Treatment of heart failure with preserved ejection fraction

Abstract

The treatment of diastolic dysfunction in a subject and, in particular, compositions for use and methods in treating diastolic dysfunction are disclosed. In one aspect, the disclosure provides a composition for use comprising a therapeutically effective amount of a substance that increases the level and/or activity of a crystalline protein in cardiomyocytes of the subject. In particular, the composition for use according to the disclosure comprises a therapeutically effective amount of the alpha B crystalline protein. It was found that the composition for use is capable of treating diastolic dysfunction and diastolic heart disease in subjects with failing hearts that have a higher cardiomyocyte stiffness than controls. Particularly, addition of alpha B crystalline reduced the higher stiffness of failing cardiomyocytes to the level observed in control cardiomyocytes.

Claims

1. A method of treating a subject for heart failure with preserved ejection fraction (HFpEF) characterized by diastolic dysfunction caused by diastolic stiffness of cardiomyocytes and interstitial collagen deposition, the method comprising: administering to the subject a therapeutically effective amount of a substance that increases the level and/or activity of a crystalline protein in cardiomyocytes of the subject, wherein the substance is geranylgeranylacetone (GGA) or NYK9354.

2. The method according to claim 1, wherein the crystalline protein is alpha B-crystallin.

3. The method according to claim 2, wherein the therapeutically effective amount further reduces diastolic stiffness of cardiomyocytes in the subject.

4. A method for treating a subject for heart failure with preserved ejection fraction (HFpEF) characterized by diastolic dysfunction caused by diastolic stiffness of cardiomyocytes and interstitial collagen deposition, or for aortic stenosis, or is at risk of developing said heart failure or said aortic stenosis, the method comprising: administering to the subject a composition comprising a therapeutically effective amount of a substance that increases the level and/or activity of a crystalline protein in the subject's cardiomyocytes, wherein the substance is geranylgeranylacetone (GGA) or NYK9354.

5. The method according to claim 4, wherein the crystalline protein is a protein encoded by the crystallin alpha B (CRYAB) gene.

6. The method according to claim 5, wherein the substance mediates a post-translational modification of alpha-crystallin B chain.

7. The method according to claim 6, wherein the substance mediates the post-translational modification of alpha-crystallin B chain by phosphorylation of alpha-crystallin B chain.

8. The method according to claim 4, wherein the substance increases the level of alpha-crystallin B chain in cardiomyocytes of the subject.

9. The method according to claim 8, wherein the substance induces expression of the CRYAB gene in the cardiomyocytes of the subject.

10. The method according to claim 4, wherein the substance mediates a post-translational modification of alpha-crystallin B chain.

11. The method according to claim 10, wherein the substance mediates the post-translational modification of alpha-crystallin B chain by phosphorylation of alpha-crystallin B chain.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIGS. 1A and 1B: F.sub.passive in myocardial strips. At baseline, F.sub.passive was significantly higher in dilated cardiomyopathy (DCM) strips and aortic stenosis (AS) strips compared to donor (FIG. 1A). After extracting thick and thin filaments, the remaining F.sub.passive can be attributed to the E-matrix and this was higher in DCM and AS than in donor (FIG. 1B). *P<0.05 DCM vs Donor; #P<0.05 AS vs Donor.

(2) FIGS. 2A and 2B: F.sub.passive caused by the E-matrix. AS strips had significantly higher E-matrix based F.sub.passive than donor in stretches exceeding 15% from slack length (FIG. 2A). At all stages of muscle strain, F.sub.passive caused by the E-matrix was higher in DCM strips than in donor (FIG. 2B). #P<0.05 AS vs Donor; *P<0.05 DCM vs Donor.

(3) FIGS. 3A and 3B: F.sub.passive caused by titin. In AS strips, titin-based F.sub.passive was higher than in donor strips at 10-20% stretches, but not at the highest stretches (FIG. 3A). At all stages of muscle strain, F.sub.passive caused by titin was higher in DCM strips than in donor (FIG. 3B). #P<0.05 AS vs Donor; *P<0.05 DCM vs Donor.

