The present invention relates to the field of therapeutics. Specifically, the present invention relates to homoarginine or to a metabolic precursor thereof for use in a method for the prevention or treatment of myocardial failure. An aim of modern medicine is to provide personalized or individualized treatment regimens. Those are treatment regimens which take into account a patient's individual needs or risks. Individualized treatment regimens offer benefits for the individual patient as well as for society as a whole. For the individual patient, personalized treatment avoids excessive therapy while ensuring that necessary measures are taken. As every therapy may cause undesired harmful side effects, the avoidance of unnecessary therapies saves the patient from potentially harmful side effects. On the other hand, the identification of patients with special needs ensures that these individuals receive the appropriate treatment. For the health system as a whole, the avoidance of unnecessary therapies allows for a more economic use of resources. Previous studies have shown that low serum concentrations of the cationic amino acid homoarginine are associated with endothelial dysfunction due to decreased nitric oxide availability and increased risk of all-cause and overall cardiovascular mortality (WO 2011/098519 Al). In addition, homoarginine seems to be a risk factor for sudden cardiac death (SCD) in dialysis patients and for fatal stroke in patients undergoing coronary angiography.

Homoarginine has previously been suggested to exert protective cardiovascular effects, putatively mediated by regulating the synthesis of the endothelial derived relaxing factor (EDRF), nitric oxide (NO). Data from the LUdwigshafen Risk and Cardiovascular Health (LURIC) study have shown that low homoarginine is a significant and independent risk factor for all-cause and cardiovascular mortality (Maerz et al, 2010, Circulation 122: 967-975). These findings among patients referred for coronary angiography were replicated in a cohort of dialysis patients with type 2 diabetes participating in the 4D study ("Die Deutsche Diabetes Dialyse Studie"). More detailed analyses on specific causes of death revealed an association of low homoarginine levels with fatal stroke (Pilz et al, 2011, Stroke 42: 1132-1134) and sudden cardiac death (Drechsler et al, 2011, Eur J Heart Fail. 13: 852-859) as well as a positive correlation of homoarginine with angiographic left ventricular ejection fraction and an inverse correlation with N-terminal pro-B-type natriuretic peptide (Pilz et al, 2011, Heart. 97: 1222-1227). Preliminary data from other study cohorts confirm the associations of homoarginine deficiency with adverse outcomes which might point to previously unknown pathophysiologic processes that could be relevant for the diagnosis and treatment of life- threatening heart and kidney diseases. There is always a need for improved therapeutics in the field of cardiovascular diseases.

In the context of the present invention, it has surprisingly been found that homoarginine improves myocardial function and affects cardiac remodeling. Accordingly, in a first aspect, the present invention relates to homoarginine or to a metabolic precursor thereof for use in a method for the prevention or treatment of myocardial failure.

Preferably, the present invention relates to homoarginine for use in a method for the prevention or treatment of myocardial failure.

Homoarginine is a cationic amino acid, which is derived from lysine and mainly synthesized in the kidney by transaminidation of its precursor (Ryan et al, 1969, Arch Biochem Biophys 131 : 521-526; Ryan and Wells, 1964, Science 144: 1122-1127). Studies have shown that homoarginine serves as a precursor of nitric oxide (NO) by increasing the intracellular concentration of L-arginine, which is the main substrate for NO synthase (Knowles et al, 1989, Proc Natl Acad Sci U S A 86: 5159-5162; Yang and Ming, 2006, Curr Hypertens Rep 8: 54-59).

The current understanding of homoarginine metabolism is, however, in its infancy. Previous studies suggest that homoarginine is mainly derived from the kidney and may thus be associated with renal function. Beyond this, an association of homoarginine with bone metabolism has recently been characterized (Pilz et al., Osteroporos Int. 2012). In the context of the present invention, the term "homoarginine" generally refers to a chemical compound as described by formula (I) below. Preferably, homoarginine is L-homoarginine.

Determining the amount of homoarginine relates to measuring the amount or concentration, preferably semi-quantitatively or quantitatively. Measuring can be done directly or indirectly. Direct measuring relates to measuring the amount or concentration of homoarginine based on a signal which is obtained from the amino acid itself and the intensity of which directly correlates with the number of molecules of the amino acid present in the sample. Such a signal - sometimes referred to herein as intensity signal - may be obtained, e.g., by measuring an intensity value of a specific physical or chemical property of the amino acid. Indirect measuring includes measuring of a signal obtained from a secondary component (i.e. a component not being the amino acid itself) or a biological read out system, e.g., measurable cellular responses, ligands, labels, or enzymatic reaction products.

In the context of the present invention, it has been surprisingly found that administration of homoarginine is effective in the improvement of myocardial function and positively affects cardiac remodeling, thereby reducing ventricular dysfunction which is known as one reason for myocardial failure. Consequently, the present invention provides effective agents for the prevention and/or for the treatment of myocardial failure. Based on the finding that administered homoarginine is capable of improving diverse biological parameters of myocardial function in a living organism (see example section), the present invention now provides an effective tool for preventing and treating myocardial failure.

