Age-associated alterations of cardiac structure and function in the female F344xBN rat heart

Animals

All procedures were performed in accordance with the Guide for the Care and Use of Laboratory Animals as approved by the Council of the American Physiological Society and the Animal Use Review Board of Marshall University. All procedures were conducted in strict accordance with the Public Health Service Animal Welfare Policy. Virgin adult (6-month), aged (26-month), and very aged (30-month) female Fischer 344/NNiaHSd × Brown Norway/BiNia F1 hybrid (F344xBN) rats were obtained from the National Institute for Aging and housed two per cage in an Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC) approved vivarium. Animal ages were chosen based on survivability curves from the National Institute of Aging to approximate females in the third, seventh, and eighth decades of life. Given that previous data have demonstrated a complete loss of cyclicity at 16 months of age, the estrous phase was not monitored (PMID 17460359). Housing conditions consisted of a 12 h light–12 h dark cycle and temperature was maintained at 22 ± 2 °C. Animals were provided with food and water ad libitum. Rats were allowed to recover from shipment for at least 2 weeks before experimentation during which time the animals were carefully observed and weighed weekly. Any of the rats showing signs of failure to thrive, such as precipitous weight loss, disinterest in environment, or unexpected gait alterations were excluded from the study.

Echocardiographic procedures

Rats (6, 26, and 30 months) were anesthetized using a cocktail of ketamine (40 mg/kg) and xylazine (10 kg/mg) which was injected into the intraperitoneal cavity in order to perform echocardiograms. Echocardiographic procedures were done as previously described by Walker and colleagues (Walker et al. 2006). In order to prevent disturbances of ultrasound waves the ventral thorax was shaved and the rats were placed either on their backs or left side and covered with ultrasonic transmission gel. A Phillips 5500 Echocardiographic system with a 12-MHz transducer was used to take two-dimensional echocardiographic measurements, two-dimensional guided M-mode, Doppler M-mode, and parasternal long- and shot-axis views. Parasternal long- and short-axis views were used to determine two-dimensional cardiac structural measurements. The echocardiographic views were then used to position the M-mode echocardiographic line. Valvular blood flow velocities were evaluated using pulse wave Doppler with the probe toward the apex (x-axis) and the depth along the y-axis (long axis procedure). Positioning the probe toward the left ventricle and across the heart made during short axis procedures made it possible to evaluate wall structure in order to calculate ejection fraction and fractional shortening during systole. A digital echocardiographic analysis system was used to analyze M-mode displays. All echocardiography procedures and parameters were measured by the same echocardiogram technician to limit inter-observer variability.

Left ventricular mass (LVM) was calculated according to the following equation on the basis of previous reports demonstrating a good correlation (r = 0.78, SEE=0.124, p < 0.0001) between calculated and postmortem LV mass (Litwin et al. 1994, 1995).

where IVSd is the left ventricular septal thickness diastole, LVIDd is the left ventricular internal dimension during diastole, and PWTd is the posterior wall thickness diastole (Aigbe et al. 2012).

Serum collection

During tissue collection blood was collected by cardiac puncture into a BD Vacutainer serum collection tube. The blood was centrifuged at 800 × g for 15 min to separate serum. The serum was collected and used to measure serum parameters using an Abaxis VetScan analyzer (Abaxis, Union City, CA, USA). The following parameters were determined non-fasting serum glucose (GLU), alanine aminotransferase (ALT), alkaline phosphatase (ALP), blood urea nitrogen (BUN), albumin (ALB), calcium (Ca²⁺), creatinine (CRE), amylase (AMY), globulin (GLOB), potassium (K⁺), sodium (Na⁺), phosphorus (PHOS), total bilirubin (TBIL), and total protein (Drexler et al. 2003).

Heart collection

Rats were anesthetized with an intraperitoneal injection of ketamine (40 mg/kg) and xylazine (10 mg/kg) and supplemented as necessary for reflexive response. Before heart collection, a three lead electrocardiogram (EKG) was performed on all animals using the Biopac Student Lab software (BIOPAC Systems, Inc., Microsoft). After completion of the EKG, the heart was removed after a midline laparotomy and placed in Krebs–Ringer bicarbonate buffer (KRB) containing the following: 118 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl₂, 1.2 mM KH₂PO₄, 1.2 mM MgSO₄, 24.2 mM NaHCO₃, and 10 mM α-d-glucose (pH 7.4) equilibrated with 5 % CO₂/95 % O₂ and maintained at 37 ° C. Isolated hearts were quickly massaged to remove any blood from the ventricles, cleaned of connective tissue, weighed, and immediately snap frozen in liquid nitrogen.