(4) FIGS. 4A-4C: F.sub.passive in single myocytes. FIG. 4A: In donor cardiomyocytes, F.sub.passive significantly increased after administration of alkaline phosphatase and increased further after performing a prestretch in an acidic environment. After in vitro administration of α-B crystallin, F.sub.passive normalized again to baseline. FIG. 4B: In AS cardiomyocytes, no significant change in F.sub.passive was observed after incubation with alkaline phosphatases (AP). After the prestretch in pH 6.6, F.sub.passive increased significantly compared to baseline, but in vitro treatment with α-B crystallin dropped F.sub.passive to a level significantly lower than baseline. FIG. 4C: In DCM cardiomyocytes, incubation with AP had no effect on F.sub.passive, but performing a prestretch in an acidic environment did increase passive stiffness significantly. After in vitro treatment with α-B crystallin, F.sub.passive dropped to a level significantly below baseline. *P<0.05 AP vs Baseline; #P<0.05 pH 6.6+Prestretch vs AP; ‡ P<0.05 α-B crystallin vs pH 6.6+Prestretch; § P<0.05 α-B crystallin vs Baseline.

(5) FIGS. 5A and 5B: Effects of Acidosis and Prestretch on Fpassive in single AS myocytes. Incubating single AS cardiomyocytes in pH 6.6 did not change F.sub.passive (FIG. 5A). Performing a prestretch to ˜2.6 μm SL increased F.sub.passive, but only significantly during higher stretches (FIG. 5B). *P<0.05 vs AS Baseline.

(6) FIGS. 6A-6C: Effects of α-B crystallin at Baseline and after Acidosis and Prestretch on Fpassive in single AS cardiomyocytes. Incubation with α-B crystallin decreased F.sub.passive significantly (FIG. 6A). In single AS cardiomyocytes that underwent prestretch alone (FIG. 6B) or a prestretch in combination with pH 6.6 (FIG. 6C), α-B crystallin decreased F.sub.passive significantly as well. *P<0.05 vs AS Baseline.

(7) FIG. 7: The amino acid sequence (SEQ ID NO:1) corresponding to a full length human alpha-crystallin B chain protein (CRYAB), natural variants of this protein exist.

(8) FIG. 8: The nucleic acid sequence (SEQ ID NO:2) corresponding to a full length CRYAB cDNA of a human CRYAB.

(9) FIG. 9: Confocal laser microscopy for α-B crystallin in donor and aortic stenosis (AS). Confocal laser microscopic images were obtained from left ventricular (LV) myocardium of donor and AS patients with immunohistochemical visualization of cell membranes (Column A), nuclei (Column B), and α-B crystallin (Column C). In myocardium of AS patients, intensity of α-B crystallin expression (Column C) was higher than in donor with visualization in the merged images (Column D) of subsarcolemmal aggresomes especially in the vicinity of capillaries (white arrows). The latter suggests insufficient covering by α-B crystalline of sarcomeric proteins like titin in AS.

(10) FIG. 10: Diagram showing respective responses of donor and aortic stenosis (AS)/dilated cardiomyopathy (DCM) cardiomyocytes to administration of alkaline phosphatase, prestretch, and α-B crystallin.

DETAILED DESCRIPTION

(11) Human Samples

(12) Aortic stenosis (AS) patients (N=7) had symptomatic, severe AS without concomitant coronary artery disease. Biopsy specimens from this group were procured from endomyocardial tissue resected from the septum (Morrow procedure) during aortic valve replacement. Dilated cardiomyopathy (DCM) specimens were procured from LV biopsies (N=3) or from explanted hearts from end-stage heart failure patients (N=3). DCM patients had no significant coronary stenosis and their biopsies showed no inflammation or infiltration. Control samples were obtained from explanted donor hearts (N=8).

