Adenylyl Cyclase 6 Improves Calcium Uptake and LV Function in Aged Hearts

Abstract

Objectives

This study sought to test the hypothesis that activation of adenylyl cyclase 6 (AC6) expression in cardiac myocytes improves Ca²⁺ uptake and LV function in aging mice.

Background

Aging hearts exhibit impaired β-adrenergic receptor signaling and left ventricular (LV) dysfunction.

Methods

Twenty-month-old mice with cardiac-directed and regulated AC6 expression were randomized into two groups, and AC6 expression was activated in one group (AC6-On) but not the other (AC6-Off). One month later, LV function and sarcoplasmic reticulum Ca²⁺ uptake were assessed.

Results

AC6 expression was associated with increased LV contractility, as reflected by ejection fraction (P=0.02), rate of pressure development (P=0.002), and the slope of LV end-systolic pressure-volume relationship (P=0.04). No changes in LV weight to tibial length ratio, LV fibrosis, and expression of fetal genes (atrial natriuretic factor, α-skeletal muscle actin, and β-myosin heavy chain) and collagens were observed between AC6-On and AC6-Off groups. However, LV samples from AC6□On mice showed increases in: isoproterenol-stimulated cAMP production (P=0.04), PKA activity (P<0.0004), phosphorylation of phospholamban (at Ser16 site; P=0.04) and cardiac troponin I (at Ser23/24 sites; P=0.01), velocity of sarcoplasmic reticulum Ca²⁺ uptake (P<0.0001), and SERCA2a affinity for Ca²⁺ (P<0.0001). Finally, we found that AC6 expression increased sarcoplasmic reticulum Ca²⁺ storage in cardiac myocytes isolated from 23-month-old rats. In contrast, AC6 expression in 7-month-old mice did not change LV function and Ca²⁺ uptake.

Conclusions

These results indicate that activation of cardiac AC6 expression improves impaired function of aged hearts, through improved Ca²⁺ uptake.

Introduction

Cardiac senescence is associated with reduced left ventricular (LV) function (1–4), and impaired cardiac β-adrenergic receptor (βAR) responsiveness (5). In humans, LV diastolic and systolic function in response to βAR stimulation progressively decrease after the age of 20 years. At age 80, LV contractile reserve is less than half of what it was at age 20 (6). In addition, congestive heart failure (CHF) is a common disease of the elderly (7–9), and older patients with CHF have a particularly poor prognosis (8).

Adenylyl cyclase (AC) is the effector molecule for βAR signaling (10), playing a pivotal role in contractile responsiveness, cardiac relaxation, and LV diastolic function (11). AC catalyzes ATP to generate cAMP, a second messenger that is required for many intracellular events (12). Reduced βAR responsiveness in aging hearts occurs in the presence of increased plasma catecholamine levels (13), underscoring the abnormality in AC signaling (5,14). Indeed, impaired LV cAMP production is associated with decreased cardiac AC content in hearts from animals of advanced age (15,16). However, the precise mechanism by which AC regulates cardiac function in aging hearts is not known, and is the focus of the current investigation.

Cardiac-directed expression of AC type 6 (AC6) increases LV function in CHF (17,18), where impaired βAR-AC signaling and impaired Ca²⁺ uptake are prominent (19). Increased cardiac cAMP production and improved sarcoplasmic reticulum (SR) Ca²⁺ uptake are of mechanistic importance for the beneficial effect of AC6 on failing hearts (18,20). In this study, we used a transgenic line with cardiac-directed tetracycline-regulated AC6 expression (21,22) to test the hypothesis that activation of cardiac AC6 expression improves LV function by increasing cardiac cAMP production and correcting Ca²⁺ uptake impairment in hearts from older mice.

