- Email: [email protected]
Effect of exercise on passive myocardial stiffness in mice with diastolic dysfunction
Accepted Manuscript Effect of exercise on passive myocardial stiffness in mice with diastolic dysfunction
Rebecca E. Slater, Joshua G. Strom, Henk Granzier PII: DOI: Reference:
S0022-2828(17)30088-3 doi: 10.1016/j.yjmcc.2017.04.006 YJMCC 8538
To appear in:
Journal of Molecular and Cellular Cardiology
Received date: Revised date: Accepted date:
3 April 2017 24 April 2017 27 April 2017
Please cite this article as: Rebecca E. Slater, Joshua G. Strom, Henk Granzier , Effect of exercise on passive myocardial stiffness in mice with diastolic dysfunction. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Yjmcc(2017), doi: 10.1016/j.yjmcc.2017.04.006
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Effect of Exercise on Passive Myocardial Stiffness in Mice with Diastolic Dysfunction
Department of Cellular and Molecular Medicine, University of Arizona, Tucson,
IP
1
T
Rebecca E Slater 1 ; Joshua G Strom1 , and Henk Granzier 1
CR
AZ 85721; Sarver Molecular Cardiovascular Research Program, University of Arizona,
PT
ED
M
AN
US
Tucson, AZ 85721.
CE
Address for Correspondence: Henk Granzier, PhD
AC
Department of Molecular and Cellular Medicine MRB 325. 1656 E Mabel Street. University of Arizona, Tucson, AZ-85724-5217 Voice: (520)6263641; FAX:(520)6267600; Email:[email protected]
1
ACCEPTED MANUSCRIPT Abstract Heart failure with preserved ejection fraction (HFpEF) is a complex syndrome, characterized by increased diastolic stiffness and a preserved ejection fraction, with no effective treatment options. Here we studied the therapeutic potential of exercise for improving diastolic function in a mouse model with HFpEF-like symptoms, the TtnΔIAjxn mouse model. TtnΔIAjxn mice
IP
T
have increased diastolic stiffness and reduced exercise tolerance, mimicking aspects of HFpEF observed in patients. We investigated the effect of free-wheel running exercise on diastolic
CR
function. Mechanical studies on cardiac muscle strips from the LV free wall revealed that both
US
TtnΔIAjxn and wildtype (WT) exercised mice had a reduction in passive stiffness, relative to sedentary controls. In both genotypes, this reduction is due to an increase in the compliance of
AN
titin whereas ECM-based stiffness was unaffected. Phosphorylation of titin’s PEVK and N2B spring elements were assayed with phospho-site specific antibodies.
Exercised mice had
M
decreased PEVK phosphorylation and increased N2B phosphorylation both of which are
ED
predicted to contribute to the increased compliance of titin. Since exercise lowers the heart rate we examined whether reduction in heart rate per se can improve passive stiffness by
PT
administering the heart-rate-lowering drug ivabradine. Ivabradine lowered heart rate in our study but it did not affect passive tension, in neither WT nor TtnΔIAjxn mice. We conclude that
CE
exercise is beneficial for decreasing passive stiffness and that it involves beneficial alterations in
AC
titin phosphorylation.
2
ACCEPTED MANUSCRIPT INTRODUCTION Heart Failure with preserved Ejection Fraction (HFpEF) is a complex syndrome with diastolic stiffening that leads to higher diastolic pressures and insufficient filling, the culmination of which eventually leads to decreased cardiac reserve, and pulmonary edema[1, 2]. The prevalence of HFpEF is increasing and is predicted to exceed 8% of persons older than 65
T
Unlike heart failure with reduced ejection fraction (HFrEF),
IP
by the year 2020[3].
pharmacological approaches to treating HFpEF do not currently exist and clinical trials focused
CR
on HFpEF have had neutral results [4, 5].
US
Physical training has been reported to have beneficial effects on diastolic function under a range of conditions and earlier studies [6, 7] suggest that exercise could be a potential therapy
AN
for HFpEF through decreasing diastolic stiffness. The Exercise Training in Diastolic Heart Failure trial(Ex-DHF) studied HFpEF patients and concluded that exercise training improved
M
exercise capacity and diastolic function[8]. However, the mechanistic basis of how exercise
ED
affects passive stiffness in HFpEF is not well established. Recent studies have indicated that rodents have decreased phosphorylation at titin’s PEVK element following exercise and that this
PT
hypo-phosphorylation might play a role in lowering titin-based stiffness [6, 9, 10]. Although this data suggests that exercise is a promising therapy for lowering diastolic stiffness, mechanistic
CE
studies and passive stiffness measurements are required. In the present study, we investigated the effects of exercise on the passive stiffness of LV wall
AC
muscle strips and evaluated both extracellular matrix (ECM) and titin-based stiffness. Wildtype (WT) mice were studied as were TtnΔIAjxn mice that have the I-A junction of titin removed resulting in increased diastolic stiffness and reduced exercise tolerance, mimicking aspects of HFpEF observed in patients[11]. A beneficial effect of exercise on passive stiffness was found in both genotypes (WT and TtnΔIAjxn) and extraction studies showed that the effect was derived from an increase in titin compliance. Additionally, we performed phosphorylation studies and focused on the phosphorylation status of the PEVK element, in which an increase in 3
ACCEPTED MANUSCRIPT phosphorylation has been shown to increase stiffness [12], and the N2B element, in which an increase in phosphorylation has been shown to decrease stiffness[13]. Because it is known that exercise reduces heart rate[14] we also studied the effect of the heart-rate lowering drug ivabradine.
Ivabradine is a selective inhibitor of the cardiac pacemaker If-current [15],
effectively lowering heart rate[16] with recent work that suggests that ivabradine improves
IP
T
diastolic function [17, 18]. Hence we included in our work a study of the effects of ivabradine on
AC
CE
PT
ED
M
AN
US
CR
passive stiffness.
4
ACCEPTED MANUSCRIPT MATERIALS
Mice Male 4-month-old mice on a C57BL/6J background were used for all experiments. TtnΔIAjxn mice had the I-A junction region of titin removed, for details see Granzier et al [11]. All
IP
T
experiments were approved by the University of Arizona Institutional Animal Care and Use Committee and followed the U.S. National Institutes of Health Using Animals in Intramural
CR
Research guidelines for animal use.
US
Exercise
For the exercise protocol, a free rotating running wheel exercise protocol was used as previously
AN
described [19]. Mice were housed individually in a cage that contained an 11.5cm-diameter
M
wheel with a 5.0cm wide running surface (6208; PetSmart; Phoenix, AZ) equipped with a digital magnetic counter (BC600, Sigma Sport, Olney Il) activated by wheel rotation. All animals were
ED
given water and standard rodent feed ad libitum. Daily exercise values for time and distance run were recorded and the average speed was calculated from these recorded values. Sedentary
PT
animals were housed individually in cages without running wheels. 28 days was selected in
CE
order to allow an initial acclimatization period of ~1 week followed by enough time to reach steady-state exercise levels and ensuing adaptations in protein expression.
AC
Ivabradine Treatment
Ivabradine (Waterstone Technology) was added to the drinking water for twenty-eight days. In order to get a correct dosage, mice where weighed and water intake tracked and ivabradine was dissolved in water to achieve a dose of 20 mg/kg BW/day. (The addition of the drug did not affect water intake.) Ivabradine is water soluble so controls received water alone. Heart rate was measured using the tail-cuff measurement system (MC4000 MultiChannel, Hatteras Instruments, Cary, NC). Briefly, mice are placed on a warmed holder and the tail is thread 5
ACCEPTED MANUSCRIPT through a balloon cuff and placed under a LED. Pulse is measured as a LED passes light through the tail and a photodiode detects changes in absorption due to blood flow with each beat. Mice were allowed to acclimate to the system for ~10 minutes before measurements were taken each day. In order to account for stress on the mice due to the system, measurements were
IP
T
then taken for three consecutive days each week with the last day taken as the measurement.
CR
Echocardiography
Echocardiography was performed as previously described [20-22] on ivabradine treated WT and
US
TtnΔIAjxn mice as well as untreated controls of both genotypes. Mice were anesthetized with 13% isoflurane (USP, Phoenix), then placed in dorsal recumbence on a heated platform for
AN
echocardiography. Transthoracic echo images were obtained with a Vevo 2100 High-Resolution
M
Imaging System (Visual-Sonics, Toronto, Canada) using the model MS550D scan head designed for murine cardiac imaging. Care was taken to avoid animal contact and excessive pressure
ED
which could induce bradycardia. Body temperature was maintained at 37°C. Ima ging was performed at a depth setting of 1 cm. Images were collected and stored as a digital cine loop for
PT
off-line calculations. Doppler and functional calculations were obtained according to American
CE
Society of Echocardiography guidelines. Passive LV filling peak velocity, E (cm/sec), and atrial contraction flow peak velocity, A (cm/sec), were acquired from the images of mitral valve
AC
Doppler flow from tilted parasternal long axis views. A sweep speed of 100 mm/sec was used for studies. The heart rate of animals during the echocardiographic study was maintained in the range of 500 -550 beats/min. Measures of diastolic function, specifically the relationship between early and late filling (E/A ratio) and how quickly the flow velocity declines in early diastole (E-wave deceleration time), are reported.
6
ACCEPTED MANUSCRIPT Muscle Mechanics Mice were placed under isoflurane anesthesia, cervically dislocated, and the ribcage was rapidly opened to access the heart. One ml of HEPES pH 7.4(in mM: NaCL, 133.5; KCl, 5; NaH2PO4, 1.2; MgSO4,1.2; HEPES, 10) solution containing and additional 30mM KCl and 3 mM BDM was injected into the LV through the apex, after which time the heart noticeably relaxed and ceased
IP
T
pumping. The heart was removed and the LV was isolated from the other chambers. The apex and base were removed from the LV leaving a cross-sectional slice approximately 2mm thick.
CR
The septum and RV attachment regions were discarded and the LV free wall tissue was placed in
US
relaxing solution (in mM; 20 BES, 10 EGTA, 6.56 MgCl2, 5.88 NaATP, 1 DTT, 46.35 Kpropionate, 15 creatine phosphate, pH 7.0). Endocardial fibers (apical to base orientation) were
AN
dissected and discarded. Once visualized, mid-myocardial fibers (circumferential orientation) were carefully removed and skinned in fresh relaxing solution pH 7.0 with 1% Triton-X-100
M
(Pierce, IL, USA) overnight at 4°C and protease inhibitors (phenylmethylsulfonyl fluoride
ED
(PMSF), 0.5 mM; leupeptin, 0.04mM; E64, 0.01mM), then washed for one hour with relaxing solution. Phosphatase inhibitor cocktail 2 (P5726, Sigma-Aldrich) 1 ml was added into 100 ml of
PT
relaxing solution. In experiments shown in supplementary figure 3 papillary muscles were also carefully dissected immediately after placing the LV in relaxing solution. Following a wash with
CE
relaxing solution, 150-250 μm (diameter) strips were dissected and aluminum clips were placed
AC
on both ends of the preparation. The aluminum clips were attached to a strain gauge force transducer and high-speed length motor and the preparation was submersed in relaxing solution. A laser diffraction system was used to measure the sarcomere length by taking advantage of Bragg’s law (dsinθ=mλ) to calculate the lattice spacing (d) based on the wavelength (λ), peak order (m), and angle (θ). The length and cross-sectional area (CSA) of the fiber then measured in order to normalize measured forces to CSA. Once the fiber was mounted and measured, a stretch-hold-release protocol was used in which a skinned fiber dissected from the circumferentially aligned free wall was stretched to a given sarcomere length(SL) within the 7
ACCEPTED MANUSCRIPT physiological range (2.0 to 2.3 µm) at 100%/s, held for 90 seconds, and subjected to a sinusoidal frequency sweep from 0.1 Hz to 100 Hz at +/-2%. The sinusoidal frequency sweep was used to quantify the viscous and elastic modulus. Viscous and elastic moduli are defined as (σ/ε)sin(θ) and (σ/ε)cos(θ) respectively where σ is stress, ε is strain, and θ is the phase shift. In order to quantify the relative contributions of titin and ECM, skinned muscle strips were extracted using
IP
T
first 0.6 M KCl in relaxing solution followed by 1.0 M KI in relaxing solution to depolymerize the thick and thin filaments respectively, removing the anchoring points of titin but keeping ECM
CR
intact[23, 24].
US
In this study, both skinned papillary muscles as well as muscle strips from the LV free wall at the mid-level myocardium where fibers are circumferentially aligned were used. Papillary
AN
muscles are a popular specimen for mechanical studies as they are generally easy to dissect and thought to reflect the properties of the heart wall. However, this assumption might not be valid
M
as ECM expression differs throughout the heart [25, 26]. LV free wall strips, while more difficult
ED
to dissect, are thus thought to be more representative of the surrounding myocardium. In order to test if there was indeed a difference between papillary muscles and LV free wall fibers from
PT
both regions of the heart were dissected and their properties compared.
CE
The TtnΔIAjxn is a previously published mouse model in which the I -band/A-band (IA) junction of titin was removed[11], see Supplemental Figure S1A. To further establish its
AC
suitability for the present work, passive stiffness was measured using the muscle mechanics protocol as described above. The sinusoidal frequency sweep protocol was also used to quantify the viscous and elastic modulus. Total, titin, and ECM-based passive tension were increased in the TtnΔIAjxn mice (Supplemental Figure S1B-D). Viscous and elastic moduli were calculated as the ratio of stress to strain multiplied by the sine or cosine of the phase shift respectively. Both the viscous and elastic moduli were higher in the TtnΔIAjxn compared to WT at all frequencies ranging from 0.1 Hz to 100 Hz (Supplemental Figure S2). Hence we confirmed the previously
8
ACCEPTED MANUSCRIPT published work showing that the TtnΔIAjxn is a suitable model for increased passive stiffness seen in HFpEF and we used this model to test the effect of exercise on passive stiffness when baseline stiffness is elevated.