In the context of the present invention, the term "prevention" or "treatment" means that not only symptoms of the disease are relieved but that also the disease itself is treated or prevented. In a preferred embodiment, the term "treatment" means improving the prognosis of said disease.

The term "myocardial failure" as used herein generally refers to a dysfunction of the cardiovascular system, in particular to a dysfunction in form of ventricular dysfunction. Myocardial failure according to the present invention includes, but is not limited to, myocardial failure as a consequence of coronary heart disease, arterial hypertension, aortic stenosis, myocarditis, cardiomyopathy, and/or inherited abnormalities of the myocardium. From a clinical point of view, these diseases are characterized by a systolic or diastolic dysfunction. Methods of how to diagnose and determine myocardial failure in a patient are well known to the skilled person and include, for example, electrocardiogram, chest X-ray, echocardiography and/or the analysis of endogenously expressed biomarkers indicating myocardial performance. Accordingly, in a preferred embodiment, myocardial failure relates to ventricular dysfunction, preferably as a consequence of coronary heart disease, arterial hypertension, aortic stenosis, myocarditis, cardiomyopathy, and/or inherited abnormalities of the myocardium.

Even more preferably, the myocardial failure relates to a systolic or diastolic dysfunction.

Coronary heart disease and arterial hypertension are known as the major causes for myocardial failure in the Western industrialized countries. Presumably, at least 50 % of all diagnosed myocardial failure is based on either coronary heart disease or arterial hypertension and/or on a combination thereof. Approximately 5 % of all myocardial failure symptoms are caused by aortic stenosis. Inherited abnormalities of the myocardium are rather a rare cause of myocardial failure. Myocardial failure is generally characterized by and results in a clinical follow up of cardiac remodelling.

Accordingly, in a preferred embodiment, the myocardial failure is characterized by cardiac remodelling.

The term "cardiac remodelling" as used herein generally means a chronic change and/or alteration of myocardial tissue including, but not limited to, collagen proliferation, the increase of collagen fibers, hypertrophy of myocytes, preferably the hypertrophy of myocytes in vital areas of the myocardium, the development of a ventricular dilatation and/or the development of a systolic or diastolic dysfunction.

In yet another preferred embodiment, the myocardial failure is characterized by ventricular remodelling. The term "ventricular remodelling" as used herein includes any change in size, shape, structure and physiology of the heart after injury to the myocardium. The injury is typically due to acute myocardial infarction, but may also be due to a number of causes that result in increased pressure or volume overload on the heart. After the insult occurs, a series of histopatho logical and structural changes occur in the ventricular myocardium that lead to progressive decline in ventricular performance. Ultimately, ventricular remodelling may result in diminished contractile (systolic) function and reduced stroke volume. Ventricular remodelling according to the present invention also includes ventricular hypertrophy. Even more preferably, the ventricular remodelling is characterized by ventricular hypertrophy.

The term "ventricular hypertrophy" as used herein generally refers to an increase in size of the patient's heart ventricle(s). In the context of the present invention, ventricular hypertrophy preferably means an increase in size of the patient ^' s left ventricle, but this increase in size may also relate to patient ^' s right ventricle. Hence, ventricular hypertrophy according the present invention also relate to an increase in size of the patient ^' s right heart ventricle. In the context of the present invention, ventricular hypertrophy is preferably characterized by an increased collagen proliferation in form of, for example, an increased number of collagen fibers and/or an increased number and size of myocytes, thereby resulting in a hypertrophy of myocytes or of the myocardium.

As found in the context of the present invention, an increase of the endogenous level of homoarginine results in a change of diverse biological parameters that are indicative for myocardial function. In particular, it has been found that an increased level of homoarginine improves myocardial function and affects cardiac remodelling.

Hence in a preferred embodiment, the method of the present invention results in a change of at least one biological parameter indicating myocardial performance.

In a preferred embodiment, an indicator of myocardial performance means an indicator of myocardial remodeling and/or myocardial fibrosis. The term "biological parameter" as used herein generally includes any gene expression product which is differentially expressed, i.e. up regulated or down regulated in presence or absence of a certain condition, disease, or complication. The biological parameter of the present invention, however, can also refer to one or more diagnostic criteria of a patient, such as, for example, the patient ^' s heart beat, the patient ^' s age, and/or the patient ^' s ventricular ejection volume. Usually, a biological parameter of the present invention is referred to as a biomarker and may be defined as a nucleic acid (including, for example, mR A), a protein, peptide, or small molecule compound. The amount of a suitable biological parameter can indicate the presence or absence of a condition, a disease, or a complication, and thus allows for a particular diagnosis and therapy. Biological parameters indicating cardiovascular function including their analysis are well known in the art and may be preformed by commercially available routine assay systems, including, but not limited to, immunoassays, antibody approaches, histological analysis, fluorescence approaches and/or all methods of gene profiling and/or gene expression analysis including, for example, Western Blots, Northern Blots, quantitative PCR and real time (RT)-PCR approaches (Schnabel et al, 2010. Eur Heart J, 31 : 3024-3031).

