Cardiac Physiology in Transgenic Mice
Abstract—By use of gene targeting and/or transgenesis, it is now possible to make defined changes in genes whose functions underlie mammalian cardiovascular function. Because of technical and economic considerations, these experiments are largely confined to the mouse. Genetic modification of the loci responsible for aspects of cardiac development, differentiation, and function via gene targeting, as well as modulation of the cardiac protein complement using transgenesis, has begun to provide mouse models of cardiac hypertrophy, dilated cardiomyopathy, and hypertrophic cardiomyopathies. In order to use these animal models fully and explore their phenotypes at the whole organ and whole animal levels, the extension of cardiovascular physiological methodologies to the mouse is imperative. Techniques for exploring aspects of cardiovascular function are well developed for larger animal models, but their modification for the small size of the mouse heart and for the animal’s rapid cardiac cycle has proven to be a formidable challenge, requiring the combined efforts of the molecular biology, physiology, and cardiology communities. We review here the ability of present-day technology to obtain reproducible data on murine cardiac function at the whole organ and animal levels.
Animal models of cardiovascular disease have been a staple tool of the cardiologist and cardiovascular physiologist for years. However, until very recently, the models available were either a result of a fortuitous genetic event, a surgical intervention, or a pharmacological regimen. As manipulation of the mammalian genome becomes more routine, it has become possible to generate animal models for studying cardiovascular function and disease such that the trait responsible for the perturbation is precisely defined at the genetic level. By using gene targeting via homologous recombination in murine embryonic stem cells, a specific gene can be either ablated or modified, and the mouse carrying the targeted locus can be generated for subsequent studies.1 Thus, a gene suspected of playing an important role in cardiac development, function, or pathogenesis can either be “knocked out” or modified in a particular domain or even in a single amino acid residue, and the resultant phenotype can be studied over the animal’s lifetime in the absence of any acute pharmacological or surgical intervention. Similarly, transgenesis can be used to effectively remodel the protein complement of the cardiac compartment, and the resultant phenotypes can be studied in a longitudinal fashion. Because these animal models are the result of a stable genetic modification, the primary genetic etiology of any changes that are observed is known. Furthermore, these models are stable as the mutations are propagated through the germline. If the genetic modifications result in a cardiovascular phenotype, this leads to the creation of a new animal whose numbers can be expanded and subsequently disseminated to the cardiovascular community for a variety of assays and studies.
These technologies resulted largely from advances in molecular genetics. However, the typical molecular geneticist is generally ill-equipped to deal with many of the assays needed to define the resultant phenotype(s), particularly when the cardiovascular system is affected. Molecularly oriented studies dealing with transcript determinations, gene actions and interactions, protein-protein interactions, and signal transduction pathways, although obviously valuable in their own rights, fail to define over the lifetimes of the animals the changes that occur at the whole organ and whole animal levels—aspects of biology that have been traditionally the focus of physiologists.
Because of technical and economic considerations, the mouse is almost always the model of choice in which to carry out these genetic modifications. However, from the physiologist’s point of view, the small size of the animal and the short cardiac cycle (heart rates of 500 to 790 bpm) render the murine system relatively refractory to the typical repertoire of physiological analyses that have been used to determine normal and abnormal cardiovascular function. The general subject of physiological methodologies applied to engineered mice has been reviewed recently in a comprehensive fashion.2 The purpose of this mini review is to summarize how the two formerly disparate fields of molecular genetics and physiology have developed common interests in producing and analyzing transgenic and gene-targeted mouse models and how both in vivo and ex vivo physiological analyses that have previously been restricted to large animal models can now be applied productively to analyses of the genetically engineered murine cardiovascular system.
Ex Vivo Heart Perfusion Techniques
With the advent of transgenic and gene-targeted manipulations of the cardiac contractile apparatus, it became obvious that analyses of whole organ function would be needed to analyze important aspects of any resultant phenotypes. Techniques were developed in which the well-characterized isolated perfused heart apparatus was miniaturized such that parameters of myocardial contractility could be reproducibly assessed by using either a Langendorff (retrograde, nonworking) or left-sided working heart preparation.3 In the Langendorff preparation, the heart is excised from an anesthetized mouse and perfused in a retrograde fashion with 37°C oxygenated Krebs’ solution. A small caliber polyethylene catheter is passed into the LA via a pulmonary vein, advanced across the mitral valve into the LV, and pushed through the LV apex until the flanged end of the catheter rests against the LV endocardium. The other end of the catheter is attached to a pressure transducer. Continuous heart rate, aortic pressure, and LV systolic and diastolic pressures are measured, and contractility/relaxation, as measured by the first derivative of instantaneous intraventricular pressure (±dP/dt), is determined. The ability of the isolated heart to respond to pharmacological agents such as β-adrenergic agonists can be assessed by the addition of the agent to the perfusate.
After the establishment of stable retrograde perfusion, the preparation can be switched to a working heart preparation by introducing a catheter into the LA via the pulmonary vein and perfusing the isolated heart in an anterograde fashion. The working heart preparation is technically more difficult to work with than the Langendorff preparation but has the advantage of measuring work-related indices of whole heart function. Since the isolated working heart is being continuously loaded, cardiac work capacity can be determined by changing the flow rate into the LA. The Starling length-tension relationship can then be analyzed. Like the retrograde Langendorff heart, the working heart preparation allows one to determine the cardiac response to β-adrenergic agonists, which can be assessed at baseline loading conditions and repeated as load is increased.4 Real-time measurements of parameters such as heart rate, mean aortic pressure, left intraventricular pressure, systolic pressure, diastolic pressure, end-diastolic pressure, peak pressure, relaxation time, ±dP/dt, duration of contraction and relaxation, and atrial pressure can all be obtained.
