INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

GENETICALLY ENGINEERED MICE as specific cardiovascular disease models have become invaluable instruments for understanding cardiac pathophysiology (7). Techniques such as transverse aortic constriction (33), induction of ischemia-reperfusion, and myocardial infarction (19) have been adapted to the smaller dimensions in the mouse and are now readily employed in this species. However, the physiological cardiovascular phenotype of the mouse strain C57BL/6, which is most abundantly used as the background strain of genetically engineered mice, is still incompletely characterized. Most notably, developmental changes in left ventricular (LV) dimensions and mass, heart rate (HR), blood pressure, and contractility of ventricular myocytes need to be further evaluated. It was the primary aim of this study to check for age- and gender-dependent changes of cardiovascular parameters in C57BL/6 mice and for the underlying mechanisms. These data are of particular interest for comparison of homozygously bred mutant mice with C57BL/6 as the wild type rather than for comparison of mutant mice (/ or +/) bred from heterozygous parents with their homozygous +/+ littermates. Littermates would unequivocally provide the closest genetic background and best control but are not always available in sufficient numbers.

Because transgenic mice or mice mutated by targeted gene disruption are usually rare early after generation, it is desirable to analyze cardiac size and functional parameters without invasive procedures (10). High-resolution echocardiography (EC) allows noninvasive assessment of LV dimensions and mass (18, 31). However, the reproducibility of this technique is somewhat limited by the small size of the mouse and the high HR. Although LV myocardial mass in EC correlates with necropsy heart weight (5), the correlation of LV myocardial mass in the developing heart determined by EC with that determined by extremely reliable histomorphometry (HM) has not been systematically studied. Therefore, the second aim of this study was to evaluate the accuracy of EC to study cardiac dimensions in mice at defined ages and compare three common calculation algorithms for LV myocardial mass based on one- or two-dimensional EC recordings with the corresponding HM reference measurements.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal housing and body weight. All animal procedures were approved by the local committee for animal care and conducted in conformity with the "Guiding Principles for Research Involving Animals and Human Beings" established by the American Physiological Society. C57BL/6 mice of either sex (Charles River; Sulzfeld, Germany) were housed at 24°C with a 12:12-h light-dark cycle. C57BL/6 mice (35 male and 34 female) were prospectively included in this study and euthanized on postnatal days 21, 30, 50, 70, and 111 (P21, P30, P50, P70, and P111, respectively). A senescence group (6 male and 4 female) was investigated on postnatal day 580 (P580). Arterial blood pressure measurements, electrocardiography (ECG) recordings, LV muscle mass (LVM) evaluation, and ECs were performed on P21, P30, P50, P70, P111, and P580. Body weight was determined from P21 to P50 each 3rd day, from P50 to P75 each 5th day, and from P75 to P111 each 14th day. The senescence group was weighed on P580.

Arterial blood pressure. Mice were anesthetized with enflurane (3% for induction or surgical procedures and 0.8-1% for blood pressure measurements) in 70% nitrous oxide and 30% oxygen by face mask. The degree of anesthesia was adapted to minimize cardiodepressive effects. Thus the HR during anesthesia was maintained in the physiological range of conscious mice under resting conditions (21, 27). The core temperature was maintained at 37°C using a feedback-controlled heating pad. Mean arterial blood pressure (MABP), systolic blood pressure (SBP), and diastolic blood pressure (DBP) were recorded in the left carotid artery with an 8-cm-long fluid-filled polyethylene-10 catheter (pressure transducer, Braun; Melsungen, Germany; and Servomed blood pressure monitor, Hellige; Freiburg, Germany).

Electrocardiography. Lead II surface ECG was recorded simultaneously to arterial blood pressure recordings with a Servomed ECG monitor (Hellige). HR, sinus cycle length (SCL), PQ, QRS, and QT periods were evaluated as previously described (16). QRS was measured from the Q-wave onset to the return of the S-wave to the isoelectric line, and the QT interval was defined from the Q-wave onset to the final return of the uni- or bipolar T-wave to the isoelectric line (Fig. 1). The QT was rate corrected (QT_c) according to Mitchell (28).