Electrocardiographic analysis

EKGs collected from the aging female rats were analyzed using BioPac Student Lab PRO software. All animals were evaluated for any electrical anomalies in all three leads. Lead II was used to evaluate changes in EKG parameters, six complete cardiac cycles from the Lead II tracing were used to obtain data for comparison of ventricular acceleration time (VAT), heart rate, ST interval, T amplitude, QRS interval, QT interval, PR interval, T duration, T–T interval, R + S amplitude, S amplitude, R amplitude, Q amplitude, and P amplitude (Fig. 3a). Mean electrical axis was calculated from leads I and III after calibration of 2 cm/mV (Tsika et al. 2010) using the following formula derived by Singh and Athar (Singh and Sajjad 2003).

Heart rate was determined by the average intervals between R waves on lead II.

Determination of myocyte cross-sectional area (CSA) and histological analysis

An IEC Minotome Cryostat was used to section (8 μm) frozen hearts (n = 4) on poly-lysine coated slides. To determine morphology, heart sections were stained with hematoxylin and eosin stain. Picrosirius red staining was employed to examine collagen. The collagen area fraction of the picrosirius red stained tissues was determined using Image J software.

Immunostaining for dystrophin (NCL-DYS2, Novocastra Vector Laboratories, Burlingame, CA, USA) and connexin-43 (#3512, Cell Signing Technology, Danvers, MA, USA) was visualized by immunofluorescence as outlined by the antibody manufacturer. A phosphate-buffered saline (PBS) containing 0.5 % Tween-20 (PBS-T) at pH 7.5 was used to wash sections three times before incubating for 30 min in a blocking solution (5 % BSA). In a humidified chamber at 24 °C, the dystrophin antisera diluted in PSB-T (antibody dilution of 1:100) was added to sections for 1 h. After incubation the sections were washed three time with PBS, and incubated again for 30 min in a humidified chamber at 24 °C with a secondary antibody solution containing an FITC anti-rabbit IgG (1:200) and DAPI (1.5 μg/ml) in order to visualize cell nuclei. Before mounting sections were washed a final time with PBS. The epifluorescence of specimens was visualized using an Olympus fluorescence microscope (Olympus, Melville, NY, USA) fitted with a 40x objective. Images were recorded digitally using A CCD camera and the Olympus MicroSuite™ Basic from Olympus America (Olympus) were used to digitally record and analyze images, respectively. The CSA of the cardiac myocytes was determined by measuring the area within the dystrophin positive staining.

Statistics

Results are given as mean ± SEM. Statistical analyses were performed using Sigma Stat 3.5 statistical software (Systat Software, Inc.). Age comparisons between echocardiographic structural, functional parameters, and morphologic indices were evaluated by one-way ANOVA with the Student–Newman–Keuls post hoc test for parametric normally distributed data or Kruskal–Wallis one-way analysis of variance on Ranks with a Dunn's post hoc test for none parametric distributions. Linear regression analysis was performed with dependent variables against the independent variables age, correlation were ranked as low correlation (0.3 to 0.7), moderate correlation (0.5 to 0.7), high correlation (0.7 to 0.9), very high correlation (0.9 to 1) between parameters (supplemental data). The level of significance accepted a priori was ≤0.05.

Cardiac function is preserved in the aging female F344xBN rat

Compared to 6-month-old animals, heart and body weight was higher at 26 months (72 ± 2 vs. 95 ± 2 mg, p < 0.001; 235.3 ± 1.7 vs. 296.4 ± 6.5 g; p < 0.05) and 30 months (104 ± 4 mg, p < 0.05); 235.3 ± 1.7 vs. 315.6 ± 8.6 g; p < 0.05) (Table 1). Compared to 6-month-old animals, cardiac myocyte CSA was 20 ± 1.9 % and 28 ± 2.3 % higher in the 26- and 30-month-old animals, respectively (p < 0.05) (Fig. 1). Histological analysis using picrosirius red staining demonstrated a significant age-related decrease in collagen reactivity (Fig. 1) (p < 0.05). Immunohistochemical staining of the aging female heart suggested alterations in Cx43 localization from the cell ends to the lateral margins (Fig. 2a). Alignment of the Cx43 staining appears to be uniform, localized, and linear within the 6-month tissues sections, as would be expected because of Cx43 localization along the intercalated disk. The linearity of the staining appears to become less uniform and diffuse with aging which may indicate alteration in Cx43 distribution within the cardiacmyocyte. In addition to changes in Cx43 immunoreactivity, aging also appeared to be associated with changes in cardiac rhythm. In particular, aging appeared to be characterized by the presence of hyper-acute T waves, a long PR interval, a long P wave duration, tall P waves, and significant Q waves (Fig. 3b, Table 2). No evidence of arrhythmias was observed in any age group.