The isolated perfused heart preparation was first used to examine the functional consequences of altering the normal cardiac MyHC protein complement by pharmacological induction of hyperthyroidism and hypothyroidism.3 The hypothyroidism-induced MyHC isoform switch from V1 to V3 caused a depression in myocardial performance that was evident in altered time to peak pressure, time to half-relaxation, and ±dP/dt, consistent with the long-standing hypothesis that the MyHC isoform is an important determinant of muscle function. Contraction and relaxation indicators at minimal preloads and afterloads showed that the hyperthyroid mouse values significantly exceeded the control mice values and that the hypothyroid mice values were significantly below the normal values. The Starling function curves were greatly reduced in hypothyroid animals, and there was no difference in the slope of the pressure and volume loads.
The isolated working heart has now been used to examine a large number of genetically engineered mouse hearts. These include transgenic models in which components of the β-adrenergic system have been overexpressed5 or in which isoform switches in the myosin light chain complement have been made.6 Hearts in which nonfunctional cardiac MyHC or cardiac actin genes have been created via “gene targeting” have also been analyzed,7 8 as has a mouse model for FHC.9 Although this list is by no means inclusive, the data gathered from these studies demonstrate that the overall contractile function of murine cardiac muscle can be determined. A number of laboratories are now detecting very subtle changes in cardiac contractility and the ability of the heart to respond to altered preload and afterload conditions; these data are instrumental in defining the functional correlates of the genetic changes that have been made.
One of the clearest examples of the value for this integrated approach is illustrated by studies dealing with phospholamban, a small (52–amino acid) phosphoprotein that in the 1980s was implicated in regulating the activity of the sarcoplasmic reticulum Ca2+-ATPase pump on the basis of extensive in vitro analyses. To determine the action(s) of this protein within the whole animal context, the gene was ablated, and cardiac function was determined using the isolated perfused working heart.4 Phospholamban deficiency resulted in hyperdynamic cardiac functional parameters with significantly increased ±dP/dt values and enhanced ventricular filling. Because both the homozygotes and heterozygotes were viable, a dose-response curve could be constructed, and decreases in phospholamban expression correlated well with increases both in heart contractility and in the affinity of the pump for Ca2+.
Recently, the capacity to obtain isovolumic measurements in the isolated working heart has been developed (M.M. LeWinter, unpublished data, 1997). This offers the advantage of making functional measurements under controlled loading conditions with independent control of the coronary perfusion pressure. The technical challenge was to develop a miniaturized balloon for insertion into the LV from which reproducible measurements could be made. By using a thin high-density polyethylene balloon with an unstressed volume of 0.02 to 0.03 mL, LV volume could be both measured and reproducibly varied, allowing end-systolic pressure-volume relationships to be determined. The end-systolic pressure-volume relationship can also be coupled with oxygen consumption (V̇o2) measurements in order to characterize the V̇o2-pressure-volume relationship. By using these parameters, estimates of the relationship between chemical energy input and mechanical energy output of the contractile machinery, a measure of contractile efficiency, could be made, in addition to estimates of nonmechanical energy consumption.
In Vivo Measurements
The isolated heart preparations cannot be used to study the interdependent physiology of the cardiohumoral vascular system. To circumvent this difficulty, an in vivo open-chest model was developed and was first used to study the physiological effects of transgenic ventricular expression of p21ras driven by a myosin light chain promoter.10 The expression of p21ras resulted in the development of cardiac hypertrophy, as detected by molecular and histological analyses. To study cardiac function in vivo, the mouse was anesthetized, the chest was opened, and a bilateral vagotomy was performed. A 2F high-fidelity micromanometer catheter was introduced into the LA and subsequently advanced into the LV for pressure measurement. Continuous LV systolic and diastolic pressures, aortic pressure, and ±dP/dt were recorded. At a sampling rate of 1000 Hz, dP/dtmin could be measured and subsequently used for determining τ, the time constant of isovolumic LV pressure decay, an index frequently used to describe human ventricular relaxation but not previously described in mice. The measurements were first taken at baseline and then repeated during isoproterenol infusion to compare responses to β-adrenergic stimulation. Compared with control mice, homozygous p21ras mice showed no difference in peak LV systolic pressure, +dP/dt, or LV end-diastolic pressure. However, they did show a significantly prolonged τ value, indicating impaired LV relaxation or diastolic dysfunction.
The open-chest system of analysis can also provide information about transgenically improved cardiac function. The full-length murine α-MyHC promoter was used to drive cardiac-specific expression of the human β2-AR cDNA.5 As an adjunct to molecular, biochemical, and isolated atrial tissue mechanics, LV function was assessed in vivo. Since this study involved transgenic alteration of AR expression, cardiac response to boluses of isoproterenol was determined in addition to baseline studies. Baseline LV end-diastolic pressure and mean aortic pressure were similar in transgenic and control animals, but heart rate, contractility (+dP/dtmax), and relaxation (−dP/dtmax) were significantly increased in the transgenic mice. Although control animals demonstrated increases in LV contractility with isoproterenol, the transgenic mice did not, indicating a maximally activated β-AR system. Indeed, at the highest isoproterenol dose, the mean aortic pressure decreased in the transgenic mice. This was probably due to peripheral vasodilatory effects of β-AR stimulation without any compensatory increase in cardiac output possible from the already maximally activated transgenic heart.
