survival cellular signalling events. Using these cells as a supplement to experiments in isolated perfused hearts subjected to ischemia – reperfusion, it has been possible to show that opioid-induced cardioprotection is dependent on the JAK/STAT pathway: JAK2 as a mediator of STAT3, Akt, and GSK-3beta. H9c2 is an embryonal rat-heart ventricle derived cell line. These cells have been used to characterize ischemia-induced apoptosis; ischemia induced heat shock protein expression changes as well as the detailed role of p38MAP kinase in cell injury due to simulated ischemia [32,33]. Myotubes or human myoblasts (Girardi cells) have also been used to study ischemic preconditioning. Cells can be obtained from American Type Culture Collection or European Collection of Cell Cultures.
In silico models There is limited tradition for in silico models in ischemic heart research. One exception is the electrophysiological modelling of the consequences of changes in ion balance with ischemia. Important input in these models includes potassium loss to the extra-cellular space, intracellular proton accumulation and opening of ATP dependent potassium channels, cellular electrical uncoupling and delayed conduction, diastolic calcium overload and increase in cellular sodium [34]. A paper by Keener [35] describes fibrillation of the heart as a consequence of spatial heterogeneity during regional ischemia. Using general arguments and numerical simulations with generic models of excitable media, they conclude that a spatial region with an elevated resting potential surrounded by a spatial region wherein action potentials are shortened can drive a break up instability, leading to the rapid initiation of a fibrillatory state. 268
www.drugdiscoverytoday.com
Vol. 3, No. 3 2006
New interesting possibilities will emerge with the ongoing advancements of proteomics and of knowledge on regulatory cellular enzymes and biochemistry. One example related to myocardial ischemia is the modelling of human cardiac mitochondrial metabolic network [36]. In this study, four different conditions were investigated: normal condition, ischemia (25% reduction in oxygen supply), diabetes and diet restriction (low fat, high glucose). Interestingly, their network-balancing model indicated that, while, as expected, oxidation of fatty acids is severely limited with the reduction in oxygen supply, the flux of fatty acids into phospholipids increased in this situation. Therapeutic interventions proposed to increase ATP availability under ischemia such as glucose – potassium – insulin treatment (GIK), keton body supply and increase in pyrovate dehydrogenase activity were also tested. Surprisingly, the theoretical modelling revealed no increased ATP availability, indicating a need for more clarifying experiments.
Model comparison In vitro and ex vivo models give information about intrinsic myocardial and vascular (in the case of isolated hearts) factors in the absence of confounding factors (Table 1). In in vivo models these factors should be part of the investigation. The autonomic nerve system, the central nervous system, bloodborn elements like hormones, cytokines and blood born cells are studied as well as the interaction between the heart and other organs. There has been a tendency to perform experiments in animals of only one gender (especially when using rats or mice), and, for various reasons, male sex has been preferred in many laboratories. Recently, research has been focusing on the question of a possible difference in ischemia-reperfusion tolerance between genders and especially on the molecular mechanisms of observed difference. Interestingly Wang et al. 2006 were able to show that recovery of contractile function after 20 min of acute ischemia in the isolated mice heart was improved by the presence of estrogen receptor alfa [37]. They subsequently showed that this improved recovery was associated more with ERK-phosphorylation, than with JNK- phosphorylation and proapoptotic enzyme caspase 3. Although, in the study, there was a gender difference in some inflammatory markers in the examined hearts (TNF-a, IL-1b and IL6), this did not correlate with the improved recovery of contractile function. This way the use of both genders in animal experimental studies can also serve as an experimental tool or a natural model for understanding the mechanisms of ischemia-reperfusion injury. The selection of model endpoints depends on the aim of a study and the hypothesis under investigation. This is illustrated when investigating the ischemic syndromes STUNNING, HIBERNATION and ischemic preconditioning [38,39] (glossary).
Vol. 3, No. 3 2006
Drug Discovery Today: Disease Models | Cardiovascular diseases
Table 1. Comparison summary table In vivo
In vitro
In silico
Close to clinical situation
Mechanisms intrinsic to the heart
Transgenic animals (mice) Permits investigation of influence of extracardiac factors
Well-controlled models Large number of experiments possible
Permit a large number of scenarios to be tested
Commercial cell based molecular genetics techniques can be used
Save research animals and laboratory costs
Often harbour factors not related to the heart which can not be controlled Changes induced by surgery, instrumentation and anaesthetics modify outcome
Difficult to transfer results to the situation in human medicine Species limitations
Severely restricted by current knowledge (model input)
Heart surgery models Coronary collateral growth Hibernating myocardium
Cell signalling in ischemic myocardium Mitochondrial function Heart metabolism
Preclinical testing Cardiovascular hemodynamics
Ischemic and hypoxic cell death Combined in vivo–ex vivo examination
Access
Literature
Literature
Literature
Patents
n/a
n/a
n/a
References
[1,3–6,8–13]
[14–26]
[27,28]
Pros
Cons
Best use of model
Adult human cardiomyocyte cultures are not available
Studies combining chronic in vivo treatment or interventions followed by ex vivo/in vitro experiments represent an important strategy when investigating myocardial ischemia. Experimental animals can be treated in vivo with pharmacological agents with proposed beneficial effect, with different diets or subjected to other interventions for days or weeks before heart harvesting. Thereafter, the heart or cells isolated from this heart can be subjected to ischemia and tolerance can be tested. This creates a time window for gene expression changes and structural changes in the heart, and when followed by ex vivo examination the consequences of the treatment intrinsic to the heart can be examined specifically without interference from other organ systems via blood or autonomic nerves. And last but not least, the animal ethical perspectives always have to be taken into consideration as well as national regulation (i.e. European Convention for the Protection of Vertebrate Animals used for Experimental and other Scientific Purposes CETS No.: 123) (Links).
Translation to the human situation The role of age, gender, coronary anatomy (collateral flow), innate and adopted immunity and inflammation, diet and coexisting diseases must not be underestimated when translating results to humans. Myocardial ischemia is a disease of the aging heart in humans, but many animal experiments are conducted in younger animals. With respect to the human heart, an important example is the anatomy of the coronary circulation. The healthy human heart has few if any collaterals. However, a (partly) blocked vessel may lead to development of a collateral circulation: small capillary-like branches of the artery develop into larger vessels over time
Electrophysiology (well established) Metabolomics, ATP flux controlling network (new)
in response to tissue ischemia. Thus, treatment promoting angiogenesis and neovascularisation used successfully animal hearts might be of little value for an elderly patient with chronic heart disease who has already experienced a strong endogenous stimulation of collateral vessel growth. Cardioprotection by ischemic preconditioning is an example where a huge overlap of results obtained in different species makes it reasonable to transfer results to the human heart as research has demonstrated. But with new experimentally developed cardioprotective treatments like postconditioning the question about validation across different species have to be examined again. The human genome can now be compared with the genome of other species (especially laboratory animals like mice and rats). This makes transfer to the human situation possible when it comes to research addressing a single gene product. Post-translational modification of gene products is important for the final function of many proteins and is often dependent on many different genes. In this area knowledge is still limited when it comes to species differences. With complex disease processes like chronic ischemic heart disease transfer from animal models to the human situation is often difficult.