T
Titin Isoform Expression
IP
After 28 days of exercise or ivabradine treatment, LV tissues were rapidly dissected and flash
CR
frozen in liquid nitrogen and solubilized between glass pestles cooled in liquid nitrogen. Tissues were primed at -20oC for a minimum of 20 min, then suspended in 50% urea buffer ([in mol/L]
US
8 Urea, 2 Thiourea, 0.05 Tris-HCl, 0.075 Dithiothreitol with 3% SDS and 0.03% Bromophenol blue pH 6.8) and 50% glycerol with protease inhibitors ([in mmol/L ] 0.04 E64, 0.16 Leupeptin
AN
and 0.2 PMSF) at 60oC for 10 min. Then the samples were centrifuged at 13000 rpm for 5 min,
M
aliquoted and flash frozen in liquid nitrogen and stored at -80oC [27]. Using agarose gel electrophoresis total titin, N2B titin, N2BA titin, and T2 were analyzed. Briefly, the solubilized
ED
samples were electrophoresed on 1% agarose gels using a vertical SDS-agarose gel system (Hoefer)[28, 29]. Gels were run at 15 mA per gel for 3 h and 20 min, then stained using
PT
Coomassie brilliant blue (Acros organics), scanned using a commercial scanner (Epson 800,
CE
Epson Corporation, Long Beach CA) and analyzed with One-D scan (Scanalytics Inc, Rockville MD). Each sample was loaded in a range of five volumes and the integrated optical density
AC
(IOD) of titin and MHC were determined as a function of loading volume. The slope of the linear relationship between IOD and loading was obtained for each protein to quantify expression ratios.[9]
Phosphorylation Solubilized samples were run on a 0.8% agarose gel in a vertical gel electrophoresis chamber. Gels run at 15 mA per gel for 3 h and 20 min were then transferred onto PVDF membranes
9
ACCEPTED MANUSCRIPT (Immobilon-FL, Millipore) using a semi-dry transfer unit (Trans-Blot Cell, Bio-Rad, Hercules CA). Blots were then probed with primary antibodies (described below) followed by secondary antibodies conjugated with fluorescent dyes with infrared excitation spectra (Biotium Company, Hayward CA). Blots were scanned using an Odyssey Infrared Imaging System (Li-COR Biosciences, Lincoln NE) and the images were analyzed using Li-COR software. Phosphorylation
IP
T
of the S11878 (GL Biochem Shanghai) and S12022 PKC(Genescript) sites in the PEVK element of titin was probed with primary phospho-specific antibodies that have been described
CR
previously[30]. Phosphorylation of the S4010 sites was probed via phospho-specific antibodies
US
(GL Biochem, Shanghai). Signal was normalized to total protein transferred quantified by Ponceau S(Sigma) staining (S11878, S12022). Ponceau S scans staining (S11878, S12022) or by
AN
primary antibody co-labeling (S4010) with the Z1Z2 titin antibody. Ponceau S was analyzed in One-D scan [9]. Samples were run in triplicate and an average of three technical replicates was
ED
M
taken as a biological sample.
PT
Statistical analysis
Data was analyzed (GraphPad Prism) 2-way ANOVA (Figures 1, 4, 6, 8) nonlinear (Figures 2, 3,
CE
7, S1-3) or linear (Figure 5) regression as appropriate. For regression, a statistical analysis was performed to test whether the data sets were best fit by one or two equations. The regression
AC
was performed with an extra sum-of-squares F test to determine whether there was evidence that the data sets differed from each other. The simpler model (i.e., single equation) was preferred for p-values > 0.05.
10
ACCEPTED MANUSCRIPT Results Exercise. Motivated by previous work that suggested that exercise may be beneficial for patients with HFpEF [8] we studied the effect of free wheel running exercise on diastolic function using both WT and TtnΔIAjxn mice. After an initial acclimatization of 1 week, WT mice
T
ran for ~6 hours each night and TtnΔIAjxns ran for an average of ~4 hours each night; towards
IP
the end of the 28-day regimen both groups ran ~4 hours each night. (Figure 1A). TtnΔIAjxn
CR
mice exhibited a reduction in both speed and distance (Figure 1B, C) at both the early and late time points although deficits became less pronounced with time. While TtnΔIAjxn mice exhibit
US
exercise deficiency compared to the WT, they do voluntarily exercise making it possible to
Muscle Mechanics.
AN
compare the effects of exercise on passive stiffness.
In pilot studies we compared the commonly used papillary muscle to
M
muscle strips dissected from the LV wall (see Methods for details). This revealed that papillary
ED
muscles had significantly higher passive tensions compared to LV wall muscle strips, regardless of genotype, and that this is largely due to increased ECM stiffness whereas titin-based tension
PT
is similar in the two preparation types (Supplemental Figure 3).
Thus muscle mechanical
CE
studies that seek insights that can be related to chamber stiffness require the use of LV wall muscle, as was done in our study. We first measured in LV wall muscle strips maximal active
AC
tension at a sarcomere length of 2.0 µm. No significant differences in active tension were observed due to exercise in either genotype (in mN/mm 2) WT Sed: 33.0±2.0, TtnΔIAjxn Sed: 32±2.0, WT Ex: 26.4±2.0, TtnΔIAjxn Ex: 28.0±2.6. Then we measured passive tensions up to a sarcomere length (SL) of 2.3µm.
The total passive tension was found to be reduced after
exercise in both genotypes (Figure 2A, D). The average total tension at SL=2.3 µm in the WT was reduced by 20±5% due to exercise compared to a 17±2% reduction in the KO; this difference was not statistically significant. No significant changes were observed in ECM-based passive stiffness in either genotype (Figure 2B and E). Exercise lowered titin-based tension in both 11
ACCEPTED MANUSCRIPT genotypes (Figure 2C and F). Titin passive tension decreased by 31±9% in the WT and 27±5% in the TtnΔIAjxn; this difference between the genotypes was not statistically significant. In the WT and TtnΔIAjxn, both viscous and elastic moduli were significantly reduced in exercised animals compared to sedentary controls (Figure 3). In order to compare between genotypes, the decrease across all measured frequencies was averaged. The elastic modulus decrease was not
IP
T
statistically different between groups: the elastic modulus of WT mice decreased by an average of 55±2% while the TtnΔIAjxn decreased by an average of 54±4%. Interestingly, a small but
CR
significantly different decrease in viscous modulus was seen after exercise: 46±1% decrease in
US
the WT compared to 50±2% decrease in the TtnΔIAjxn(p<0.01).
AN
Exercise-induced adaptation in Titin A stand-out finding of our study is that exercise lowers titin-based passive stiffness and to study the underlying mechanism we performed a titin No significant differences between genotype and/or exercise were found in
M
protein analysis.
ED
total titin expression, T2 expression (a titin degradation production) or N2B/N2BA isoform expression (Figure 4A and B). This indicates that the stiffness differences are due to alternate
PT
mechanisms. Titin has multiple phosphorylation sites which modulate stiffness[9, 19, 31, 32]. Here we tested the phospho-sites in titin’s PEVK region using phospho-specific antibodies.
CE
Western blots revealed no differences between genotypes and no differences in S12022 phosphorylation in response to exercise (Figure 4C and D). Interestingly an exercise-specific
AC
effect was seen in S11878 phosphorylation. Namely, Western blot analysis showed hypophosphorylation of the S11878 site after exercise in both TtnΔIAjxn mice and WT mice (Figure 4C and D). The mean decrease in signal for PS11878 was greater in the WT compared to the TtnΔIAjxn; the signal in the exercise compared to sedentary was decreased by ~40% in the WT compared to ~22% in the TtnΔIAjxn, but this difference did not reach statistical significance. Additionally, an exercise-specific effect was seen in S4010 phosphorylation. Western blot
12
ACCEPTED MANUSCRIPT analysis revealed hyper-phosphorylation of the S4010 site after exercise in both WT and TtnΔIAjxn mice (Figure 5D, C) with a mean increase of ~18% and ~20% respectively. In summary, exercise reduces titin-based passive tension in both WT and TtnΔIAjxn animals. No changes in titin expression or isoforms were observed due to exercise but exercise resulted in a hypo-phosphorylation of titin's S11878 site in the PEVK region and a hyper-phosphorylation of
IP
T
titin’s S4010 site in the N2B region.
CR
Ivabradine. Ivabradine is a funny channel inhibitor that acts to reduce heart rate [17, 18] and because exercise is known to lower heart rate[14] the effects of heart rate reduction on diastolic
US
function in WT and TtnΔIAjxn mice were studied. Mice treated with ivabradine had significantly lower heart rates compared to water alone regardless of genotype (Figure 5). Echocardiography
AN
revealed a reduced deceleration time of the Ewave in TtnΔIAjxn mice relative to WT mice,
M
consistent with the increased titin stiffness of this mouse and an earlier study[11], but no effect of ivabradine was seen in either TtnΔIAjxn or WT mice (Figure 6). E/E’ was trending lower
ED
(p=0.06) in TtnΔIAjxn compared to WT but ivabradine had no effect in either genotype.
PT
Likewise, muscle mechanics revealed no significant differences due to ivabradine treatment. Again, a genotype effect did exist, TtnΔIAjxn mice had higher passive tension than WT
CE
counterparts, but the drug had no effect on either genotype (Figure 7). Consistent with these observations, no changes in isoform expression nor in phosphorylation was seen between
AC
groups (Figure 8). In summary, ivabradine was effective at lowering the heart rate of mice but there were no measurable effects on diastolic function at the whole organ level (via Echo) nor at the muscle tissue level (via muscle mechanics), nor at the titin protein level (gel analysis).
13
ACCEPTED MANUSCRIPT Discussion Previous work has suggested that exercise may be beneficial for patients with HFpEF. The Exercise Training in Diastolic Heart Failure (Ex-DHF), a small 64 patient pilot study, concluded that exercise training improved exercise capacity and diastolic function in HFpEF patients[8].
We first demonstrated a location-based dependence of passive stiffness
IP
cardiac muscle.
T
Here TtnΔIAjxn and WT mice were used to test the effect of exercise on passive stiffness of
CR
measurements within the LV; specifically, papillary muscles have higher ECM-based tension than wall tissue and therefore we focused our work on solely wall muscle strips. These studies
US
revealed that voluntary exercise decreases titin-based tension in a phosphorylation-dependent manner in both WT and TtnΔIAjxn mice while heart-rate reduction (without exercise) revealed
AN
no stiffness differences in either genotype. Below we discuss these findings in detail.
M
TtnΔIAjxn Mouse and papillary vs. wall muscle. We report a significant difference in
ED
passive tension measurements within the LV depending on the tissue used for measurements. Specifically, the papillary muscles have higher passive tension than circumferentially aligned
PT
wall tissue, due to an increased ECM contribution; titin-based passive tension is largely unaffected by location (Supplementary Figure 3). These findings were consistent across
CE
genotypes. This result is of interest because myocardial stiffness depends mainly on titin and ECM with titin stiffness dominating at shorter sarcomere lengths and ECM taking over at longer
AC
lengths[33]. Both are known to be altered in disease states including HFpEF; it has been proposed that titin-based changes are more prominent early in the disease progression while ECM becomes more important at more advanced stages[34]. Understanding the relative contribution of these two components to total stiffness requires measurements on isolated muscle that is representative for the LV wall. We demonstrate that the location of the muscle that is studied is an important consideration in experimental design when the relative
14
ACCEPTED MANUSCRIPT contribution of titin to total stiffness is of interest and ECM stiffness values will be related to that of the LV chamber. The previously published TtnΔIAjxn mouse is a mechanical analog of elevated diastolic stiffness seen in heart failure with preserved ejection fraction[11]. Initial characterization showed that the
T
TtnΔIAjxn mouse exhibited increased chamber stiffness during diastole as shown through an
IP
increased diastolic stiffness parameter (β), and a decreased deceleration time of the E wave
CR
(Figure 6A). Here, these findings are confirmed via muscle mechanics and it additionally demonstrates that the TtnΔIAjxn exhibits exercise intolerance as seen in HFpEF patients
US
(Figure 1). No significant difference in active tension of muscle strips was observed between TtnΔIAjxn and WT mice indicating no major changes to myofilament cross-sectional area and
AN
that therefore the passive tension differences that we measured are likely due to a change intrinsic to the myofibrils. Furthermore, our work indicates that the viscosity of the TtnΔIAjxn
M
myocardium is significantly higher than that of WT myocardium. Possible mechanisms for
ED
increased viscosity include the increased strain on the spring region of titin resulting in increased interaction between the PEVK region and actin, or increased immunoglobulin
PT
unfolding both of which are known sources of viscosity[35-37]. Overall, the TtnΔIAjxn model is
CE
a suitable mechanical analog for increased passive stiffness seen in HFpEF. Exercise and titin. In order to test the effects of exercise, a protocol was used in which mice
AC
were able to freely access a running wheel. Passive stiffness was decreased in the exercised mice and this decrease is due to a reduction in titin-based stiffness with no detectable effect on ECM. Altered titin isoform expression ratios have been reported in patients with heart disease [3841] and Nagueh et al. have suggested that in patients an elevated N2BA/N2B ratio is associated with improved exercise tolerance [42]. However, the N2BA/N2B ratio was not altered in either WT or TntΔIAJxn mice due to exercise.