Although the open-chest and isolated perfused heart models can provide very detailed information about cardiac function, these two modalities cannot directly address in situ hemodynamics, since the interrelated forces of cardiac contractility, intrathoracic pressures, and pericardial constraints of the intact animal are affected or even removed. Insertion of a 25-gauge needle through the chest wall into the LV enabled Iwase et al11 to obtain pressure measurements in the closed-chest animal using a 1.8F Millar micromanometer. The high-fidelity 1.8F and 2.0F Millar Mikro-Tip manometers have now been used by a number of groups to obtain accurate hemodynamic measurements in a closed-chest preparation.12 13 Gottshall et al12 used these techniques to study a transgenic mouse strain that expressed ras under the control of a myosin light chain 2 promoter. Using echocardiography to selectively and noninvasively screen the general transgenic population, they were able to select a subpopulation for breeding in order to obtain a population with an increased frequency of detectable hypertrophy. Closed-chest hemodynamic measurements showed that although the LV end-systolic and end-diastolic pressures did not differ significantly from the pressures in the control animals, there was a significant intraventricular pressure gradient in a number of the ras mice, a result consistent with flow measurements obtained using Doppler echocardiography, which showed a high-velocity jet in the mid-LV or outflow tract.12
The basic methodology for the closed-chest preparation has been described in some detail.13 After induction of anesthesia, control of body temperature and airway access were accomplished, and the right femoral artery was cannulated with a length of pulled polyethylene tubing, which was advanced into the descending aorta. The aortic catheter was connected to a fixed-dome pressure transducer to measure arterial BP. The right femoral vein was likewise cannulated; this catheter was connected to a CMA 100-mL syringe pump for infusion of cardiotonic medications, including isoproterenol and propranolol. The right carotid artery was cannulated with a 2F Millar Mikro-Tip transducer, which has a reported frequency response that is flat to 10 000 Hz, thus allowing accurate measurements of pressure at the extremely rapid heart rate of the mouse. Parameters assessed included heart rate, mean arterial pressure, systolic and diastolic LV pressure, and LV end-diastolic pressure, all measured from the pressure wave forms. Contractility (+dP/dt) could be calculated from the LV pressure wave form, and time to peak pressure and time to half-relaxation were determined, among other indices. Compared with control mice, the hyperthyroid animals demonstrated a 40% increase in indices of contractility, whereas the hypothyroid group had a 40% decrease in contractility. Relaxation indices were similarly altered. The isoproterenol dose-response relationship was augmented in the hyperthyroid group compared with the control group and attenuated in the hypothyroid group compared with the wild-type control group.
The closed-chest model has been used to assess the functional consequences of cardiac overexpression of a human β2-AR polymorphism in the mouse14 and to describe the functional consequences of a dilated cardiomyopathy resulting from chronic overexpression of a retinoic acid receptor in the heart.15 Despite the relatively complex experimental apparatus, the ability to assess integrated cardiovascular function at both baseline conditions and during infusion of cardiotonic agents in the whole animal offers a valuable complement to both the isolated working heart and open-chest preparations.
In anticipation of expanded transgenic manipulation of cardiac ion channels, in vivo studies of murine cardiac EP were developed in C57Bl/6J mice.16 After induction of anesthesia and endotracheal intubation, a surface six-lead ECG was recorded. The chest was opened with a midline sternotomy, the pericardium was reflected, and a total of four temporary pacing wires (stainless steel Teflon-coated wires) were attached, one each on the exposed RV and LV and two on the RA. In some animals, the chest was closed after externalization of the wires. Unipolar and bipolar ECGs were recorded from the RV, LV, and RA, and pacing thresholds for each lead were determined. A modified EP stimulator was used to pace at coupling intervals as short as 17 milliseconds to provide the rapid stimulation rates required in the mouse. Standard ECG time intervals including PR, QRS, QT, QTc, JT, JTc, and R-R were calculated. Sinus node recovery time, refractory periods of the atrium and atrioventricular node, and retrograde ventriculoatrial conduction were likewise determined with a protocol analogous to that used for human EP studies. A subgroup of the study population underwent intraperitoneal administration of either procainamide or quinidine, Vaughan Williams class I antiarrhythmic agents. After observation for 20 minutes, these mice had a full EP study repeated with a six-lead ECG recorded every 5 minutes. This elegant study demonstrated the feasibility of conducting epicardial EP studies in mice. In general, all parameters normally obtained in a human EP study were obtainable in mice, although a His-bundle electrogram could not be recorded from the epicardial surface. Additionally, Bazett’s correction of the QT interval for heart rate did not appear valid in mice, possibly because of the failure of the square-root denominator in Bazett’s formula to hold at the rapid cycle length of the mouse heart. Comparison with previously published data obtained from other strains of mice suggests that there may be strain-dependent EP differences that must be taken into consideration when generating normative EP data.
These techniques have recently been applied to a genetically engineered model of FHC, in which the normal α-MyHC allele was targeted such that a single amino acid residue responsible for FHC in humans was reproduced in the animal model.9 These animals exhibited aspects of the human phenotype and were subsequently subjected to an EP study. As was noted for other aspects of the phenotypic presentation of the disease, EP abnormalities were variable. In some of the FHC mice, ventricular ectopy could be induced using programmed stimulation and β-agonists. Bigeminal rhythm and ventricular couplets/triplets could be detected in some of the animals as contrasted with the control mice subjected to the identical experimental protocols. A 12-lead EP study also showed that the FHC mice had an altered, rightward ECG axis and prolonged sinus recovery times.17
Noninvasive Physiological Studies
All of the investigational techniques described above are inherently limited by their invasive nature; ie, the mouse does not survive the analysis. Since cardiovascular disease is a dynamic process characterized by periods of compensation and decompensation, analytical methods that can be repeated over time are especially attractive. One of the most common tools in clinical cardiology is the echocardiogram, a noninvasive procedure that provides the clinician with a wealth of information about the structural and functional characteristics of the heart under study. The expansion of this technique into small animal models such as the mouse has been somewhat hindered by the rapid heart rate of these animals, which often exceeds the resolution capabilities of readily available ultrasound transducers. For example, using a 7-MHz transducer, two-dimensional echocardiographic images are obtained at a rate of ≈30 frames per second. If the heart rate of the mouse is 600 bpm or 10 beats per second, there would be only 3.3 frames recorded per beat, with each frame representing ≈1/3 of the cardiac cycle.