(13) Force Measurements in Small Muscle Strips

(14) Small muscle strips (150-450 μm in diameter and 800-1900 μm in length) were dissected from biopsy specimens (n=24 for AS; n=16 for DCM; n=20 for Donor). After incubation for 1 hour in relaxing solution supplemented with 0.2% TritonX-100 to remove all membrane structures, strips were attached between a force transducer and length motor in a relaxing solution (in mmol/L: free Mg, 1; KCl, 100; EGTA, 2; Mg-ATP, 4; imidazole, 10; pH 7.0). Strips were gently stretched till slack length, i.e., the minimal length at which passive tension (F.sub.passive) is being built up. As a test of cell viability, each muscle strip was transferred from relaxing to maximally activating solution (pCa4.5), at which isometric force developed. After stabilization for 5 minutes in relaxing solution, strips were stretched to 10, 15, 20 and 25% relative to slack length and F.sub.passive was measured at each stage of muscle strain. Subsequently, thick and thin filaments were extracted by immersing the preparation in relaxing solution containing 0.6 mol/L KCl (45 minutes at 20° C.) followed by relaxing solution containing 1.0 mol/L KI (45 minutes at 20° C.)..sup.1-3 After the extraction procedure, the muscle bundles were stretched again and the F.sub.passive remaining after KCl/KI treatment was ascribed to the extracellular matrix (E-matrix). At each stage of muscle strain, F.sub.passive following extraction was subtracted from baseline value to yield F.sub.passive attributable to cardiomyocyte-stiffness, or titin.

(15) Initial stretches with intact muscle strips, in which both cardiomyocytes and E-matrix contribute to F.sub.passive, showed that DCM and AS strips developed higher forces during all stretches compared to donors (FIG. 1A). The extraction protocol decreased F.sub.passive in all groups (FIG. 1B).

(16) At all stages of muscle strain, F.sub.passive caused by the E-matrix was higher in DCM strips than in donors (FIG. 2B). AS strips had significantly higher E-matrix based F.sub.passive than donor in stretches exceeding 15% from slack length (FIG. 2A).

(17) F.sub.passive related to cardiomyocyte stiffness, which is attributable to titin, was higher over the whole range of stretches in DCM and the range of stretches from 10-20% in AS (FIGS. 3A and 3B).

(18) The difference in F.sub.passive related to titin cannot be explained by isoform shifts because both DCM and AS express more of the compliant N2BA titin isoform. To further analyze the difference in F.sub.passive related to titin, isolated cardiomyocytes were subjected to treatment with AP to discern the contribution of phosphorylation and to conditions prevailing in failing myocardium such as high stretch and hypoperfusion-related acidosis.

(19) Force Measurements in Isolated Cardiomyocytes

(20) Force measurements were performed on single demembranated cardiomyocytes (n ranged from 9 to 15 for each group and experimental protocol) as described previously..sup.4, 5 Cardiomyocytes were isolated from donor, AS, and DCM hearts. Briefly, samples were thawed in relaxing solution (in mmol/L: free Mg, 1; KCl, 100; EGTA, 2; Mg-ATP, 4; imidazole, 10; pH 7.0), mechanically disrupted and incubated for 5 minutes in relaxing solution supplemented with 0.5% TRITON® X-100. The cell suspension was washed 5 times in relaxing solution. Single cardiomyocytes were selected under an inverted microscope and attached with silicone adhesive between a force transducer and a piezoelectric motor. Cardiomyocyte F.sub.passive was measured in relaxing buffer at room temperature within a sarcomere length (SL) range between 1.8 and 2.4 μm. Force values were normalized to cardiomyocyte cross-sectional area calculated from the diameter of the cells, assuming a cylindrical shape. As a test of cell viability, each cardiomyocyte was also transferred from relaxing to maximally activating solution (pCa4.5), at which isometric force developed. Once a steady state force was reached, the cell was shortened within 1 ms to 80% of its original length to determine baseline force. Only cells developing active forces >20 kN/m.sup.2 were included in the analysis. Thereafter, cardiomyocytes were incubated in relaxing solution supplemented with alkaline phosphatases (AP) (2000 U/mL; New England Biolabs), 6 mmol/L dithiothreitol (MP Biochemicals) for 40 minutes at 20° C. and F.sub.passive was measured again at SL 1.8-2.4 μm. Subsequently, cardiomyocytes were stretched to ˜2.6 μm SL, incubated in relaxing buffer with pH 6.6, and held in the stretched state for 15 minutes (prestretch). Thereafter, cardiomyocytes were returned to slack length and stabilized for 5 minutes before recording the F.sub.passive with an identical stretch protocol from SL 1.8-2.4 μm in the low pH buffer. Finally, the pH 6.6 buffer was supplemented with recombinant human α-B crystallin at 0.1 mg/ml and F.sub.passive was measured in the presence of α-B crystallin again at SL 1.8-2.4 μm.