Conclusion A broad spectrum of different experimental models is needed for the study of myocardial ischemia ranging from advanced instrumentation of large animals to imaging of mitochondria in cardiomyocytes. We need to be able to distinguish between events intrinsic to the heart on one side, and the integrated whole body response to myocardial ischemia on the other side. In addition, within the heart tissue ischemia affects both cardiomyocytes and the non-myocyte components in the www.drugdiscoverytoday.com
269
Drug Discovery Today: Disease Models | Cardiovascular diseases
Vol. 3, No. 3 2006
10
Links European Society of Cardiology, Euro Heart Survey Publications web page include information about cardiovascular diseases in Europa in 2006: http://www.escardio.org/knowledge/ehs/ publications/ World Health Organisation web page for health topics and cardiovascular disease provide links to information about global burden and prevention: http://www.who.int/topics/ cardiovascular_diseases/en/ The web page from Karolinska Institute, Stockholm, Sweden is an example of an university based information pool for cardiovascular diseases and disorders with multiple links to other pages: http://www.mic.ki.se/Diseases/C14.html Information about the European Convention for the Protection of Vertebrate Animals used for Experimental and other Scientific Purposes CETS No.: 123) is found at: http://www.coe.int/T/E/ Legal_affairs/Legal_co-operation/Biological_safety,_use_of_animals/ Laboratory_animals/ Information about available cell cultures can be found at European Collection of Cell Cultures: http://www.ecacc.org.uk/, and American Type Culture Collection: http://www.biotech.ist.unige.it/ cldb/descat1.html The University of Michigan: Transgenic animal web: http://www.med.umich.edu/tamc/links.html
11
12
13
14
15
16
17
18
heart. Given the accumulation of knowledge about the ischemic process it is time to ask for transfer of this knowledge into clinical use in human medicine. Advancement in imaging techniques, further development of genetically modified animal models and functional genomics, combined with the concepts of metabolomics might provide the necessary tools to answer the request.
References 1 Nithipatikom, K. et al. (2006) Effects of selective inhibition of cytochrome P-450 v-hydroxylases and ischemic preconditioning in myocardial protection. Am. J. Physiol. 290, H500–H505 2 Granville, D.J. et al. (2004) Reduction of ischemia and reperfusion-induced myocardial damage by cytochrome P450 inhibitors. PNAS 101, 1321–1326 3 Silva, G.V. et al. (2005) Mesenchymal Stem Cells Differentiate into an Endothelial Phenotype, Enhance Vascular Density, and Improve Heart Function in a Canine Chronic Ischemia Model. Circulation 111, 150–156 4 Liu, Y. et al. (2006) Effects of basic fibroblast growth factor microspheres on angiogenesis in ischemic myocardium and cardiac function: analysis with dobutamine cardiovascular magnetic, resonance tagging. Eur. J. Cardiotorac. Surg. 30, 103–108 5 Lyseggen, E. et al. (2005) Myocardial train analysis in acute coronary occlusion. a tool to assess myocardial viability and reperfusion. Circulation 112, 3901–3910 6 Steensrud, T. et al. (2004) Replacing potassium with nicorandil in cold St. Thomas’ hospital cardioplegia improves preservation of energetics and function in pig hearts. Ann. Thorac. Surg. 77, 1391–1397 7 Tsang, A. et al. (2005) Myocardial postconditioning: reperfusion injury revisited. Am. J. Physiol. 289, H2–H7 8 Zhao, Z.Q. et al. (2003) Inhibition of myocardial injury by ischemic postconditioning during reperfusion: comparison with ischemic preconditioning. Am. J. Physiol. 285, H579–H588 9 Schwartz, L.M. et al. (2006) Ischemic postconditioning during reperfusion activates Akt and ERK without protecting against lethal myocardial ischemia-reperfusion injury in pigs. Am. J. Physiol. 290, H1011–H1018
270
www.drugdiscoverytoday.com
19
20
21
22
23 24
25 26
27
28
29
30
31
Philipp, S. et al. (2006) Postconditioning protects rabbit hearts through a protein kinase C-adenosine A2b receptor cascade. Card. Res. 70, 308–314 Krenz, M. et al. (2001) Acute ethanol exposure fails to elicit preconditioning-like protection in in situ rabbit hearts because of its continued presence during ischemia. J. Am. Coll. Cardiol. 37, 601–607 Ytrehus, K. et al. (1994) Rat and rabbit heart infarction: effect of anesthesia, perfusate, risk zone, and methods of infarct sizing. Am. J. Physiol. 267, H2383–H2390 Tsuchida, A. et al. (1994) The anti-infarct effect of an adenosine A1 selective agonist is diminished after prolonged infusion as is the cardioprotective effect of ischemic preconditioning in rabbit heart. J. Mol. Cell Cardiol. 26, 303–311 Qvigstad, E. et al. (2005) Dual Serotonergic Regulation of Ventricular Contractile Force Through 5-HT2A and 5-HT 4 Receptors Induced in the Acute Failing Heart. Circ. Res. 97, 268–276 Tada, H. et al. (2006) Endogenous erythropoietin system in nonhematopoietic lineage cells plays a protective role in myocardial ischemia/ reperfusion. Card. Res. 71, 466–477 Nishihara, M. et al. (2006) Erythropoietin affords additional cardioprotection to preconditioned hearts by enhanced phosphorylation of glycogen synthase kinase-3b. Am. J. Physiol. 291, H748–H755 Yang, Z. et al. (2002) Cardiac overexpression of A1-adenosine receptor protects intact mice against myocardial infarction. Am. J. Physiol. 282, H949–H955 Shiomi, T. et al. (2004) Overexpression of gluthatione peroxidise prevents left ventricular remodelling and failure after myocardial infarction in mice. Circulation 109, 544–549 Jayakumar, J. et al. (2001) Heart shock protein 70 gene transfection protects mitochondrial and ventricular function against ischemiareperfusion injury. Circulation 104, 303–307 Tanaka, M. et al. (2004) Overexpression of human copper/zinc superoxide dismutase (SOD1) suppress ischemia-reperfusion injury and subsequent development of graft coronary artery disease in murine cardiac grafts. Circulation 110, 200–206 Sovershaev, M.A. et al. (2006) Preconditioning by 17{beta}-estradiol in isolated rat heart depends on PI3-K/PKB pathwat, PKC and ROS. Am. J. Physiol. 291, H1554–H1562 Penna, C. et al. (2006) Post-conditioning reduces infarct size in the isolated rat heart: Role of coronary flow and pressure and the nitric oxide/cGMP pathway. Basic Res. Cardiol. 101, 168–179 Sutherland, F.J. et al. (2003) Mouse isolated perfused heart: Characteristics and cautions. Clin. Exp. Pharmacol. Physiol. 30, 867–878 Aasum, E. et al. (2003) Age-Dependent changes in metabolism, contractile function, and ischemic sensitivity in hearts from db/db mice. Diabetes 52, 434–441 Willems, L. et al. (2004) Age-related changes in ischemic tolerance in male and female mouse hearts. J. Mol. Cell Cardiol. 38, 245–256 Shanmuganathan, S. et al. (2005) Mitochondrial permeability transition pore as a target for cardioprotection in the human heart. Am. J. Physiol. 289, H237–H242 Juhaszova, M. et al. (2004) Glycogen synthase kinase-3b mediates convergence of protection signaling to inhibit the mitochondrial permeability transition pore. J. Clin. Invest. 113, 1535–1549 Okada, T. et al. (2005) Integrated pharmacological preconditioning and memory of cardioprotection: role of protein kinase C and phosphatidylinositol 3-kinase. Am. J. Physiol. 289, H761–H767 Claycomb, W.C. et al. (1998) HL-1 cells: A cardiac muscle cell line that contracts and retains phenotypic characteristics of the adult cardiomyocyte. Proc. Natl. Acad. Sci. U S A 95, 2979–2984 Gross, E.R. et al. (2006) The JAK/STAT pathway is essential for opioidinduced cardioprotection: JAK2 as a mediator of STAT3, Akt, and GSK-3b. Am. J. Physiol. 291, H827–H834 Seymour, E.M. et al. (2003) HL-1 myocytes exhibit PKC and KATP channel-dependent delta opioid preconditioning. J. Surg. Res. 114, 187–194
Vol. 3, No. 3 2006
32 33
34
35