No age-associated changes in E–E′ ratio were found in the female F344xBN rats. However, E′ was significantly reduced in the 30-month age group when compared to 6- and 26-month age groups (Fig. 4; p < 0.05). Left ventricular isovolumic relaxation time was significantly higher in 26-month-old animals (0.036 ± 0.005 s) compared to that seen in 6-month-old animals (0.015 ± 0.002 s; p < 0.05). Mitral valve deceleration time demonstrated no significant change with age. Mitral valve Emax significantly lower at 26 (61.5 ± 1.5 cm/s) and 30 months (55.4 ± 1.4 cm/s) when compared to that seen in the 6-month-old animals (78.0 ± 3.3 cm/s; p < 0.05). Mitral valve Amax levels were lower at 30 months (35.1 ± 1.6 cm/s) when compared to 6 months (41.5 ± 1.3 cm/s; p < 0.05). No significant changes were found in MV E/A ratio with age (Table 3).

Compared to 6-month-old animals, ejection fraction was significantly higher at 26 months (82 ± 1.0 % vs. 74 ± 0.9 %; p < 0.05). Fractional shortening was lower at 30 months (41.5 ± 1.7 %) when compared to that observed in 26-month-old animals (45.6 ± 1.0 %; p < 0.05). End systolic volume was significantly lower at 26 months (0.082 ± 0.005 ml) when compared to 6-month-old animals (0.143 ± 0.017 ml; p < 0.05). However, heart rate was unaltered with age (Table 4).

Aging increases septal and posterior wall thickness and valvular insufficiency in the female F344xBN heart

Compared to 6-month-old animals, left ventricular septal thickness (IVS) during systole was higher in 26-month-old animals (0.193 ± 0.008 vs. 0.253 ± 0.003 cm; p < 0.05). Left ventricular internal dimension during systole (LVIDs) significantly increased at 26 months (0.331 ± 0.009 cm) compared to 6 months (0.378 ± 0.017 cm; p < 0.05). During diastole LVID demonstrated no significant change. Left ventricular posterior wall thickness (LVPW) during systole and diastole was significantly increased at 30 months (LVPWs: 0.288 ± 0.009 cm; LVPWd: 0.189 ± 0.008 cm) compared to 6 months (LVPWs: 0.193 ± 0.002 cm; LVPWd: 0.153 ± 0.010 cm; p < 0.05). No changes were found in right ventricular dimension during diastole or left ventricular mass with age (Table 5).

The incidence of tricuspid, mitral, and pulmonary valve insufficiency was higher in 26- and 30-month-old female hearts relative to that observed in 6-month-old animals. The presence of aortic insufficiency with age was only seen in the 30-month age group (Fig. 4).

Aging is associated with alterations in serum glucose and electrolytes in the female F344xBN rat

In the aging female F344xBN non-fasting serum glucose levels were significantly decreased in 30 month-old (323.7 ± 13.3) compared to 6-month-old (372.8 ± 16.0) female F344xBN rats. Serum levels of ALP, BUN, PHOS, and potassium significantly decreased at 26 and 30 months compared to that observed in the 6-month-old animals. TP, GLOB, and CRE increased significantly at 30 months (6.3 ± 0.1, 2.12 ± 0.1, and 0.4 ± 0.03, respectively) compared to 6 months (5.8 ± 0.05, 1.56 ± 0.12, and 0.3 ± 0.002, respectively). In the 26-month-old female F344xBN rat serum calcium (11.2 ± 0.1) levels were significantly increased compared to 6-month-old female rats (10.7 ± 0.09). No significant changes were found in serum ALB, ALT, sodium, and TBIL with age in the female F344xBN. Significant evaluation of calcium to phosphate (Ca⁺/PHOS) and sodium to potassium (Na⁺/K⁺) were observed in the 26- and 30-month age groups when compared to 6-month-old animals (p < 0.05). Compared to 6-month-old animals, BUN/CRE and the albumin to globulin ration (ALB/GLOB) were lower in the 26- and 30-month age groups (Table 1) (p < 0.05).