(21) In a second set of experiments, single AS cardiomyocytes underwent either incubation at pH 6.6 alone or a prestretch in relaxing buffer with pH 6.6 to ˜2.6 μm SL, followed by the same stretch protocol from SL 1.8-2.4 μm.

(22) Finally, in a third set of experiments, F.sub.passive was measured in single AS cardiomyocytes from SL 1.8-2.4 μm before and after incubation with α-B crystallin, before and after incubation with α-B crystallin and a prestretch to ˜2.6 μm SL, and before and after incubation with α-B crystallin, a prestretch to ˜2.6 μm SL and incubation at pH 6.6.

(23) In donor cardiomyocytes (FIG. 4A), the F.sub.passive-SL curve shifted upward after administration of alkaline phosphatase (AP), which dephosphorylates titin. In contrast, the F.sub.passive-SL curve failed to shift in AS and DCM cardiomyocytes (FIGS. 4B and 4C), consistent with preexisting hypophosphorylation of titin. The F.sub.passive-SL curve of donor cardiomyocytes after AP was still lower than the F.sub.passive-SL curve after AP in AS and DCM cardiomyocytes. This suggests mechanisms other than hypophosphorylation to contribute to the high diastolic stiffness observed in AS and DCM cardiomyocytes. The additional effects of prestretch and acidic pH were, therefore, investigated. The F.sub.passive-SL curve shifted further upward in donor cardiomyocytes after performing a prestretch and imposing an acidic pH. After administration of α-B crystallin, the curve returned to an intermediate position between baseline and AP.

(24) In AS cardiomyocytes (FIG. 4B), no significant change in the F.sub.passive-SL curve was observed after incubation with AP, consistent with a pre-existing hypophosphorylation of titin. After the prestretch and pH 6.6, the F.sub.passive-SL curve shifted upward compared to baseline. After treatment with α-B crystallin, the F.sub.passive-SL curve dropped to a position that was significantly lower than baseline and comparable to the baseline position of donor cardiomyocytes. This finding implies presence at baseline of prestretch- and pH-induced changes in AS myocardium, which could be corrected by administration of α-B crystallin.

(25) The same series of experiments in single DCM cardiomyocytes show similar findings (FIG. 4C) as in AS cardiomyocytes: incubation with AP had no effect on the F.sub.passive-SL curve, but performing a prestretch in an acidic environment shifted the curve significantly upward. After in vitro treatment with α-B crystallin, the F.sub.passive-SL curve fell to a position that was significantly below baseline and comparable to the baseline position of donor cardiomyocytes. This again implies presence at baseline of prestretch- and pH-induced changes in DCM cardiomyocytes, which could be corrected by administration of α-B crystallin.

(26) The relative importance of prestretch and pH 6.6 were analyzed in single AS cardiomyocytes. In the absence of prior administration of AP, lowering the pH to 6.6 had no effect on F.sub.passive (FIG. 5A), but a prestretch to ˜2.6 μm SL significantly shifted the F.sub.passive-SL curve upward (FIG. 5B).

(27) Incubation with α-B crystallin shifted the F.sub.passive-SL curve significantly downward compared to baseline (FIG. 6A). Prestretch or prestretch in combination with pH 6.6 yielded a similar result (FIGS. 6B and 6C). These findings suggest that prestretch-induced changes significantly contributed to the high stiffness of AS cardiomyocytes, both at baseline and after prestretch. The finding that high stiffness in the cardiomyocytes is reversed by addition of α-B crystallin suggests there is an insufficient availability of α-B crystalline in AS cardiomyocytes to neutralize stretch-induced effects on titin distensibility both at baseline and following prestretch. These results show that increasing the availability of alpha-crystallin B chain protein in cardiomyocytes that demonstrate a pre-existing stiffness is a means for lowering the stiffness of the cardiomyocytes, thereby forming a suitable treatment method for diastolic heart failure.

