ABHD5 AND PARTIAL HDAC4 FRAGMENTS AND VARIANTS AS A THERAPEUTIC APPROACH FOR THE TREATMENT OF CARDIOVASCULAR DISEASES

Abstract

The present invention relates to Abhydrolase containing domain 5 (ABHD5) and N-terminal fragments of HDAC4 (HDAC4-NT) and variants of the aforementioned peptides for the treatment and prevention of heart failure. The present invention further provides vectors for the cardiomyocyte-specific expression of said peptides and a test system comprising ABHD5 for the identification of novel compounds which are useful for the treatment of heart failure.

Claims

1.-17. (canceled)

18. An elongated and/or multimerized variant of the N-terminal fragment of histone deacetylase 4 (HDAC4-NT).

19. The variant of HDAC4-NT of claim 18, which has at least 80% sequence identity over the entire length of the protein according to SEQ ID NO:13 or 14.

20. The variant of HDAC4-NT of claim 18, which is an elongation variant comprising the amino acid positions 2 to 202, 1 to 220, 1 to 216, 1 to 212 or 1 to 208 of human HDAC4 as defined by SEQ NO: 11.

21. The variant of HDAC4-NT of claim 18, wherein the variant is a multimerization variant comprising at least 2, at least 3, at least 4, or at least 5 repeats of HDAC4-NT.

22. A vector comprising a nucleic acid encoding HDAC4-NT or a variant thereof.

23. The vector of claim 22, selected from the group consisting of plasmids, cosmids, phages, viruses and artificial chromosomes.

24. The vector of claim 22, selected from the group consisting of adenovirus vectors, adeno-associated virus (AAV) vectors, alphavirus vectors, herpes virus vectors, arena virus vectors, measles virus vectors, pox virus vectors, NYVAC, avipox vectors, vesicular stomatitis virus vectors, retrovirus vectors, lentivirus vectors, viral like particles, and bacterial spores.

25. The vector of claim 24, which is an AAV vector.

26. The vector of claim 25, selected from the group consisting of AAV type 6, type 1, type 5, type 9 and type 2.

27. The vector of claim 22 wherein HDAC4-NT or the variant thereof is expressed under control of the human troponin promoter and a recognition site for micro-RNA 122 between the promoter and the terminator sequence so that said recognition site becomes part of the transcript produced from this vector.

28. A method of treating of preventing heart failure, the method comprising administering an effective amount of a nucleic acid encoding HDAC4-NT or a variant thereof or a vector comprising a nucleic acid encoding HDAC4-NT or a variant thereof, to a patient.

29. A method of treating myocardial remodeling during heart failure, the method comprising administering an effective amount of a nucleic acid encoding HDAC4-NT or a variant thereof or a vector comprising a nucleic acid encoding HDAC4-NT or a variant thereof, to a patient.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0172] FIG. 1: Principle of invention and improvement of current technologies. AAV-9 is an example as an use in vivo. (A) gain of function of the protease abhd5 to increase the quantity of cardioprotective HDAC4-NT. (B) Gene transfer of HDAC4-NT, optimized HDAC4-NT or HDAC4-NT related constructs via adeno associated virus (AAV). NT: N-terminus; abhd5: 1-acylglycerol-3-phosphate O-acyltransferase;

[0173] FIG. 2: HDAC4 aa 2-201 were cloned into an AAV-9 virus (NT). NT was applied to animals 6 weeks before TAC surgery (indicated as a lines in the timecourse of echos, A). Controls received Luciferase cloned into AAV-9 (LUC). In consecutive echos, fs decreased after TAC surgery whereas NT treated animals were protected from reduced function, n=5, **p<0.01. Arrow indicates timepoint of echo in figure (B). Heartweight/Bodyweight ratio was normalized (n=5; **p<0.01) (C)

[0174] FIG. 3: In part, dyregulated genes were normalized, such as myh6 (A) and nppb (D). Myh7 (B) and nppa (C) were not normalized by NT-treatment (n=5). Values indicate relative expression level normalized to WT sham group; ±SEM *p<0.05.