The term "indicator of myocardial performance" as used herein means any biological parameter which provides information about the medical condition of the patient's heart function. In particular, an indicator of mycocardial performance according to the present invention means a biological parameter that provides information about a patient's cardiovascular function, in particular the patient's myocardial function. Examples of biological parameters indicating myocardial performance in a patient include, but are not limited to, endogenously expressed biomarkers such as natriuretic peptides and beta-myosin heaving chain (β-MHC), copeptin, C-terminal-pro-endothelin-1, mid-regional-pro- adrenomedullin (MR-proADM), atrial natriuretic factor (ANF), mid-regional-pro-atrial natriuretic peptide (MR-proANP), brain natriuretic peptide (BNP), and N-terminal-pro-B-type natriuretic peptide (NT-proBNP). Moreover, in the context of the present invention, it has been found that homoarginine affects the proliferation of collagen fibers and, therefore, the progress of fibrosis (see example section). Hence, a biological parameter indicating myocardial failure according to the present invention may also refer to the steady state level and/or to the progression of fibrosis. An indicator of myocardial performance according to the present invention may also refer to the patient's ventricular ejection fraction (in % of volume per volume) which is a measurement of the percentage of blood leaving the patient ^' s heart each time it contracts. In particular, the term "ejection fraction" refers to the percentage of blood that is pumped out of a filled ventricle with each heartbeat. Because the left ventricle is the heart's main pumping chamber, ejection fractions are usually measured only in the left ventricle (LV). A normal left ventricular ejection fraction is considered to be in the range of 55 to 70 %. Under certain circumstances, in particular in patients with cardiovascular diseases in form of myocardial failure, the ejection fraction may decrease.

According,y tn the context of the present invention, a biological parameter indicating myocardial performance is preferably defined by the patient ^' s ventricular ejection fraction in % of volume per volume. Techniques for measuring the patient's ventricular ejection fraction are well known in the art and include, for example, echocardiogram, cardiac catheterization, magnetic resonance imaging (MRI), computerized tomography (CT), or nuclear medicine scan. In the context of the present invention, the ventricular ejection fraction (VEF) is categorized into normal (> 55 %), mildly reduced (45-54 %), moderately reduced (30-44 %), and severely reduced (< 30 %). In a preferred embodiment, the patient's ventricular ejection fraction is a left ventricular ejection fraction.

Accordingly, in a preferred embodiment, the biological parameter indicating myocardial performance is defined by the concentration of at least one endogenously expressed biomarker, by the steady state level or progression of fibrosis and/or by the patient ^' s ventricular ejection fraction.

Methods of how to determine the steady state level or progression of fibrosis are well known to the skilled artisan and include, for example, cardiac magnetic resonance imaging (MRI), myocardial histology or the determination of systemic markers of myocardial fibrosis, including, but limited to, galectin-3.

In another preferred embodiment, the method of prevention or treatment according to the present invention results in a down-regulation of the concentration of at least one endogenously expressed biomarker, in a reduction of fibrosis and/or in an increase of the patient's ventricular ejection fraction.

In yet another preferred embodiment, the method results in a down-regulation of the concentration of at least one endogenously expressed biomarker, whereby this down- regulation is characterized by a decrease in concentration of at least 30 %, 40 % , 50 %, 60 %, 70 %, 80 %, or 90 %.

That is, different endogenously expressed biomarker may differently decrease in concentration as a result of the present method, and thereby different levels of down- regulation may be observed. As found in the context of the present invention (see example section), the down-regulation of beta-myosin heavy chain (β-MHC) resulted in a decrease of concentration of at least 90 % in response to homoarginine, the down-regulation of atrial natriuretic factor (ANF) resulted in a decrease of concentration of at least 40 %, while the down-regulation of brain natriuretic peptide (BNP) resulted in a decrease of concentration of at least 50 %.