This finding is consistent both with the findings of
Hidalgo, et al.[6] and another 6-week voluntary running study in rats in which LV titin levels
15
ACCEPTED MANUSCRIPT were studied[43]. Thus, our study further establishes that the effect of exercise on titin stiffness in rodent models using 4-6 week exercise protocols do not involve titin isoform switching. However, in both WT and TntΔIAjxn mice we did find a reduction in PEVK phosphorylation, specifically a reduction in phosphorylation of S11878. The PEVK region of titin is known to be phosphorylated by PKC[9] and Hidalgo, et al. reported a reduction in PKC protein level in
IP
T
the LV of exercised mice[6]. This reduction is consistent with the hypo-phosphorylation of the
CR
PEVK element observed here. Both the S11878 and S12022 sites can be phosphorylated by PKC and it has been previously proposed that PKC has a stronger affinity towards S11878 than
US
S12022 due to differences in amino acid composition around the sites [44]. The exerciseinduced hypo-phosphorylation of S11878 but not S12022 in the mice studied here may be due to
AN
this differential effect. If these mice had a high basal phosphorylation level of S11878 and low basal phosphorylation of S12022 due to the stronger affinity in the former, a decrease in
M
PKCactivity may have a measurable effect on the S11878 site but not in the already low
ED
S12022.
PT
Phosphorylation of titin’s N2B element was also altered in both WT and TntΔIAjxn mice. Specifically, we observed an increase in phosphorylation of the S4010, a known PKA site which
CE
is known to reduce the stiffness of titin [45]. Exercise induces the release of catecholamines [46] which activates the β-adrenergic receptor pathway resulting in an increase in cAMP and, in turn,
AC
activation of PKA[47]. Thus, PKA phosphorylation of titin after exercise increases in both WT and TtnΔIAjxn.
A reduced degree of hypo-phosphorylation of S11878 after exercise was trending in TtnΔIAjxn mice but it did not reach statistical significance. Although WT mice ran for the same amount of time as TtnΔIAjxn they did run faster and cover more distance thus it is possible that phosphorylation level is a function of exercise intensity. It is worth noting that exercise protocols used here were not normalized -- that is to say, mice were allowed to exercise at will 16
ACCEPTED MANUSCRIPT and therefore exercised at varying durations and speeds. Because only total distance ran was recorded it is possible the different genotypes ran in different ways or at different times, i.e. in numerous short bursts vs long sustained runs or at different times throughout the day. These different types of exercise could affect phosphorylation status of cardiac titin. Interestingly, in the current study, exercise deficiency in the TtnΔIAjxn mice during the earlier time points (e.g.
IP
T
day 7-12) trends greater than the deficiency at later time points (e.g. day 21-28) indicating that decrease in titin stiffness due to exercise may, in turn, increase exercise capacity in the impaired
CR
animals. While other exercise protocols -- such as treadmill running or swimming -- may be
US
better to achieve the same level of exercise across groups they introduce the confounding factor of stress to the animal which in turn may affect phosphorylation results. Further work is needed
AN
to fully characterize the effect of different types of exercise on LV function in these models. Exercise studies from both our lab and others have previously seen changes in titin
M
phosphorylation due to exercise. A study by Müller, et al reported that acutely exercised rats (15
ED
minutes on a treadmill) exhibited increased titin stiffness relative to untrained controls in contrast to the decreased stiffness seen here[10]. This was also due to changes in
PT
phosphorylation -- specifically a 40% decrease in the Ser4099 site in the N2B element and an
CE
increase in Ser11878 phosphorylation in the PEVK [10]. However, these studies were performed after a single bout of exercise and thus measure the effect of acute rather than chronic exercise
AC
Additionally, variables such as age [48], stress[49], and time of day[50-52] when mice were sacrificed could alter initial phosphorylation and may account for these observed differences. Here we found hypo-phosphorylation of titin’s PEVK element due to chronic exercise. Hypo-phosphorylation of the PEVK element is known to reduce passive stiffness[9]. Additionally, we observed hyper-phosphorylation of titin’s N2B element after chronic exercise, an effect also known to reduce passive stiffness[13, 49]. Thus, these results are consistent with
17
ACCEPTED MANUSCRIPT the reduction in passive stiffness seen in exercised animals.
Viscosity. Total and titin-based passive stiffness were reduced in the exercised animals compared to sedentary controls. Interestingly, while similar decreases in passive tension and
T
elastic modulus were seen, the viscous modulus reduction was slightly but significantly greater
IP
in TtnΔIAjxn mice than in WT. The TtnΔIAjxn exhibits increased viscosity compared to WT
CR
possibly due to the increased strain on the spring region of titin resulting in increased interaction between the PEVK region and actin, a known source of viscosity [35, 36], or increases
US
in immunoglobulin unfolding[37]. The reduction of passive tension due to exercise may reduce the effect of these differences in the TtnΔIAjxn mice and thus cause a greater viscosity decrease
AN
in the TtnΔIAjxn compared to the WT. More in-depth study of the viscoelasticity of the
M
TtnΔIAjxn mouse is needed to fully explain this observation. Overall, the passive stiffness reduction we observed is consistent with the decreased PEVK
ED
phosphorylation and increased N2B phosphorylation. This work supports exercise as an effective tool for modulating passive stiffness and provides additional evidence for the beneficial
CE
PT
effects of exercise in HFpEF.
Heart Rate Reduction. Because exercise training is known to reduce resting heart rate[14],
AC
we tested whether heart-rate reduction would have a similar effect of exercise in the mice. The drug ivabradine was used due to both heart-rate lowering properties and evidence of titin isoform switching after ivabradine treatment in mice[17]. Here, we found that ivabradine did successfully lower the heart rate of mice of both genotypes; however, no significant stiffnessbased effects were observed. A possible explanation is that the TtnΔIAjxn model is a mild model of HFpEF for ivabradine to reveal differences. It may be that in more severe diastolic dysfunction increasing the time of diastole could, at least in part, compensate for the hea rt's 18
ACCEPTED MANUSCRIPT inability to relax by allowing the stiff heart more time to fill. In a previous study, Reil, et al compared control, diabetic (db/db) mice, and db/db mice treated with ivabradine for 4 weeks[17]. They reported that db/db mice had typical features of HFpEF as measured by pressure-volume analysis and that ivabradine treatment improved diastolic function in db/db mice. Additionally, they showed increased levels of N2B titin in db/db mice, known to increase
IP
T
passive stiffness, which was normalized by treatment with ivabradine. In our study, TtnΔIAjxns did not have an abnormal isoform ratio to begin with and thus ivabradine treatment may not be
CR
enough to change the equilibrium of titin isoforms. Furthermore, the db/db mice had severely
US
impaired relaxation[17], likely a major factor in their diastolic dysfunction phenotype, and thus may be well suited to the ivabradine mechanism resulting in increased filling time. Together,
AN
these studies indicate that while heart rate reduction may be a viable option for some
M
phenotypes, blanket use to treat HFpEF is not likely to result in improved outcomes in all cases.
ED
We conclude that exercise is an effective method for decreasing passive stiffness in both WT and TtnΔIAjxn mice and that this effect might be due to PEVK and N2B phosphorylation. This is of
patients [30, 40, 41].
PT
particular interest given that phosphorylation of titin is known to be dysregulated in HFpEF Zile, et al. reported that patients with HFpEF had deranged titin
CE
phosphorylation and, importantly that the PEVK’s S11878 was hyper-phosphorylated [30]. Furthermore, Zile, et al. found human HFpEF patients had increased N2B phosphorylation.
AC
Here, we demonstrated that exercise decreases PEVK phosphorylation and additionally decreases N2B phosphorylation. Thus exercise may be an effective therapeutic option for HFpEF and expansion of trials such as the Ex-DHF should be considered.
19
ACCEPTED MANUSCRIPT Acknowledgements We are grateful to our current and former lab members (in particular Chandra Saripalli, Luann Wyly, and Javier Perez) who contributed to this work and to the University of Arizona Genetic Engineering of Mouse Models (GEMM) and Mouse Phenotyping Core Facilities.
NIH
AC
CE
PT
ED
M
AN
US
CR
IP
T
HL062881 (HG) and HL118524 (HG), (H.L.) funded this research.
20
ACCEPTED MANUSCRIPT Captions Figure 1: Average distance, time, and speed of free wheel running exercise for WT and TtnΔIAjxn mice at two different time ranges. In the 7-12 days range (left) TtnΔIAjxn ran significantly less distance each night(A) less time(B) and slower (C) than WT. In the 21-28 range (right) TtnΔIAjxn ran significantly less distance each night(A) but ran an equal amount of time(B) at a lower speed (C) than WT.
CR
IP
T
Figure 2: Passive tension of LV wall muscle strips isolated from WT (A-C) and TtnΔIAjxn mice (D-F). Total passive tension was decreased in exercise compared to sedentary in both WT(A) and TtnΔIAjxn(D) mice. ECM-based tension was unchanged in WT B) and TtnΔIAjxn(E). Titin-based tension was decreased in exercise compared to sedentary in both genotypes at the higher sarcomere lengths (C,F). Thus exercise decreases the total and titinbased tension in both genotypes.
M
AN
US
Figure 3: Viscous and elastic moduli of LV muscle strips isolated from WT and TtnΔIAjxn sedentary and exercised mice. Viscous moduli (VM) and elastic moduli (EM) are obtained from a sinusoidal frequency sweep and are calculated as the ratio of the stress over the strain times the sine(VM) or cosine(EM) of the phase shift(see methods). Absolute VM and EM are measured in mN/mm2 ; here values obtained at a sarcomere length of 2.3 µm are plotted normalized to the mean value of sedentary mice at the lowest frequency (this value is set to 1). A,B) Viscous modulus was decreased in WT and TtnΔIAjxn with exercise (p<0.001). C,D) Elastic modulus was decreased in WT and TtnΔIAjxn with exercise (p<0.001).
CE
PT
ED
Figure 4: Titin Isoform expression and PKC site phosphorylation in WT and TtnΔIAjxn mice with or without exercise. A, B) Isoform expression (N2BA/N2B ratio), total titin (relative to MHC) and the titin degradation product T2 (relative to full-length titin, T1) were independent of genotype and exercise. B, C) PS12022 had no differences due to genotype or exercise while PS11878 was hypo-phosphorylated in both genotypes after exercise and PS4010 was hyper-phosphorylated in both genotypes after exercise. Figure 5: Heart rate reduction by ivabradine. Heart rate was found to be reduced in ivabradine treated groups compared to non-treated in both WT and TtnΔIAjxn.
AC
Figure 6. Effect of ivabradine on diastolic function. Pulse wave echocardiography. A. MV DT was significantly reduced in the TtnΔIAjxn mice compared to WT. However, treatment with ivabradine has no effect within each genotype. B. Neither genotype nor treatment had a significant effect on the E/A ratio. C. E/E’ was trending lower (note scale) in TtnΔIAjxn mice compared to WT(p=0.06) but treatment with ivabradine had no effect. Figure 7: Passive tension of LV wall muscle strips isolated from WT and TtnΔIAjxn mice with and without ivabradine. Total passive tension was unaffected by ivabradine treatment in both WT and TtnΔIAjxn (A,D). ECM-based(B,E) and titin-based(C,F) were likewise unaffected by ivabradine. Gray squares: -minus Ivabradine; black circles: + ivabradine (data points largely overlap).
21
ACCEPTED MANUSCRIPT Figure 8: Titin Isoform expression and PKC site phosphorylation in WT and TtnΔIAjxn mice with or without ivabradine. A,B) Isoform expression (N2BA/N2B ratio), total titin (relative to MHC) and the titin degradation product T2 (relative to full-length titin, T1) were independent of genotype and treatment. B,C) PS12022, PS11878, PS4010 likewise had no differences due to genotype or treatment
Supplemental Figure Captions
CR
IP
T
Figure S1: Passive tension of LV muscle strips isolated from WT and TtnΔIAjxn mice A) Total passive tension was increased in TtnΔIAjxn (p<0.001). B) ECM-based tension was also increased in TtnΔIAjxn. C) Titin-based tension was significantly (p<0.001) increased in TtnΔIAjxn.
M
AN
US
Figure S2: Viscous and elastic moduli of LV muscle strips isolated from WT and TtnΔIAjxn mice Viscous moduli (VM) and elastic moduli (EM) are obtained from a sinusoidal frequency sweep as the ratio of the stress over the strain times the sine(VM) or cosine(EM) of the phase shift(see methods). Absolute VM and EM are measured in mN/mm2 ; here values obtained at a sarcomere length of 2.3 µm on un-extracted fibers are plotted normalized to the value of the untreated sample at the lowest frequency(this value is set to 1) A) Viscous modulus was increased in TtnΔIAjxn (p<0.001). B) Elastic modulus was increased in TtnΔIAjxn mice (p<0.001). (Black symbols: WT; Grey: TtnΔIAjxn mice. )
AC
Disclosures None
CE
PT
ED
Figure S3: Passive tension of papillary muscle and LV wall muscle strips isolated from WT (A-C) and TtnΔIAjxn mice (D-F). Total passive tension was increased in papillary muscles compared to wall in both WT(A) and TtnΔIAjxn(D). ECM-based tension was increased in papillary muscles compared to wall tissue in both WT(B) and TtnΔIAjxn(E). Titinbased tension was unaffected in the WT but slightly but significantly higher in papillary muscles of the TtnΔIAjxn compared to wall of the same genotype (C,F). Thus papillaries over-represent the ECM-based tension relative to wall muscle but titin-based tension is largely the same.