(28) Immunofluorescence Staining and Confocal Scanning Laser Microscopy.

(29) Frozen human heart tissue was sectioned to a thickness of 5 μm using a cryostat (Leica). The sections were fixed with 3% paraformaldehyde, permeabilized with 0.05% TWEEN® 20 and immunostained using goat anti-α-B crystallin (Santa Cruz) diluted 1 in 100 in PBS+1% BSA (immunohistochemical grade; Vector Laboratories). Antigoat conjugated to Alexa 555 (Thermofisher) was used to visualize α-B crystallin. Membranes were stained using WGA conjugated with Alexa 647 (Thermofisher) diluted 1 in 100 in PBS. Nuclei were visualized using Picogreen reagent (Thermofisher) diluted 1 in 10 000 in PBS. Confocal scanning laser microscopy was performed on a Leica TCS SP8 STED 3X (Leica Microsystems). Picogreen, Alexa 555, and Alexa 647 were irradiated with a pulsed white light laser at 502, 553, and 631 nm, respectively. A 63× oil objective with NA 1.4 Numerical Aperture was used to image the sample. Detection of the fluorescent signal was performed with gated Hybrid Detectors. Finally, the images were deconvolved using Huygens Professional (Scientific Volume Imaging).

(30) Scanning Laser Microscopy

(31) Confocal laser microscopical images were obtained from LV myocardium of donor and AS patients with immunohistochemical visualization of cell membranes, nuclei, and α-B crystallin (FIG. 9). In myocardium of AS patients, intensity of α-B crystallin expression was higher than in donor with appearance of α-B crystallin-containing aggresomes18 (FIG. 9), which were especially prominent in the subsarcolemma close to the capillaries (white arrows in FIG. 9). The latter suggests signals originating from the microvascular endothelium to be involved in the subsarcolemmal mobilization of α-B crystallin in failing cardiomyocytes.

(32) Statistical Analysis

(33) Differences between groups were analyzed with an unpaired, two-tailed Student t test. Differences within groups were measured by repeated measures analysis of variance. All analyses were performed using Prism software (GraphPad Software Inc., version 6.0).

(34) High diastolic stiffness of failing human myocardial strips and cardiomyocytes were investigated and the following was observed: (1) high diastolic stiffness of cardiomyocytes significantly contributes to the overall stiffness of LV myocardial strips of AS and DCM patients; (2) dephosphorylation with AP shifts the diastolic Fpassive-SL relation upward in donor but not in AS or DCM cardiomyocytes; (3) after dephosphorylation, exposure to prestretch causes an upward shift of the diastolic Fpassive-SL relations in AS and DCM cardiomyocytes and a further upward shift of the diastolic Fpassive-SL relation in donor cardiomyocytes; (4) subsequent administration of α-B crystallin shifts the diastolic Fpassive-SL relations downward in donor, AS, and DCM cardiomyocytes to a position that coincides with the baseline diastolic Fpassive-SL relation of donor cardiomyocytes and falls below the baseline diastolic Fpassive-SL relation in AS and DCM cardiomyocytes. This finding is consistent with α-B crystallin, providing protection against stretch-induced damage to titin in failing AS and DCM cardiomyocytes.