Tantini, B. et al. (2006) Involvement of polyamines in apoptosis of cardiac myoblasts in a model of simulated ischemia. J. Mol. Cell Cardiol. 40, 775–782 Tanno, M. et al. (2003) Diverse mechanisms of myocardial p38 mitogenactivated protein kinase activation: evidence for MKK-independent activation by a TAB1-associated mechanism contributing to injury during myocardial ischemia. Circ. Res. 93, 254–261 Nygren, A. et al. (2006) Measurements of electrophysiological effects of components of acute ischemia in langendorff-perfused rat hearts using voltage-sensitive dye mapping. J. Cardiovasc. Electrophysiol. 17, S113–S123 Keener, J.P. (2003) Model for the onset of fibrillation following coronary artery occlusion. J. Cardiovasc. Electrophysiol. 14, 1225–1232
Drug Discovery Today: Disease Models | Cardiovascular diseases
36
37 38 39 40
Thiele, I. et al. (2005) Candidate metabolic network states in human mitochondria. impact of diabetes, ischemia, and diet. J. Biol. Chem. 280, 11683–11695 Wang, M. et al. (2006) Estrogen receptor-a mediates acute myocardial protection in females. Am. J. Physiol. 290, H2204–H2209 Bolli, R. et al. (1999) Molecular and cellular mechanisms of myocardial stunning. Physiol. Rev. 79, 609–634 Downey, J.M. et al. (2005) We think we see a pattern emerging here. Circ. 111, 120–121 Das, D.K. et al. (2005) Mitochondrial function in cardiomyocytes: target for cardioprotection. Curr. Opin. Anaestesiol. 18, 77–82
www.drugdiscoverytoday.com
271
DRUG DISCOVERY
TODAY
DISEASE
MODELS
Vol. 3, No. 3 2006
Editors-in-Chief Jan Tornell – AstraZeneca, Sweden Andrew McCulloch – University of California, San Diego, USA
Cardiovascular diseases
Models of myocardial ischemia Kirsti Ytrehus Department of Medical Physiology, Institute of Medical Biology, Faculty of Medicine, University of Tromsø, Breivika 9037, Norway
The heart is highly dependent on aerobic metabolism. Given the widespread occurrence of coronary atherosclerosis in the human, myocardial ischemia is a huge
Section Editor: Steve Pogwizd – University of Illinois at Chicago, Chicago, USA
clinical problem. Myocardial ischemic is a molecular disease of cardiomyocytes, but also, as a consequence of reduced heart function, becomes a disease of the whole organism. Experimental models range from large animal models designed to mimic the human clinical situation, to gene modified mice, isolated hearts and cell cultures. Introduction Heart disease includes a large number of different diseases such as genetically derived defects in ion channels, congenital disease, heart valve disorders, cardiomyopathies, myocardial infarction and heart failure. In human medicine, it is assumed that the combat of ischemic heart disease and its consequences will be a worldwide challenge for years to come. Ischemic heart disease is listed among the degenerative diseases and diseases of an aging population. Some of the available Internet resources describing the impact of ischemic heart disease are listed in Links. To establish useful models for ischemic heart disease, we not only need models of disease mechanisms to understand how and why injury and malfunction occur (Box 1), but also need models for further development of interventions like surgery, revascularization and tissue engineering as well as good models for large scale drug testing. When investigating molecular mechanisms, mice models have been increasingly popular in the latest years owing to the possibility of studying genemodified animals and this area is rapidly expanding. Genetically engineered mice strains with single gene knockout and E-mail address: K. Ytrehus ([email protected]) 1740-6757/$ ß 2006 Elsevier Ltd. All rights reserved.
DOI: 10.1016/j.ddmod.2006.10.013
the development of heart-specific conditional knockout techniques are continually expanding our understanding of the gene products regulating susceptibility towards ischemia. Furthermore, the use of cell imaging and cell culture technique has given a new momentum to ischemia research. Techniques previously only used within cell proliferation and cancer research have proven useful. The ischemic heart and ischemic heart disease involves an organ with severe dysfunction and injury at the molecular level, but with great consequences for the integrated organism. This is the background for the diversity of experimental methods needed in this research field.
In vivo models Experimental models aimed at studying myocardial ischemia in the integrated organism are important in human medicine given the limited possibilities for well-controlled human studies. Whether the investigator wants to be close to the situation in human medicine or studies the influence of extra-cardiac factors in ischemic heart disease, the model of choice will be an in vivo model. This also includes development of new clinical methods, instruments and pharmacological agents. Most in vivo experimental models of myocardial ischemia have been established in the dog, pig, rabbit, rat or mouse. Although other species have also been used, the accumulated knowledge of the most often used laboratory animals in models of myocardial ischemia cannot be underscored. In vivo models are usually divided into chronic or acute models. In addition, myocardial ischemia can be studied in 263
Drug Discovery Today: Disease Models | Cardiovascular diseases
Glossary Hibernation: is defined as reversible reduction in contractile function due to reduced coronary perfusion. The original definition relating to chronic ischemia is extended to also include acute hibernation. Hibernation may represent an endogenous regulation resulting in reduction in function to reduce oxygen demand and preserve cellular integrity. With chronic hibernation normalization of ultrastructure and recovery of function may be delayed after normalization of blood flow. Because collateral flow development is sparse in the pig heart, this species is often used for hibernation studies. Ischemic postconditioning: is protection against ischemiareperfusion injury caused by one or multiple very brief, transient ischemic episode (30–60 s) followed by reperfusion (30–60 s) at early reperfusion after a prolonged ischemic insult. The protective mechanism that limits injury is rapidly activated and maintained for several hours, but it can only be activated close in time to the start of reperfusion. Postconditioning can be instituted in vivo as well as ex vivo in isolated perfused hearts. Ischemic preconditioning: is protection against ischemic injury (mainly infarction) caused by one or multiple, transient ischemic episode before a prolonged ischemic insult. Five to ten minutes transient ischemia is usually needed for preconditioning to be triggered. This results in acute (1–2 h) as well as delayed (24–72 h) protection. Preconditioning as phenomenon has proven to be very robust with respect to experimental models and species. Acute preconditioning can be instituted in the in vivo situation as well as in ex vivo isolated perfused hearts. Stunning: refers to reversible post-ischemic dysfunction in the setting of normalised coronary perfusion. The term came into use at the beginning of the 1980s and gained increasing interest after reperfusion became clinically possible. This also led to the understanding that stunning was not only a laboratory phenomenon but occurred also in humans subjected to either therapeutic or spontaneous reperfusion. The stunning phenomenon was extensively reviewed by Mareban and Bolli in 1999 [38] who also stressed the importance of the following statement: stunning is sometimes inappropriately applied to situations in which the persistence of contractile abnormalities in postischemic tissue is due to other causes (such as myocellular death, persistent ischemia or nonischemic injury). It is important to note that the stunning phenomenon is observed and can be investigated both in vivo and in vitro but that the extent and duration of the stunning are highly modeldependent. It is assumed that dysfunction is due to ROS production and decreased myofilamental calcium responsiveness. The ROS-calcium relationship at ischemia-reperfusion seems model dependent and it follows that it will be difficult to design universal treatment regimes.