Materials and Methods

Animals

The Animal Use and Care Committee of the VA San Diego Healthcare System, in accordance with NIH and AAALAC guidelines, approved this study. Twenty-month-old transgenic mice with cardiac myocyte-specific tetracycline-regulated (tet-off) AC6 expression (22) were used for echocardiography, in vivo physiology, and biochemistry studies. Suppression of AC6 transgene expression is complete until tetracycline is removed from the water supply, and chronic tetracycline treatment did not affect cardiac function. Mice were randomized into two groups. AC6 expression was activated (by removing tetracycline suppression) in one group (AC6-On), and was continuously suppressed by tetracycline in the other group (AC6-Off). These mice were studied one month after activation (or continued inactivation) of AC6 transgene expression. For comparison, echocardiography, in vivo physiology, and biochemistry studies were also performed on 7-month-old tetracycline-regulated (tet-off) AC6 mice. For Ca²⁺ transients study, cardiac myocytes were isolated from 23-month-old Sprague Dawley rats (Harlan Laboratories, Indianapolis, IN, USA).

Echocardiography

Echocardiography was performed under light anesthesia as previously reported (23). Data were acquired and analyzed without knowledge of group identity.

LV in vivo Physiological Studies

A 1.4F conductance-micromanometer catheter was used to measure LV pressure and volume to assess the LV end-systolic pressure-volume relationship (ESPVR) as previously reported (18). Data were acquired and analyzed without knowledge of group identity.

Necropsy and LV Fibrosis Assessment

Body and LV weights (including septum) and tibial lengths were recorded. A short axis midwall LV ring was formalin-fixed and paraffin embedded. LV sections were stained with picrosirius red, and collagen fractional area was quantified using NIH ImageJ software (24).

Biochemistry Studies

Total RNA extraction and quantitative RT-PCR were performed as previously reported (24). LV samples were homogenized and used for Western blotting as described previously (23). AC activity, PKA activity, caspase 3/7 activity in LV samples were measured as reported (18).

Ca²⁺ Uptake

LV tissues were homogenized, and ATP-dependent initial rate of SR Ca²⁺ uptake was measured by modified Millipore filtration technique as reported (Tang et al., 2004).

Ca²⁺ Transients

Cardiac myocytes were isolated from adult rats as previously described (25), plated on laminin-coated 25 mm glass coverslips, and infected with adenovirus encoding green fluorescent protein or murine AC6 (400 vp/cell). Forty hours after infection, cells were loaded with the Ca²⁺-sensitive fluorescent indicator Fura-2 AM (3 µM) and intracellur Ca²⁺ concentration was monitored using Digital Fluorescence Imaging System (Intracellular Imaging, Cincinnati, OH, USA) as described previously (23). To assess SR Ca²⁺ load, caffeine-induced Ca²⁺ release was initiated by addition of 10 mM caffeine to Tyrode's solution. The peak amplitude of Ca²⁺ transients was calculated from the baseline and the transient rise after caffeine treatment. Data were acquired and analyzed without knowledge of group identity.

Statistical Analysis

Results are shown as mean ± SE. Group differences were compared using unpaired, 2-tailed Student's t-test. The null hypothesis was rejected when P<0.05.

Results

AC Activity in Aging Hearts

To confirm that aging is associated with decreased AC activity in the heart (5,14,16,26), we measured LV cAMP production in 7- and 20-month-old mice. There were a 43% reduction in basal (P=0.0001), a 56% reduction in isoproterenol-stimulated (P=0.04), and a 58% reduction in NKH477-stimulated LV cAMP production (P=0.0001) in 20- vs 7-month-old mice (Fig. 1A). We also found a 59% reduction of mRNA expression of AC6, a major cardiac AC isoform, in LV samples from 20- vs 7-month-old mouse hearts (Fig. 1B).

Fig. 1. Aging was associated with reduced LV AC activity and AC6 expression.