[0175] FIG. 4: Validation experiment of the siRNA-screen confirmed results from the siRNA screen. Co-expression of flag tagged HDAC4 with myc tagged PKA leads to a cleavage of HDAC4 (first three lanes). The cleavage is not longer present when abhd5 is knocked down by using different siRNA in the last three lanes (siRNAs labeled with 1-3). Quantification of the western blots confirmed the results (p<0.05).

[0176] FIG. 5: To proof the results from our loss of function approach, we overexpressed abhd5 in cardiomyocytes (last three lanes). Overexpressed flag tagged HDAC4 was cleaved whenever gfp tagged abhd5 was coexpressed. IB: immunoblot; Ad: adenovirus construct.

[0177] FIG. 6: Overexpression of abhd5 rescues FCS induced MEF2 luciferase activity. Neonatal rat ventricular cardiomyocytes were treated with 10% FCS for 24 h. HDAC4 alone was not able to inhibit this activation, whereas coexpression of abhd5 was able to blunt the MEF2 activation back on a basic value. This effect was not seen when cells were treated with a EGFP tagged control virus. MEF2-Luc: myosin enhancer factor 2 luciferase construct; FCS: fetal bovine serum; Ad: Adenovirus; #p<0.05 treatment vs. no treatment; $p<0.05 Ad-abhd5 treated vs. cells treated with FCS 10%.

[0178] FIG. 7: Cardiomyocyte hypertrophy was significant reduced when abhd5 was expressed. A FCS induced hypertrophy was completely blunted whereas EGFP control virus did not show any beneficial effects on cardiomyocytes. NRVM: neonatal rat ventricular myocytes; FCS: felta bovine serum; #p<0.05 cell size of groups as indicated compared to control group without FCS; $p<0.05 cell size of abhd5 treated group compared to FCS treated group.

[0179] FIG. 8: (A) Overexpression of ABHD5 leads to proteolysis of HDAC4. Flag tagged HDAC4 was overexpressed in neonatal rat cardiomycytes with ABHD5, Western blot was performed with an antibody, recognizing Flag. Successful overexpression of ABHD5-GFP was confirmed with a western blot, detecting GFP. (B) HDAC4 and ABHD5 were cloned into a plasmid bearing constitutively active promoter (pNOP) and Galactose inducible promoter (pGAL) respectively. By overexpression either together or with backbone vector in Saccharomyces cerevisiae we found cleavage of HDAC4, confirmed by westernblot analysis. This experiment confirms that ABHD5 cleaves HDAC4 directly without the recruitment of factors only present in mammalian cells. (C) By co-treatment of mice with insulin and a beta receptor agonist (ISO) we found an increase of ABHD5 expression and consequently an increase of endogenous HDAC4 proteolysis. We show here exemplary western blots, performed with antibodies recognizing endogenous N-terminal HDAC4 and endogenous ABHD5. Westernblot against GAPDH confirmed equally protein loading.

[0180] FIG. 9: (A) The transcription factor myocyte enhancer factor 2 (MEF2) is sufficiently repressed by overexpression of ABHD5. MEF2-luciferase reporter was overexpressed in neonatal rat cardiomyocytes and stimulated with either endothelin-1 (ET-1) or (B) fetal calf serum (FCS) for 24 h. Co-expression of HDAC4 and ABHD5 leads to a cleavage of HDAC4 and consequently to a repression of MEF2 luciferase activity. Equally expression of HDAC4, ABHD5 and equally protein loading was confirmed by westernblot analysis as indicated. Values are shown as mean±SEM; n>3/group; **p<0.05.

[0181] FIG. 10: ABHD5 counteracts cardiomyocyte hypertrophy in vitro. We stimulated neonatal rat cardiomyocyte with the prohypertorphic agents endothelin-1 (ET-1) and fetal calf serum (FCS) for 24 h as indicated. By adenoviral overexpression of ABHD5, hypertrophic response was blunted. (B) Cardiomyocyte size was quantified by counting 3 fields of view (>100 cardiomyocytes per field). Values are sown as mean±SEM; **p<0.05.