22
ACCEPTED MANUSCRIPT References
AC
CE
PT
ED
M
AN
US
CR
IP
T
[1] Borlaug BA, Paulus WJ. Heart failure with preserved ejection fraction: pathophysiology, diagnosis, and treatment. Eur Heart J. 2010;32:670-9. [2] Borlaug BA. The pathophysiology of heart failure with preserved ejection fraction. Nat Rev Cardiol. 2014;11:507-15. [3] van Heerebeek L, Paulus WJ. Understanding heart failure with preserved ejection fraction: where are we today? Neth Heart J. 2016;24:227-36. [4] Redfield MM, Borlaug BA, Lewis GD, Mohammed SF, Semigran MJ, LeWinter MM, et al. PhosphdiesteRasE-5 Inhibition to Improve CLinical Status and EXercise Capacity in Diastolic Heart Failure (RELAX) Trial: Rationale and Design. Circ Heart Fail. 2012;5:653-9. [5] Gladden JD, Linke WA, Redfield MM. Heart failure with preserved ejection fraction. Pflugers Arch. 2014;466:1037-53. [6] Hidalgo C, Saripalli C, Granzier HL. Effect of exercise training on post-translational and posttranscriptional regulation of titin stiffness in striated muscle of wild type and IG KO mice. Arch Biochem Biophys. 2014;552-553:100-7. [7] Zile MR, Bourge RC, Redfield MM, Zhou D, Baicu CF, Little WC. Randomized, double-blind, placebo-controlled study of sitaxsentan to improve impaired exercise tolerance in patients with heart failure and a preserved ejection fraction. JACC Heart Fail. 2014;2:123-30. [8] Edelmann F, Gelbrich G, Düngen H-D, Fröhling S, Wachter R, Stahrenberg R, et al. Exercise training improves exercise capacity and diastolic function in patients with heart failure with preserved ejection fraction: results of the Ex-DHF (Exercise training in Diastolic Heart Failure) pilot study. J Am Coll Cardiol. 2011;58:1780-91. [9] Hidalgo C, Hudson B, Bogomolovas J, Zhu Y, Anderson B, Greaser M, et al. PKC phosphorylation of titin's PEVK element: a novel and conserved pathway for modulating myocardial stiffness. Circ Res. 2009;105:631-8, 17 p following 8. [10] Müller AE, Kreiner M, Kötter S, Lassak P, Bloch W, Suhr F, et al. Acute exercise modifies titin phosphorylation and increases cardiac myofilament stiffness. Front Physiol. 2014;5:449. [11] Granzier HL, Hutchinson KR, Tonino P, Methawasin M, Li FW, Slater RE, et al. Deleting titin's I-band/A-band junction reveals critical roles for titin in biomechanical sensing and cardiac function. Proc Natl Acad Sci U S A. 2014;111:14589-94. [12] Hudson BD, Hidalgo CG, Gotthardt M, Granzier HL. Excision of titin's cardiac PEVK spring element abolishes PKCalpha-induced increases in myocardial stiffness. J Mol Cell Cardiol. 2010;48:972-8. [13] Yamasaki R. Protein Kinase A Phosphorylates Titin's Cardiac-Specific N2B Domain and Reduces Passive Tension in Rat Cardiac Myocytes. Circulation Research. 2002;90:1181-8. [14] D’Souza A, Bucchi A, Johnsen AB, Logantha SJRJ, Monfredi O, Yanni J, et al. Exercise training reduces resting heart rate via downregulation of the funny channel HCN4. Nature Communications. 2014;5:3775. [15] Thollon C, Division Pathologies Cardiaques et Vasculaires IdRS, 11, rue des Moulineaux, 92150 Suresnes, France, Cambarrat C, Division Pathologies Cardiaques et Vasculaires IdRS, 11, rue des Moulineaux, 92150 Suresnes, France, Vian J, Division Pathologies Cardiaques et Vasculaires IdRS, 11, rue des Moulineaux, 92150 Suresnes, France, et al. Electrophysiological effects of S 16257, a novel sino‐atrial node modulator, on rabbit and guinea‐pig cardiac preparations: comparison with UL‐FS 49. British Journal of Pharmacology. 2016;112:37-42. [16] Bois P, Bescond J, Renaudon B, Lenfant J. Mode of action of bradycardic agent, S 16257, on ionic currents of rabbit sinoatrial node cells. Br J Pharmacol. 1996;118:1051-7. [17] Reil J-C, Hohl M, Reil G-H, Granzier HL, Kratz MT, Kazakov A, et al. Heart rate reduction by If-inhibition improves vascular stiffness and left ventricular systolic and diastolic function in a mouse model of heart failure with preserved ejection fraction. Eur Heart J. 2013;34:2839-49. 23
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
[18] Fang Y, Debunne M, Vercauteren M, Brakenhielm E, Richard V, Lallemand F, et al. Heart rate reduction induced by the if current inhibitor ivabradine improves diastolic function and attenuates cardiac tissue hypoxia. J Cardiovasc Pharmacol. 2012;59:260-7. [19] Chung CS, Hutchinson KR, Methawasin M, Saripalli C, Smith JE, 3rd, Hidalgo CG, et al. Shortening of the elastic tandem immunoglobulin segment of titin leads to diastolic dysfunction. Circulation. 2013;128:19-28. [20] Methawasin M, Strom JG, Slater RE, Fernandez V, Saripalli C, Granzier H. Experimentally Increasing the Compliance of Titin Through RNA Binding Motif-20 (RBM20) Inhibition Improves Diastolic Function In a Mouse Model of Heart Failure With Preserved Ejection Fraction. Circulation. 2016;134:1085-99. [21] Mohammed SF, Ohtani T, Korinek J, Lam CSP, Larsen K, Simari RD, et al. Mineralocorticoid accelerates transition to heart failure with preserved ejection fraction via "nongenomic effects". Circulation. 2010;122:370-8. [22] Bull M, Methawasin M, Strom J, Nair P, Hutchinson K, Granzier H. Alternative Splicing of Titin Restores Diastolic Function in an HFpEF-Like Genetic Murine Model (TtnΔIAjxn). Circ Res. 2016;119:764-72. [23] Wu Y, Cazorla O, Labeit D, Labeit S, Granzier H. Changes in titin and collagen underlie diastolic stiffness diversity of cardiac muscle. J Mol Cell Cardiol. 2000;32:2151-62. [24] Granzier HL, Irving TC. Passive tension in cardiac muscle: contribution of collagen, titin, microtubules, and intermediate filaments. Biophys J. 1995;68:1027-44. [25] Istratoaie O, OfiTeru AM, Nicola GC, Radu RI, Florescu C, Mogoanta L, et al. Myocardial interstitial fibrosis - histological and immunohistochemical aspects. Rom J Morphol Embryol. 2015;56:1473-80. [26] Fan D, Takawale A, Lee J, Kassiri Z. Cardiac fibroblasts, fibrosis and extracellular matrix remodeling in heart disease. Fibrogenesis Tissue Repair. 2012;5:15. [27] Lahmers S, Wu Y, Call DR, Labeit S, Granzier H. Developmental control of titin isoform expression and passive stiffness in fetal and neonatal myocardium. Circ Res. 2004;94:505-13. [28] Warren CM, Krzesinski PR, Greaser ML. Vertical agarose gel electrophoresis and electroblotting of high-molecular-weight proteins. Electrophoresis. 2003;24:1695-702. [29] Warren CM, Jordan MC, Roos KP, Krzesinski PR, Greaser ML. Titin isoform expression in normal and hypertensive myocardium. Cardiovasc Res. 2003;59:86-94. [30] Zile MR, Baicu CF, Ikonomidis JS, Stroud RE, Nietert PJ, Bradshaw AD, et al. Myocardial stiffness in patients with heart failure and a preserved ejection fraction: contributions of collagen and titin. Circulation. 2015;131:1247-59. [31] Hidalgo CG, Chung CS, Saripalli C, Methawasin M, Hutchinson KR, Tsaprailis G, et al. The multifunctional Ca(2+)/calmodulin-dependent protein kinase II delta (CaMKIIdelta) phosphorylates cardiac titin's spring elements. J Mol Cell Cardiol. 2013;54:90-7. [32] Hamdani N, Krysiak J, Kreusser MM, Neef S, Dos Remedios CG, Maier LS, et al. Crucial role for Ca2(+)/calmodulin-dependent protein kinase-II in regulating diastolic stress of normal and failing hearts via titin phosphorylation. Circ Res. 2013;112:664-74. [33] LeWinter MM, Granzier H. Cardiac titin: a multifunctional giant. Circulation. 2010;121:213745. [34] Franssen C, Gonzalez Miqueo A. The role of titin and extracellular matrix remodelling in heart failure with preserved ejection fraction. Neth Heart J. 2016;24:259-67. [35] Chung CS, Bogomolovas J, Gasch A, Hidalgo CG, Labeit S, Granzier HL. Titin-actin interaction: PEVK-actin-based viscosity in a large animal. J Biomed Biotechnol. 2011;2011:310791. [36] Chung CS, Methawasin M, Nelson OL, Radke MH, Hidalgo CG, Gotthardt M, et al. Titin based viscosity in ventricular physiology: an integrative investigation of PEVK-actin interactions. J Mol Cell Cardiol. 2011;51:428-34.
24
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
[37] Anderson BR, Bogomolovas J, Labeit S, Granzier H. Single molecule force spectroscopy on titin implicates immunoglobulin domain stability as a cardiac disease mechanism. J Biol Chem. 2013;288:5303-15. [38] Neagoe C, Kulke M, del Monte F, Gwathmey JK, de Tombe PP, Hajjar RJ, et al. Titin isoform switch in ischemic human heart disease. Circulation. 2002;106:1333-41. [39] Makarenko I, Opitz CA, Leake MC, Neagoe C, Kulke M, Gwathmey JK, et al. Passive stiffness changes caused by upregulation of compliant titin isoforms in human dilated cardiomyopathy hearts. Circ Res. 2004;95:708-16. [40] Borbely A. Cardiomyocyte Stiffness in Diastolic Heart Failure. Circulation. 2005;111:774-81. [41] Borbely A, Falcao-Pires I, van Heerebeek L, Hamdani N, Edes I, Gavina C, et al. Hypophosphorylation of the Stiff N2B titin isoform raises cardiomyocyte resting tension in failing human myocardium. Circ Res. 2009;104:780-6. [42] Nagueh SF, Shah G, Wu Y, Torre-Amione G, King NM, Lahmers S, et al. Altered titin expression, myocardial stiffness, and left ventricular function in patients with dilated cardiomyopathy. Circulation. 2004;110:155-62. [43] Bellafiore M, Cappello F, Palumbo D, Macaluso F, Bianco A, Palma A, et al. Increased expression of titin in mouse gastrocnemius muscle in response to an endurance-training program. Eur J Histochem. 2007;51:119-24. [44] Hudson B, Hidalgo C, Saripalli C, Granzier H. Hyperphosphorylation of mouse cardiac titin contributes to transverse aortic constriction-induced diastolic dysfunction. Circ Res. 2011;109:858-66. [45] Yamasaki R, Wu Y, McNabb M, Greaser M, Labeit S, Granzier H. Protein kinase A phosphorylates titin's cardiac-specific N2B domain and reduces passive tension in rat cardiac myocytes. Circ Res. 2002;90:1181-8. [46] Zouhal H, Jacob C, Delamarche P, Gratas-Delamarche A. Catecholamines and the effects of exercise, training and gender. Sports Med. 2008;38:401-23. [47] Kruger M, Linke WA. Protein kinase-A phosphorylates titin in human heart muscle and reduces myofibrillar passive tension. J Muscle Res Cell Motil. 2006;27:435-44. [48] Loffredo FS, Nikolova AP, Pancoast JR, Lee RT. Heart Failure with Preserved Ejection Fraction: Molecular Pathways of the Aging Myocardium. Circ Res. 2014;115:97-107. [49] Fukuda N, Wu Y, Nair P, Granzier HL. Phosphorylation of titin modulates passive stiffness of cardiac muscle in a titin isoform-dependent manner. J Gen Physiol. 2005;125:257-71. [50] Podobed PS, Alibhai FJ, Chow CW, Martino TA. Circadian regulation of myocardial sarcomeric Titin-cap (Tcap, telethonin): identification of cardiac clock-controlled genes using open access bioinformatics data. PLoS One. 2014;9:e104907. [51] Kinases and phosphatases in the mammalian circadian clock. 2011;585:1393–9. [52] Portaluppi F, Tiseo R, Smolensky MH, Hermida RC, Ayala DE, Fabbian F. Circadian rhythms and cardiovascular health. Sleep Med Rev. 2011;16:151-66.
25
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
Figure 1
26
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
Figure 2
27
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
Figure 3
28
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
Figure 4
29
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
Figure 5
30
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
Figure 6
31
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
Figure 7
32
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
Figure 8
33
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
Figure 9
34
ACCEPTED MANUSCRIPT Highlights
The therapeutic potential of exercise for improving diastolic function in a mouse model with HFpEF-like symptoms, the TtnΔIAjxn mouse model was studied.
Exercise lowered passive stiffness at the LV chamber and isolated muscle levels
Exercise altered the phosphorylation status of titin and this is predicted to contribute to the increased compliance of titin.
CE
PT
ED
M
AN
US
Exercise targets titin and improves diastolic health.