(A) Basal, Isoproterenol (+GTPγs), and NKH477-stimulated cAMP production were reduced in LV samples form 20-month-old vs 7-month-old mice. (B) Quantitative real-time RT-PCR showed reduced LV expression of AC6 mRNA in 20-month-old vs 7-month-old mice. Probability values (values above bars) are from Student's t-test (unpaired, 2-tailed). Error bars denote 1 SE; numbers in bars indicate group size. Iso, Isoproterenol + GTPγs; NKH, NKH477 (a water-soluble forskolin derivative)

Echocardiography

Aging was associated with a decline in LV ejection fraction (7-month-old: 80±3%, n=13; 20-month-old: 57%±10%, n=17; P<0.0001). Both 7- and 20-month-old mice with cardia□directed and regulated AC6 expression were randomized, and AC6 expression was activated in AC6-On group but not the AC6-Off group. There were no group differences (AC6-On group vs AC6-Off group) in any of the echocardiographic measures prior to activation of AC6 expression. However, in 20-month-old mice, activation of AC6 expression increased LV ejection fraction (Table 1). These mice also showed reduced LV end-systolic dimension after AC6 expression was activated (P<0.05), consistent with improved LV ejection fraction. Other measures showed no group differences. In 7-month-old mice, activation of cardiac AC6 expression did not change LV ejection fraction and any of the other echocardiographic measures (Table 1).

Table 1.

Echocardiography Measurements

	7-month-Old			20-month-Old
	AC6-Off (7)	AC6-On (6)	P	AC6-Off (8)	AC6-On (9)	P
HR (beats/min)	597±16	607±14	ns	490±20A	515±16A	ns
EDD (mm)	3.9±0.5	3.6±0.6	ns	4.0±0.2C	3.8±0.1C	ns
ESD (mm)	2.3±0.4	2.0±0.3	ns	2.8±0.1C	2.3±0.2C	<0.05
EF (%)	80±2	81±5	ns	59±3B	70±4C	<0.02
PWTh (mm)	0.6±0.02	0.6±0.03	ns	0.9±0.03B	0.9±0.02B	ns
IVSTh (mm)	0.6±0.01	0.6±0.02	ns	0.9±0.04B	0.9±0.02B	ns

HR, heart rate; EDD, end-diastolic diameter; ESD, end-systolic diameter; EF, ejection fraction; PWTh, posterior wall thickness; IVSTh, interventricular septum wall thickness; ns, not significant;

^AP<0.002 vs same group at 7 months;

^BP<0.001 vs same group at 7 months;

^Cnot significantly different vs same group at 7 months. Transgenic mice at ages 7 and 20 months were randomized into 2 groups, respectively.

AC6 expression was activated in one group (AC6-On) but not the other (AC6-Off) for 1 month. Values represent mean ± SE. Probability values shown are from unpaired, 2-tailed Student's t-test.

LV Contractile Function

Activation of cardiac-directed AC6 expression increased the slope of ESPVR by 2.6-fold in 20-month-old mice (AC6-Off: 1.8±0.6 mmHg/μL, n=10; AC6-On: 4.6±1.0 mmHg/μL, n=9; P=0.04; Fig. 2A, B). AC6 expression in 20-month-old mice was also associated with increased rate of LV pressure development (LV +dP/dt) (P=0.002; Fig. 2C), and increased stroke work (P=0.04; Fig. 2D). There were no group differences in stroke volume (AC6-Off: 19±2 μL, n=10; AC6-On: 20±3 μL, n=10; P=0.78) and cardiac output (AC6-Off: 7±1 mL/min, n=10; AC6-On: 10±2 mL/min, n=10; P=0.20) in 20-month-old mice. Heart rate was similar in both groups (AC6-Off: 444±31 bpm, n=10; AC6-On: 469±27 bpm, n=10; P=0.55). However, in 7-month-old mice, activation of AC6 expression did not affect the slope of ESPVR (AC6-Off: 4.3±0.4 mmHg/μL, n=7; AC6-On: 4.0±0.6 mmHg/μL, n=6; P=0.69) and +dP/dt (AC6-Off: 5448±453 mmHg/s, n=7; AC6-On: 7285±1073 mmHg/s, n=6; P=0.12).

Fig. 2. Cardiac-directed AC6 expression increased LV function in 20-month-old mice.