[0182] FIG. 11: Transgenic overexpression of ABHD5 in vivo protects from cardiac hypertrophy and heart failure. We generated transgenic animals, overexpressing ABHD5 under the control of the αMHC promoter. 3 weeks after induction of pathological cardiac remodeling by transaortic constriction (TAC), we found less hypertrophy indicated by a reduction of heartweight/bodyweight ratio (HW/BW) and heartweight/tibia length ratio (HW/TL). Left ventricular function was improved in transgenic animals, shown as ejection fraction and fractional shortening. Classical pathological genes, such as atrial natriuretic peptide (ANP) and brain natriuretic peptide (BNP) were less induced in ABHD5 transgenic animals. Values are shown as mean±SEM; *p<0.05; **p<0.01; ***p<0.001.

[0183] FIG. 12: HDAC4-NT reverses pathological induction of metabolic genes. We measured 4 weeks after induction of pathological cardiac remodeling by transaortic constriction (TAC) the enzymatic activity of key enzymes of the glycolysis. Pathological induction of glycolytic enzymes was reversed when HDAC4 was overexpressed by using the previous described novel AAV-construct. We show here a hierarchical clustering of enzymatic activity of key enzymes as indicated. Color key shows difference from the mean.

[0184] FIG. 13: HDAC4 2-202 inhibits MEF2 luciferase activity as well. To test whether different mutations of HDAC4 lead to inhibition of MEF2, we overexpressed HDAC4 2-202 by using an adenovirus system in neonatal rat cardiomyocytes and treated them with the prohypertrophic agent fetal calf serum (FCS). By overexpression of either HDAC4 2-201 or 2-202 endogenous MEF2 activity was sufficient inhibited. We found HDAC4 2-202 was less efficient compared to HDAC4 2-201 in inhibiting MEF2 luciferase activity. In case of pharmacological interventions, this could be an eligible aim since transient or moderate inhibition of MEF2 might be more useful in the clinical setting. Therefore, we claim HDAC4 and HDAC4 mutants as potential therapeutic tools for inhibition of MEF2 and consecutive pathological hypertrophy or pathological cardiac remodeling.

[0185] FIG. 14: HDAC4 aa 2-201 were cloned into an AAV-9 virus (NT). NT was applied to animals 6 weeks before TAC surgery. Controls received Luciferase cloned into AAV-9 (LUC). Expression analysis of hearts revealed a specific pattern of genes that was completely normalized in animals that received AAV-NT. These genes are listed beside the heatmap. All of these genes were upregulated in TAC (p<0.0001) and normalized as compared to the AAV-LUC treated sham group (+20%).

[0186] FIG. 15: Forced treadmill running was performed twice a day during a 4 week running training program. Shown is the total distance that WT and HDAC4-KO mice run during the entire training program within 4 weeks (A) and the daily distance until exhaustion (B). AAV-NT expression in HDAC4 KO animals was able to rescue the phenotype of reduced exercise performance. Values are shown as mean+SEM (HDAC4-KO and WT n=8; AAV-NT n=3). Echocardiography was performed directly after exercise and revealed a reduced fractional shortening in HDAC4-KO animals, values are shown as mean+SEM (n=8/group; *p<0.05). Westernblot against N-terminal HDAC4 shows and increased HDAC4-NT production after 2 weeks of exercise (D). GAPDH is shown as a loading control.

[0187] The following examples are merely intended to illustrate the invention. They shall not limit the scope of the claims in any way.

EXAMPLES

[0188] Materials and Methods

[0189] Transthoracic echocardiography. Echocardiography was performed using a Sonos 5500 with a S12 transducer (12 MHz). The echocardiographer was blinded with respect to the treatment group. Mice were shaved and left ventricular parasternal short-axis views were obtained in M-mode imaging at the papillary muscle level. Three consecutive beats were used for measurements of left ventricular end-diastolic internal diameter (LVEDD) and left ventricular end-systolic internal diameter (LVESD). Fractional shortening (FS) was calculated as FS %=[(LVEDD−LVESD)/LVEDD]×100%.

[0190] Generation of an adeno associated virus (HDAC4-AAV). HDAC4 aa2-201 was cloned into a double-stranded AAV-vector downstream of a CMV-enhanced short (260 bp) myosin light chain promoter (CMVenh/MLC260). AAV9 vectors were produced with the three plasmid transfection method.