View larger version (10K): [in this window] [in a new window]   Fig. 1.   Representative lead II surface electrocardiogram (ECG) from an adult C57Bl/6 mouse at postnatal day 111 (P111) with an explanation of the ECG parameter definition. The heart rate (HR) in the example shown is 580 beats/min. The T-wave is attached to the S-wave and is often bipolar.

Echocardiography. For comparison, EC and HM were performed on identical mice on the same day. EC was performed using a commercially available ultrasound device equipped with a new linear array transducer operating at an emission frequency of 15 MHz with frame rates up to 280 Hz (HDI-5000, Philips Medical Systems; Bothell, WA). Animals were anesthetized, and the HR was monitored as described above. Parasternal short- and long-axis views were obtained in two-dimensional B-mode. At least 20 cardiac cycles were obtained in B-mode imaging for each view, and each imaging plane was acquired three times to assess reproducibility. One-dimensional M-mode imaging was performed two-dimensionally guided in the parasternal short-axis view at the level of the papillary muscle. Parasternal short-axis views were divided into six segments, and long-axis views were divided into seven segments (4). Imaging was considered adequate if the endocardial and epicardial borders could be properly visualized in five or more segments. The endocardial borders were manually traced on the innermost endocardial edge while the epicardial borders were defined by tracing along the first bright pixel adjacent to myocardial tissue (5). LV end-diastolic and end-systolic internal chamber diameters (LVEDD and LVESD, respectively) and wall thickness was assessed from M-mode traces (leading edge to trailing edge). End-diastolic measurements were obtained at the peak of the R-wave, whereas end-systolic measurements (M-mode and B-mode) were obtained at the time of minimum internal chamber dimensions.

(1)

(2)

(3)

LVM histomorphometry. Hearts were excised after blood pressure and ECG recordings and perfused in a Langendorff apparatus with Ca²⁺-free Tyrode solution (in mM: 135 NaCl, 4 KCl, 1 MgCl₂, 2 HEPES, and 2.6 EGTA; pH 7.4 and perfusion pressure 5 kPa) to yield myocardial relaxation and to maintain the heart under reproducible conditions for freezing in liquid N₂. Transversal cryostat sections (20 µm thick) were taken in 1-mm steps parallel to the atrioventricular groove starting 0.5 mm from the apex. After the sections were stained with hematoxylin and eosin (H-E), magnified and calibrated images of the sections were generated with a charge-coupled device camera. Images were analyzed morphometrically 1) for myocardial wall thickness at the same position as that used for in vivo EC and 2) for LV mass using three-dimensional reconstruction (Fig. 2) with image analysis software (Scion Image 4; Frederick, MD). Tibia length was measured to calculate individual heart weight-to-tibia length ratios.

View larger version (100K): [in this window] [in a new window]   Fig. 2.   A: representative image of hematoxylin-eosin-stained heart section stack viewed from the ventricle apex toward valves. The frontal section represents the position of in vivo echocardiography (EC) 4.5 mm from the apex, whereas the more apical sections of this three-dimensional file are omitted. The left ventricular (LV) cavum is colored red, and the right ventricular (RV) cavum is colored blue. B: EC M-mode sweep obtained in the corresponding imaging plane of the same heart in vivo. C: short-axis view in the EC B-mode; D: long-axis view in the EC B-mode. E and F: position of the measured areas and distances. A1 and A2, epicardial and endocardial short-axis areas; L, length in the long-axis view; a, full major radius; b, minor axis radius of the internal diameter of the LV at the tip of the papillary muscle (as assessed in the long-axis view); d, truncated major radius.