Because of higher sampling frequency, motion mode, or M-mode, echocardiography has thus far been the mainstay of murine studies. In contrast to two-dimensional analysis, M-mode analysis is capable of a sampling rate of ≈1000/s. Determination of LVMe was shown to be feasible in a necropsy-validated study.18 Mice of varying ages were weighed and anesthetized before two-dimensionally guided M-mode evaluation, which was performed with the mice in the left decubitus or supine position using a 7.5-MHz transducer, taking care to avoid compression of the thorax (which distorts ventricular geometry) by the transducer. M-mode analysis of the LV was obtained in the short axis and recorded on strip-chart paper at a speed of 100 mm/s. Measurements to within 0.1-mm precision were made of the anterior and inferior wall thickness and LV end-diastolic dimension. By using an uncorrected cube approximation and a specific gravity of myocardium of 1.055 mg/mm3, cardiac mass was calculated. All mice were then killed within 4 hours of the echocardiogram. After removal of the RV free wall, major blood vessels, and atria, the LV (including interventricular septum) was weighed. By use of a least squares linear regression analysis, the correlation of LVMe to the true LVM determined at necropsy was found to be excellent, with a slope of 0.96 and r=.94, exceeding the correlation of true LVM to either body weight or animal age.
Two-dimensionally guided M-mode echocardiography has been used to study a number of experimentally induced cardiac pathologies, including transgenic expression of myf5 and banding of the abdominal aorta, both of which result in cardiac hypertrophy.19 By using a methodology similar to that described above, myf5 mice displayed hypertrophied, but somewhat dilated, LVs. The correlation of true LVM to LVMe was higher than the correlation of true LVM to body mass. However, the correlation was less strong in the myf5 mice compared with the wild-type mice, possibly because of concomitant LV dilatation and hypertrophy in the genetically altered group. A smaller group of mice that had undergone aortic banding showed concentric hypertrophy without LV dilatation. The banded mice had significantly increased interventricular septal thickness and posterior wall thickness. Although the number of banded animals was too small for a separate linear regression analysis, the data fell on the linear regression line for normal mice that compared true LVM with LVMe.
Echocardiography has also been used to assess LV dilatation in mice with a surgically created arteriovenous fistula, LV hypertrophy in transgenic mice overexpressing the H-ras gene, and LV contractility in mice overexpressing the β2-AR.20 Iwase et al,21 using a 9-MHz rather than a 7.5-MHz transducer, were able to obtain high-resolution images that enabled them to accurately determine LV internal dimensions, fractional shortening, and ejection fractions in both control and transgenic animals that were expressing the cardiac stimulatory G-protein α subunit. By adjusting the anesthesia protocols, these measurements could made on both the control and transgenic cohorts at similar heart rates. The data demonstrated that chronic sympathetic stimulation resulted in a cardiomyopathy in the older mice, characterized by an elevated mortality rate, depressed baseline LV function (independent of heart rate), and an increased frequency of premature ventricular beats, resulting in arrhythmias. Because of its relative simplicity and noninvasive nature, murine echocardiography is now widely used, and a number of companies are planning to release, in the near future, dedicated hardware/software packages for murine analysis. Additionally, a number of groups are designing new echo transducers. Although still in the experimental phases, these new tools should dramatically increase the resolution of the technique for the murine heart, and 60-μ lateral and 30-μ axial resolutions have been reported in mouse embryos by use of a 50-MHz ultrasound imaging system.22
Mirroring the current use of echocardiography in the human fetus, murine echocardiography is being extended for use in fetal mice. Although a wealth of information exists about the anatomic development of the mouse heart, there are little data concerning early hemodynamics. Invasive imaging techniques, in which the embryos are exposed via sequential regional hysterectomies and subsequently video-imaged at 60 Hz, enable one to measure chamber dimensions at both end systole and end diastole and to calculate inflow and outflow velocities as early as embryonic day 10.5.23 In an attempt to develop a less invasive procedure, a 7.5-MHz transducer was used to perform fetal mouse Doppler echocardiography on both normal mice and mice with trisomy 16 (analogous to trisomy 21 in humans).24 Embryos were studied noninvasively from embryonic days 10 to 19, with the trisomic embryos studied on embryonic days 11 to 14 of gestation. No abnormalities could be detected in the normal mice, whereas 25% of the trisomic mice had atrioventricular valve regurgitation. As in human patients with trisomy 21, the spectrum of cardiac defects in the trisomic mice ranged from none to severe abnormalities. It is likely that with further refinement of echocardiographic technique and equipment, fetal echocardiography will become a more feasible method for the study of early cardiac hemodynamics in both normal and genetically altered mice. Additionally, as higher frequency transducers become more widely available, two-dimensional echocardiography capabilities will improve.