AC
CR
IP
T
with the effect largely due to an increase in titin compliance.
35
Rebecca E. Slater, Joshua G. Strom, Henk Granzier PII: DOI: Reference:
S0022-2828(17)30088-3 doi: 10.1016/j.yjmcc.2017.04.006 YJMCC 8538
To appear in:
Journal of Molecular and Cellular Cardiology
Received date: Revised date: Accepted date:
3 April 2017 24 April 2017 27 April 2017
Please cite this article as: Rebecca E. Slater, Joshua G. Strom, Henk Granzier , Effect of exercise on passive myocardial stiffness in mice with diastolic dysfunction. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Yjmcc(2017), doi: 10.1016/j.yjmcc.2017.04.006
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Effect of Exercise on Passive Myocardial Stiffness in Mice with Diastolic Dysfunction
Department of Cellular and Molecular Medicine, University of Arizona, Tucson,
IP
1
T
Rebecca E Slater 1 ; Joshua G Strom1 , and Henk Granzier 1
CR
AZ 85721; Sarver Molecular Cardiovascular Research Program, University of Arizona,
PT
ED
M
AN
US
Tucson, AZ 85721.
CE
Address for Correspondence: Henk Granzier, PhD
AC
Department of Molecular and Cellular Medicine MRB 325. 1656 E Mabel Street. University of Arizona, Tucson, AZ-85724-5217 Voice: (520)6263641; FAX:(520)6267600; Email:[email protected]
1
ACCEPTED MANUSCRIPT Abstract Heart failure with preserved ejection fraction (HFpEF) is a complex syndrome, characterized by increased diastolic stiffness and a preserved ejection fraction, with no effective treatment options. Here we studied the therapeutic potential of exercise for improving diastolic function in a mouse model with HFpEF-like symptoms, the TtnΔIAjxn mouse model. TtnΔIAjxn mice
IP
T
have increased diastolic stiffness and reduced exercise tolerance, mimicking aspects of HFpEF observed in patients. We investigated the effect of free-wheel running exercise on diastolic
CR
function. Mechanical studies on cardiac muscle strips from the LV free wall revealed that both
US
TtnΔIAjxn and wildtype (WT) exercised mice had a reduction in passive stiffness, relative to sedentary controls. In both genotypes, this reduction is due to an increase in the compliance of
AN
titin whereas ECM-based stiffness was unaffected. Phosphorylation of titin’s PEVK and N2B spring elements were assayed with phospho-site specific antibodies.
Exercised mice had
M
decreased PEVK phosphorylation and increased N2B phosphorylation both of which are
ED
predicted to contribute to the increased compliance of titin. Since exercise lowers the heart rate we examined whether reduction in heart rate per se can improve passive stiffness by
PT
administering the heart-rate-lowering drug ivabradine. Ivabradine lowered heart rate in our study but it did not affect passive tension, in neither WT nor TtnΔIAjxn mice. We conclude that
CE
exercise is beneficial for decreasing passive stiffness and that it involves beneficial alterations in
AC
titin phosphorylation.
2
ACCEPTED MANUSCRIPT INTRODUCTION Heart Failure with preserved Ejection Fraction (HFpEF) is a complex syndrome with diastolic stiffening that leads to higher diastolic pressures and insufficient filling, the culmination of which eventually leads to decreased cardiac reserve, and pulmonary edema[1, 2]. The prevalence of HFpEF is increasing and is predicted to exceed 8% of persons older than 65
T
Unlike heart failure with reduced ejection fraction (HFrEF),
IP
by the year 2020[3].
pharmacological approaches to treating HFpEF do not currently exist and clinical trials focused
CR
on HFpEF have had neutral results [4, 5].
US
Physical training has been reported to have beneficial effects on diastolic function under a range of conditions and earlier studies [6, 7] suggest that exercise could be a potential therapy
AN
for HFpEF through decreasing diastolic stiffness. The Exercise Training in Diastolic Heart Failure trial(Ex-DHF) studied HFpEF patients and concluded that exercise training improved
M
exercise capacity and diastolic function[8]. However, the mechanistic basis of how exercise
ED
affects passive stiffness in HFpEF is not well established. Recent studies have indicated that rodents have decreased phosphorylation at titin’s PEVK element following exercise and that this
PT
hypo-phosphorylation might play a role in lowering titin-based stiffness [6, 9, 10]. Although this data suggests that exercise is a promising therapy for lowering diastolic stiffness, mechanistic
CE
studies and passive stiffness measurements are required. In the present study, we investigated the effects of exercise on the passive stiffness of LV wall
AC
muscle strips and evaluated both extracellular matrix (ECM) and titin-based stiffness. Wildtype (WT) mice were studied as were TtnΔIAjxn mice that have the I-A junction of titin removed resulting in increased diastolic stiffness and reduced exercise tolerance, mimicking aspects of HFpEF observed in patients[11]. A beneficial effect of exercise on passive stiffness was found in both genotypes (WT and TtnΔIAjxn) and extraction studies showed that the effect was derived from an increase in titin compliance. Additionally, we performed phosphorylation studies and focused on the phosphorylation status of the PEVK element, in which an increase in 3
ACCEPTED MANUSCRIPT phosphorylation has been shown to increase stiffness [12], and the N2B element, in which an increase in phosphorylation has been shown to decrease stiffness[13]. Because it is known that exercise reduces heart rate[14] we also studied the effect of the heart-rate lowering drug ivabradine.
Ivabradine is a selective inhibitor of the cardiac pacemaker If-current [15],
effectively lowering heart rate[16] with recent work that suggests that ivabradine improves
IP
T
diastolic function [17, 18]. Hence we included in our work a study of the effects of ivabradine on
AC
CE
PT
ED
M
AN
US
CR
passive stiffness.
4
ACCEPTED MANUSCRIPT MATERIALS
Mice Male 4-month-old mice on a C57BL/6J background were used for all experiments. TtnΔIAjxn mice had the I-A junction region of titin removed, for details see Granzier et al [11]. All
IP
T
experiments were approved by the University of Arizona Institutional Animal Care and Use Committee and followed the U.S. National Institutes of Health Using Animals in Intramural
CR
Research guidelines for animal use.
US
Exercise
For the exercise protocol, a free rotating running wheel exercise protocol was used as previously
AN
described [19]. Mice were housed individually in a cage that contained an 11.5cm-diameter
M
wheel with a 5.0cm wide running surface (6208; PetSmart; Phoenix, AZ) equipped with a digital magnetic counter (BC600, Sigma Sport, Olney Il) activated by wheel rotation. All animals were
ED
given water and standard rodent feed ad libitum. Daily exercise values for time and distance run were recorded and the average speed was calculated from these recorded values. Sedentary
PT
animals were housed individually in cages without running wheels. 28 days was selected in
CE
order to allow an initial acclimatization period of ~1 week followed by enough time to reach steady-state exercise levels and ensuing adaptations in protein expression.
AC
Ivabradine Treatment
Ivabradine (Waterstone Technology) was added to the drinking water for twenty-eight days. In order to get a correct dosage, mice where weighed and water intake tracked and ivabradine was dissolved in water to achieve a dose of 20 mg/kg BW/day. (The addition of the drug did not affect water intake.) Ivabradine is water soluble so controls received water alone. Heart rate was measured using the tail-cuff measurement system (MC4000 MultiChannel, Hatteras Instruments, Cary, NC). Briefly, mice are placed on a warmed holder and the tail is thread 5
ACCEPTED MANUSCRIPT through a balloon cuff and placed under a LED. Pulse is measured as a LED passes light through the tail and a photodiode detects changes in absorption due to blood flow with each beat. Mice were allowed to acclimate to the system for ~10 minutes before measurements were taken each day. In order to account for stress on the mice due to the system, measurements were
IP
T
then taken for three consecutive days each week with the last day taken as the measurement.
CR
Echocardiography
Echocardiography was performed as previously described [20-22] on ivabradine treated WT and
US
TtnΔIAjxn mice as well as untreated controls of both genotypes. Mice were anesthetized with 13% isoflurane (USP, Phoenix), then placed in dorsal recumbence on a heated platform for
AN
echocardiography. Transthoracic echo images were obtained with a Vevo 2100 High-Resolution
M
Imaging System (Visual-Sonics, Toronto, Canada) using the model MS550D scan head designed for murine cardiac imaging. Care was taken to avoid animal contact and excessive pressure
ED
which could induce bradycardia. Body temperature was maintained at 37°C. Ima ging was performed at a depth setting of 1 cm. Images were collected and stored as a digital cine loop for
PT
off-line calculations. Doppler and functional calculations were obtained according to American
CE
Society of Echocardiography guidelines. Passive LV filling peak velocity, E (cm/sec), and atrial contraction flow peak velocity, A (cm/sec), were acquired from the images of mitral valve
AC
Doppler flow from tilted parasternal long axis views. A sweep speed of 100 mm/sec was used for studies. The heart rate of animals during the echocardiographic study was maintained in the range of 500 -550 beats/min. Measures of diastolic function, specifically the relationship between early and late filling (E/A ratio) and how quickly the flow velocity declines in early diastole (E-wave deceleration time), are reported.
6
ACCEPTED MANUSCRIPT Muscle Mechanics Mice were placed under isoflurane anesthesia, cervically dislocated, and the ribcage was rapidly opened to access the heart. One ml of HEPES pH 7.4(in mM: NaCL, 133.5; KCl, 5; NaH2PO4, 1.2; MgSO4,1.2; HEPES, 10) solution containing and additional 30mM KCl and 3 mM BDM was injected into the LV through the apex, after which time the heart noticeably relaxed and ceased
IP
T
pumping. The heart was removed and the LV was isolated from the other chambers. The apex and base were removed from the LV leaving a cross-sectional slice approximately 2mm thick.
CR
The septum and RV attachment regions were discarded and the LV free wall tissue was placed in
US
relaxing solution (in mM; 20 BES, 10 EGTA, 6.56 MgCl2, 5.88 NaATP, 1 DTT, 46.35 Kpropionate, 15 creatine phosphate, pH 7.0). Endocardial fibers (apical to base orientation) were
AN
dissected and discarded. Once visualized, mid-myocardial fibers (circumferential orientation) were carefully removed and skinned in fresh relaxing solution pH 7.0 with 1% Triton-X-100
M
(Pierce, IL, USA) overnight at 4°C and protease inhibitors (phenylmethylsulfonyl fluoride
ED
(PMSF), 0.5 mM; leupeptin, 0.04mM; E64, 0.01mM), then washed for one hour with relaxing solution. Phosphatase inhibitor cocktail 2 (P5726, Sigma-Aldrich) 1 ml was added into 100 ml of
PT
relaxing solution. In experiments shown in supplementary figure 3 papillary muscles were also carefully dissected immediately after placing the LV in relaxing solution. Following a wash with
CE
relaxing solution, 150-250 μm (diameter) strips were dissected and aluminum clips were placed
AC
on both ends of the preparation. The aluminum clips were attached to a strain gauge force transducer and high-speed length motor and the preparation was submersed in relaxing solution. A laser diffraction system was used to measure the sarcomere length by taking advantage of Bragg’s law (dsinθ=mλ) to calculate the lattice spacing (d) based on the wavelength (λ), peak order (m), and angle (θ). The length and cross-sectional area (CSA) of the fiber then measured in order to normalize measured forces to CSA. Once the fiber was mounted and measured, a stretch-hold-release protocol was used in which a skinned fiber dissected from the circumferentially aligned free wall was stretched to a given sarcomere length(SL) within the 7
ACCEPTED MANUSCRIPT physiological range (2.0 to 2.3 µm) at 100%/s, held for 90 seconds, and subjected to a sinusoidal frequency sweep from 0.1 Hz to 100 Hz at +/-2%. The sinusoidal frequency sweep was used to quantify the viscous and elastic modulus. Viscous and elastic moduli are defined as (σ/ε)sin(θ) and (σ/ε)cos(θ) respectively where σ is stress, ε is strain, and θ is the phase shift. In order to quantify the relative contributions of titin and ECM, skinned muscle strips were extracted using
IP
T
first 0.6 M KCl in relaxing solution followed by 1.0 M KI in relaxing solution to depolymerize the thick and thin filaments respectively, removing the anchoring points of titin but keeping ECM
CR
intact[23, 24].
US
In this study, both skinned papillary muscles as well as muscle strips from the LV free wall at the mid-level myocardium where fibers are circumferentially aligned were used. Papillary
AN
muscles are a popular specimen for mechanical studies as they are generally easy to dissect and thought to reflect the properties of the heart wall. However, this assumption might not be valid
M
as ECM expression differs throughout the heart [25, 26]. LV free wall strips, while more difficult
ED
to dissect, are thus thought to be more representative of the surrounding myocardium. In order to test if there was indeed a difference between papillary muscles and LV free wall fibers from
PT
both regions of the heart were dissected and their properties compared.
CE
The TtnΔIAjxn is a previously published mouse model in which the I -band/A-band (IA) junction of titin was removed[11], see Supplemental Figure S1A. To further establish its
AC
suitability for the present work, passive stiffness was measured using the muscle mechanics protocol as described above. The sinusoidal frequency sweep protocol was also used to quantify the viscous and elastic modulus. Total, titin, and ECM-based passive tension were increased in the TtnΔIAjxn mice (Supplemental Figure S1B-D). Viscous and elastic moduli were calculated as the ratio of stress to strain multiplied by the sine or cosine of the phase shift respectively. Both the viscous and elastic moduli were higher in the TtnΔIAjxn compared to WT at all frequencies ranging from 0.1 Hz to 100 Hz (Supplemental Figure S2). Hence we confirmed the previously
8
ACCEPTED MANUSCRIPT published work showing that the TtnΔIAjxn is a suitable model for increased passive stiffness seen in HFpEF and we used this model to test the effect of exercise on passive stiffness when baseline stiffness is elevated.