(35) Cardiomyocyte Versus Extracellular Matrix Stiffness

(36) High diastolic stiffness of failing human cardiomyocytes was a significant contributor to overall stiffness of LV myocardial strips of AS and DCM patients ≤20% stretch in AS and ≤25% stretch in DCM. Use of dissected myocardial strips precluded visualization of sarcomeres, and strip lengthening was, therefore, expressed as percentage of stretch with respect to slack length, that is, the minimal length at which Fpassive started to develop (FIGS. 1A and 1B). At 25% stretch, the contribution of cardiomyocyte stiffness to overall stiffness no longer differed between donor and AS cardiomyocytes but continued to differ between donor and DCM cardiomyocytes (FIGS. 3A and 3B). This could relate to less constraint by the extracellular matrix in DCM despite raised collagen volume fraction in both AS and DCM. The latter could be consistent with different distribution and homeostasis of myocardial fibrosis in AS and DCM: in DCM, there is focal replacement fibrosis, whereas in AS, there is diffuse reactive fibrosis and, as reflected by plasma biomarker elevations, fibrinolytic mechanisms are present in DCM in contrast to mainly profibrotic mechanisms in AS. These findings illustrate the importance of concentric versus eccentric remodeling for the constraint imposed by the extracellular matrix on the cardiomyocytes.

(37) Cardiomyocyte Stiffness and Titin Dephosphorylation.

(38) Altered cardiomyocyte stiffness can result from isoform shifts of the giant cytoskeletal protein titin, from post-translational modifications of titin such as phosphorylation, formation of disulfide bonds, carbonylation, and s-glutathionylation, or from stretch-induced titin modification. Because of higher expression of the compliant N2BA isoform in AS and DCM, titin isoform shifts do not contribute to the observed rise of cardiomyocyte stiffness observed in AS and DCM cardiomyocytes in the present disclosure. Because of altered activity in failing myocardium of different kinases such as protein kinase A, protein kinase C, protein kinase G, calcium/calmodulin-dependent kinase II, and extracellular signal regulated kinase, altered phosphorylation of titin by these kinases was suspected to be involved in the raised stiffness of failing human cardiomyocytes. The present disclosure observed treatment with AP to raise diastolic stiffness in donor cardiomyocytes but not in AS and DCM cardiomyocytes (FIG. 10). This implies a pre-existing imbalance of titin phosphorylation in AS and DCM cardiomyocytes with either reduced phosphorylation of sites that increase titin elasticity or increased phosphorylation of sites that decrease titin elasticity.

(39) Post-translational modifications of titin other than phosphorylation have recently been implicated in altered cardiomyocyte stiffness. These mechanisms include, among others, modification of the titin molecule induced by excessive physical stretch. A previous study indeed showed stretch induced mechanical unfolding of immunoglobulin domains of titin to expose cryptic cysteines to S-glutathionylation, which interfered with the ability of titin to refold and left titin in a more extensible state. In acidic pH, the reverse was observed, namely, a prestretch-induced reduction of titin extensibility..sup.7 This is especially relevant to failing myocardium, which is exposed to both high filling pressures and jeopardized coronary perfusion. The present study, therefore, imposed prestretch and acidic pH on failing human cardiomyocytes.

(40) Cardiomyocyte Stiffness and Prestretch

(41) In the present study, cardiomyocytes were subjected to a prestretch protocol, which consisted of a 15-minute stretch period at 2.6 μm followed by a 5-minute stabilization period at slack length. This prestretch protocol was executed in Ph 6.6 in donor, AS, and DCM cardiomyocytes. After dephosphorylation with AP, exposure to prestretch caused an upward shift of the diastolic Fpassive-SL relations in AS and DCM cardiomyocytes and a further upward shift of the diastolic Fpassive-SL relation in donor cardiomyocytes (FIGS. 4A-4C). The identical position of all diastolic Fpassive-SL relations after prestretch argues in favor of previous stretch-induced damage being involved in the baseline elevation of diastolic stiffness of AS and DCM cardiomyocytes. The prestretch-induced upward shift of the diastolic Fpassive-SL relations in AS and DCM cardiomyocytes was indeed smaller than the prestretch induced upward shift of the diastolic Fpassive-SL relation in donor cardiomyocytes (FIG. 10). This yielded an identical position of all diastolic Fpassive-SL relations because in AS and DCM cardiomyocytes, a smaller shift was superimposed on baseline stretch-induced damage, whereas in donor cardiomyocytes, prestretch elicited a larger shift because of absent baseline stretch-induced damage.