The closed-chest method of assessing cardiac function in the intact mouse has been augmented by the inclusion of transthoracic echocardiography for simultaneous measurements of LV dimension and pressure in transgenic mice.15 A combination of the two analytical tools, intact animal catheterization and transthoracic echocardiography, allowed the generation of pressure-dimension loops similar to the pressure-volume loops used in the study of cardiovascular function in larger mammals. Ventricular pressure-dimension loops under increased (phenylephrine administration) or decreased (nitroprusside administration) afterload conditions were obtained by digitizing the ventricular pressure wave form, septum, and posterior wall.
These techniques were used to analyze the dose-response relationship of a transgene that induced a model of dilated cardiomyopathy in the mouse.15 In this experimental model, the full-length murine β-MyHC promoter was used to drive expression of a constitutively active retinoic acid receptor in the developing ventricle. Mice with high levels of expression showed remarkable dilatation of the LV accompanied by LA enlargement and intra-atrial thrombi and calcification, presumably from stasis of blood within that chamber. Analyses using echocardiography alone were able to detect both LV and RV dilation. The severity of dilation varied with transgene copy number in a linear fashion. For all of the affected lines, the area circumscribed by the pressure-dimension loops, an index of total LV stroke work, was reduced compared with control values. Load-independent LV contractility, determined by the relationship of the velocity of fiber shortening to end-systolic wall stress, was significantly decreased, with a parallel shift in the relationship down and to the left, a significantly lower y-intercept, and no difference in slope. These analyses provide an assessment of LV function, independent of the confounding effects of alterations in preload and afterload.
Magnetic Resonance Imaging
MRI is well established as a valuable diagnostic tool for measuring compartment volumes, internal wall thicknesses, mass, and shape of the heart at different anatomic levels. Siri et al25 applied gated (diastolic) MRI at four different anatomic levels, obtaining two-dimensional “slices” of the LV in intact anesthetized mice and validated the LVMs against the LV values determined by gravimetric means. Independently, Kubota et al26 also established the validity of MRI in the mouse by using it to determine the overall architecture, volume, mass, and ejection fractions in mice that overexpress tumor necrosis factor-α. These hearts were dilated and showed reduced ejection fractions. Using MRI, one can obtain, simultaneously, both morphological and functional data. The data acquisition can easily be gated to the ECG and respiratory signals such that the image can be placed exactly at a known point in the cardiac cycle. Another significant advantage of this technique compared, for example, with M-mode echocardiography is that data can be gathered at multiple levels essentially simultaneously, such that a more accurate picture of overall chamber function and morphology can be obtained. The disadvantage is the very significant costs involved in purchasing, maintaining, and operating the equipment.
Untethered Measurements in the Conscious Mouse
Heart rate is another important parameter of cardiovascular physiology amenable to noninvasive assessment. Because of unavoidable changes in heart rate with handling of experimental animals, a system of heart rate analysis without appreciable physical contact is highly desirable. The use of telemetry systems for long-range heart rate monitoring in the mouse has been discussed for at least 30 years, although the development of microchip technology and microcomputing capabilities has only recently made feasible the use of implantable telemetry devices.27 The telemetry system consists of a transmitter implanted into the peritoneal cavity, a receiver that is placed under the animal’s cage, and a data acquisition system. Weighing ≈4 g, the transmitter occupies a volume of <1.9 mL and has no operative mortality during placement if the operator is moderately skilled. With careful placement of the transmitter leads, an ECG comparable to limb lead II can be obtained. The raw analog data are digitized and converted to heart rate by a custom software package.
Telemetry was used to probe heart rate variability in transgenic mice overexpressing the human β1-AR.28 The heart rates of the transgenic mice did not differ from those in nontransgenic control mice, but heart rate variability was decreased. No arrhythmias were observed, and life span was unaffected. Heart rate has also been studied in mice overexpressing the short isoform of Gsα in the heart.29 The heart rate in these animals was significantly elevated compared with that in control animals, and again, heart rate variability in the conscious untethered animals was significantly decreased. This lack of variability extended even to the normal variations usually observed during an animal’s normal circadian rhythm.
Telemetry has also been used as an adjunct to treadmill exercise in transgenic mice overexpressing the ventricular isoform of the regulatory myosin light chain.30 In this case, a treadmill stress exercise protocol was developed to noninvasively assess cardiac function (see below). Telemetry was used not only to screen for baseline differences in heart rate but also to monitor for cardiovascular conditioning, evidenced by a progressive decrease in resting heart rate corresponding to improved exercise tolerance, a well-known phenomenon in humans. In this experimental scenario, the transgenic and nontransgenic mice had similar heart rates at rest and during maximal treadmill exercise.
Arterial BP is yet another important parameter of cardiovascular function. Although elegant studies exist in which surgical procedures have been used to obtain mean arterial pressures, regional hemodynamics, and vascular resistance, a routine BP measurement is normally all that is required on a day-to-day basis during an extended longitudinal study. Ideally, BP should be determined in an unanesthetized, unrestrained animal. However, this is not yet possible. In the mouse, a computer-driven tail-cuff blood pressure that establishes a routine of cuff inflation and deflation has been shown to be highly correlative with intra-arterial pressure assessment.31 The system measures BP by means of automated inflation of a tail cuff with simultaneous monitoring of the oscillation of blood in the mouse’s tail detected photoelectrically. The BP is determined to be the cuff inflation pressure at which the oscillatory wave form falls to a programmable percentage of its baseline. Tail-cuff pressures were obtained after a 7-day training period to acclimate the mice to the restraining device and cuff inflation. After the noninvasive BP measurements, a surgically implanted catheter was placed at the junction of the left carotid artery and aorta and tunneled subcutaneously to exit at the nape of the neck. The catheter was then coiled into a flat button-shaped elastomer sewn to the skin between the scapulae. Intra-arterial blood pressure was determined the day after catheter implantation by use of a pressure transducer. The wave forms were analyzed for peak and trough points, and the mean arterial pressure was calculated as the summation of all the data points divided by the number of data points acquired. There was a strong correlation (r=.86) between tail-cuff systolic pressure and mean arterial pressure, although the heart rates were higher, as determined during tail-cuff BP measurements. The noninvasive tail-cuff method should thus prove to be a practical alternative method of BP analysis.