conscious or anaesthetized state. The majority of studies are acute ischemia in the anaesthetized animal. This requires understanding of the impact of the anaesthetic agent in use and inclusion of sham operated animals and timed controls. In the clinical situation spontaneous or intentional ischemia is regional (coronary occlusion, angioplasty) or global (lethal arrhythmias or coronary bypass surgery). Correspondingly, experimental models are aimed at simulating either global ischemia or regional ischemia.
Dog models Historically, experiments with myocardial ischemia in the anaesthetized dog have led to basic understanding of heart function and overall hemodynamic changes during regional ischemia. Techniques for measurements of oxygen consump264
www.drugdiscoverytoday.com
Vol. 3, No. 3 2006
Box 1. Ischemic injury – a cell perspective Contractile failure – ATP and CP loss – Diastolic calcium overload – Cytosolic proton accumulation – Increase in cellular free phosphate Arrhythmias – Partial depolarization of membrane potential – Conduction delay – Shortening of the action potential Cardiomyocyte cell death (necrosis, apoptosis, oncosis) – Mitochondrial permeability transition – Cytosolic calcium overload – Caspases – Hypercontraction, sarcolemma rupture
tion and substrate metabolism under ischemia combined with estimates of heart work were originally established in the in situ canine heart. Regional ischemia and infarct models were developed for testing potential cardioprotective compounds. An example demonstrating the use of established knowledge about infarction and tissue lipid metabolism in the dog heart combined with newly raised questions is the use of specific cytochrome P-450 (CYP) antagonists for cardioprotection [1]. The role of arachidonic acid metabolites in myocardial ischemia has been questioned throughout the years, more recently in conjunction with the clinical use of cyclooxygenase COX II blockers and the increasing understanding of the diversity of the various CYP products. Nithipatikom et al. 2006 [1] used regional ischemia in anaesthetized dogs to investigate the role of cytochrome P450-hydroxylases and 20-HETE in an infarct model. A reduced level of 20-HETE was associated with reduction in infarct size. COX and lipoxygenase enzymes metabolize arachidonic acid (AA) to 5-, 12- and 15-hydroxyeicosatetraenoic acid (HETE), prostaglandins, prostacyclin, thromboxane and leukotrienes. The third enzymatic pathway for the metabolism of AA is by cytochrome P-450 (CYP) to epoxyeicosatrienoic acids (EETs), dihydroxyeicosatetraenoic acids (DiHETEs) and 19- and 20-HETE, and also other HETEs and reactive oxygen species (ROS). Gottlieb and co-workers indicated in their studies that EETs are beneficial and 20-HETEs detrimental in ischemia. It is known that HETEs induce vasoconstriction [2]. Endothelium produced EET are also described as endothelial derived hyperpolarization factors (EDHF). A second model is a chronic model with a surgically implanted ameroid constrictor placed around the left coronary artery. This results in gradual narrowing of the artery lumen and corresponding gradual reduction in ejection fraction and other contractile parameters of the affected myocardium over a few weeks. The dog heart has a great potential for coronary collateral vessel growth and therefore this model is convenient for studying the regulation of clinically
Vol. 3, No. 3 2006
important physiological adaptive angiogenesis. Therapeutic intervention like supply of endothelial stem cell [3] or angiogenic factors like the cytokine vascular endothelial growth factor (VEGF) and the basic fibroblast growth factor have also been successfully tested in this situation [4]. Thirdly, Lyseggen et al. [5] used an open chest dog heart model to document echocardiographic indices of potential use during evaluation of reperfused myocardium. The study aim was to find means to evaluate post-ischemic viability and therefore whether reperfusion was successful. Measurements based on echocardiography, which can be used in a clinical setting were compared with sonomicrometry, intraventricular pressure measurements, tissue staining for necrosis and coronary perfusion measurements using radionucleotid microspheres for microvascular perfusion. They tested if area of detectable contractile activity as well as dyskinetic or akinetic areas with surviving tissue could be classified as viable with their specific echo measurement. The study postulated that the passive dynamic properties of non-necrotic but akinetic tissue could be distinguished from necrotic tissue owing to the necrotic oedema by echo technique, and in the experimental model they confirmed this was the case.
Porcine models Mainly there are two categories of models of myocardial ischemia in the pig heart. The first one is related to clinical cardiac surgery, advanced instrumentation and a need for mimicking the situation in human surgery. The second one includes models of acute or chronic regional ischemia like models of acute ISCHEMIC PRECONDITIONING or chronic hibernating myocardium (Glossary). Acute ischemic preconditioning is cardioprotection induced by short lasting ischemic episodes before more severe ischemic injury, for example 5 min ischemia followed by 5 min reperfusion before ischemic injury. Relation between heart work and oxygen consumption in the ischemic and postischemic state is important when examining new therapeutic interventions. One aim of new therapeutics would be to increase heart work without increasing oxygen consumption out of proportions. In the setting of hypothermic cardioplegia, it was shown that addition of nicorandil, a potassium K-ATP channel opener and NO-donor, tended to improve mechanoenergetic efficiency when compared to a traditional cardioplegic potassium chloride solution [6]. Recent in vivo studies by Vinten-Johansen and co-workers in the dog heart have put the focus at the early phase of reperfusion [7,8] (Glossary). Transient and repetitive ischemia of 30-s duration applied during early reperfusion (so called postconditioning) has been reported to result in a significant reduction of reperfusion injury both in vivo and in vitro. Postconditioning can be applied clinically in conjunction with therapeutic reperfusion in contrast to clinical use of ischemic preconditioning which would require treatment before the disease is evident. Although promising due
Drug Discovery Today: Disease Models | Cardiovascular diseases
to potential clinical use, more research is needed to clarify under which circumstances postconditioning occurs and what the mechanisms behind the phenomenon are. In an open chest pig heart study of myocardial infarction by 30 min regional ischemia, classical ischemic preconditioning was demonstrated, but the postconditioning protocol failed to protect [9]. This study combined tissue sampling for western blot detection of cell signalling activation of key kinases in the cell survival pathways (Akt, p70s kinase, ERK), these signalling proteins were comparably affected in both the preconditioned and the post-conditioned group whereas only the preconditioned group had significant reduction in infarct size.