Sarcomere shortening. For the recording of sarcomere shortening, isolated ventricular myocytes were prepared from female mice of four age groups (±SD; group 1: P21 ± 1, n = 4; group 2: P93 ± 4, n = 4; group 3: P169 ± 4, n = 7; and group 4: P362 ± 2, n = 4). Ventricular myocytes were isolated enzymatically as previously described for guinea pigs, rats, and rabbits (24). Briefly, animals were euthanized and the hearts excised. Hearts were prepared in Tyrode solution with EGTA instead of CaCl₂ (in mM: 135 NaCl, 4 KCl, 1 MgCl₂, 2 HEPES, 2.6 EGTA, and 10 glucose and 1 mg/ml BSA; pH 7.4) and mounted in a Langendorff perfusion system. Pressure was adjusted to 0.05 bar and temperature to 36°C. Hearts were perfused with Tyrode solution with EGTA for 5 min and, subsequently, with high-K⁺ solution for 5 min [in mM: 4 NaCl, 10 KCl, 130 K-glutamate, 1 MgCl₂, 0.05 CaCl₂, 2 HEPES, and 10 glucose and 1 mg/ml BSA; pH 7.4 (KOH)] before enzymes were added to the high-K⁺ solution. The hearts were perfused first for 8-10 min with trypsin (1,000 Bae units/40 ml, Roche Molecular Biochemicals; Mannheim, Germany) and second for 10-13 min with collagenase (type L, 25 mg in 40 ml, Sigma; St. Louis, MO) and finally sectioned in small parts, which were directly transferred into normal Tyrode solution (composition as described above) but with 1.8 mM CaCl₂ instead of EGTA plus trypsin inhibitor (0.167 mg/ml, Sigma). In this solution, the pieces were disintegrated completely by stirring with glass rods. The solution was passed through a filter (125 µm mesh width) and centrifuged gently. The cells were taken from the pellet. For the isolation of myocytes from the hearts of P21 mice, enzyme concentrations and perfusion times were reduced by one-half. Isolated cells were kept in oxygenated standard Tyrode solution (in mM: 135 NaCl, 4 KCl, 1.8 CaCl₂, 1 MgCl₂, 2 HEPES, and 11 glucose and 1 mg/ml BSA; pH 7.4) at 22°C up to 6 h until use. Sarcomere shortening of ventricular myocytes was recorded with a video imaging system and SarcLen software (IonOptix; Milton, MA). The regular striation pattern of the sarcomeres is analyzed by fast Fourier transformation. Sarcomere shortening shifts the power spectrum peak, which corresponds to the absolute sarcomere length (23). The video system was mounted to an inverted microscope (Zeiss IM 35; Jena, Germany; lens, Neofluar ×40, 0.75) equipped with an experimental chamber with permanent perfusion of Tyrode solution heated to 36°C. Contractions were induced by bipolar external stimuli (0.4 ms, 30 V, SD9, Grass; Quincy, MA). Stimuli were applied in 20-s pulse trains interrupted by a 30-s stimulation pause. The stimulation protocol was as follows: 0.5, 10, 1, 8, 2, 6, and 4 Hz.

(4)

Statistical analysis. Values are presented as means ± SE. The SE of the estimate (SEE) is given for linear regression. Statistical comparisons were performed by unpaired Student's t-test (two-tail criterion) or two-way ANOVA, followed by post hoc Student's t-tests for multiple comparisons. P < 0.05 was considered significant. Bland-Altman analysis (bias plot) was performed to assess the agreement between EC and HM (bias ± 2SD) (1). Linear regression and ANOVA were calculated with GraphPad Prism 3.03 (GraphPad Software; San Diego, CA).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Developmental changes of body weight and tibia length. The body weights of male and female mice increased continuously from 10.2 ± 0.3 and 9.8 ± 0.3 g at P21 to 33.2 ± 1.2 and 29.4 ± 1.8 g at P580, respectively. In female mice, the body weight increased slower until P60. Between P60 and P580, the body weights of adult male and female mice increased at a comparable rate by 40.1% and 44.1%, respectively. The body weight of male mice was higher at any given time (Fig. 3A). Tibia length developed without gender differences, but growth slowed down after P70 (Fig. 3B). The heart weight-to-body weight ratio remained constant but was higher in female mice except for P580 (5.00 ± 0.19 in females vs. 4.25 ± 0.09 in males, P < 0.006; Fig. 3C), whereas the heart weight-to-tibia length ratio increased until P50 and did not show gender differences (Fig. 3D).