A major clinical determinant of cardiovascular function is a patient’s ability to perform exercise, and a number of clinical exercise paradigms exist involving either bicycle or treadmill ergometry. However, the laboratory mouse is generally a sedentary animal in a minimally physiologically stressful environment. In order to challenge the cardiovascular system such that subtleties in cardiovascular reserve or function might occur, controlled exercise regimens have been established. Swimming was first used to measure the response of adult female C57/Bl6 mice to a chronic conditioning regimen.32 Groups of twelve 8-week-old mice were made to swim in water tanks 15 cm deep held at 30°C to 32°C with a surface area of 225 cm2. The exercise sessions initially lasted 20 minutes and occurred twice daily as the mice became more adept. The duration of swimming was increased in 10-minute increments until the conclusion of the study, when the twice daily sessions lasted 90 minutes. The mean heart rate response to swimming decreased significantly over a 4-week study period.
Compared with sedentary control mice, which were handled daily but not subjected to swimming, the conditioned mice showed a 10% increase in heart weight and a 16% increase in the ratio of heart weight to body weight. Succinate dehydrogenase activity was significantly increased in the soleus muscles of the conditioned mice compared with control mice, but there was no increase in cardiac norepinephrine content, no difference in cardiac myofibrillar ATPase activity, and no shift in MyHC isoform expression. However, it is difficult to control some of the quantitative aspects of the swimming regimen. Since there are a number of mice swimming simultaneously, it is possible for a mouse to stay afloat not by active swimming but rather by simply wedging against its neighbor; alternatively, mice may climb onto each other in an effort to avoid drowning. Thus, there is no way to accurately quantify each individual mouse’s exercise capability or degree of exhaustion, since they are put into the tank in large groups. It is also difficult to control and vary the intensity of the exercise or develop a stress versus conditioning protocol.
As a complementary protocol, treadmill exercise provides a quantifiable method for exercise stress testing that provides performance data for individual mice. Incremental treadmill regimens can be used to uncover cardiovascular phenotypes that under normal conditions are not apparent. Chronic physiological measurements are now possible on an unrestrained nonanesthetized animal, and V̇o2, V̇co2,33 and heart rate (see above) can all be measured in real time. When this regimen is used, the acute murine cardiovascular responses to exercise appear to closely parallel the human responses, with heart rate, V̇o2, and V̇co2 increasing as workload increases and exhaustion is reached.33 This paradigm was used to examine the exercise capabilities of transgenic mice in which the atrial myosin light chain complement was replaced by the ventricle-specific protein.30 A four-lane motor-driven treadmill with adjustable belt speed and posterior shock grid was modified to include an infrared detection system allowing accurate quantification of the number of times during an exercise session that the mouse contacted the shock grid, indicating slowed running and fatigue. The treadmill regimen began with a 2-week acclimation period, during which time exercise duration, treadmill speed, and treadmill incline were gradually increased. After the training period, the mice exercised twice weekly at 20 m/min at a 7° incline for 50 minutes. The speed of the treadmill was increased over the 5-week study period to 27 m/min. Heart weights and contractile apparatus isoform content were relatively unaffected, and no hypertrophy occurred. Cardiovascular conditioning as assessed by resting heart rate was also minimal, indicating that an effective stress regimen had been effected. Compared with the wild-type exercised animals, the transgenic mice performed less well at higher treadmill speeds.
Summary and Future Directions
As disease models become a focus of genetic remodeling of the murine heart, it will become even more essential to move beyond the standard molecular, biochemical, and cytological analyses and explore the phenotypes at the whole organ and whole animal levels. Standard physiological analyses can be productively applied to animal models in which the genetic etiology is known but the pathogenesis of the phenotype remains obscure. With noninvasive modalities, whole organ function can be examined in the intact animal and repeated over time to examine compensatory and decompensatory changes. The recent miniaturization of analytical techniques and the relative ease with which the mouse genome may be manipulated combine to make the study of molecular cardiology practical in mice.
The murine physiological analyses outlined above and summarized in the Table⇓ have only recently been developed and have not been widely disseminated throughout the cardiovascular community. In particular, the invasive and ex vivo procedures are routinely used only by a few groups, and intergroup variability, which would require identical mouse models being jointly studied by the different centers capable of carrying out these highly specialized procedures, is unknown. However, within a particular group, once a procedure becomes well established, it generates highly reproducible data. In particular, the working heart and closed-chest procedures produce data that show minimal scatter, and highly significant differences can be established with relatively small groups (four to eight pairs) of transgenic and nontransgenic mice. For all of the analyses referred to in the above sections, strict attention to standard experimental practices are critical for minimizing variability. This includes the use of sex- and age-matched transgenic and nontransgenic littermates, with both groups being housed for their entire lives under identical conditions. Grouping data from “dirty” and “clean (microisolator-housed)” mice is not advised. Ideally, all physiological studies should be carried out with the investigators operating in a blinded fashion and with both transgenic and nontransgenic animals being analyzed the same day. Interobservational error can be significant and needs to be controlled. In our experience, the personnel need to be dedicated to certain procedures. If these precautions are taken, all of the analyses outlined above can yield highly reliable data and, if a large number of genetically engineered animals are contemplated, are worth developing on site, despite the significant commitments in training, personnel, and capital costs for the equipment and/or its maintenance. At the minimum, the electrophysiological, blood pressure, and echocardiographic analyses are relatively straightforward and require minimal setup costs, and a large number of animals can be rapidly processed and used in longitudinal studies.