Rabbit models In vivo models of myocardial ischemia in the rabbit can be acute models as well as chronic models. The reader is referred to papers by Downey and co-workers describing the use of the rabbit heart for in vivo studies of regional myocardial ischemia, ischemic preconditioning, post-conditioning and pharmacological cardio-protection [10–13] (Glossary). One advantage of in vivo rabbit models is the possibility to transfer results from the in vitro isolated perfused heart into the in vivo setting without change of animal species. This advantage is also found in rat and mice models. Another advantage is that through special feeding regimes atherosclerotic conditions can be induced in the rabbit.
Rat The chronic in vivo rat model of regional ischemia and infarction has been used for more than three decades. When used without reperfusion, this model is mostly suited for postinfarction remodelling and failure, and for the investigation of scar tissue and inflammation related to tissue repair [14]. For the study of infarct size limitation, reperfusion is obligate because the rat heart (like the rabbit and mice heart) has no collaterals; occlusion of blood flow to an area eventually leads to loss of myocardial cells in that area. With standardized reperfusion and quantification of infarct size, cell death delay protocols for cardio-protection are easy to establish in vivo in a rat model [12]. Possibilities for hemodynamic measurements are limited in the rat heart compared to the complex instrumentation that is possible in larger animals and also during human heart-surgery. However, a major improvement has come with the newest noninvasive imaging modalities, namely echo Doppler technique, NMR imaging and spectroscopy, and PET, which make possible the continuous in vivo monitoring of metabolic activity and also allows the study of the distribution of specific intra and/or extracellular tracers.
Mice Several experimental models developed for rats have now been adjusted for use in mice. This involves miniaturization www.drugdiscoverytoday.com
265
Drug Discovery Today: Disease Models | Cardiovascular diseases
of equipment and, in most cases, optical aid to identify structures. The anatomical location of the left coronary artery is constant between individuals in mice (as well as in rats). Established technique for intubation and artificial ventilation makes chronic experiments with regional ischemia – reperfusion possible in mice because the animal can recover after surgery. Within the field of transgenic mice research, Internet resources have also been developed by commercial animal supply companies, universities and other research institutions – one example being the one provided by The University of Michigan: Transgenic animal web (see Links). An interesting finding with respect to myocardial ischemia has been the role of erythropoietin in non-hematopoetic tissue. Tada et al. [15] were able to show that endogenous nonhematopoetic erythropoietin receptors modulate ischemic injury. Selective knock out of these receptors in mice resulted in worsening of ischemia outcome. Correspondingly, studies have shown that the addition of erythropoietin protects [16] and combined with the protection afforded by ischemic preconditioning has a cumulative protective effect through inhibition of GSK3beta activity. Across species and experimental models, adenosine released from ischemic tissue plays a very central role in modulating the ischemic process in the heart. Adenosine binds to receptors on cells in the vessel wall, cardiomyocytes, as well as on other cell types in the heart and blood stream and is also a precursor in adenine nucleotide synthesis. Adenosine given in short episodes before myocardial ischemia mimics the effect of ischemic preconditioning (pharmacological preconditioning). This effect is based on stimulation of the A1 or A3 adenosine receptor in the myocardium. However, continuous use of adenosine receptor stimulation by adenosine or receptor subtype specific analogues for pharmacological preconditioning fail to maintain the heart in a protected state thus limiting potential clinical use [13]. The role of adenosine receptor agonism as a cardioprotective principle was confirmed in a study using a transgenic mice model of moderate adenosine A1 receptor overexpression [17]. When subjected to a standardized 45-min regional ischemia and 24-h reperfusion, infarcts were significantly smaller in hearts from transgenic mice compared to wild type. Other genes have also been reported to induce a cardioprotective phenotype when over-expressed in rat or mice, for example heat shock protein 70, and the antioxidant enzymes manganese superoxide dismutase and glutathione peroxidase [18–20]. It is important to distinguish changes directly induced by selective gene modification from those secondary adaptive changes. When results on modulation of ischemia by use of specific inhibitors or activators in wild type animals are confirmed in dedicated transgenic animal models this provide a strong case for a mechanistic connection with the gene product in question. 266
www.drugdiscoverytoday.com
Vol. 3, No. 3 2006
Ex vivo and in vitro models In the present text we will include ex vivo models under the in vitro heading, referring to organ (heart), tissue (muscle) and primary cells isolated from animals.
Isolated perfused hearts With a steady increase over the past 20–30 years, isolated perfused hearts are now extensively used in the study of myocardial ischemia (Fig. 1). The simplicity of the technique and the relatively low cost, usually permits larger number of experiments and more hypotheses to be tested compared with the more complex in vivo models. The isolated perfused rat heart is used for biochemical tissue analyses, various
Figure 1. (a) Isolated perfused hearts have been instrumental in understanding the basic principles of cardio-protection, in understanding cell signalling and last but not least in understanding time dependent metabolic changes in the heart muscle during limited oxygen supply. The Langendorff perfusion technique dates back to more than 100 years, and is still used extensively in basic heart research worldwide. Although many modifications of this technique are possible, basically hearts are mounted by aorta on a perfusion apparatus and buffer delivered retrogradely in the aorta line and into the coronaries above the closed aortic valve by constant flow or constant pressure. Global or graded ischemia is produced by clamping the supply line. Hypoxic perfusion is also used. The working heart model was developed and further modified by the work of Morgan, Neely and others almost 50 years ago. When the heart is perfused in working mode, buffer is delivered into the left atrium, and is pumped trough the aortic valve by the heart itself, corresponding to the in vivo situation. The heart is responsible for its own oxygen and substrate supply in the working mode and spontaneous recovery of contractile function after global ischemia is often lost after short time ischemia. With the Langendorff perfusion technique more prolonged ischemia can be tested without complete loss of measurable contractile function. (b) Regional ischemia in the isolated perfused heart is obtained by ligation of the left coronary artery or one of its main braches depending on species. In the rat and mice heart, the anatomy of the artery is constant. In the rabbit (this figure), the artery divides into several branches close to the aortic root. Normally perfused myocardium, ischemic risk zone and infarct are indicated by difference in colours (normally perfused myocardium: red and pink; ischemic risk zone: violet; infarct: blue).