(A) LV pressure-volume loops, generated by inferior vena caval occlusion, are shown. (B) Activation of AC6 expression increased slope of ESPVR in AC6-On mice compared to AC6-Off mice. (C) AC6-On mice showed increased +dP/dt compared with AC6-Off mice. (D) Activation of AC6 expression increased stroke work in hearts from 20-month-old mice. (E) There was a substantial increase in −dP/dt in AC6-On mouse hearts. Probability values shown are from Student's t-test (unpaired, 2-tailed). Error bars denote 1 SE; numbers in bars indicate group size.

LV Relaxation

Activation of cardiac-directed AC6 expression in 20-month-old mice decreased LV −dP/dt (P=0.03; Fig. 2E). In contrast, AC6 expression in 7-month-old mice did not affect LV −dP/dt (AC6-Off: −5260±425 mmHg/s, n=7; AC6-On: −6827±833 mmHg/s, n=6; P=0.11).

LV Hypertrophy

Activation of cardiac-directed AC6 expression in 20-month-old mice did not change LV weight to tibial length ratio (AC6-Off: 7.2±0.6 mg/mm, n=10; AC6-On: 7.5±0.5 mg/mm, n=9; P=0.71) or cross sectional cardiac myocyte area in LV sections (AC6-Off: 649±35 μm², n=6; AC6-On: 663±21 μm², n=6; P=0.74). AC6-Off and AC6-On mice showed similar mRNA expression of fetal genes: atrial natriuretic factor (AC6-Off: 100±22%, n=6; AC6-On: 86±15%, n=9; P=0.61), α-skeletal muscle actin (AC6-Off: 100±20%, n=6; AC6-On: 126±24%, n=9; P=0.45), and β-myosin heavy chain (AC6-Off: 100±19%, n=6; AC6-On: 84±32%, n=9; P=0.71). In addition, expression of FHL1, a regulator of LV hypertrophy, showed no group difference (AC6-Off: 100±9%, n=6; AC6-On: 113±12%, n=9; P=0.47). Similarly, activation of cardiac AC6 expression in 7-month-old mice did not change LV weight to tibial length ratio (AC6-Off: 6.4±0.4 mg/mm, n=7; AC6-On: 6.2±0.3 mg/mm, n=6; P=0.71), and mRNA expression of fetal genes and FHL1. Although activation of AC6 expression in 7-month-old mice did not change cross sectional cardiac myocyte area in LV sections (AC6-Off: 489±21 μm², n=6; AC6-On: 465±27 μm², n=6; P=0.50), aging was associated with an increase (7 month: 489±21 μm², n=6; 20 months: 649±35 μm², n=6; P=0.003).

Ca²⁺ Uptake

We compared ATP-dependent initial SR Ca²⁺ uptake in LV homogenates from 20-month-old AC6-Off and AC6-On mice. Activation of AC6 increased velocity of Ca²⁺ uptake in the presence of 0.2 μM and 2 μM Ca²⁺, approximate intracellular Ca²⁺ concentration range in cardiac myocytes (27) (Fig. 3A). AC6 expression also increased SERCA2a affinity for Ca²⁺ (Fig. 3B). Increased Ca²⁺ uptake and SERCA2a affinity for Ca²⁺ would be expected to increase cardiac contractile function as we observed. In contrast, activation of AC6 expression in 7-month-old mice did not change velocity of Ca²⁺ uptake (Fig. 3C) and SERCA2a affinity for Ca²⁺ (Fig. 3D).

Fig. 3. Activation of cardiac AC6 expression increased Ca²⁺ uptake in 20-month-old, but not in 7-month-old mice.

(A) Cardiac-directed AC6 expression improved SR Ca²⁺ uptake in the presence of 0.2 and 2 μM Ca²⁺ in LV samples from 20-month-old mice (n=6 for each group). (B) Cardiac-directed AC6 expression increased SERCA2a affinity for Ca²⁺ in LV samples from 20-month-old mice. (C) AC6 expression did not affect SR Ca²⁺ uptake in the presence of 0.2 μM (P=0.33) and 2 μM Ca²⁺ (P=1.00) in LV samples from 7-month-old mice (n=6 for each group). (D) AC6 expression did not affect SERCA2a affinity for Ca²⁺ in LV samples from 7-month-old mice (P=0.72). Probability values are from Student's t-test (unpaired, 2-tailed). Error bars denote 1 SE; numbers in bars indicate group size.