As murine modeling of the cardiovascular system increases, it is important to remember the possible limitations of murine cardiovascular physiology, where the length of the cardiac cycle is 1/10 that of the human. It is apparent, even at this early stage, that the murine models of human cardiac disease, although exhibiting important facets of the phenotype, fail to demonstrate a complete spectrum of the characteristic human pathologies.9 Numerous differences between the human and murine electrophysiological and contractile protein isoform complements exist and undoubtedly underlie the functional differences in normal cardiac function as well as the responses to specific genetic mutations. The study of cardiovascular disease using genetically manipulated models is thus likely to begin to use larger animals, such as rabbits and pigs. Transgenic techniques are established for these animals,35 and the models offer the obvious advantages of larger size and a cardiovascular system that is more closely related to that of the human. They are well suited for experiments that require repeated measures over time, such as measuring outcomes of therapeutic modalities. Both invasive and noninvasive modes of cardiovascular study become less formidable, since technology applicable to human infants weighing as little as 1.5 kg can be applied to a 4-kg rabbit. The obvious disadvantages to large-animal transgenics include increased cost, more complex animal husbandry requirements, and ethical issues. However, large-animal studies confirming findings in at least some of the mouse models will be critical for validating the applicability of the murine experimental models to human cardiovascular disease.
These caveats aside, the mouse will remain the system of choice for germline manipulation of cardiovascular function. The technology of genetic modification is developing rapidly, and proof of principal for both cardiac-specific gene targeting and conditional expression of a transgene in the heart have been demonstrated.36 37 Organ-specific gene targeting will allow one to study gene modifications that have been made only in the cardiac compartment, freeing the analyses from the possible confounding effects the mutations may have had on other interacting organ systems. Conditional transgenesis will allow an investigator to discern not only the effects of a transgene’s action on cardiovascular physiology but also whether the heart can recover once those effects are removed. The heart’s response to a “transgenic bolus” and its plasticity in being able to recover has tremendous potential in defining the events that underlie the compensatory and decompensatory phases of cardiovascular disease. Not only will one be able to model a particular disease state and determine the effects of pharmacological and other treatment modalities on its progression, but the innate ability of the heart to recover from the “disease” can be explored. The rapid dissemination of the already developed techniques for studying murine cardiovascular physiology and continued improvements promise an exciting and productive period for cardiovascular science.
Selected Abbreviations and Acronyms
|FHC||=||familial hypertrophic cardiomyopathy|
|LA||=||left atrium (atrial)|
|LV, RV||=||left and right ventricle (ventricular)|
|LVMe||=||LVM by echocardiography|
|MRI||=||magnetic resonance imaging|
|MyHC||=||myosin heavy chain|
Studies carried out in the authors’ laboratories were supported by National Institutes of Health grants HL-56370, HL-41496, HL-52318, HL-22619, and HL-56620; by the Marion Merrell-Dow foundation (Dr Robbins); and by a National Institutes of Health training grant T32 HL-07825 (Dr James).
- Received August 27, 1997.
- Accepted November 24, 1997.
- © 1998 American Heart Association, Inc.
Christensen G, Wang YB, Chien KR. Physiological assessment of complex cardiac phenotypes in genetically engineered mice. Am J Physiol. 1997;41:H2513–H2524.
Grupp IL, Subramaniam A, Hewett TE, Robbins J, Grupp G. Comparison of normal, hypodynamic, and hyperdynamic mouse hearts using isolated work-performing heart preparations. Am J Physiol. 1993;265:H1401–H1410.
Luo W, Grupp IL, Harrer J, Ponniah S, Grupp G, Duffy JJ, Doetschman T, Kranias EG. Targeted ablation of the phospholamban gene is associated with markedly enhanced myocardial contractility and loss of β-agonist stimulation. Circ Res. 1994;75:401–409.
Milano CA, Allen LF, Rockman HA, Dolber PC, McMinn TR, Chien KR, Johnson TD, Bond RA, Lefkowitz RJ. Enhanced myocardial function in transgenic mice overexpressing the beta 2-adrenergic receptor. Science. 1994;264:582–586.
Palermo J, Gulick J, Colbert M, Fewell J, Robbins J. Transgenic remodeling of the contractile apparatus in the mammalian heart. Circ Res. 1996;78:504–509.
Kumar A, Crawford K, Close L, Madison M, Abbas Z, Lorenz J, Doetschman T, Pawlowski S, Duffy J, Neumann J, Robbins J, Boivin G, O’Toole B, Lessard JL. Rescue of cardiac actin deficient mice by enteric actin. Proc Natl Acad Sci U S A. 1997;94:4406–4411.
Geisterfer-Lowrance AA, Christe M, Conner DA, Ingwall JS, Schoen FJ, Seidman CE, Seidman JG. A mouse model of familial hypertrophic cardiomyopathy. Science. 1996;272:731–734.
Hunter JJ, Tanaka N, Rockman HA, Ross JJ, Chien KR. Ventricular expression of a MLC-2v-ras fusion gene induces cardiac hypertrophy and selective diastolic dysfunction in transgenic mice. J Biol Chem. 1995;270:23173–23178.