Age-associated alterations in heart rhythm the female F344xBN heart

Unlike the aging male F344xBN rats no evidence of arrhythmias were detected with age in the female F344xBN (Walker et al. 2006). Nonetheless, our data suggest that aging in the female F344xBN is associated with changes in heart rhythm including increases in the VAT, ST interval, T amplitude, QRS interval, QT interval, T duration, Q amplitude, P amplitude, and a shift in the mean electrical axis with age (Table 4), increase in heart weight, myocyte CSA, and LVM. Alterations in heart rhythm intervals such as increased PR interval, P wave amplitude, and QRS complex are oftentimes indicative of myocardial disease (Bestetti and Oliveira 1990). Previous studies have shown that aged male rats (20 months) had no difference in R–R interval, P wave, as well as QRS length; however, PR and QT interval were increased with age (da Silva 2002). Our current findings support the existence of alterations in cardiac conductance but provide little information as to the underlying mechanism. The elevation in VAT and QRS interval suggest that aging might not only affect depolarization but that it may also be associated with changes in the ability of the heart to undergo repolarization. Correlation analysis demonstrated a high correlation between heart weight to MV Amax velocity (0.752) and MV Amax velocity to CSA (0.827) (Tables S1–S4). Because the nature of correlation does not denote causation, it is important to note that further research is required to delineate the nature of these correlations and to better elucidate the causation of the age-associated changes listed in this study. Similarly, whether the changes in VAT and QRS interval we observed in the current study are due to differences in ion handling, ion channel density, alterations how electrical signals may propagate through the myocardium or if they reflect age-associated increases in myocyte CSA or chamber dimensions (Fig. 1, Table 3) is currently unclear and will require additional experimentation.

Heat rhythm propagation is dependent on cell-to-cell coupling between cardiomyocytes and that this coupling is dependent, at least in part, by ion channels and gap junctions. Previous studies have shown that age-related changes in impulse propagation may be related to abnormalities in the pattern of ventricular activation (Rossi et al. 2008). Recent data has suggested that alterations in heart rhythm intervals during aging may be associated with alterations in Cx43 expression/activation/heterogeneity, fibrosis, and hypertrophy (Rossi et al. 2008; Hayashi et al. 2002; Xiao et al. 2007; Howarth et al. 2008; Jansen et al. 2012). Other work examining Cx43 has demonstrated increases in atrial fibrillation in patients exhibiting increased heterogeneity of Cx43 distribution (Nattel et al. 2007). These possibilities appear to be consistent with our data, where we observed to be qualitative alterations in Cx43 localization with aging in the female F344xBN heart (Fig. 2). Future experiments to directly investigate if or how changes in Cx43 protein abundance and localization may affect cardiac rhythm in the aging female F344xBN rat model may be helpful in delineating whether Cx43 may play a role in the aging female heart.

In addition, we also examined if age-related changes in heart rhythm were associated with alterations in blood chemistry. It has been suggested that electrolyte disorders can affect the ion currents in the heart and that such changes might be related to the development of cardiac arrhythmias (El-Sherif and Turitto 2011). For example, previous studies have shown that decreased plasma sodium, potassium, and ionized calcium levels are associated with a higher risk of arrhythmias in hemodialysis patients (Berta et al. 2012; Tong and Hou 2006). Although very little is known about the role of serum electrolytes in influencing cardiac function during aging, alterations in serum electrolytes have been associated with increased all-cause mortality among patients suffering from coronary heart disease (Grandi et al. 2012). For the most part, we found that age-related changes in blood chemistry appeared to be within the normal range. Recent work has suggested that decreases in plasma potassium may be related to the development of tachyarrhythmias (Sabir et al. 2008). Whether the age-associated reduction in potassium levels seen in this study, although within the normal physiological range, might play a role in the changes seen within the EKG data and will require further investigation.

In conclusion, this study provides reference values for cardiac structure and function in the aging female F344xBN heart. Taken together, our data suggest that the female, unlike the male, F344xBN rat demonstrates subtle age-associated changes in cardiac structure, function, and conductance. The age-associated increase in heart wall thickness in the absence of fibrosis and increase heterogeneity of Cx43 distribution may partially explain the age-associated alterations in heart rhythm intervals. The possibility of altered serum ion levels may also contribute to reduced conductance. Further study is needed to better understand the mechanistic basis of how aging may affect cardiac structure and function in the female F344xBN rat.