Accordingly, the indicator of myocardial performance may also be defined by the patient ^' s endogenous blood concentration of natriuretic peptides. The term "natriuretic peptides" generally refers all known atrial natriuretic peptides (ANP) and brain natriuretic peptides (BNP) described in the art (Bonow 1996, Circulation 93, 1946), including, for example, pre- proANP, proANP, NT -pro ANP, ANP and pre-proBNP, proBNP, NT-proBNP, and BNP. The pre-pro peptide (134 amino acids in the case of pre-proBNP) comprises a short signal peptide, which is enzymatically cleaved releasing the pro peptide (108 amino acids in the case of proBNP). The pro peptide is further cleaved into an N-terminal pro peptide (NT -pro peptide, 76 amino acids in case of NT -proBNP) and the active hormone (32 amino acids in the case of BNP, 28 amino acids in the case of ANP). Preferably, natriuretic peptides according to the present invention are NT-proANP, ANP, and, more preferably, NT -proBNP, BNP, and variants thereof. ANP and BNP are the active hormones and have a shorter half-life than their respective inactive counterparts, NT -pro ANP and NT -proBNP. BNP is metabolised in the blood, whereas NT -proBNP circulates in the blood as an intact molecule and as such is eliminated renally. The in- vivo half-life of NTproBNP is 120 min longer than that of BNP, which is 20 min (Smith 2000, J Endocrinol 167, 239). Preanalytics are more robust with NTproBNP allowing easy transportation of the sample to a central laboratory (Mueller 2004, Clin Chem Lab Med 42, 942). Blood samples can be stored at room temperature for several days or may be mailed or shipped without recovery loss. In contrast, storage of BNP for 48 hours at room temperature or at 4° Celsius leads to a concentration loss of at least 20 % (Wu 2004, Clin Chem 50, 867). Therefore, depending on the time-course or properties of interest, either measurement of the active or the inactive forms of the natriuretic peptide can be advantageous. Preferred natriuretic peptides according to the present invention are NT- proBNP or variants thereof. The human NT-proBNP, is a polypeptide comprising, preferably, 76 amino acids in length corresponding to the N-terminal portion of the human NT-proBNP molecule. An alternatively preferred natriuretic peptide is the mid-regional-pro-atrial natriuretic peptide (MR-proANP). Methods for measurement natriuretic peptides have been described (Ala-Kopsala 2004, Clin Chem 50, 1576), and the concentration of the natriuretic peptides is preferably calculated in pg/ml, but may also be indicated by any other suitable concentration format, such as, for example, pmol/dl. For example, a concentration of brain natriuretic peptide (BNP) of < 50 pg/ml and/or a concentration of N-terminal pro-B-type natriuretic peptide (NT-proBNP) of < 300 pg/ml excludes myocardial failure with high probability (Stork et al, Dtsch Med Wochenschr 2008; 133:636-641). Accordingly, in another preferred embodiment of the present invention, the endogenously expressed biomarker is selected from the group consisting of natriuretic peptides and beta- myosin heavy chain (β-MHC).

More preferably, the natriuretic peptides are selected from the group consisting of atrial natriuretic factor (ANF), brain natriuretic peptide (BNP) and N-terminal pro-B-type natriuretic peptide (NT-proBNP).

In yet another preferred embodiment of the present invention, the patient's ventricular ejection fraction is increased to at least 35 % of volume per volume, preferably to at least 40 % of volume per volume, more preferably to 45-54 % of volume per volume, and most preferably to at least 55 % of volume per volume. As already outlined, methods of how to determine an increase in the patient ^' s ventricular ejection fraction, in particular in the patient's left ventricular ejection fraction, are well known in the art and to the skilled person, in particular in a clinical context. In the context of the present invention, it has been demonstrated that homoarginine has an effect on biological parameter indicating myocardial performance. Hence, it can be envisaged that also any metabolic precursor of homoarginie can have the same effect. One well known precursor of homoarginine is L-lysine. Accordingly, in a preferred embodiment, the aforementioned metabolic precursor is L-lysine.

L-Lysine is an a-amino acid with the chemical formula H02CCH(NH2)(CH2)4NH2. It is an essential amino acid, as it is not synthesized in animals, hence it must be ingested as lysine or lysine-containing proteins. In plants and bacteria, it is synthesized from aspartic acid (aspartate). Lysine is a base. The ε-amino group often participates in hydrogen bonding and as a general base in catalysis. Common posttranslational modifications include methylation of the ε-amino group, giving methyl-, dimethyl-, and trimethyllysine. The latter occurs in calmodulin. Other posttranslational modifications at lysine residues include acetylation and ubiquitination. Collagen contains hydroxylysine which is derived from lysine by lysyl hydroxylase. O-Glycosylation of lysine residues in the endoplasmic reticulum or Golgi apparatus is used to mark certain proteins for secretion from the cell. Lysine is metabolised in mammals to give acetyl-CoA, via an initial transamination with a-ketoglutarate. The bacterial degradation of lysine yields cadaverine by decarboxylation. Allysine is a derivative of lysine, used in the production of elastin and collagen. It is produced by the actions of the enzyme lysyl oxidase on lysine in the extracellular matrix and is essential in the crosslink formation that stabilizes collagen and elastin. L-Lysine is a necessary building block for all proteins in the body. L-Lysine plays a major role in calcium absorption; building muscle protein; recovering from surgery or sports injuries; and the body's production of hormones, enzymes, and antibodies. Lysine can be modified through acetylation, methylation, ubiquitination, sumoylation, neddylation, biotinylation, and carboxylation which tend to modify the function of the protein of which the modified lysine residue(s) are a part. Furthermore, the present invention relates to the use of homoarginine or of a metabolic precursor thereof for the preparation of a medicament for the prevention or the treatment of myocardial failure.