[0191] RNA analysis. Total RNA was isolated from ventricular tissue using TRIzol (Invitrogen, Germany). Total RNA was digested with DNase, and cDNA synthesis from 500 ng of RNA was carried out using a SuperScript first-strand synthesis system for RT-PCR (Invitrogen). Quantitative real-time PCR (qPCR) was performed with Universal ProbeLibrary (Roche) by using TaqMan Universal PCR Mastermix (Applied Biosystems) and detection on a 7500 Fast Cycler (Applied Biosystems) as described previously [1].

[0192] Transverse aortic constriction. TAC to a 27 gauge stenosis was performed in 9-10 week-old male black six mice (charles river), mice as described previously [2]. AAV vectors were intravenously injected into the tail vein of male adult mice as 150-200 sL bolus using a sterile syringe and 29-gauge needle. Animals were euthanized by cervical dislocation. Organs were dissected and rapidly frozen in liquid nitrogen.

[0193] Western blotting. Proteins from heart tissue and cultured cardiomyocytes were isolated, and Western blot analysis was performed according to protocols described previously [1]. Primary antibodies used were anti-flag (santa cruz), anit-myc (santa cruz), anti-gfp (abcam). Primary antibody incubation was followed by corresponding horseradish peroxidase (HRP)-conjugated secondary anti-mouse and anti-rabbit antibodies and ECL detection. Relative protein levels were detected by densitometry using the Image J program.

[0194] Histology. Hematoxylin and cosin (H&E) and Masson's trichrome stainings were performed as previously described [3]. Cardiomyocyte size was assessed on H&E-stained sections by using Image J software (http://rsb.info.nih.gov/ij/). More than 200 randomly chosen cardiomyocytes from each group were analyzed to measure cross-sectional cardiomyocyte area. To quantify cardiac fibrosis, 20 trichrome-stained sections (magnification 20×) from the left ventricle were randomly selected, and morphometric analysis by using Image J was performed. Photographs were acquired with an Olympus SZH zoom stereo dissection scope with an Optronics DEI-750 CCD digital camera. All data were analyzed by a single observer blinded to the mouse genotypes.

[0195] Results:

[0196] Overexpression of N-terminal HDAC4 via an adeno associated virus (AAV) In mice is cardioprotective. HDAC4-NT is able to inhibit the transcription factor myocyte enhancer factor 2 (MEF2). [4] MEF2 is thought to be involved in pathological cardiac remodeling. [5] We therefore hypothesized that HDAC4-NT could have beneficial and cardioprotective effects in vivo. By cloning HDAC4-NT into a cardiotrophic AAV substrain (AAV9) under the control of a cardiomyocyte specific promoter (CMVenh-MLC260), we were able to transduce cardiomyocytes in vivo via a single tail vein injection. Mice were injected 4 weeks before they were exposed to transthoracic aortic constriction (TAC)-surgery as a model for cardiac stress and the development of heart failure. By doing so, animals that were treated with AAV-HDAC4-NT showed reduced cardiac hypertrophy, improved cardiac function and normalized gene regulation from genes that are known to play an important role in pathological cardiac remodeling (FIG. 2). Cardiac fibrosis that was developed by control-mice was diminished when animals were treated with HDAC4-NT.

[0197] We further aimed to investigate the endogenous role of HDAC4-NT. In wildtype animals, HDAC4-NT production is increased after physiological exercise. We generated conditional HDAC4-knockout animals, lacking HDAC4 in cardiomyocytes only and exposed these animals to running exercise. HDAC4-KO animals showed a reduced exercise tolerance with a reduced left ventricular function after running exercise. Reduced exercise performance was rescued when HDAC4-KO were treated with AAV-HDAC4-NT, indicating that HDAC4-NT is crucial for sustained cardiac function after cardiac stress (FIG. 15). By further carefully characterization of the animals we did not found any harmful effects or any side effects that were linked to AAV-HDAC4-NT treatment.