(42) Without being bound by theory, it is believed that the origin of the baseline stretch-induced damage in AS and DCM cardiomyocytes relates to external stretch on cardiomyocytes or to internal stretch within cardiomyocytes. The former relates to elevated LV filling pressures at rest or during exercise. The latter is consistent with either a modified Z-disc structure or with the previously observed widening of the Z-disc. Z-disc widening results from reduced elasticity of cytoskeletal proteins, which from both sides pull at and open up adjacent Z lines. In AS and DCM cardiomyocytes, internal stretch and stretch-induced damage could have resulted from the aforementioned imbalance of titin phosphorylation.

(43) In contrast to a previous study, separate imposition of pH 6.6 failed to induce an upward shift of the diastolic Fpassive-SL relation (FIG. 5A). The upward shift of the diastolic Fpassive-SL relation after combined administration of prestretch and acidic pH, therefore, seemed to be solely related to preceding sarcomere stretch. Furthermore, omission of previous treatment with AP also did not influence the combined effect of prestretch and acidic pH (FIG. 5B).

(44) Cardiomyocyte Stiffness and α-B Crystallin

(45) α-B crystallin protects cardiomyocytes against stretch-induced damage in acidic pH (FIGS. 4A-4C). The present disclosure also administered α-B crystallin to AS and DCM cardiomyocytes. In contrast to donor cardiomyocytes, α-B crystallin not only corrected the combined effects of prestretch and acidic pH but also reversed the baseline upward displacement of the diastolic Fpassive-SL relation (FIG. 10). This finding was consistent with baseline involvement of previous stretch-induced titin damage in AS and DCM cardiomyocytes and was confirmed in a separate series of experiments, in which α-B crystallin shifted the diastolic Fpassive-SL relation downward without any previous or concomitant intervention (FIGS. 6A-6C). In these experiments, the magnitude of the downward displacement of the diastolic Fpassive-SL relation was similar in the absence (FIG. 6A) or presence of foregoing interventions (FIGS. 6B and 6C).

(46) In AS and DCM cardiomyocytes, α-B crystallin lowered diastolic stiffness well below baseline values as previously reported after administration of protein kinase A or protein kinase G PKA or PKG. This supports overlapping effects of titin phosphorylation and stretch-induced titin aggregation possibly because of pre-existing stretch-induced titin aggregation obstructing phosphorylation at sites that specifically increase titin elasticity. This finding has important therapeutic implications as it implies limited efficacy of drugs that increase PKA or PKG activity for treatment of diastolic LV dysfunction related to high cardiomyocyte stiffness and could relate to the failure of dobutamine to improve diastolic LV dysfunction and of phosphodiesterase 5 inhibitors to improve exercise tolerance or hemodynamics in HFPEF.

(47) The present disclosure observed upregulation and subsarcolemmal localization of α-B crystallin in AS and DCM cardiomyocytes. Because of the close vicinity of capillaries (white arrows in FIG. 7), the localization of α-B crystallin in subsarcolemmal aggresomes was consistent with signals from the microvascular endothelium being involved in their formation. The subsarcolemmal localization also suggested that endogenous α-B crystallin was diverted from the sarcomeres and, therefore, failed to exert its protective action on titin distensibility, which was, however, restored after administration of exogenous α-B crystallin. The latter finding supports future therapeutic efforts to raise concentration of α-B crystallin in failing myocardium through direct administration of α-B crystallin, through administration of α-B crystallin analogues or through administration of heat shock protein-inducing drugs such as geranylgeranylacetone or NYK9354.

(48) High cardiomyocyte stiffness significantly contributed to overall myocardial stiffness in AS and DCM. High cardiomyocyte stiffness resulted from titin phosphorylation failing to improve cardiomyocyte stiffness and from previous stretch-induced aggregation of titin, both of which were corrected by administration of α-B crystallin. Diastolic LV dysfunction in heart failure could, therefore, benefit from treatment with α-B crystallin.

(49) For the purpose of clarity and a concise description, features are described herein as part of the same or separate aspects and preferred embodiments thereof, however, it will be appreciated that the scope of the disclosure may include embodiments having combinations of all or some of the features described.

(50) The following aspects are aspects of the disclosure:

(51) Aspect 1. A composition for use in treatment of diastolic dysfunction in a subject, wherein the composition comprises a therapeutically effective amount of a substance that increases the level and/or activity of a crystalline protein in cardiomyocytes of the subject.