View larger version (20K): [in this window] [in a new window]   Fig. 3.   A: development of body weight (BW) in male and female C57BL/6 mice. B: development of tibia length in male and female mice. C: calculated heart weight (HW)-to-BW ratio (HW/BW). D: calculated HW-to-tibia length ratio. Values are means ± SE of 8-15 mice.

Developmental changes of arterial blood pressure. SBP, DBP, MABP, and HR increased until P50 in either sex and then remained constant until P580 (Fig. 4 and Table 1). Arterial blood pressure did not differ between male and female mice at any age.

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Mean arterial blood pressure (BP) of male and female C57BL/6 mice over the life span from P21 to P580. BP was measured using a liquid-filled catheter in the left common carotid artery connected to a pressure transducer. Values are means ± SE of 8-15 mice.

Developmental changes of HR, atrioventricular conduction, and repolarization. HR increased from 396 ± 16 beats/min in P21 to 570 ± 31 beats/min in P70 mice, whereas the SCL was shorter in young mice, accordingly. PQ decreased from P21 to P50 and remained constant from then on. QRS remained constant during the life span (Table 1). QT decreased from P21 to P50, consistent with the parallel increase in HR, whereas the QT_c values, which were calculated using a modified Bazett equation to account for the high resting HR in mice (28), remained constant (Table 1).

Developmental changes of LV wall dimensions determined by EC and HM. LV wall thickness determined by B-mode-guided M-mode acquisition correlated strongly with HM (y = 0.89x + 0.052, R = 0.79, P < 0.0001, SEE: 0.078 mm; Fig. 5, A-C). No significant differences were found between anteroseptal and posterior wall dimensions in EC or HM. Wall dimensions could be accurately assessed at all ages, even at P5 (data not shown).

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Comparison of LV wall dimensions (A-C) and LV muscle mass (LVM; D-F) determined by or calculated from EC or histomorphometry (HM). A and D: absolute LV wall thickness (A) or calculated LV myocardial mass (D). Values are means ± SE of 6-12 mice. B: correlation of EC measurements in B-mode-guided M-mode acquisitions on the y-axis with HM measurements on the x-axis (y = 0.89x  0.052, R = 0.79, P < 0.0001). C and F: Bland-Altman blots comparing EC- and HM-based determinations of wall dimensions (C) or LVM calculations (F). E: correlation of EC measurement-based LVM calculation (area-length algorithm) on the y-axis with HM measurement-based LVM calculation on the x-axis (y = 1.11x  6.45, R = 0.93, P < 0.001).

Developmental changes of LVM calculated from EC and HM. Linear regression analysis revealed the strongest correlation between LVM_HM and diastolic LVM_Area-Length (y = 1.11x  6.45, R = 0.93, P < 0.001; Fig. 5E). The SEE was calculated as 7.01 mg for LVM_Area-Length. LVM_PENN underestimated LVM substantially; thus correlation between LVM_HM and LVM_PENN was weaker (y = 0.84x  10.05, R = 0.84, P < 0.001). Although the truncated-ellipsoid method overestimated LVM systematically compared with the HM reference, a strong positive linear correlation was also found for this algorithm (y = 1.41x  3.70, R = 0.86, P < 0.001). All systolic measurements correlated less with LVM_HM, with the highest correlation between LVM_Area-Length and LVM_HM (y = 1.42x  4.90, R = 0.87, P < 0.001).

Developmental changes of sarcomere shortening. Sarcomere shortening of isolated ventricular cardiac myocytes was monitored to compare the detected age dependency of MABP with potential changes in the myocardium. In age groups 1-4, resting sarcomere length ranged from 1.84 to 1.89 µm (group 1: 1.862 ± 0.006 µm, n = 9; group 2: 1.867 ± 0.008 µm, n = 20; group 3: 1.890 ± 0.008 µm, n = 13; and group 4: 1.843 ± 0.005 µm, n = 12; means ± SE). Significant differences could be shown between most groups (between groups 1 and 3, 1 and 4, 2 and 4, and 3 and 4), indicating an increase of sarcomere length with age in young animals, followed by a decrease in advanced age.