Iwase M, Bishop SP, Uechi M, Vatner DE, Shannon RP, Kudej RK, Wight DC, Wagner TE, Ishikawa Y, Homcy CJ, Vatner SF. Adverse effects of chronic endogenous sympathetic drive induced by cardiac Gsα overexpression. Circ Res. 1996;78:517–524.
Gottshall KR, Hunter JJ, Tanaka N, Dalton N, Becker KD, Ross J Jr, Chien KR. Ras-dependent pathways induce obstructive hypertrophy in echo-selected transgenic mice. Proc Natl Acad Sci U S A. 1997;94:4710–4715.
Lorenz J, Robbins J. Measurement of intraventricular pressure and cardiac performance in the intact closed chest anesthetized mouse. Am J Physiol. 1997;272:H1137–H1146.
Turki J, Lorenz JN, Green SA, Donnelly ET, Jacinto M, Liggett SB. Myocardial signaling defects and impaired cardiac function of a human beta 2-adrenergic receptor polymorphism expressed in transgenic mice. Proc Natl Acad Sci U S A. 1996;93:10483–10488.
Colbert MC, Hall G, Kimball TR, Witt SA, Lorenz JA, Kirby ML, Hewett TE, Klevitsky R, Robbins J. Cardiac compartment specific overexpression of a modified RAR produces dose dependent changes in systolic function, dilated cardiomyopathy and congestive heart failure in transgenic mice. J Clin Invest. 1997;100:1958–1968.
Berul CI, Aronovitz MJ, Wang PJ, Mendelsohn ME. In vivo cardiac electrophysiology studies in the mouse. Circulation. 1996;94:2641–2648.
Manning WJ, Wei JY, Katz SE, Litwin SE, Douglas PS. In vivo assessment of LV mass in mice using high-frequency cardiac ultrasound: necropsy validation. Am J Physiol. 1994;266:H1672–H1675.
Gardin JM, Siri FM, Kitsis RN, Edwards JG, Leinwand LA. Echocardiographic assessment of left ventricular mass and systolic function in mice. Circ Res. 1995;76:907–914.
Tanaka N, Dalton N, Mao L, Rockman HA, Peterson KL, Gottshall KR, Hunter JJ, Chien KR, Ross JJ. Transthoracic echocardiography in models of cardiac disease in the mouse. Circulation. 1996;94:1109–1117.
Iwase M, Uechi M, Vatner DE, Asai K, Shannon RP, Kudej RK, Wagner TE, Wight DC, Patrick TA, Ishikawa Y, Homcy CJ, Vatner SF. Cardiomyopathy induced by cardiac Gs alpha overexpression. Am J Physiol. 1997;272:H585–H589.
Srinivasan S, Artman M, Baldwin HS, Turnbull DH. High frequency ultrasound study of murine embryonic heart development in utero. Circulation. 1997;96(suppl I):I-685. Abstract.
Keller BB, MacLennan MJ, Tinney JP, Yoshigi M. In vivo assessment of embryonic cardiovascular dimensions and function in day-10.5 to -14.5 mouse embryos. Circ Res. 1996;79:247–255.
Siri FM, Jelicks LA, Leinwand LA, Gardin JM. Gated magnetic resonance imaging of normal and hypertrophied murine hearts. Am J Physiol. 1997;272:H2394–H2402.
Kubota T, McTiernan CF, Frye CS, Slawson SE, Lemster BH, Koretsky AP, Demetris AJ, Feldman AM. Dilated cardiomyopathy in transgenic mice with cardiac-specific overexpression of tumor necrosis factor-α. Circ Res. 1997;81:627–635.
Mansier P, Medigue C, Charlotte N, Vermeiren C, Coraboeuf E, Deroubai E, Ratner E, Chevalier B, Clairambault J, Carre F, Dahkli T, Bertin B, Briand P, Strosberg D, Swynghedauw B. Decreased heart rate variability in transgenic mice overexpressing atrial beta 1-adrenoceptors. Am J Physiol. 1996;271:H1465–H1472.
Uechi M, Asai K, Osaka M, Smith A, Sato N, Wagner TE, Ishikawa Y, Hayakawa H, Vatner DE, Shannon RP, Homcy CJ, Vatner SF. Depressed heart rate variability and arterial baroreflex in conscious transgenic mice with overexpression of cardiac Gsα. Circ Res.. 1998;82:416–423.
Fewell JG, Osinska H, Klevitsky R, Ng WE, Sfyris G, Bahrehmand F, Robbins J. A treadmill exercise regimen for identifying subtle cardiovascular phenotypes in transgenic mice overexpressing myosin regulatory light chain 2v. Am J Physiol. 1997;273:H1595–H1605.
Krege JH, Hodgin JB, Hagaman JR, Smithies O. A noninvasive computerized tail-cuff system for measuring blood pressure in mice. Hypertension. 1995;25:1111–1115.
Kaplan ML, Cheslow Y, Vikstrom K, Malhotra A, Geenen DL, Nakouzi A, Leinwand LA, Buttrick PM. Cardiac adaptations to chronic exercise in mice. Am J Physiol. 1994;267:H1167–H1173.
Desai KH, Sato R, Schauble E, Barsh GS, Kobilka BK, Bernstein D. Cardiovascular indexes in the mouse at rest and with exercise: new tools to study models of cardiac disease. Am J Physiol. 1997;272:H1053–H1061.
Gulick J, Hewett TE, Klevitsky R, Buck S, Moss RL, Robbins J. Transgenic remodeling of the regulatory myosin light chains in the mammalian heart. Circ Res. 1997;80:655–664
Yu Z, Redfern CS, Fishman GI. Conditional transgene expression in the heart. Circ Res. 1996;79:691–697.