Preferably, the present invention relates to the use of homoarginine for the preparation of a medicament for the prevention or the treatment of myocardial failure.

The use of homoarginine or of a metabolic precursor thereof for the preparation of a medicament according to the present invention is preferably defined by any of the embodiments as stated herein.

The term "patient", preferably, refers to a mammal, more preferably to a human, most preferably a human patient. In a preferred embodiment of the present invention, the patient is healthy with respect to diseases that increase the risk of myocardial failure. Said diseases are, preferably, coronary heart disease, arterial hypertension, renal failure, type 1 diabetes, type 2 diabetes and cardiovascular diseases such as e.g. stroke. In a further preferred embodiment of the present invention the patient suffers from chronic or acute cardiovascular diseases including myocardial failure, vertricular hypertrophy, myocardial remodelling, and acute coronary syndromes.

Moreover, the present invention relates to a pharmaceutical composition comprising homoarginine or a precursor thereof.

The term "medicament" as used herein refers, in one aspect, to a pharmaceutical composition containing homoarginine or a precursor thereof as pharmaceutical active compound, wherein the pharmaceutical composition may be used for human or non human therapy of various diseases or disorders in a therapeutically effective dose.

The pharmaceutical compositions and formulations referred to herein are administered at least once in order to treat or ameliorate or prevent a disease or condition recited in this specification. However, the said pharmaceutical compositions may be administered more than one time. Specific pharmaceutical compositions are prepared in a manner well known in the pharmaceutical art and comprise at least one active compound referred to herein above in admixture or otherwise associated with a pharmaceutically acceptable carrier or diluent. For making those specific pharmaceutical compositions, the active compound(s) will usually be mixed with a carrier or the diluent. The resulting formulations are to be adapted to the mode of administration. Dosage recommendations shall be indicated in the prescribers or users instructions in order to anticipate dose adjustments depending on the considered recipient.

A pharmaceutical composition of the present invention comprises homoarginine or a precursor thereof, and preferably, one or more pharmaceutically acceptable carrier.

The carrier(s) must be acceptable in the sense of being compatible with the other ingredients of the formulation and being not deleterious to the recipient thereof. The pharmaceutical carrier employed may be, for example, a solid, a gel or a liquid. Exemplary of solid carriers are lactose, terra alba, sucrose, talc, gelatin, agar, pectin, acacia, magnesium stearate, stearic acid and the like. Exemplary of liquid carriers are phosphate buffered saline solution, syrup, oil such as peanut oil and olive oil, water, emulsions, various types of wetting agents, sterile solutions and the like. Similarly, the carrier or diluent may include time delay material well known to the art, such as glyceryl mono-stearate or glyceryl distearate alone or with a wax. Said suitable carriers comprise those mentioned above and others well known in the art, see, e.g., Remington's Pharmaceutical Sciences, Mack Publishing Company, Easton, Pennsylvania.

The diluent(s) is/are selected so as not to affect the biological activity of the combination. Examples of such diluents are distilled water, physiological saline, Ringer's solutions, dextrose solution, and Hank's solution. In addition, the pharmaceutical composition or formulation may also include other carriers, adjuvants, or non-toxic, non-therapeutic, non- immunogenic stabilizers and the like. The pharmaceutical composition is, preferably, administered in conventional dosage forms prepared by combining the drugs with standard pharmaceutical carrier according to conventional procedures. These procedures may involve mixing, granulating and compressing or dissolving the ingredients as appropriate to the desired preparation. It will be appreciated that the form and character of the pharmaceutically acceptable carrier or diluent is dictated by the amount of active compound with which it is to be combined, the route of administration and other well-known variables. However, depending on the nature and the mode of action of an active compound the pharmaceutical composition may be administered by other routes as well.

In the context of the present invention, it is preferred that the homoarginine or the metabolic precursor thereof is administered by means of oral or parenteral administration, preferably by intramuscular, subcutaneous or intravenous administration. A therapeutically effective dose refers to an amount of the pharmaceutically active compound to be used in a pharmaceutical composition of the present invention which prevents, ameliorates or treats the symptoms accompanying a disease or condition referred to in this specification. Therapeutic efficacy and toxicity of the compound can be determined by standard pharmaceutical procedures in cell cultures or experimental animals, e.g., ED50 (the dose therapeutically effective in 50% of the population) and LD50 (the dose lethal to 50% of the population). The dose ratio between therapeutic and toxic effects is the therapeutic index, and it can be expressed as the ratio, LD50/ED50.

The dosage regimen will be determined by the attending physician and other clinical factors. As is well known in the medical arts, dosages for any one patient depends upon many factors, including the patient's size, body surface area, age, the particular compound to be administered, sex, time and route of administration, general health, and other drugs being administered concurrently. Progress can be monitored by periodic assessment. The medicament according to the present invention may, preferably, comprise drugs in addition to homoarginine or in addition to its precursor which are added to the pharmaceutical composition during its formulation. Finally, it is to be understood that the formulation of a pharmaceutical composition takes place under GMP standardized conditions or the like in order to ensure quality, pharmaceutical security, and effectiveness of the medicament.