Vol. 3, No. 3 2006
Drug Discovery Today: Disease Models | Cardiovascular diseases
died. In hearts from male db/db mice reduced ischemic tolerance developed in parallel with metabolic changes (increase in fatty acid oxidation and decrease in glucose oxidation). Interestingly, the use of peroxisome proliferator-activated receptor-alpha (PPARalfa) treatment did not affect sensitivity to ischemia-reperfusion, even though carbohydrate oxidation was increased and palmitate oxidation was decreased [24]. Age and gender dependent changes vary considerably between species, animal strains, experimental models and study endpoint used [25]. The impact of age and gender upon response to 20 min of ischemia has been investigated in mice (C57/B16 strain) using isolated perfused heart. The most significant age dependent change in post-ischemic recovery of heart function was reduced diastolic function in line with the reduction in compliance reported with aging in human medicine [25]. Figure 2. Mitochondria and myocardial ischemia. Cell signalling in acute myocardial ischemia – death versus survival signals. Mitochondria play a central role and promote cell death as well as cell survival: cell death by disrupting electron transport and energy supply, and by releasing and activating proteins that mediate apoptosis; cell survival by ability to change into a stress tolerant state in response to metabolic challenges, extra and intracellular stimuli. Cell survival signalling and death signalling is controlled by several protein kinases like MAPkinases, protein kinase Cs and glycogen synthetase kinase [27,40]. MPT, mitochondrial membrane permeability transition; ROS, reactive oxygen species.
functional measurements and evaluation of infarction size in parallel with time-limited use of pharmacological probes (non recirculating perfusion solution permits washout) (Fig. 1). We were able to show that transient pre-ischemic treatment with pharmacological concentrations of 17betaestradiol followed by 5–15 min of washout before prolonged regional ischemia induced cardio-protection, and that this was related to increase in ROS levels before regional ischemia [21], consistent with a redox signalling function of ROS resulting in activation of cell survival pathways (Fig. 2). Using groups of isolated rat hearts perfused under constant flow or under constant pressure, Penna et al. [22] showed that ISCHEMIC POSTCONDITIONING was cGMP dependent and that the treatment resulted in more robust protection when hearts are perfused at constant coronary flow. With the use of mice hearts and development of transgenic animals, the use of ex vivo perfusion techniques will remain high. The technique was reviewed by Sutherland et al. 2003 [23], and the reader is referred to this review and the corresponding list of references for information. The leptin receptor deficient db/db mouse is a well-described animal model for age dependent diabetic cardiomyopathy [24]. Using mouse isolated perfused heart, the ischemia tolerance in male hearts from a well-described natural genetic (db/db mouse) model of age dependent diabetic cardiomyopathy was stu-
Multi-cellular preparations Multi-cellular preparations are mostly papillary muscle or atria trabecula muscle harvested from different species and superfused with buffer. Suitable endpoints for the study of ischemia in these preparations are contractile force and enzyme release (troponin T) during and after simulated ischemia (usually hypoxia or anoxia). The technique has been adopted for use in human atria trabecula [26]. With a micropipette technique, single cell action potentials can be recorded from the surface of superfused muscle preparations and electrophysiological response to ischemia can be investigated. The role of ischemia induced opening of ATP dependent sarcolemmal potassium channels and action potential shortening is easily demonstrated (Box 1).
Cardiomyocytes Cardiomyocytes can be isolated acutely from hearts of all species and maintained in suspension for hours. Short time culture is also possible. Manipulation of the incubation buffer, anoxia or hypoxia, or ‘pelleting’ and sealing under paraffin can be used to simulate ischemia. By applying confocal laser scanning microscopy-based cell imaging technique to isolated adult rat cardiomyocytes in suspension mitochondrial function and integrity can be studied in intact cells. In this model, the importance of reactive oxygen species (ROS) related loss of mitochondrial membrane potential at reperfusion was demonstrated [27]. Opening of the mitochondria permeability transition pore (MPT) lead to rapid loss of the potential, and it was demonstrated that protection by preconditioning could protect against such pore opening [27]. Figure. 2 illustrates the principle role of ROS and MPT in cell death and cell survival signal transduction.
Cell culture techniques Cell culture techniques have considerably rationalized cell studies; cultures most often used are either based on cells www.drugdiscoverytoday.com
267
Drug Discovery Today: Disease Models | Cardiovascular diseases
isolated from neonatal or embryonic hearts of rats and mice [28,29] or obtained commercially as immortalized cell lines. Neonatal mice heart cell cultures can be harvested from genetically engineered mice hearts; transfection techniques as well as silencing RNA technique can be used in cell cultures. HL-1 cells are immortalized cells derived from an atrial tumour (myoma) in female mice. The cell line has been developed and extensively characterized by Claycomb and coworkers [29]. Compared to the use of primary isolation of ventricular myocytes, the advantages of having a good immortalized cell culture model for use in ischemia studies involve the following: (1) no animal use and care, (2) no costly and lengthy preparation time before any experiments can take place, (3) no variable and limited yield for highthroughput approaches and (4) no problems with heterogeneous cell population. The HL-1 cells and the following generations of these cells have been used successfully for ischemia related experiments; overall results correlate well with results obtained, for example, isolated hearts. One example of the use of this model [30,31] is the study of pharmacological preconditioning with opioid agonists and the detailed delineation of down stream survival cellular signalling events. Using these cells as a supplement to experiments in isolated perfused hearts subjected to ischemia – reperfusion, it has been possible to show that opioid-induced cardioprotection is dependent on the JAK/STAT pathway: JAK2 as a mediator of STAT3, Akt, and GSK-3beta. H9c2 is an embryonal rat-heart ventricle derived cell line. These cells have been used to characterize ischemia-induced apoptosis; ischemia induced heat shock protein expression changes as well as the detailed role of p38MAP kinase in cell injury due to simulated ischemia [32,33]. Myotubes or human myoblasts (Girardi cells) have also been used to study ischemic preconditioning. Cells can be obtained from American Type Culture Collection or European Collection of Cell Cultures.
In silico models There is limited tradition for in silico models in ischemic heart research. One exception is the electrophysiological modelling of the consequences of changes in ion balance with ischemia. Important input in these models includes potassium loss to the extra-cellular space, intracellular proton accumulation and opening of ATP dependent potassium channels, cellular electrical uncoupling and delayed conduction, diastolic calcium overload and increase in cellular sodium [34]. A paper by Keener [35] describes fibrillation of the heart as a consequence of spatial heterogeneity during regional ischemia. Using general arguments and numerical simulations with generic models of excitable media, they conclude that a spatial region with an elevated resting potential surrounded by a spatial region wherein action potentials are shortened can drive a break up instability, leading to the rapid initiation of a fibrillatory state. 268
www.drugdiscoverytoday.com
Vol. 3, No. 3 2006
New interesting possibilities will emerge with the ongoing advancements of proteomics and of knowledge on regulatory cellular enzymes and biochemistry. One example related to myocardial ischemia is the modelling of human cardiac mitochondrial metabolic network [36]. In this study, four different conditions were investigated: normal condition, ischemia (25% reduction in oxygen supply), diabetes and diet restriction (low fat, high glucose). Interestingly, their network-balancing model indicated that, while, as expected, oxidation of fatty acids is severely limited with the reduction in oxygen supply, the flux of fatty acids into phospholipids increased in this situation. Therapeutic interventions proposed to increase ATP availability under ischemia such as glucose – potassium – insulin treatment (GIK), keton body supply and increase in pyrovate dehydrogenase activity were also tested. Surprisingly, the theoretical modelling revealed no increased ATP availability, indicating a need for more clarifying experiments.