[0198] Abhd5 is a critical HDAC4 protease. To get insides about the upstream regulation of HDAC4 cleavage, we performed a siRNA-screen with a set of potential serine-proteases. By doing so, we identified a protein called 1-acylglycerol-3-phosphate O-acyltransferase (abhd5) playing a crucial role in PKA induced HDAC4 cleavage. HDAC4 cleavage was not longer present when abhd5 was knocked down by different siRNAs even when PKA was co-expressed with HDAC4 (FIG. 4). Abhd5 is characterized by typical structural features that can be found predominantly in serine proteases. However, it was not shown before, that abhd5 can indeed act as a protease. By adenoviral overexpression in neonatal rat ventricular myocytes (NRVMs) of abhd5 we were able to achieve HDAC4 cleavage without additional PKA activation or expression (FIG. 5).

[0199] Abhd5 overexpression leads to MEF2 inhibition and is cleavage dependent. By using a MEF2-luciferase assay, we found that abhd5 is able to completely normalize MEF2 activity induced by fetal calf serum (FCS), which is commonly used to induce cardiomyocyte hypertrophy. EGFP control virus did not show any beneficial effects in this system (FIG. 6)

[0200] Abhd5 Inhibits cardiomyocyte hypertrophy in vitro. We next hypothesized, that abhd5 induced HDAC4-NT is able to inhibit cardiomyocyte hypertrophy in vitro. Indeed, abhd5 was able to inhibit FCS induced cardiomyocyte hypertrophy. This effect was independent from a control virus with EGFP. (FIG. 7)

DISCUSSION

[0201] We found that HDAC4-NT is cardioprotective in mice and thus could serve as a potential pharmacological tool for the treatment of heart failure in human. This was surprising because earlier approaches to protect the heart by overexpression of a class II HDAC failed [6]. These earlier approaches used transgenic expression of full length HDACs and specific mutations of phosphosites to render HDACs signal-resistant and to force them to localize to the nucleus. However, class II HDAC mutant overexpression showed dramatic harmful effects by inducing mitochondrial dysfunction and apoptosis. In this work we provide two novel ways to circumvent harmful effects. First, we use the AAV system which allowed us to closely titrate the dosage of HDAC expression. The AAV system was already used in clinical trials (SERCA gene therapy). Second, by using an partial HDAC4 peptide we specifically seem to prevent the disease process of the heart and not essential functions such as mitochondrial function or cell survival. As a proof of concept, by expressing HDAC4 1-201 via AAVs, we show here for the first time, that this concept is a new promising therapeutic approach. Given the previous experience with the overexpression of putatively protective proteins in the same animal model, this was not expected.

[0202] Upstream Regulation could Serve as an Additional Tool:

[0203] With the identification of a critical PKA dependent protease (abhd5) we have identified the upstream signaling molecule that induced cardioprotective HDAC4 proteolysis. Abhd5 was not described as a protease and the involvement in potential cardioprotective pathways is new. Under normal conditions, abhd5 is bound to lipid droplets (LDs) and is involved in the regulation of lipolysis. [0204] 1. Backs, J., et al., CaM kinase U selectively signals to histone deacetylase 4 during cardiomyocyte hypertrophy. J Clin Invest, 2006. 116(7): p. 1853-64. [0205] 2. Backs, J., et al., The delta isoform of CaM kinase H is required for pathological cardiac hypertrophy and remodeling after pressure overload. Proc Natl Acad Sci USA, 2009. 106(7): p. 2342-7. [0206] 3. Song, K., et al., The transcriptional coactivator CAMTA2 stimulates cardiac growth by opposing class II histone deacetylases. Cell, 2006. 125(3): p. 453-66. [0207] 4. Sun, Q., et al., Role of myocyte enhancing factor 2B in epithelial myofibroblast transition of human gingival keratinocytes. Exp Biol Mod (Maywood), 2012. 237(2): p. 178-85. [0208] 5. Kim, Y., et al., The MEF2D transcription factor mediates stress-dependent cardiac remodeling in mice. J Clin Invest, 2008. 118(1): p. 124-32. [0209] 6. Czubryt, M. P., et al., Regulation of peroxisome proliferator-activated receptor gamma coactivator 1 alpha (PGC-1 alpha) and mitochondrial function by MEF2 and HDAC5. Proc Natl Acad Sci USA, 2003. 100(4): p. 1711-6.