View larger version (13K): [in this window] [in a new window]   Fig. 6.   Averaged sarcomere shortening amplitude of isolated ventricular myocytes of female mice plotted versus the stimulation frequency. A: data points refer to the four age groups (AG) investigated. The measuring points of age groups 2-4 (adult animals) were not significantly different. However, most shortening amplitudes of age group 1 differ from those of the older animals (0.5 Hz: group 1 to 3; 1 Hz: group 1 to 2 and 3; 4 Hz: group 1 to 3 and 4; 6 Hz: group 1 to 2, 3, and 4; and 8 Hz: group 1 to 2, 3, and 4). B: data of age groups 2-4 were joined and compared with age group 1. The shortening amplitudes of group 1 were significantly smaller than those of the adult animals at all frequencies except for 2 and 10 Hz. Values are means ± SE of 6-20 cells. *P < 0.05.

View larger version (33K): [in this window] [in a new window]   Fig. 7.   Averaged sarcomere shortening records after a 30-s stimulation rest. Left, shortening trains of age group 1; right, shortening trains of age group 2. Each graph displays 20 shortenings; therefore, the time scales are different. The peaks of shortening ± SE were fitted by Eq. 4. A: stimulation frequency of 0.5 Hz; age group 1, average of 10 cells; age group 2, average of 8 cells. B: stimulation frequency of 6 Hz; age group 1, average of 5 cells; age group 2, average of 4 cells. C: stimulation frequency of 10 Hz; age group 1, average of 6 cells; age group 2, average of 9 cells.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Although pathophysiological processes found in cardiac disease have been studied extensively with the use of cellular and molecular biology in mice, little is known about early postnatal heart development. Therefore, the aim of this study was to investigate the developmental changes of morphological and functional cardiac parameters including the early postnatal phase.

Determination of HR and blood pressure in mice is difficult because either parameter responds instantly to any kind of stressful situation. HR reported in adult P70-P120 Swiss-Webster mice using a telemetric blood pressure transponder ranged between 450 and 500 beats/min in awake resting mice to between 600 and 650 beats/min after light physical activity without manipulation (21). Telemetric ECG recording in conscious adult Swiss mice at light activity revealed a sinus rhythm of 630 beats/min (30). HR higher than 700 beats/min occurs only after manipulation such as weighing, restraining, or cage changing (21). Thus our HR recordings under mild enflurane anesthesia, which, in mice of a similar age, ranged from 570 beats/min in P70 to 581 beats/min in P111, are well consistent with those in awake resting animals. Deep anesthesia yielded by 3-4% enflurane or pentobarbital (0.05-0.08 mg/g body wt) or ketamine plus xylazine (0.08 mg/g plus 0.12 mg/g body wt, respectively) reduces HR substantially (<300 beats/min) due to negative chronotropic and inotropic effects (36). Mouse HR increased during postnatal development until P50, which is consistent with previous findings (34). We did not observe gender differences in HR or any other ECG parameter. PQ, PR, and QT were longer in young mice (P21 and P30), whereas QRS remained constant over the life span. For explanation of age-dependent decreases in atrioventricular conduction as well as in repolarization time, in vitro electrophysiological recordings are needed.

HM using image analysis of H-E-stained cryostat sections as applied in this study has not been used so far to determine myocardial mass. However, the analysis of cryostat sections is a common technique to precisely calculate brain infarct volume in mice (11, 12) and also allows precise determination of LV myocardial mass. Because this technique is neither affected by blood and buffer remainders nor by inconsistent ventricle preparation, it allows more reliable and reproducible ventricular myocardial mass determinations than wet heart weight.

We calculated heart weight-to-body weight ratios for standardized myocardial mass assessment. However, the index was always lower in male mice except for the maximal age tested (P580). This gender difference results from higher body weights of male mice throughout life. Whereas the heart weight-to-body weight ratio was clearly gender dependent, we also calculated heart weight-to-tibia length ratios, which are supposed to be less affected by gender. Heart weight-to-tibia length, however, increases continuously with growth and remains constant after maturity, as shown in rats (37), because the myocardial mass in growing mice of either sex increases numerically faster than the tibia length. Thus, for mice younger than P70, gender-specific heart weight-to-body weight ratios appear to be most advantageous, whereas in adult mice beyond P70 heart weight-to-tibia length appears to be a more reliable index.