Model comparison In vitro and ex vivo models give information about intrinsic myocardial and vascular (in the case of isolated hearts) factors in the absence of confounding factors (Table 1). In in vivo models these factors should be part of the investigation. The autonomic nerve system, the central nervous system, bloodborn elements like hormones, cytokines and blood born cells are studied as well as the interaction between the heart and other organs. There has been a tendency to perform experiments in animals of only one gender (especially when using rats or mice), and, for various reasons, male sex has been preferred in many laboratories. Recently, research has been focusing on the question of a possible difference in ischemia-reperfusion tolerance between genders and especially on the molecular mechanisms of observed difference. Interestingly Wang et al. 2006 were able to show that recovery of contractile function after 20 min of acute ischemia in the isolated mice heart was improved by the presence of estrogen receptor alfa [37]. They subsequently showed that this improved recovery was associated more with ERK-phosphorylation, than with JNK- phosphorylation and proapoptotic enzyme caspase 3. Although, in the study, there was a gender difference in some inflammatory markers in the examined hearts (TNF-a, IL-1b and IL6), this did not correlate with the improved recovery of contractile function. This way the use of both genders in animal experimental studies can also serve as an experimental tool or a natural model for understanding the mechanisms of ischemia-reperfusion injury. The selection of model endpoints depends on the aim of a study and the hypothesis under investigation. This is illustrated when investigating the ischemic syndromes STUNNING, HIBERNATION and ischemic preconditioning [38,39] (glossary).
Vol. 3, No. 3 2006
Drug Discovery Today: Disease Models | Cardiovascular diseases
Table 1. Comparison summary table In vivo
In vitro
In silico
Close to clinical situation
Mechanisms intrinsic to the heart
Transgenic animals (mice) Permits investigation of influence of extracardiac factors
Well-controlled models Large number of experiments possible
Permit a large number of scenarios to be tested
Commercial cell based molecular genetics techniques can be used
Save research animals and laboratory costs
Often harbour factors not related to the heart which can not be controlled Changes induced by surgery, instrumentation and anaesthetics modify outcome
Difficult to transfer results to the situation in human medicine Species limitations
Severely restricted by current knowledge (model input)
Heart surgery models Coronary collateral growth Hibernating myocardium
Cell signalling in ischemic myocardium Mitochondrial function Heart metabolism
Preclinical testing Cardiovascular hemodynamics
Ischemic and hypoxic cell death Combined in vivo–ex vivo examination
Access
Literature
Literature
Literature
Patents
n/a
n/a
n/a
References
[1,3–6,8–13]
[14–26]
[27,28]
Pros
Cons
Best use of model
Adult human cardiomyocyte cultures are not available
Studies combining chronic in vivo treatment or interventions followed by ex vivo/in vitro experiments represent an important strategy when investigating myocardial ischemia. Experimental animals can be treated in vivo with pharmacological agents with proposed beneficial effect, with different diets or subjected to other interventions for days or weeks before heart harvesting. Thereafter, the heart or cells isolated from this heart can be subjected to ischemia and tolerance can be tested. This creates a time window for gene expression changes and structural changes in the heart, and when followed by ex vivo examination the consequences of the treatment intrinsic to the heart can be examined specifically without interference from other organ systems via blood or autonomic nerves. And last but not least, the animal ethical perspectives always have to be taken into consideration as well as national regulation (i.e. European Convention for the Protection of Vertebrate Animals used for Experimental and other Scientific Purposes CETS No.: 123) (Links).
Translation to the human situation The role of age, gender, coronary anatomy (collateral flow), innate and adopted immunity and inflammation, diet and coexisting diseases must not be underestimated when translating results to humans. Myocardial ischemia is a disease of the aging heart in humans, but many animal experiments are conducted in younger animals. With respect to the human heart, an important example is the anatomy of the coronary circulation. The healthy human heart has few if any collaterals. However, a (partly) blocked vessel may lead to development of a collateral circulation: small capillary-like branches of the artery develop into larger vessels over time
Electrophysiology (well established) Metabolomics, ATP flux controlling network (new)
in response to tissue ischemia. Thus, treatment promoting angiogenesis and neovascularisation used successfully animal hearts might be of little value for an elderly patient with chronic heart disease who has already experienced a strong endogenous stimulation of collateral vessel growth. Cardioprotection by ischemic preconditioning is an example where a huge overlap of results obtained in different species makes it reasonable to transfer results to the human heart as research has demonstrated. But with new experimentally developed cardioprotective treatments like postconditioning the question about validation across different species have to be examined again. The human genome can now be compared with the genome of other species (especially laboratory animals like mice and rats). This makes transfer to the human situation possible when it comes to research addressing a single gene product. Post-translational modification of gene products is important for the final function of many proteins and is often dependent on many different genes. In this area knowledge is still limited when it comes to species differences. With complex disease processes like chronic ischemic heart disease transfer from animal models to the human situation is often difficult.
Conclusion A broad spectrum of different experimental models is needed for the study of myocardial ischemia ranging from advanced instrumentation of large animals to imaging of mitochondria in cardiomyocytes. We need to be able to distinguish between events intrinsic to the heart on one side, and the integrated whole body response to myocardial ischemia on the other side. In addition, within the heart tissue ischemia affects both cardiomyocytes and the non-myocyte components in the www.drugdiscoverytoday.com
269
Drug Discovery Today: Disease Models | Cardiovascular diseases
Vol. 3, No. 3 2006
10
Links European Society of Cardiology, Euro Heart Survey Publications web page include information about cardiovascular diseases in Europa in 2006: http://www.escardio.org/knowledge/ehs/ publications/ World Health Organisation web page for health topics and cardiovascular disease provide links to information about global burden and prevention: http://www.who.int/topics/ cardiovascular_diseases/en/ The web page from Karolinska Institute, Stockholm, Sweden is an example of an university based information pool for cardiovascular diseases and disorders with multiple links to other pages: http://www.mic.ki.se/Diseases/C14.html Information about the European Convention for the Protection of Vertebrate Animals used for Experimental and other Scientific Purposes CETS No.: 123) is found at: http://www.coe.int/T/E/ Legal_affairs/Legal_co-operation/Biological_safety,_use_of_animals/ Laboratory_animals/ Information about available cell cultures can be found at European Collection of Cell Cultures: http://www.ecacc.org.uk/, and American Type Culture Collection: http://www.biotech.ist.unige.it/ cldb/descat1.html The University of Michigan: Transgenic animal web: http://www.med.umich.edu/tamc/links.html
11
12
13
14
15
16
17
18
heart. Given the accumulation of knowledge about the ischemic process it is time to ask for transfer of this knowledge into clinical use in human medicine. Advancement in imaging techniques, further development of genetically modified animal models and functional genomics, combined with the concepts of metabolomics might provide the necessary tools to answer the request.