T
Titin Isoform Expression
IP
After 28 days of exercise or ivabradine treatment, LV tissues were rapidly dissected and flash
CR
frozen in liquid nitrogen and solubilized between glass pestles cooled in liquid nitrogen. Tissues were primed at -20oC for a minimum of 20 min, then suspended in 50% urea buffer ([in mol/L]
US
8 Urea, 2 Thiourea, 0.05 Tris-HCl, 0.075 Dithiothreitol with 3% SDS and 0.03% Bromophenol blue pH 6.8) and 50% glycerol with protease inhibitors ([in mmol/L ] 0.04 E64, 0.16 Leupeptin
AN
and 0.2 PMSF) at 60oC for 10 min. Then the samples were centrifuged at 13000 rpm for 5 min,
M
aliquoted and flash frozen in liquid nitrogen and stored at -80oC [27]. Using agarose gel electrophoresis total titin, N2B titin, N2BA titin, and T2 were analyzed. Briefly, the solubilized
ED
samples were electrophoresed on 1% agarose gels using a vertical SDS-agarose gel system (Hoefer)[28, 29]. Gels were run at 15 mA per gel for 3 h and 20 min, then stained using
PT
Coomassie brilliant blue (Acros organics), scanned using a commercial scanner (Epson 800,
CE
Epson Corporation, Long Beach CA) and analyzed with One-D scan (Scanalytics Inc, Rockville MD). Each sample was loaded in a range of five volumes and the integrated optical density
AC
(IOD) of titin and MHC were determined as a function of loading volume. The slope of the linear relationship between IOD and loading was obtained for each protein to quantify expression ratios.[9]
Phosphorylation Solubilized samples were run on a 0.8% agarose gel in a vertical gel electrophoresis chamber. Gels run at 15 mA per gel for 3 h and 20 min were then transferred onto PVDF membranes
9
ACCEPTED MANUSCRIPT (Immobilon-FL, Millipore) using a semi-dry transfer unit (Trans-Blot Cell, Bio-Rad, Hercules CA). Blots were then probed with primary antibodies (described below) followed by secondary antibodies conjugated with fluorescent dyes with infrared excitation spectra (Biotium Company, Hayward CA). Blots were scanned using an Odyssey Infrared Imaging System (Li-COR Biosciences, Lincoln NE) and the images were analyzed using Li-COR software. Phosphorylation
IP
T
of the S11878 (GL Biochem Shanghai) and S12022 PKC(Genescript) sites in the PEVK element of titin was probed with primary phospho-specific antibodies that have been described
CR
previously[30]. Phosphorylation of the S4010 sites was probed via phospho-specific antibodies
US
(GL Biochem, Shanghai). Signal was normalized to total protein transferred quantified by Ponceau S(Sigma) staining (S11878, S12022). Ponceau S scans staining (S11878, S12022) or by
AN
primary antibody co-labeling (S4010) with the Z1Z2 titin antibody. Ponceau S was analyzed in One-D scan [9]. Samples were run in triplicate and an average of three technical replicates was
ED
M
taken as a biological sample.
PT
Statistical analysis
Data was analyzed (GraphPad Prism) 2-way ANOVA (Figures 1, 4, 6, 8) nonlinear (Figures 2, 3,
CE
7, S1-3) or linear (Figure 5) regression as appropriate. For regression, a statistical analysis was performed to test whether the data sets were best fit by one or two equations. The regression
AC
was performed with an extra sum-of-squares F test to determine whether there was evidence that the data sets differed from each other. The simpler model (i.e., single equation) was preferred for p-values > 0.05.
10
ACCEPTED MANUSCRIPT Results Exercise. Motivated by previous work that suggested that exercise may be beneficial for patients with HFpEF [8] we studied the effect of free wheel running exercise on diastolic function using both WT and TtnΔIAjxn mice. After an initial acclimatization of 1 week, WT mice
T
ran for ~6 hours each night and TtnΔIAjxns ran for an average of ~4 hours each night; towards
IP
the end of the 28-day regimen both groups ran ~4 hours each night. (Figure 1A). TtnΔIAjxn
CR
mice exhibited a reduction in both speed and distance (Figure 1B, C) at both the early and late time points although deficits became less pronounced with time. While TtnΔIAjxn mice exhibit
US
exercise deficiency compared to the WT, they do voluntarily exercise making it possible to
Muscle Mechanics.
AN
compare the effects of exercise on passive stiffness.
In pilot studies we compared the commonly used papillary muscle to
M
muscle strips dissected from the LV wall (see Methods for details). This revealed that papillary
ED
muscles had significantly higher passive tensions compared to LV wall muscle strips, regardless of genotype, and that this is largely due to increased ECM stiffness whereas titin-based tension
PT
is similar in the two preparation types (Supplemental Figure 3).
Thus muscle mechanical
CE
studies that seek insights that can be related to chamber stiffness require the use of LV wall muscle, as was done in our study. We first measured in LV wall muscle strips maximal active
AC
tension at a sarcomere length of 2.0 µm. No significant differences in active tension were observed due to exercise in either genotype (in mN/mm 2) WT Sed: 33.0±2.0, TtnΔIAjxn Sed: 32±2.0, WT Ex: 26.4±2.0, TtnΔIAjxn Ex: 28.0±2.6. Then we measured passive tensions up to a sarcomere length (SL) of 2.3µm.
The total passive tension was found to be reduced after
exercise in both genotypes (Figure 2A, D). The average total tension at SL=2.3 µm in the WT was reduced by 20±5% due to exercise compared to a 17±2% reduction in the KO; this difference was not statistically significant. No significant changes were observed in ECM-based passive stiffness in either genotype (Figure 2B and E). Exercise lowered titin-based tension in both 11
ACCEPTED MANUSCRIPT genotypes (Figure 2C and F). Titin passive tension decreased by 31±9% in the WT and 27±5% in the TtnΔIAjxn; this difference between the genotypes was not statistically significant. In the WT and TtnΔIAjxn, both viscous and elastic moduli were significantly reduced in exercised animals compared to sedentary controls (Figure 3). In order to compare between genotypes, the decrease across all measured frequencies was averaged. The elastic modulus decrease was not
IP
T
statistically different between groups: the elastic modulus of WT mice decreased by an average of 55±2% while the TtnΔIAjxn decreased by an average of 54±4%. Interestingly, a small but
CR
significantly different decrease in viscous modulus was seen after exercise: 46±1% decrease in
US
the WT compared to 50±2% decrease in the TtnΔIAjxn(p<0.01).
AN
Exercise-induced adaptation in Titin A stand-out finding of our study is that exercise lowers titin-based passive stiffness and to study the underlying mechanism we performed a titin No significant differences between genotype and/or exercise were found in
M
protein analysis.
ED
total titin expression, T2 expression (a titin degradation production) or N2B/N2BA isoform expression (Figure 4A and B). This indicates that the stiffness differences are due to alternate
PT
mechanisms. Titin has multiple phosphorylation sites which modulate stiffness[9, 19, 31, 32]. Here we tested the phospho-sites in titin’s PEVK region using phospho-specific antibodies.
CE
Western blots revealed no differences between genotypes and no differences in S12022 phosphorylation in response to exercise (Figure 4C and D). Interestingly an exercise-specific
AC
effect was seen in S11878 phosphorylation. Namely, Western blot analysis showed hypophosphorylation of the S11878 site after exercise in both TtnΔIAjxn mice and WT mice (Figure 4C and D). The mean decrease in signal for PS11878 was greater in the WT compared to the TtnΔIAjxn; the signal in the exercise compared to sedentary was decreased by ~40% in the WT compared to ~22% in the TtnΔIAjxn, but this difference did not reach statistical significance. Additionally, an exercise-specific effect was seen in S4010 phosphorylation. Western blot
12
ACCEPTED MANUSCRIPT analysis revealed hyper-phosphorylation of the S4010 site after exercise in both WT and TtnΔIAjxn mice (Figure 5D, C) with a mean increase of ~18% and ~20% respectively. In summary, exercise reduces titin-based passive tension in both WT and TtnΔIAjxn animals. No changes in titin expression or isoforms were observed due to exercise but exercise resulted in a hypo-phosphorylation of titin's S11878 site in the PEVK region and a hyper-phosphorylation of
IP
T
titin’s S4010 site in the N2B region.
CR
Ivabradine. Ivabradine is a funny channel inhibitor that acts to reduce heart rate [17, 18] and because exercise is known to lower heart rate[14] the effects of heart rate reduction on diastolic
US
function in WT and TtnΔIAjxn mice were studied. Mice treated with ivabradine had significantly lower heart rates compared to water alone regardless of genotype (Figure 5). Echocardiography
AN
revealed a reduced deceleration time of the Ewave in TtnΔIAjxn mice relative to WT mice,
M
consistent with the increased titin stiffness of this mouse and an earlier study[11], but no effect of ivabradine was seen in either TtnΔIAjxn or WT mice (Figure 6). E/E’ was trending lower
ED
(p=0.06) in TtnΔIAjxn compared to WT but ivabradine had no effect in either genotype.
PT
Likewise, muscle mechanics revealed no significant differences due to ivabradine treatment. Again, a genotype effect did exist, TtnΔIAjxn mice had higher passive tension than WT
CE
counterparts, but the drug had no effect on either genotype (Figure 7). Consistent with these observations, no changes in isoform expression nor in phosphorylation was seen between
AC
groups (Figure 8). In summary, ivabradine was effective at lowering the heart rate of mice but there were no measurable effects on diastolic function at the whole organ level (via Echo) nor at the muscle tissue level (via muscle mechanics), nor at the titin protein level (gel analysis).
13
ACCEPTED MANUSCRIPT Discussion Previous work has suggested that exercise may be beneficial for patients with HFpEF. The Exercise Training in Diastolic Heart Failure (Ex-DHF), a small 64 patient pilot study, concluded that exercise training improved exercise capacity and diastolic function in HFpEF patients[8].
We first demonstrated a location-based dependence of passive stiffness
IP
cardiac muscle.
T
Here TtnΔIAjxn and WT mice were used to test the effect of exercise on passive stiffness of
CR
measurements within the LV; specifically, papillary muscles have higher ECM-based tension than wall tissue and therefore we focused our work on solely wall muscle strips. These studies
US
revealed that voluntary exercise decreases titin-based tension in a phosphorylation-dependent manner in both WT and TtnΔIAjxn mice while heart-rate reduction (without exercise) revealed
AN
no stiffness differences in either genotype. Below we discuss these findings in detail.
M
TtnΔIAjxn Mouse and papillary vs. wall muscle. We report a significant difference in
ED
passive tension measurements within the LV depending on the tissue used for measurements. Specifically, the papillary muscles have higher passive tension than circumferentially aligned
PT
wall tissue, due to an increased ECM contribution; titin-based passive tension is largely unaffected by location (Supplementary Figure 3). These findings were consistent across
CE
genotypes. This result is of interest because myocardial stiffness depends mainly on titin and ECM with titin stiffness dominating at shorter sarcomere lengths and ECM taking over at longer
AC
lengths[33]. Both are known to be altered in disease states including HFpEF; it has been proposed that titin-based changes are more prominent early in the disease progression while ECM becomes more important at more advanced stages[34]. Understanding the relative contribution of these two components to total stiffness requires measurements on isolated muscle that is representative for the LV wall. We demonstrate that the location of the muscle that is studied is an important consideration in experimental design when the relative
14
ACCEPTED MANUSCRIPT contribution of titin to total stiffness is of interest and ECM stiffness values will be related to that of the LV chamber. The previously published TtnΔIAjxn mouse is a mechanical analog of elevated diastolic stiffness seen in heart failure with preserved ejection fraction[11]. Initial characterization showed that the
T
TtnΔIAjxn mouse exhibited increased chamber stiffness during diastole as shown through an
IP
increased diastolic stiffness parameter (β), and a decreased deceleration time of the E wave
CR
(Figure 6A). Here, these findings are confirmed via muscle mechanics and it additionally demonstrates that the TtnΔIAjxn exhibits exercise intolerance as seen in HFpEF patients
US
(Figure 1). No significant difference in active tension of muscle strips was observed between TtnΔIAjxn and WT mice indicating no major changes to myofilament cross-sectional area and
AN
that therefore the passive tension differences that we measured are likely due to a change intrinsic to the myofibrils. Furthermore, our work indicates that the viscosity of the TtnΔIAjxn
M
myocardium is significantly higher than that of WT myocardium. Possible mechanisms for
ED
increased viscosity include the increased strain on the spring region of titin resulting in increased interaction between the PEVK region and actin, or increased immunoglobulin
PT
unfolding both of which are known sources of viscosity[35-37]. Overall, the TtnΔIAjxn model is
CE
a suitable mechanical analog for increased passive stiffness seen in HFpEF. Exercise and titin. In order to test the effects of exercise, a protocol was used in which mice
AC
were able to freely access a running wheel. Passive stiffness was decreased in the exercised mice and this decrease is due to a reduction in titin-based stiffness with no detectable effect on ECM. Altered titin isoform expression ratios have been reported in patients with heart disease [3841] and Nagueh et al. have suggested that in patients an elevated N2BA/N2B ratio is associated with improved exercise tolerance [42]. However, the N2BA/N2B ratio was not altered in either WT or TntΔIAJxn mice due to exercise.