In yet another preferred embodiment of the present invention, the homoarginine or the metabolic precursor thereof is administered for a period of several weeks, months or years. Preferably, the homoarginine or the metabolic precursor thereof is administered for a period of at least 2, 4 or 6 weeks, or for a period of at least 2, 4 or 6 months, or for a period of at least 2, 4, or 6 years. Any variations of these time periods in days, weeks or months are envisaged in the context of the present invention, depending on the patient's need thereof. Preferably, the administration according to the present invention is in form of a long-term administration such as, for example, the administration for a time period of several years. In yet another preferred embodiment of the present invention, the homoarginine or the metabolic precursor thereof is administered in conjunction with at least one additional treatment of myocardial failure.

Several conventional treatments of myocardial failure are known to the skilled person (Hoppe et al., 2005, Z Kardiol 94: 488-509). These include, but are not limited to, inhibitors of the angiotensin converting enzyme (ACE), inhibitors of the neutral endopeptidase (NEP), inhibitors of diverse matrix metalloproteases (MMPs), beta blockers, ATi receptor blockers, diuretic agents and aldosterone antagonists. Hence, in a preferred embodiment of the present invention, the additional treatment of myocardial failure is in form of administering one or more inhibitor(s) of the angiotensin converting enzyme (ACE), one or more inhibitor(s) of the neutral endopeptidase (NEP), one or more inhibitors of matrix metalloproteases (MMPs), one or more beta blocker(s), one or more ATi receptor blocker(s), one or more diuretic agent(s) and/or one ore more aldosterone antagonist(s).

Even more preferably, the matrix metalloprotease (MMP) is selected from the group consisting of matrix metalloprotease 1 (MMP1), matrix metalloprotease 2 (MMP2) and matrix metalloprotease 9 (MMP9).

In the context of the present invention, it can also be envisaged that the level of homoarginine or the level of the metabolic precursor thereof is regulated endogenously, preferably by means of regulation gene expression, more preferably by the regulation of genes encoding enzymes involved in the generation or catabolism of homoarginine. Hence, in the context of the present invention, it is envisaged that this regulation is mediated by a medicine or an endogenous compound or an enzyme that is involved in the arginine metabolic pathway and/or metabolism, in particular in the metabolic pathway and/or the biosynthesis of homoarginine within the organism.

Accordingly, in yet another preferred embodiment of the present invention, the aforementioned metabolic precursor is a medicine, an endogenous compound or an enzyme involved in the homoarginine metabolic pathway. Even more preferably, the medicine, endogenous compound or the enzyme is involved in the up-regulation of endogenous serum or tissue levels of homoarginine.

Diverse enzymes which fulfil these criteria have been identified in the context of the present invention (see example section).

Accordingly, in a preferred embodiment, the metabolic precursor is a gene product or a regulator of a gene product associated with the CPSl locus on chromosome 2, the AGXT2 locus on chromosome 5, the GATM locus on chromosome 15 and/or the MED23/ARG1 locus on chromosome 6.

The term "regulator of a gene product" as used herein generally means any gene product, medicine, endogenous compound or enzyme which is involved in or participates to the regulation of gene expression with respect to these gene loci identified herein. Even more preferably, the metabolic precursor is selected from the group of enzymes consisting of L-arginine: glycine amidinotransferase (AGAT), carbamoyl phosphatase synthetase I (CPSl), mitochondrial alanine-glyoxlate aminotransferase and arginase (ARG1).

The following example is only intended to illustrate the present invention and shall not limit the scope of the invention in any way. FIGURES

Figure 1A, IB: Transthoracic echocardiographic measurements revealed an improvement in cardiac function four weeks following aortic banding surgery and homoarginine treatment. EF and FS were significantly higher in the 80 mg HA-treatment group after 28 days as compared with untreated animals with AB (*p < 0,05 vs. AS w/o HA).

Figure 2A: Rats within the 40 mg + 80 mg AB-group showed a decrease in collagen area fraction of more than 59 % (*p < 0.05; **p< 0.05 vs. AS w/o HA). Figure 2B, 2C: Two exemplary sirius red stainings of rats with HA-treatment (4A) and untreated animals (4B) 28 days after aortic banding surgery.

Figure 3A: Quantitative measurement of the myocyte size 28 days after AB. Rats with aortic banding experienced a more than 14 % decrease in their myocyte size after homoarginine treatment (*p < 0.05; **p< 0.05 vs. AS w/o HA). Sham-operated rats showed a reduction of 4,8 % after application of 80 mg HA (#p < 0,05 vs. Sham w/o HA). Figure 3B, 3C: Exemplary hematoxylin eosin stainings from a representative section in HA-treated (6A) and untreated animals (6B). Figure 4 A, 4B, 4C: RT-qPCR with ANF, BNP and β-MHC normalized to the house-keeping gene HPRT revealed a downregulation.