(52) Aspect 2. Composition for use according to aspect 1, wherein the substance increases the level and/or activity of a crystalline protein that reduces a diastolic stiffness of the cardiomyocytes of the subject.

(53) Aspect 3. Composition for use according to aspect 1 or aspect 2, wherein the crystalline protein that is increased in level and/or activity by the substance is a protein encoded by the crystalline alpha B (CRYAB) gene.

(54) Aspect 4. Composition for use according to any one of the foregoing aspects, wherein the substance increases the level of alpha B chain crystalline protein in cardiomyocytes of the subject.

(55) Aspect 5. Composition for use according to aspect 4, wherein the substance is alpha B chain crystalline protein of SEQ ID NO:1, or a protein that has an amino acid sequence identity of at least 80%, preferably at least 90%, more preferably at least 95%, most preferably 98% or 99% with the alpha B chain crystalline protein, which protein is capable of reducing a diastolic stiffness of cardiomyocytes of the subject.

(56) Aspect 6. Composition for use according to aspect 4, wherein the substance induces expression of the CRYAB gene in the cardiomyocytes of the subject.

(57) Aspect 7. Composition for use according to aspect 6, wherein the substance is geranylgeranylacetone (GGA) or NYK9354.

(58) Aspect 8. Composition for use according to any one of aspect 1 to 3, wherein the substance mediates a post-translational modification of alpha B chain crystalline, preferably by phosphorylation of the alpha B chain crystalline protein.

(59) Aspect 9. Composition for use according to any one of the foregoing aspects, wherein the subject to be treated suffers from diastolic heart failure or heart failure with preserved ejection fraction.

(60) Aspect 10. Composition for use according to any one of the foregoing aspects, wherein the subject to be treated suffers from aortic stenosis and/or dilated cardiomyopathy.

(61) Aspect 11. Method of treating diastolic dysfunction, diastolic heart failure or heart failure with preserved ejection fraction in a subject, comprising administering to the subject a therapeutically effective amount of a substance that increases the level and/or activity of a crystalline protein in cardiomyocytes of the subject.

REFERENCES

(62) 1. Granzier H. L., and T. C. Irving. Passive tension in cardiac muscle: contribution of collagen, titin, microtubules, and intermediate filaments. Biophys. J. 1995; 68(March): 1027-1044. doi: 10.1016/S0006-3495(95)80278-X. 2. Hamdani N., C. Franssen, A. Lourenco, et al. Myocardial titin hypophosphorylation importantly contributes to heart failure with preserved ejection fraction in a rat metabolic risk model. Circ. Heart Fail. 2013; 6(6):1239-49. doi:10.1161/CIRCHEARTFAILURE.113.000539. 3. Zile M. R., C. F. Baicu, J. Ikonomidis, et al. Myocardial Stiffness in Patients with Heart Failure and a Preserved Ejection Fraction: Contributions of Collagen and Titin.; 2015. doi:10.1161/CIRCULATIONAHA.114.013215. 4. Borbély A., J. van der Velden, Z. Papp, et al. Cardiomyocyte stiffness in diastolic heart failure. Circulation. 2005; 111(6):774-81. doi:10.1161/01.CIR.0000155257.33485.6D. 5. Van Heerebeek L., A. Borbély, H. W. M. Niessen, et al. Myocardial structure and function differ in systolic and diastolic heart failure. Circulation. 2006; 113(16):1966-73. doi:10.1161/CIRCULATIONAHA.105.587519. 6. Borbély A., I. Falcao-Pires, L. van Heerebeek, et al. Hypophosphorylation of the Stiff N2B titin isoform raises cardiomyocyte resting tension in failing human myocardium. Circ. Res. 2009; 104(6):780-6. doi:10.1161/CIRCRESAHA.108.193326. 7. Kötter S., A. Unger, N. Hamdani, et al. Human myocytes are protected from titin aggregation-induced stiffening by small heat shock proteins. J. Cell. Biol. 2014; 204(2):187-202. doi:10.1083/jcb.201306077.