References 1 Nithipatikom, K. et al. (2006) Effects of selective inhibition of cytochrome P-450 v-hydroxylases and ischemic preconditioning in myocardial protection. Am. J. Physiol. 290, H500–H505 2 Granville, D.J. et al. (2004) Reduction of ischemia and reperfusion-induced myocardial damage by cytochrome P450 inhibitors. PNAS 101, 1321–1326 3 Silva, G.V. et al. (2005) Mesenchymal Stem Cells Differentiate into an Endothelial Phenotype, Enhance Vascular Density, and Improve Heart Function in a Canine Chronic Ischemia Model. Circulation 111, 150–156 4 Liu, Y. et al. (2006) Effects of basic fibroblast growth factor microspheres on angiogenesis in ischemic myocardium and cardiac function: analysis with dobutamine cardiovascular magnetic, resonance tagging. Eur. J. Cardiotorac. Surg. 30, 103–108 5 Lyseggen, E. et al. (2005) Myocardial train analysis in acute coronary occlusion. a tool to assess myocardial viability and reperfusion. Circulation 112, 3901–3910 6 Steensrud, T. et al. (2004) Replacing potassium with nicorandil in cold St. Thomas’ hospital cardioplegia improves preservation of energetics and function in pig hearts. Ann. Thorac. Surg. 77, 1391–1397 7 Tsang, A. et al. (2005) Myocardial postconditioning: reperfusion injury revisited. Am. J. Physiol. 289, H2–H7 8 Zhao, Z.Q. et al. (2003) Inhibition of myocardial injury by ischemic postconditioning during reperfusion: comparison with ischemic preconditioning. Am. J. Physiol. 285, H579–H588 9 Schwartz, L.M. et al. (2006) Ischemic postconditioning during reperfusion activates Akt and ERK without protecting against lethal myocardial ischemia-reperfusion injury in pigs. Am. J. Physiol. 290, H1011–H1018
270
www.drugdiscoverytoday.com
19
20
21
22
23 24
25 26
27
28
29
30
31
Philipp, S. et al. (2006) Postconditioning protects rabbit hearts through a protein kinase C-adenosine A2b receptor cascade. Card. Res. 70, 308–314 Krenz, M. et al. (2001) Acute ethanol exposure fails to elicit preconditioning-like protection in in situ rabbit hearts because of its continued presence during ischemia. J. Am. Coll. Cardiol. 37, 601–607 Ytrehus, K. et al. (1994) Rat and rabbit heart infarction: effect of anesthesia, perfusate, risk zone, and methods of infarct sizing. Am. J. Physiol. 267, H2383–H2390 Tsuchida, A. et al. (1994) The anti-infarct effect of an adenosine A1 selective agonist is diminished after prolonged infusion as is the cardioprotective effect of ischemic preconditioning in rabbit heart. J. Mol. Cell Cardiol. 26, 303–311 Qvigstad, E. et al. (2005) Dual Serotonergic Regulation of Ventricular Contractile Force Through 5-HT2A and 5-HT 4 Receptors Induced in the Acute Failing Heart. Circ. Res. 97, 268–276 Tada, H. et al. (2006) Endogenous erythropoietin system in nonhematopoietic lineage cells plays a protective role in myocardial ischemia/ reperfusion. Card. Res. 71, 466–477 Nishihara, M. et al. (2006) Erythropoietin affords additional cardioprotection to preconditioned hearts by enhanced phosphorylation of glycogen synthase kinase-3b. Am. J. Physiol. 291, H748–H755 Yang, Z. et al. (2002) Cardiac overexpression of A1-adenosine receptor protects intact mice against myocardial infarction. Am. J. Physiol. 282, H949–H955 Shiomi, T. et al. (2004) Overexpression of gluthatione peroxidise prevents left ventricular remodelling and failure after myocardial infarction in mice. Circulation 109, 544–549 Jayakumar, J. et al. (2001) Heart shock protein 70 gene transfection protects mitochondrial and ventricular function against ischemiareperfusion injury. Circulation 104, 303–307 Tanaka, M. et al. (2004) Overexpression of human copper/zinc superoxide dismutase (SOD1) suppress ischemia-reperfusion injury and subsequent development of graft coronary artery disease in murine cardiac grafts. Circulation 110, 200–206 Sovershaev, M.A. et al. (2006) Preconditioning by 17{beta}-estradiol in isolated rat heart depends on PI3-K/PKB pathwat, PKC and ROS. Am. J. Physiol. 291, H1554–H1562 Penna, C. et al. (2006) Post-conditioning reduces infarct size in the isolated rat heart: Role of coronary flow and pressure and the nitric oxide/cGMP pathway. Basic Res. Cardiol. 101, 168–179 Sutherland, F.J. et al. (2003) Mouse isolated perfused heart: Characteristics and cautions. Clin. Exp. Pharmacol. Physiol. 30, 867–878 Aasum, E. et al. (2003) Age-Dependent changes in metabolism, contractile function, and ischemic sensitivity in hearts from db/db mice. Diabetes 52, 434–441 Willems, L. et al. (2004) Age-related changes in ischemic tolerance in male and female mouse hearts. J. Mol. Cell Cardiol. 38, 245–256 Shanmuganathan, S. et al. (2005) Mitochondrial permeability transition pore as a target for cardioprotection in the human heart. Am. J. Physiol. 289, H237–H242 Juhaszova, M. et al. (2004) Glycogen synthase kinase-3b mediates convergence of protection signaling to inhibit the mitochondrial permeability transition pore. J. Clin. Invest. 113, 1535–1549 Okada, T. et al. (2005) Integrated pharmacological preconditioning and memory of cardioprotection: role of protein kinase C and phosphatidylinositol 3-kinase. Am. J. Physiol. 289, H761–H767 Claycomb, W.C. et al. (1998) HL-1 cells: A cardiac muscle cell line that contracts and retains phenotypic characteristics of the adult cardiomyocyte. Proc. Natl. Acad. Sci. U S A 95, 2979–2984 Gross, E.R. et al. (2006) The JAK/STAT pathway is essential for opioidinduced cardioprotection: JAK2 as a mediator of STAT3, Akt, and GSK-3b. Am. J. Physiol. 291, H827–H834 Seymour, E.M. et al. (2003) HL-1 myocytes exhibit PKC and KATP channel-dependent delta opioid preconditioning. J. Surg. Res. 114, 187–194
Vol. 3, No. 3 2006
32 33
34
35
Tantini, B. et al. (2006) Involvement of polyamines in apoptosis of cardiac myoblasts in a model of simulated ischemia. J. Mol. Cell Cardiol. 40, 775–782 Tanno, M. et al. (2003) Diverse mechanisms of myocardial p38 mitogenactivated protein kinase activation: evidence for MKK-independent activation by a TAB1-associated mechanism contributing to injury during myocardial ischemia. Circ. Res. 93, 254–261 Nygren, A. et al. (2006) Measurements of electrophysiological effects of components of acute ischemia in langendorff-perfused rat hearts using voltage-sensitive dye mapping. J. Cardiovasc. Electrophysiol. 17, S113–S123 Keener, J.P. (2003) Model for the onset of fibrillation following coronary artery occlusion. J. Cardiovasc. Electrophysiol. 14, 1225–1232
Drug Discovery Today: Disease Models | Cardiovascular diseases
36
37 38 39 40
Thiele, I. et al. (2005) Candidate metabolic network states in human mitochondria. impact of diabetes, ischemia, and diet. J. Biol. Chem. 280, 11683–11695 Wang, M. et al. (2006) Estrogen receptor-a mediates acute myocardial protection in females. Am. J. Physiol. 290, H2204–H2209 Bolli, R. et al. (1999) Molecular and cellular mechanisms of myocardial stunning. Physiol. Rev. 79, 609–634 Downey, J.M. et al. (2005) We think we see a pattern emerging here. Circ. 111, 120–121 Das, D.K. et al. (2005) Mitochondrial function in cardiomyocytes: target for cardioprotection. Curr. Opin. Anaestesiol. 18, 77–82
www.drugdiscoverytoday.com
271