The best agreement between HM three-dimensional integration and EC-based mass calculation algorithms was found for the area-length algorithm. Correlation between EC data and HM data was even better than for wall dimensions. This discrepancy could be expected because mass calculations that rely on measurements of areas rather than distances are not as sensitive to measurement artifacts caused by incorrect placement of the measurement cursor (20).

We found an excellent correlation between wall dimensions by EC and HM. Best agreement was found between end-diastolic EC data and HM data obtained from hearts fixated in diastole. Repeated measurements delivered reproducible records. Epicardial and endocardial structures could be clearly differentiated both in M-mode and B-mode imaging. Papillary muscle and the mitral valve could be easily identified. Endocardial border definition was possible in the majority of all imaging sequences and allowed reliable measurements even in small hearts.

HM measurements in vitro revealed larger cavity dimensions than in vivo EC. Hearts prepared for HM were perfused with a Ca²⁺ chelating buffer at 5 kPa, resulting in maximal dilatation, whereas in beating hearts diastolic relaxation is always incomplete.

The data presented in this paper demonstrate the feasibility of high-resolution EC in young, adult, and elderly mice, allowing the assessment of developmental changes in LV dimensions and mass. The high-frequency ultrasound transducer used in this study allows vertical resolution of up to 90-100 µm in B-mode imaging. Because of HR between 400 and 600 beats/min in mice, high frame rates in EC are required for measurements in defined phases of the cardiac cycle. The ultrasound system used provides frame rates up to 280 Hz, which provides sufficient temporal resolution to study functional changes adequately (31). Temporal and spatial resolution of EC are therefore within the range of high-resolution MRI (35). EC as a real-time imaging technique may be advantageous over MRI in some scenarios because it does not require proper image triggering by ECG, which frequently hampers imaging by MRI.

Arterial blood pressure increased during development until P50 and remained constant after this age. Increasing blood pressure during maturation is a novel finding in mice, but a well-known phenomenon in human cardiovascular development (32). Gender differences were not detected. SBP (127.5-130.2 mmHg), MABP (110.4-115.7 mmHg), and DBP (93.4-102.4 mmHg) values found in adult C57BL/6 mice (P50-P111) were consistent with telemetric measurements reported for adult male Swiss-Webster mice during light activity (110-124 mmHg) (21, 27). The 24-h average MABP reported for conscious C57Bl/6 at P38 (110 mmHg) (25) ranges between our MABP measurements in P30 (97.5 mmHg) and P50 (110.4 mmHg). Enflurane initially increases cytosolic Ca²⁺ concentration ([Ca²⁺]_i) in vitro in isolated cardiac myocytes by inhibition of the Na⁺/Ca²⁺ exchanger (15) or by increased Ca²⁺ release from the sarcoplasmic reticulum (5). However, >1.2% enflurane as well as isoflurane reduce blood pressure and HR in mice concentration dependently (unpublished data), indicating that in vivo, during the steady state, anesthesia [Ca²⁺]_i-decreasing effects, i.e., inhibition of Ca²⁺ influx through L-type Ca²⁺ channels (2) or depletion of intracellular Ca²⁺ stores, prevail over [Ca²⁺]_i-increasing effects, i.e., release from intracellular stores and/or decreased Ca²⁺ outward transport.

Blood pressure is determined by both cardiac output (HR × stroke volume) and total peripheral resistance. Because HR was lower in P21 and P30 compared with adult animals, we decided to also monitor myocardial contractility. Cardiac contractility was recorded as sarcomere shortening in isolated myocytes during zero load shortening. Sarcomere shortening is superior to cell shortening, which is a more commonly recorded parameter, because it is independent of cell size and thus allows direct comparison between cells (29). Zero load shortening of isolated myocytes is monitored widely as assessment for cardiac contractility (8). Concerning the basic contractile characteristics like force-frequency relation and postrest behavior, the results of zero load shortenings are comparable to force recordings of cardiac muscle strips, e.g., changes in the shortening-frequency relation also appeared in the force-frequency relation (3).