In addition, we isolated cardiac myocytes from 23-month-old rats, infected these cells with adenovirus encoding AC6, and measured intracellular Ca²⁺ concentration under basal and caffeine-stimulated conditions (Fig. 4A). There was no group difference in basal intracellular Ca²⁺ concentration in cardiac myocytes (Fig. 4B). However, there was a 1.5-fold increase in peak amplitude of caffeine-stimulated Ca²⁺ transients associated with AC6 expression in cardiac myocytes from aging rats (Fig. 4C). These results indicate that AC6 expression increases Ca²⁺ storage in aging cardiac myocytes.

Fig. 4. Increased Ca²⁺ storage in cardiac myocytes from 23-month-old rats following AC6 gene transfer.

(A) Representative Ca²⁺ transients recorded from cardiac myocytes stimulated with caffeine. Left Panel: cardiac myocyte infected with adenovirus encoding green fluorescent protein (CON); Right Panel: cardiac myocyte infected with adenovirus encoding AC6 (AC6). (B) The bar graph shows no group difference in basal intracellular Ca²⁺ concentration in cardiac myocytes. Data were derived from 67 CON cells and 77 AC6 cells (blinded; P=0.40). (C) AC6 expression increased peak amplitude of caffeine-stimulated Ca²⁺ transients recorded from cardiac myocytes. Data were derived from 41 CON cells and 51 AC6 cells (blinded). Probability values shown are from Student's t-test (unpaired, 2-tailed). Error bars denote 1 SE.

AC6 Expression and cAMP Production

In 20-month-old mice, cardiac-directed AC6 transgene expression was associated with a 10-fold increase in LV AC6 protein content (Fig. 5A, B). LV expression of other cardiac AC isoforms (AC2, AC3, AC4, AC5, AC7, and AC9) was unchanged in these mice (Fig. 5C). Increased AC6 protein content was associated with a 1.7-fold increase in isoproterenol-stimulated cAMP production (P<0.01) and a 3-fold increase in forskolin-stimulated cAMP production (P<0.002) in LV samples from 20-month-old mice, and basal LV cAMP production was not changed (Fig. 5D). Despite unchanged protein expression of cAMP-dependent protein kinase (PKA) catalytic subunit (Fig. 5E), PKA activity was increased after activation of AC6 expression (Fig. 5F), indicating a role of AC6-increased cAMP production in regulating PKA activity in the aging heart.

Fig. 5. Cardiac-directed AC6 expression activated cAMP signaling in 20-month-old mouse hearts.

(A) Representative Western blot shows marked increase in AC6 protein content in LV samples from AC6-On mice. (B) The bar graph summarizes AC6 content in LV samples from AC6-Off and AC6-On mice. (C) AC6 expression did not affect expression of other AC isoforms in the heart. (D) AC6 expression was associated with increased LV cAMP generation capacity. NKH477, a water-soluble forskolin derivative. (E) No group difference was observed in protein content of PKA catalytic subunit (P=0.30). (F) AC6-On mice showed increased PKA activity. Probability values are from Student's t-test (unpaired, 2-tailed). Error bars denote 1 SE; numbers in bars indicate group size. du, densitometric unit