Figure 5: Manhattan plot of GWAS meta-analysis with inverse normal transformed residuals of a regression analysis of homoarginine adjusted for age, sex, bmi and principle components as phenotype. Each dot represents one SNP present in the meta-analysis dataset and it's y-axis coordinate indicates significance level in association with homoarginine.

Figure 6: Regional plot of the CPS1 region on chromosome 2. Each dot represents one SNP present in the meta-analysis dataset and its y-axis coordinate indicates significance level in association with homoarginine level. For each SNP, linkage disequilibrium measured as r2 with the lead SNP is indicated by the color scheme. Genome recombination rates are shown by blue lines and genetic annotation is shown at the bottom. Figure 7: Regional plot of the AGXT2 region on chromosome 5. Each dot represents one SNP present in the meta-analysis dataset and its y-axis coordinate indicates significance level in association with homoarginine level. For each SNP, linkage disequilibrium measured as r2 with the lead SNP is indicated by the color scheme. Genome recombination rates are shown by blue lines and genetic annotation is shown at the bottom.

Figure 8: Regional plot of the GATM region on chromosome 15. Each dot represents one SNP present in the meta-analysis dataset and its y-axis coordinate indicates significance level in association with homoarginine level. For each SNP, linkage disequilibrium measured as r2 with the lead SNP is indicated by the color scheme. Genome recombination rates are shown by blue lines and genetic annotation is shown at the bottom.

Figure 9: Regional plot of the ARG1 region on chromosome 6. Each dot represents one SNP present in the meta-analysis dataset and its y-axis coordinate indicates significance level in association with homoarginine level. For each SNP, linkage disequilibrium measured as r2 with the lead SNP is indicated by the color scheme. Genome recombination rates are shown by blue lines and genetic annotation is shown at the bottom.

Figure 10: eQTL expression analysis of the association of rsl346268 (proxy of rsl 153858) with the expression of GATM and nearby genes in artery tissue, cells from whole blood and monocytes in TVS.

Figure 11: Manhattan plot of GWAS in LURIC with inverse normal transformed residuals of a regression analysis of homoarginine adjusted for age, sex, bmi and principle components as phenotype. Each dot represents one SNP present in the dataset and it's y-axis coordinate indicates significance level in association with homoarginine.

Figure 12: Manhattan plot of GWAS in YFS with inverse normal transformed residuals of a regression analysis of homoarginine adjusted for age, sex, bmi and principle components as phenotype. Each dot represents one SNP present in the dataset and it's y-axis coordinate indicates significance level in association with homoarginine.

Figure 13: Regional plots of the CPS1, AGXT2 and GATM regions in LURIC and YFS, respectively. Each dot represents one SNP present in the dataset and its y-axis coordinate indicates significance level in association with homoarginine level. For each SNP, linkage disequilibrium measured as r2 with the lead SNP is indicated by the color scheme. Genome recombination rates are shown by blue lines and genetic annotation is shown at the bottom. Figure 14: Manhattan plot of GWAS in LURIC with inverse normal transformed residuals of a regression analysis of homoarginine adjusted for age, sex, bmi and principle components as phenotype with additional adjustment for the lead SNPs from the meta-analysis (rs7422339, rs37369 and rsl 153858). Each dot represents one SNP present in the dataset and it's y-axis coordinate indicates significance level in association with homoarginine.

Figure 15: QQ plots of GWAS results in LURIC (A), YFS (B) and meta-analysis (C), respectively.

EXAMPLES Example 1

The influence of homoarginine on myocardial function and remodelling was investigated in rats undergoing aortic banding.

Methods and experimental program

Twenty-one rats with a mean body weight of 299 g were subjected equally to aortic banding (AB) or sham operation (SH). Rats from both groups were gavaged with 0 mg, 40 mg and 80 mg (mg/kg body weight) homoarginine once a day for a period of one month. Already two weeks after treatment the mean serum values of homoarginine rised from 1 ,0 up to 274 mg/L in the 80 mg AB-group. After 4 weeks echocardiographic and invasive measurements were performed under anaesthesia with isofluran and the rats were euthanized. The heart was arrested by an injection of saturated potassium chloride solution and heart weight, liver weight, lung weight and tibia length was determined. After dissection and weighing of the LV, samples of the myocardium were taken and snap frozen for biochemical analyses or fixed in formalin for further histological evaluation. The gavaged animals were compared with unfed animals concerning macroscopic findings, echocardiography, histology and gene expression levels of molecular hypertrophic markers. Results

The amounts of orally applied homoarginine showed a significant positive correlation with the echocardiographic ejection fraction (+14 %), fractional shortening (+20 %), left ventricular weight/ tibia length quotient (+20 %) and left ventricle weight (+20 %) in the 80 mg AB-group compared with unfed animals with aortic stenosis (p < 0.05). Furthermore high doses of homoarginine in the 80 mg AB-group were inversely associated with the collagen area fraction (-59 %), myocyte cross-sectional area (-14 %), maximum arterial LV-pressure loops (-13 %) and the molecular hypertrophic markers atrial natriuretic factor (ANF), brain natriuretic peptide (BNP) and beta-myosin heavy chain (β-MHC) (p < 0.05). Especially in the analysis of β-MHC rats within the 80 mg AB-group experienced a marked down-regulation of 97 % in comparison to unfed animals with aortic stenosis. BNP was down-regulated by 44 % and ANF by 59 %. Our results in the 40 mg AB-group showed similar data as compared with the 80 mg AB-group. The only correlation of homoarginine within the sham group was a decrease of 4.8 % in myocyte size of the 80 mg-group vs. untreated sham rats (p < 0.05).