This finding is consistent both with the findings of
Hidalgo, et al.[6] and another 6-week voluntary running study in rats in which LV titin levels
15
ACCEPTED MANUSCRIPT were studied[43]. Thus, our study further establishes that the effect of exercise on titin stiffness in rodent models using 4-6 week exercise protocols do not involve titin isoform switching. However, in both WT and TntΔIAjxn mice we did find a reduction in PEVK phosphorylation, specifically a reduction in phosphorylation of S11878. The PEVK region of titin is known to be phosphorylated by PKC[9] and Hidalgo, et al. reported a reduction in PKC protein level in
IP
T
the LV of exercised mice[6]. This reduction is consistent with the hypo-phosphorylation of the
CR
PEVK element observed here. Both the S11878 and S12022 sites can be phosphorylated by PKC and it has been previously proposed that PKC has a stronger affinity towards S11878 than
US
S12022 due to differences in amino acid composition around the sites [44]. The exerciseinduced hypo-phosphorylation of S11878 but not S12022 in the mice studied here may be due to
AN
this differential effect. If these mice had a high basal phosphorylation level of S11878 and low basal phosphorylation of S12022 due to the stronger affinity in the former, a decrease in
M
PKCactivity may have a measurable effect on the S11878 site but not in the already low
ED
S12022.
PT
Phosphorylation of titin’s N2B element was also altered in both WT and TntΔIAjxn mice. Specifically, we observed an increase in phosphorylation of the S4010, a known PKA site which
CE
is known to reduce the stiffness of titin [45]. Exercise induces the release of catecholamines [46] which activates the β-adrenergic receptor pathway resulting in an increase in cAMP and, in turn,
AC
activation of PKA[47]. Thus, PKA phosphorylation of titin after exercise increases in both WT and TtnΔIAjxn.
A reduced degree of hypo-phosphorylation of S11878 after exercise was trending in TtnΔIAjxn mice but it did not reach statistical significance. Although WT mice ran for the same amount of time as TtnΔIAjxn they did run faster and cover more distance thus it is possible that phosphorylation level is a function of exercise intensity. It is worth noting that exercise protocols used here were not normalized -- that is to say, mice were allowed to exercise at will 16
ACCEPTED MANUSCRIPT and therefore exercised at varying durations and speeds. Because only total distance ran was recorded it is possible the different genotypes ran in different ways or at different times, i.e. in numerous short bursts vs long sustained runs or at different times throughout the day. These different types of exercise could affect phosphorylation status of cardiac titin. Interestingly, in the current study, exercise deficiency in the TtnΔIAjxn mice during the earlier time points (e.g.
IP
T
day 7-12) trends greater than the deficiency at later time points (e.g. day 21-28) indicating that decrease in titin stiffness due to exercise may, in turn, increase exercise capacity in the impaired
CR
animals. While other exercise protocols -- such as treadmill running or swimming -- may be
US
better to achieve the same level of exercise across groups they introduce the confounding factor of stress to the animal which in turn may affect phosphorylation results. Further work is needed
AN
to fully characterize the effect of different types of exercise on LV function in these models. Exercise studies from both our lab and others have previously seen changes in titin
M
phosphorylation due to exercise. A study by Müller, et al reported that acutely exercised rats (15
ED
minutes on a treadmill) exhibited increased titin stiffness relative to untrained controls in contrast to the decreased stiffness seen here[10]. This was also due to changes in
PT
phosphorylation -- specifically a 40% decrease in the Ser4099 site in the N2B element and an
CE
increase in Ser11878 phosphorylation in the PEVK [10]. However, these studies were performed after a single bout of exercise and thus measure the effect of acute rather than chronic exercise
AC
Additionally, variables such as age [48], stress[49], and time of day[50-52] when mice were sacrificed could alter initial phosphorylation and may account for these observed differences. Here we found hypo-phosphorylation of titin’s PEVK element due to chronic exercise. Hypo-phosphorylation of the PEVK element is known to reduce passive stiffness[9]. Additionally, we observed hyper-phosphorylation of titin’s N2B element after chronic exercise, an effect also known to reduce passive stiffness[13, 49]. Thus, these results are consistent with
17
ACCEPTED MANUSCRIPT the reduction in passive stiffness seen in exercised animals.
Viscosity. Total and titin-based passive stiffness were reduced in the exercised animals compared to sedentary controls. Interestingly, while similar decreases in passive tension and
T
elastic modulus were seen, the viscous modulus reduction was slightly but significantly greater
IP
in TtnΔIAjxn mice than in WT. The TtnΔIAjxn exhibits increased viscosity compared to WT
CR
possibly due to the increased strain on the spring region of titin resulting in increased interaction between the PEVK region and actin, a known source of viscosity [35, 36], or increases
US
in immunoglobulin unfolding[37]. The reduction of passive tension due to exercise may reduce the effect of these differences in the TtnΔIAjxn mice and thus cause a greater viscosity decrease
AN
in the TtnΔIAjxn compared to the WT. More in-depth study of the viscoelasticity of the
M
TtnΔIAjxn mouse is needed to fully explain this observation. Overall, the passive stiffness reduction we observed is consistent with the decreased PEVK
ED
phosphorylation and increased N2B phosphorylation. This work supports exercise as an effective tool for modulating passive stiffness and provides additional evidence for the beneficial
CE
PT
effects of exercise in HFpEF.
Heart Rate Reduction. Because exercise training is known to reduce resting heart rate[14],
AC
we tested whether heart-rate reduction would have a similar effect of exercise in the mice. The drug ivabradine was used due to both heart-rate lowering properties and evidence of titin isoform switching after ivabradine treatment in mice[17]. Here, we found that ivabradine did successfully lower the heart rate of mice of both genotypes; however, no significant stiffnessbased effects were observed. A possible explanation is that the TtnΔIAjxn model is a mild model of HFpEF for ivabradine to reveal differences. It may be that in more severe diastolic dysfunction increasing the time of diastole could, at least in part, compensate for the hea rt's 18
ACCEPTED MANUSCRIPT inability to relax by allowing the stiff heart more time to fill. In a previous study, Reil, et al compared control, diabetic (db/db) mice, and db/db mice treated with ivabradine for 4 weeks[17]. They reported that db/db mice had typical features of HFpEF as measured by pressure-volume analysis and that ivabradine treatment improved diastolic function in db/db mice. Additionally, they showed increased levels of N2B titin in db/db mice, known to increase
IP
T
passive stiffness, which was normalized by treatment with ivabradine. In our study, TtnΔIAjxns did not have an abnormal isoform ratio to begin with and thus ivabradine treatment may not be
CR
enough to change the equilibrium of titin isoforms. Furthermore, the db/db mice had severely
US
impaired relaxation[17], likely a major factor in their diastolic dysfunction phenotype, and thus may be well suited to the ivabradine mechanism resulting in increased filling time. Together,
AN
these studies indicate that while heart rate reduction may be a viable option for some
M
phenotypes, blanket use to treat HFpEF is not likely to result in improved outcomes in all cases.
ED
We conclude that exercise is an effective method for decreasing passive stiffness in both WT and TtnΔIAjxn mice and that this effect might be due to PEVK and N2B phosphorylation. This is of
patients [30, 40, 41].
PT
particular interest given that phosphorylation of titin is known to be dysregulated in HFpEF Zile, et al. reported that patients with HFpEF had deranged titin
CE
phosphorylation and, importantly that the PEVK’s S11878 was hyper-phosphorylated [30]. Furthermore, Zile, et al. found human HFpEF patients had increased N2B phosphorylation.
AC
Here, we demonstrated that exercise decreases PEVK phosphorylation and additionally decreases N2B phosphorylation. Thus exercise may be an effective therapeutic option for HFpEF and expansion of trials such as the Ex-DHF should be considered.
19
ACCEPTED MANUSCRIPT Acknowledgements We are grateful to our current and former lab members (in particular Chandra Saripalli, Luann Wyly, and Javier Perez) who contributed to this work and to the University of Arizona Genetic Engineering of Mouse Models (GEMM) and Mouse Phenotyping Core Facilities.
NIH
AC
CE
PT
ED
M
AN
US
CR
IP
T
HL062881 (HG) and HL118524 (HG), (H.L.) funded this research.
20
ACCEPTED MANUSCRIPT Captions Figure 1: Average distance, time, and speed of free wheel running exercise for WT and TtnΔIAjxn mice at two different time ranges. In the 7-12 days range (left) TtnΔIAjxn ran significantly less distance each night(A) less time(B) and slower (C) than WT. In the 21-28 range (right) TtnΔIAjxn ran significantly less distance each night(A) but ran an equal amount of time(B) at a lower speed (C) than WT.
CR
IP
T
Figure 2: Passive tension of LV wall muscle strips isolated from WT (A-C) and TtnΔIAjxn mice (D-F). Total passive tension was decreased in exercise compared to sedentary in both WT(A) and TtnΔIAjxn(D) mice. ECM-based tension was unchanged in WT B) and TtnΔIAjxn(E). Titin-based tension was decreased in exercise compared to sedentary in both genotypes at the higher sarcomere lengths (C,F). Thus exercise decreases the total and titinbased tension in both genotypes.
M
AN
US
Figure 3: Viscous and elastic moduli of LV muscle strips isolated from WT and TtnΔIAjxn sedentary and exercised mice. Viscous moduli (VM) and elastic moduli (EM) are obtained from a sinusoidal frequency sweep and are calculated as the ratio of the stress over the strain times the sine(VM) or cosine(EM) of the phase shift(see methods). Absolute VM and EM are measured in mN/mm2 ; here values obtained at a sarcomere length of 2.3 µm are plotted normalized to the mean value of sedentary mice at the lowest frequency (this value is set to 1). A,B) Viscous modulus was decreased in WT and TtnΔIAjxn with exercise (p<0.001). C,D) Elastic modulus was decreased in WT and TtnΔIAjxn with exercise (p<0.001).
CE
PT
ED
Figure 4: Titin Isoform expression and PKC site phosphorylation in WT and TtnΔIAjxn mice with or without exercise. A, B) Isoform expression (N2BA/N2B ratio), total titin (relative to MHC) and the titin degradation product T2 (relative to full-length titin, T1) were independent of genotype and exercise. B, C) PS12022 had no differences due to genotype or exercise while PS11878 was hypo-phosphorylated in both genotypes after exercise and PS4010 was hyper-phosphorylated in both genotypes after exercise. Figure 5: Heart rate reduction by ivabradine. Heart rate was found to be reduced in ivabradine treated groups compared to non-treated in both WT and TtnΔIAjxn.
AC
Figure 6. Effect of ivabradine on diastolic function. Pulse wave echocardiography. A. MV DT was significantly reduced in the TtnΔIAjxn mice compared to WT. However, treatment with ivabradine has no effect within each genotype. B. Neither genotype nor treatment had a significant effect on the E/A ratio. C. E/E’ was trending lower (note scale) in TtnΔIAjxn mice compared to WT(p=0.06) but treatment with ivabradine had no effect. Figure 7: Passive tension of LV wall muscle strips isolated from WT and TtnΔIAjxn mice with and without ivabradine. Total passive tension was unaffected by ivabradine treatment in both WT and TtnΔIAjxn (A,D). ECM-based(B,E) and titin-based(C,F) were likewise unaffected by ivabradine. Gray squares: -minus Ivabradine; black circles: + ivabradine (data points largely overlap).
21
ACCEPTED MANUSCRIPT Figure 8: Titin Isoform expression and PKC site phosphorylation in WT and TtnΔIAjxn mice with or without ivabradine. A,B) Isoform expression (N2BA/N2B ratio), total titin (relative to MHC) and the titin degradation product T2 (relative to full-length titin, T1) were independent of genotype and treatment. B,C) PS12022, PS11878, PS4010 likewise had no differences due to genotype or treatment
Supplemental Figure Captions
CR
IP
T
Figure S1: Passive tension of LV muscle strips isolated from WT and TtnΔIAjxn mice A) Total passive tension was increased in TtnΔIAjxn (p<0.001). B) ECM-based tension was also increased in TtnΔIAjxn. C) Titin-based tension was significantly (p<0.001) increased in TtnΔIAjxn.
M
AN
US
Figure S2: Viscous and elastic moduli of LV muscle strips isolated from WT and TtnΔIAjxn mice Viscous moduli (VM) and elastic moduli (EM) are obtained from a sinusoidal frequency sweep as the ratio of the stress over the strain times the sine(VM) or cosine(EM) of the phase shift(see methods). Absolute VM and EM are measured in mN/mm2 ; here values obtained at a sarcomere length of 2.3 µm on un-extracted fibers are plotted normalized to the value of the untreated sample at the lowest frequency(this value is set to 1) A) Viscous modulus was increased in TtnΔIAjxn (p<0.001). B) Elastic modulus was increased in TtnΔIAjxn mice (p<0.001). (Black symbols: WT; Grey: TtnΔIAjxn mice. )
AC
Disclosures None
CE
PT
ED
Figure S3: Passive tension of papillary muscle and LV wall muscle strips isolated from WT (A-C) and TtnΔIAjxn mice (D-F). Total passive tension was increased in papillary muscles compared to wall in both WT(A) and TtnΔIAjxn(D). ECM-based tension was increased in papillary muscles compared to wall tissue in both WT(B) and TtnΔIAjxn(E). Titinbased tension was unaffected in the WT but slightly but significantly higher in papillary muscles of the TtnΔIAjxn compared to wall of the same genotype (C,F). Thus papillaries over-represent the ECM-based tension relative to wall muscle but titin-based tension is largely the same.