Protein Phosphorylation

We compared PKA-induced phosphorylation of cardiac troponin I (cTnI) and phospholamban (PLN) in LV homogenates from 20-month-old AC6-Off and AC6-On mice. Activation of cardiac AC6 expression in these mice was associated with 1.9–fold increase in phosphorylation of cTnI at Ser23/24 sites (P=0.01; Fig. 6A, B). No change was found in total cTnI protein content. There was also a 1.9-fold increase in phosphorylation of PLN at Ser16 site in LV sample from 20-month-old AC6-On mice compared to AC6-Off mice (P=0.04; Fig. 6A, C), although total PLN protein content was unaltered. Western blotting showed no group differences in protein content of protein phosphatase 1 [AC6-Off: 1152±195 densitometric units (du), n=6; AC6-On: 1153±277 du, n=9; P=1.00] or protein phosphatase 2A (AC6-Off: 743±104 du, n=6; AC6-On: 604±65 du, n=9; P=0.25), indicating a selective role of AC6-increased PKA activity in regulating phosphorylation of cTnI and PLN. In addition, we found no group differences in expression of sarcoplamic reticulum Ca²⁺ pump SERCA2a (Fig. 6D), the Ca²⁺ binding protein calsequestrin (Fig. 6E), or Na⁺-Ca²⁺ exchanger 1 (Fig. 6F).

Fig. 6. Cardiac-directed AC6 expression increased phosphorylation of cTnI and PLN in 20-month-old mouse hearts.

(A) Representative Western blots show marked increase in protein content of phosphor-cTnI(Ser23/24) and phosphor-PLN(Ser16) in AC6-On mouse hearts. (B)The bar graph summarizes protein content of phospho-cTnI from Western blotting. (C) The bar graph summarizes protein content of phospho-PLN from Western blotting. (D) No group difference was observed in protein content of SERCA2a (P=0.61). (E) No group difference was seen in protein content of calsequestrin (P=0.37). (F) AC6 expression did not change mRNA expression of Na⁺-Ca²⁺ exchanger 1 (P=0.27). Probability values shown are from Student's t-test. Error bars denote 1 SE; numbers in bars indicate group size. du, densitometric unit

Extracellular Matrix

Extracellular matrix remodeling, a process that involves changes of collagen synthesis, degradation, and cross-linking, leads to increased number and size of collagen fibers within interstitial spaces. There was substantial collagen deposition in LV sections from 20-month-old mice compared to 7-month-old mice (collagen fractional area: 7-month-old: 3.1±1.1%, n=7; 20-month-old: 15.4±2.5%, n=6; P<0.001); however, activation of cardiac AC6 expression did not change collagen fractional area in 20-month-old mouse hearts (AC6-Off: 15.4±2.5%, n=6; AC6-On: 14.4±2.0%, n=9; P=0.77).

We compared mRNA content of type I and III collagens, two major constituents of the cardiac extracellular matrix, in LV samples from 20-month-old AC6-Off and AC6-On mice. Activation of AC6 expression did not change mRNA expression of collagen Iα1 (AC6-Off: 100±14%, n=6; AC6-On: 129±19%, n=9; P=0.30) or IIIα1 (AC6-Off: 100±29%, n=6; AC6-On: 85±29%, n=9; P=0.73). Quantitative RT-PCR also showed that AC6 expression did not change mRNA content of matrix metalloproteinase 2 (MMP2) (AC6-Off: 100±14%, n=6; AC6-On: 135±20%, n=9; P=0.22) or MMP 8 (AC6-Off: 100±16%, n=6; AC6-On: 87±14%, n=9; P=0.55). Similar levels of periostin mRNA, a regulator of cardiac fibrosis, were seen in LV samples from 20-month-old AC6-Off and AC6-On mice (AC6-Off: 100±16%, n=6; AC6-On: 121±19%, n=9; P=0.44). There were no group differences in mRNA expression of elastin (AC6-Off: 100±10%, n=6; AC6-On: 121±18%, n=9; P=0.69) or fibronectin (AC6-Off: 100±17, n=6; AC6-On: 89±19, n=9; P=0.39).

Cardiac Myocyte Apoptosis

Terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) was used to evaluate apoptosis. In 20-month-old mouse hearts, AC6 expression did not change apoptosis rate (data not shown). AC6-Off and AC6-On mice showed comparable LV mRNA expression of Bcl2 (AC6-Off: 100±27%, n=6; AC6-On: 104±18%, n=9; P=0.89), an important apoptosis inhibitor. Increased AC6 expression in 20-month-old mouse hearts did not change LV caspase 3/7 activity (AC6-Off: 40.3±2.4 Unit/mg, n=6; AC6-On: 37.6±2.0 Unit/mg, n=9; P=0.41). Finally, no group difference in expression of superoxidase dismutase 2 was seen after activation of AC6 expression in 20-month-old mice (AC6-Off: 100±6%, n=6; AC6-On: 93±6%, n=9; P=0.46).