The sarcomere length found in this study (between 1.84 and 1.89 µm, depending on the age group) proves that the cells used were well maintained. A clearly shorter sarcomere length of ~1.72 µm in murine cardiomyocytes has been reported by Lim et al. (22) (1.72 µm in 5-mo-old and 1.71 µm in 34-mo-old B6C3F1 hybrid mice), which may result from a different cell isolation protocol. In rats, a decrease in resting sarcomere length with age has been reported when comparing three age groups, 2 mo (1.85 µm), 6-9 mo (1.83 µm), and 24-25 mo (1.82 µm) (13). Such a distinct age dependency was not detected here, likely due to the younger age of the animals investigated in this study. However, the sarcomere length in age group 4 (362 days) was also significantly shorter than that in the younger animals.

HR from 550 to 600 beats/min, as reported here in adult mice P50, corresponds to frequencies from 9 to 10 Hz. We therefore recorded shortening up to 10 Hz, which is the highest contraction rate we could achieve in isolated murine cardiomyocytes without interference with their refractory time. We found a biphasic shortening-frequency relation with a negative correlation below 2 Hz and positive correlation above. In adult cardiomyocytes, the shortening amplitude saturated at 6 Hz. This agrees with the shortening-frequency relation of 4-mo-old C57BL/6 mice described by Lim et al. (22). A biphasic force-frequency relation was also obtained in papillary muscle of 20-wk-old Black Swiss mice (15). Maximal sarcomere shortening at 6 Hz was 4.4% here, and maximal cell shortening at 6 Hz was in the same range in the study of Lim et al. (22). Thus the values for adult cells recorded in this study appear to be reliable. In young myocytes (age group 1), we obtained a smaller shortening amplitude in combination with a less steep shortening-frequency relation. Because sarcomere shortening is independent of cell size, the observed reduced shortening is an intrinsic characteristic of immature myocytes, which will result in a reduced fractional shortening. Young myocytes proved to be more sensitive to the enzymatic isolation procedure; therefore, we reduced both the enzyme concentration and application times (see METHODS). The isolated young myocytes then exhibited healthy morphology, a sarcomere length above 1.84 µm, only a small amount of spontaneously beating cells, and could be paced up to 10 Hz. However, it cannot be excluded that the reduced sarcomere shortening in these myocytes may be influenced by the isolation procedure.

Postrest behavior of murine cardiomyocytes is intriguing, because they develop a negative staircase at low pacing frequencies that turns into a positive staircase at high frequencies. The negative staircase in hearts and tissue strips of small rodents has been well known for many decades. However, the change of a negative staircase at low pacing frequency to a positive one at high frequency has not yet been reported. The reason for this change may be an increasing role of calcium sequestration by the sarcoplasmic reticulum at high pacing frequencies, because manipulations of sarco(endo)plasmic reticulum Ca²⁺-ATPase 2 and phospholamban change the force-frequency relation (26). A frequency-dependent Ca²⁺ "sensitization" of the myofilaments may also cause this effect (14). In this context, it is interesting that positive staircases appeared only at pacing frequencies of 8 and 10 Hz in cells of age group 1, whereas the staircase turned to a positive one at 6 Hz in older animals. This may reflect the less steep shortening-frequency relation in age group 1.

In conclusion, EC allows precise noninvasive measurement of LV dimensions in P21-P580 mice and yields reliable LV mass estimates in these mice if the area-length algorithm is employed. The cardiac contribution to the arterial blood pressure increase during postnatal maturation until P50 results from 1) an increase in HR and, subsequently, cardiac output; 2) an increase in shortening amplitude at most frequencies and a steeper shortening-frequency relation; and 3) an elevated shortening amplitude that is, due to the positive shortening-frequency relation, induced by the age-dependent HR increase.