22
ACCEPTED MANUSCRIPT References
AC
CE
PT
ED
M
AN
US
CR
IP
T
[1] Borlaug BA, Paulus WJ. Heart failure with preserved ejection fraction: pathophysiology, diagnosis, and treatment. Eur Heart J. 2010;32:670-9. [2] Borlaug BA. The pathophysiology of heart failure with preserved ejection fraction. Nat Rev Cardiol. 2014;11:507-15. [3] van Heerebeek L, Paulus WJ. Understanding heart failure with preserved ejection fraction: where are we today? Neth Heart J. 2016;24:227-36. [4] Redfield MM, Borlaug BA, Lewis GD, Mohammed SF, Semigran MJ, LeWinter MM, et al. PhosphdiesteRasE-5 Inhibition to Improve CLinical Status and EXercise Capacity in Diastolic Heart Failure (RELAX) Trial: Rationale and Design. Circ Heart Fail. 2012;5:653-9. [5] Gladden JD, Linke WA, Redfield MM. Heart failure with preserved ejection fraction. Pflugers Arch. 2014;466:1037-53. [6] Hidalgo C, Saripalli C, Granzier HL. Effect of exercise training on post-translational and posttranscriptional regulation of titin stiffness in striated muscle of wild type and IG KO mice. Arch Biochem Biophys. 2014;552-553:100-7. [7] Zile MR, Bourge RC, Redfield MM, Zhou D, Baicu CF, Little WC. Randomized, double-blind, placebo-controlled study of sitaxsentan to improve impaired exercise tolerance in patients with heart failure and a preserved ejection fraction. JACC Heart Fail. 2014;2:123-30. [8] Edelmann F, Gelbrich G, Düngen H-D, Fröhling S, Wachter R, Stahrenberg R, et al. Exercise training improves exercise capacity and diastolic function in patients with heart failure with preserved ejection fraction: results of the Ex-DHF (Exercise training in Diastolic Heart Failure) pilot study. J Am Coll Cardiol. 2011;58:1780-91. [9] Hidalgo C, Hudson B, Bogomolovas J, Zhu Y, Anderson B, Greaser M, et al. PKC phosphorylation of titin's PEVK element: a novel and conserved pathway for modulating myocardial stiffness. Circ Res. 2009;105:631-8, 17 p following 8. [10] Müller AE, Kreiner M, Kötter S, Lassak P, Bloch W, Suhr F, et al. Acute exercise modifies titin phosphorylation and increases cardiac myofilament stiffness. Front Physiol. 2014;5:449. [11] Granzier HL, Hutchinson KR, Tonino P, Methawasin M, Li FW, Slater RE, et al. Deleting titin's I-band/A-band junction reveals critical roles for titin in biomechanical sensing and cardiac function. Proc Natl Acad Sci U S A. 2014;111:14589-94. [12] Hudson BD, Hidalgo CG, Gotthardt M, Granzier HL. Excision of titin's cardiac PEVK spring element abolishes PKCalpha-induced increases in myocardial stiffness. J Mol Cell Cardiol. 2010;48:972-8. [13] Yamasaki R. Protein Kinase A Phosphorylates Titin's Cardiac-Specific N2B Domain and Reduces Passive Tension in Rat Cardiac Myocytes. Circulation Research. 2002;90:1181-8. [14] D’Souza A, Bucchi A, Johnsen AB, Logantha SJRJ, Monfredi O, Yanni J, et al. Exercise training reduces resting heart rate via downregulation of the funny channel HCN4. Nature Communications. 2014;5:3775. [15] Thollon C, Division Pathologies Cardiaques et Vasculaires IdRS, 11, rue des Moulineaux, 92150 Suresnes, France, Cambarrat C, Division Pathologies Cardiaques et Vasculaires IdRS, 11, rue des Moulineaux, 92150 Suresnes, France, Vian J, Division Pathologies Cardiaques et Vasculaires IdRS, 11, rue des Moulineaux, 92150 Suresnes, France, et al. Electrophysiological effects of S 16257, a novel sino‐atrial node modulator, on rabbit and guinea‐pig cardiac preparations: comparison with UL‐FS 49. British Journal of Pharmacology. 2016;112:37-42. [16] Bois P, Bescond J, Renaudon B, Lenfant J. Mode of action of bradycardic agent, S 16257, on ionic currents of rabbit sinoatrial node cells. Br J Pharmacol. 1996;118:1051-7. [17] Reil J-C, Hohl M, Reil G-H, Granzier HL, Kratz MT, Kazakov A, et al. Heart rate reduction by If-inhibition improves vascular stiffness and left ventricular systolic and diastolic function in a mouse model of heart failure with preserved ejection fraction. Eur Heart J. 2013;34:2839-49. 23
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
[18] Fang Y, Debunne M, Vercauteren M, Brakenhielm E, Richard V, Lallemand F, et al. Heart rate reduction induced by the if current inhibitor ivabradine improves diastolic function and attenuates cardiac tissue hypoxia. J Cardiovasc Pharmacol. 2012;59:260-7. [19] Chung CS, Hutchinson KR, Methawasin M, Saripalli C, Smith JE, 3rd, Hidalgo CG, et al. Shortening of the elastic tandem immunoglobulin segment of titin leads to diastolic dysfunction. Circulation. 2013;128:19-28. [20] Methawasin M, Strom JG, Slater RE, Fernandez V, Saripalli C, Granzier H. Experimentally Increasing the Compliance of Titin Through RNA Binding Motif-20 (RBM20) Inhibition Improves Diastolic Function In a Mouse Model of Heart Failure With Preserved Ejection Fraction. Circulation. 2016;134:1085-99. [21] Mohammed SF, Ohtani T, Korinek J, Lam CSP, Larsen K, Simari RD, et al. Mineralocorticoid accelerates transition to heart failure with preserved ejection fraction via "nongenomic effects". Circulation. 2010;122:370-8. [22] Bull M, Methawasin M, Strom J, Nair P, Hutchinson K, Granzier H. Alternative Splicing of Titin Restores Diastolic Function in an HFpEF-Like Genetic Murine Model (TtnΔIAjxn). Circ Res. 2016;119:764-72. [23] Wu Y, Cazorla O, Labeit D, Labeit S, Granzier H. Changes in titin and collagen underlie diastolic stiffness diversity of cardiac muscle. J Mol Cell Cardiol. 2000;32:2151-62. [24] Granzier HL, Irving TC. Passive tension in cardiac muscle: contribution of collagen, titin, microtubules, and intermediate filaments. Biophys J. 1995;68:1027-44. [25] Istratoaie O, OfiTeru AM, Nicola GC, Radu RI, Florescu C, Mogoanta L, et al. Myocardial interstitial fibrosis - histological and immunohistochemical aspects. Rom J Morphol Embryol. 2015;56:1473-80. [26] Fan D, Takawale A, Lee J, Kassiri Z. Cardiac fibroblasts, fibrosis and extracellular matrix remodeling in heart disease. Fibrogenesis Tissue Repair. 2012;5:15. [27] Lahmers S, Wu Y, Call DR, Labeit S, Granzier H. Developmental control of titin isoform expression and passive stiffness in fetal and neonatal myocardium. Circ Res. 2004;94:505-13. [28] Warren CM, Krzesinski PR, Greaser ML. Vertical agarose gel electrophoresis and electroblotting of high-molecular-weight proteins. Electrophoresis. 2003;24:1695-702. [29] Warren CM, Jordan MC, Roos KP, Krzesinski PR, Greaser ML. Titin isoform expression in normal and hypertensive myocardium. Cardiovasc Res. 2003;59:86-94. [30] Zile MR, Baicu CF, Ikonomidis JS, Stroud RE, Nietert PJ, Bradshaw AD, et al. Myocardial stiffness in patients with heart failure and a preserved ejection fraction: contributions of collagen and titin. Circulation. 2015;131:1247-59. [31] Hidalgo CG, Chung CS, Saripalli C, Methawasin M, Hutchinson KR, Tsaprailis G, et al. The multifunctional Ca(2+)/calmodulin-dependent protein kinase II delta (CaMKIIdelta) phosphorylates cardiac titin's spring elements. J Mol Cell Cardiol. 2013;54:90-7. [32] Hamdani N, Krysiak J, Kreusser MM, Neef S, Dos Remedios CG, Maier LS, et al. Crucial role for Ca2(+)/calmodulin-dependent protein kinase-II in regulating diastolic stress of normal and failing hearts via titin phosphorylation. Circ Res. 2013;112:664-74. [33] LeWinter MM, Granzier H. Cardiac titin: a multifunctional giant. Circulation. 2010;121:213745. [34] Franssen C, Gonzalez Miqueo A. The role of titin and extracellular matrix remodelling in heart failure with preserved ejection fraction. Neth Heart J. 2016;24:259-67. [35] Chung CS, Bogomolovas J, Gasch A, Hidalgo CG, Labeit S, Granzier HL. Titin-actin interaction: PEVK-actin-based viscosity in a large animal. J Biomed Biotechnol. 2011;2011:310791. [36] Chung CS, Methawasin M, Nelson OL, Radke MH, Hidalgo CG, Gotthardt M, et al. Titin based viscosity in ventricular physiology: an integrative investigation of PEVK-actin interactions. J Mol Cell Cardiol. 2011;51:428-34.
24
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
[37] Anderson BR, Bogomolovas J, Labeit S, Granzier H. Single molecule force spectroscopy on titin implicates immunoglobulin domain stability as a cardiac disease mechanism. J Biol Chem. 2013;288:5303-15. [38] Neagoe C, Kulke M, del Monte F, Gwathmey JK, de Tombe PP, Hajjar RJ, et al. Titin isoform switch in ischemic human heart disease. Circulation. 2002;106:1333-41. [39] Makarenko I, Opitz CA, Leake MC, Neagoe C, Kulke M, Gwathmey JK, et al. Passive stiffness changes caused by upregulation of compliant titin isoforms in human dilated cardiomyopathy hearts. Circ Res. 2004;95:708-16. [40] Borbely A. Cardiomyocyte Stiffness in Diastolic Heart Failure. Circulation. 2005;111:774-81. [41] Borbely A, Falcao-Pires I, van Heerebeek L, Hamdani N, Edes I, Gavina C, et al. Hypophosphorylation of the Stiff N2B titin isoform raises cardiomyocyte resting tension in failing human myocardium. Circ Res. 2009;104:780-6. [42] Nagueh SF, Shah G, Wu Y, Torre-Amione G, King NM, Lahmers S, et al. Altered titin expression, myocardial stiffness, and left ventricular function in patients with dilated cardiomyopathy. Circulation. 2004;110:155-62. [43] Bellafiore M, Cappello F, Palumbo D, Macaluso F, Bianco A, Palma A, et al. Increased expression of titin in mouse gastrocnemius muscle in response to an endurance-training program. Eur J Histochem. 2007;51:119-24. [44] Hudson B, Hidalgo C, Saripalli C, Granzier H. Hyperphosphorylation of mouse cardiac titin contributes to transverse aortic constriction-induced diastolic dysfunction. Circ Res. 2011;109:858-66. [45] Yamasaki R, Wu Y, McNabb M, Greaser M, Labeit S, Granzier H. Protein kinase A phosphorylates titin's cardiac-specific N2B domain and reduces passive tension in rat cardiac myocytes. Circ Res. 2002;90:1181-8. [46] Zouhal H, Jacob C, Delamarche P, Gratas-Delamarche A. Catecholamines and the effects of exercise, training and gender. Sports Med. 2008;38:401-23. [47] Kruger M, Linke WA. Protein kinase-A phosphorylates titin in human heart muscle and reduces myofibrillar passive tension. J Muscle Res Cell Motil. 2006;27:435-44. [48] Loffredo FS, Nikolova AP, Pancoast JR, Lee RT. Heart Failure with Preserved Ejection Fraction: Molecular Pathways of the Aging Myocardium. Circ Res. 2014;115:97-107. [49] Fukuda N, Wu Y, Nair P, Granzier HL. Phosphorylation of titin modulates passive stiffness of cardiac muscle in a titin isoform-dependent manner. J Gen Physiol. 2005;125:257-71. [50] Podobed PS, Alibhai FJ, Chow CW, Martino TA. Circadian regulation of myocardial sarcomeric Titin-cap (Tcap, telethonin): identification of cardiac clock-controlled genes using open access bioinformatics data. PLoS One. 2014;9:e104907. [51] Kinases and phosphatases in the mammalian circadian clock. 2011;585:1393–9. [52] Portaluppi F, Tiseo R, Smolensky MH, Hermida RC, Ayala DE, Fabbian F. Circadian rhythms and cardiovascular health. Sleep Med Rev. 2011;16:151-66.
25
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
Figure 1
26
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
Figure 2
27
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
Figure 3
28
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
Figure 4
29
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
Figure 5
30
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
Figure 6
31
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
Figure 7
32
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
Figure 8
33
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
IP
T
Figure 9
34
ACCEPTED MANUSCRIPT Highlights
The therapeutic potential of exercise for improving diastolic function in a mouse model with HFpEF-like symptoms, the TtnΔIAjxn mouse model was studied.
Exercise lowered passive stiffness at the LV chamber and isolated muscle levels
Exercise altered the phosphorylation status of titin and this is predicted to contribute to the increased compliance of titin.
CE
PT
ED
M
AN
US
Exercise targets titin and improves diastolic health.
AC
CR
IP
T
with the effect largely due to an increase in titin compliance.
35