Discussion

The most important finding of this study is that activation of cardiac AC6 expression improved aging-impaired LV contractile function and relaxation. These beneficial effects were associated with increases in cAMP production, PKA activity, and phosphorylation of PLN and cTnI. Moreover, both SR Ca²⁺ uptake and SR Ca²⁺ storage in cardiac myocytes were increased following activation of AC6 expression. Improved SR Ca²⁺ uptake appears to be the mechanism by which AC6 expression evokes these beneficial effects.

Results from our previous studies (17,18,28,29) indicate that increased cardiac AC6 has beneficial effects in the failing heart. However, increased expression of other elements of the βAR signaling pathway (e.g. β₁AR, Gαs, PKA) is associated with pronounced deleterious effects on the heart, including precipitation of heart failure, fibrosis, and increased mortality (30–33). These results indicate that AC6 is unique among βAR signaling elements in regulating cardiac function. In clinical settings, increased age is associated with reduced LV function (1–4), and elderly patients have an increased prevalence of heart failure (7–9). In this study, we tested the hypothesis that activation of AC6 expression increases function of the aging heart. We assessed LV contractile function by three different methods: LV ejection fraction, +dP/dt, and slope of ESPVR (34). Each of these measures led to the same conclusion that increased expression of cardiac AC6 improves LV contractile function in the aging heart.

Calcium plays a crucial role in controlling LV contraction and relaxation. During every heartbeat, Ca²⁺ is taken up and then released from SR. Aged hearts show dysfunctional SR Ca²⁺ uptake (35,36), which reflects reduced SR Ca²⁺ pump SERCA2a content (37,38) and reduced PLN phosphorylation (37–39). In the current study, we found that activation of AC6 expression increased PLN phosphorylation at Ser16 site in 20-month-old mice, while it did not change SERCA2a protein expression (Fig. 6). This change of PLN phosphorylation was associated with increased cAMP generating capacity and PKA activity. Most importantly, activation of cardiac AC6 expression increased SR Ca²⁺ uptake and SERCA2a affinity for Ca²⁺ in these 20-month-old mice (Fig. 3A, B). These results indicate that AC6 expression improves Ca²⁺ uptake by activating a signaling cascade leading to increased PLN phosphorylation in 20-month-old mice. Improvement of Ca²⁺ uptake may be of mechanistic importance for the observed increased LV function after AC6 expression (Fig. 2), given the fact that deletion of AC6 decreases Ca²⁺ uptake and impairs LV function (23). This notion is further supported by our findings that activation of AC6 expression did not improve SR Ca²⁺ uptake and SERCA2a affinity for Ca²⁺ in 7-month-old mouse hearts (Fig. 3C, D), where LV contractile function was unaffected.

Cardiac myocytes isolated from the hearts of 20-month-old mice may not be suitable for accurate studies for Ca²⁺ transient measurements. For example, despite our substantial experience in isolated cardiac myocytes from mice, we found >25% cell death rate and spontaneous beating in one-third of the viable cells. Because of these unavoidable problems, we elected to use instead cardiac myocytes isolated from elderly rats for Ca²⁺ transient measurements. We used rats at age of 23 months, rather than 20 months, to compensate in part for lifespan difference between mice and rats. Our data from caffeine-stimulated SR Ca²⁺ transient measurements provided direct evidence that AC6 expression increased SR storage. But more importantly, we showed that AC6 expression increased SR Ca²⁺ uptake using LV samples directly from our 20-month-old mice.

Cardiac troponin I is a sarcomeric protein directly regulating the LV function in concert with intracellular Ca²⁺ signals. In addition to decreased PLN phosphorylation (at Ser16 site), decreased cTnI phosphorylation (at Ser23/24 sites) is also a seminal alteration underlying impaired LV function of failing human hearts (40,41). The relative contributions of phosphorylation of cTnI vs PLN are not clear (42). Nevertheless, cTnI phosphorylation increases maximal tension development, crossbridge kinetics, and systolic power production (42,43). Expression of pseudophosphorylated cTnI at Ser23/24 sites increased LV contractility in mice (44). Moreover, improved LV function by cardiac AC6 expression is linked to increased cTnI phosphorylation at Ser23/24 sites in failing hearts (18). Importantly, cardiac aging is associated with decreased cTnI phosphorylation (45), suggesting cTnI phosphorylation is also a contributing factor for reduced LV function in aging hearts. In the present study, we showed that increased LV function was correlated with increased cTnI phosphorylation after activation of AC6 expression in aging hearts (Fig. 6). These results suggest that AC6 expression, in addition to its effects on SR Ca²⁺ uptake, directly regulates the property of thin filament in the contractile apparatus.

The extracellular matrix, composed mainly of fibrillar collagens (46), provides an architectural scaffold for cardiac cells and contributes to the maintenance of LV function (47,48). In this study, we confirmed that cardiac aging is associated with increased collagen deposition. However, activation of AC6 expression in 20-month-old mice did not change LV collagen deposition, indicating that the improved LV function does not likely reflect a change in LV collagen content. Additional support for this notion is our finding that AC6 expression did not change expression of type I and III collagens, the major cardiac collagen isoforms contributing 75–80% (type I) and 15–20% (type III) of total collagen content (49). Expression of MMPs and periostin was not changed by activation of AC6 expression, although aging is associated with increased expression of these important extracellular matrix proteins (50,51). In addition, our data showed that increased cardiac AC6 expression in 20-month-old mice was not associated with changes in apoptosis or reactive oxygen species. Taken together, our results suggest that activation of AC6 expression improves LV function through increased SR Ca²⁺ uptake.

Whether LV hypertrophy in aged mice affected our results is an important consideration. Although posterolateral and septal wall thicknesses at the mid-papillary muscle level were increased (0.3 mm) in aged mice (Table 1), echocardiographic estimate of LV mass, which considers more than a single plane, showed no difference between mice aged 20 months vs 7 months (data not shown). Cardiac myocyte cross-sectional area was increased, but the 13% increase in LV/TL ratio did not reach statistical significance. The disparity between cross sectional area and LV mass at necropsy vis-à-vis LV hypertrophy in 7 vs 20 month old mice may reflect LV hypertrophy associated with concomitant aging-related increases in cardiac myocyte apoptosis (52), or the limitations of assessing cardiac myocyte cross sectional area in hearts not obtained after diastolic arrest. Similarly, others have documented that, in aged rats, despite increased myocyte cross sectional area (52), LV mass and LV/TL were not increased (53). Our study shows that activation of cardiac AC6 expression, which had no independent effect on LV hypertrophy, improved aging-impaired LV contractile function and relaxation. Improved SR Ca²⁺ uptake appears to be of mechanistic importance for these beneficial effects.

Gene manipulation has recently been promoted to probe disease mechanism and test treatments designed to attenuate the progression of aging-related diseases. Transgenic lines with increased or decreased expression of a candidate gene typically are used, and, if deterioration of heart function is prevented, the candidate gene (or deletion of this gene) is said to have “rescued” the heart. However, a superior strategy, and one used in the present study, is to activate gene expression only when the adverse effect of aging is present. We accomplished this by activating AC6 expression in the setting of declining heart function in aged mice. Although additional studies will be required to determine whether activation of cardiac AC6 expression reduces mortality, our current data suggest a therapeutic potential for increased AC6 expression in aging hearts.

In conclusion, activation of cardiac-directed AC6 expression improved LV function in aged hearts. This was associated with increased cAMP production, PKA activity, and phosphorylation of PLN and cTnI. Improved SR Ca²⁺ uptake in cardiac myocytes appears to be of mechanistic importance for these beneficial effects.