1. Echocardiography

Transthoracic echocardiographic measurements revealed an improvement in cardiac function four weeks following aortic banding surgery and homoarginine treatment. EF and FS were significantly higher in the 80 mg HA-treatment group after 28 days as compared with untreated animals with AB (*p < 0.05 vs. AS w/o HA).

2. Collagen content

Homoarginine application led to a strong reduction of collagen fibers.

3. Myocyte cross-sectional area

Quantitative measurement of the myocyte size 28 days after AB demonstrated that rats with aortic banding have a more than 14 % decrease in myocyte size after homoarginine treatment. (*p < 0.05; **p< 0.05 vs. AS w/o HA).

4. Molecular analysis

The expression levels of the molecular hypertrophic markers ANF, BNP and β-MHC were decreased within animals with aortic banding and homoarginine treatment. Conclusion and future aspects

The application of homoarginine leads to a significant improvement of myocardial function and resulted in histological and molecular changes four weeks following aortic banding. This was confirmed by a reduction in fibrosis, myocyte cross-sectional area and furthermore by a down-regulation of molecular hypertrophic markers. Moreover the ejection fraction and fractional shortening as two important markers for the cardiac function improved after the 4 week treatment with the amino acid derivative. Sham-operated animals showed only in the hostological evaluation of the myocyte size a weak decrease after 80 mg HA application. The improvements found can partially be explained by previous findings suggesting that homoarginine may be involved in the metabolism of the vasodilator nitric oxide (NO) and increases NO production by serving as a precursor for nitric oxide synthase. Thereby homoarginine may be positively related to endothelial function. In addition experimental studies have shown that homoarginine exerts antihypertensive and antithrombotic effects and increases insulin production. The antihypertensive effect was confirmed by LV-pressure measurements in this experiment.

In line with these beneficial effects of homoarginine on the cardiovascular system it has been shown that low homoarginine levels are associated with endothelial dysfunction, deterioration of cardiac function and an increased risk of cardiovascular mortality in special patient groups. Further studies are needed to explore the significance of homoarginine to cardiac remodeling and cardiovascular outcome and to elucidate the underlying pathomechanisms.

Example 2

Background: Low serum levels of the amino acid derivate homoarginine have been associated with increased risk of total and cardiovascular mortality. Homoarginine deficiency may be related to renal and heart diseases but the pathophysiologic role of homoarginine and the genetic regulation of its serum levels are unknown.

Methods and Results: In 3041 patients of the Ludwigshafen Risk and Cardiovascular Health (LURIC) study referred for coronary angiography and 2074 participants of the Young Finns Study (YFS) we performed a genome-wide association study to identify genomic loci associated with homoarginine serum levels and tested for associations of identified SNPs with mortality in LURIC.

We found genome-wide significant associations with homoarginine serum levels on chromosome 2 at the CPS1 locus, on chromosome 5 at the AGXT2 locus and on chromosome 15 at the GATM locus as well as a suggestive association on chromosome 6 at the MED23/ARG1 locus. All loci harbour enzymes involved in arginine metabolism. The SNP rsl 153858 near the GATM gene was associated with mortality in a subgroup of LURIC patients less than 67 years old with a hazard ratio of 1.69 (95 %CI 1.06-2.70, P = 0.029) for individuals homozygous for the major allele compared to individuals homozygous for the minor allele in a model adjusted for conventional cardiovascular risk factors.

Conclusions: Our data provide novel insights into the genetic regulation of homoarginine metabolism. Low homoarginine or GATM activity may causally be linked to mortality. This concept needs to be tested in mechanistic studies.

METHODS

Study populations

1. Ludwigshafen Risk and Cardiovascular Health (LURIC) study

The LURIC study consists of 3,316 Caucasian patients hospitalized for coronary angiography between 1997 and 2000 at a tertiary care center in Southwestern Germany (Winkelmann et al, 2001, Pharmacogenomics 2: Sl-73). Clinical indications for angiography were chest pain or a positive non-invasive stress test suggestive of myocardial ischemia. To limit clinical heterogeneity, individuals suffering from acute illnesses other than acute coronary syndrome (ACS), chronic non-cardiac diseases and a history of malignancy within the five past years were excluded. The study was approved by the ethics committee at the "Arztekammer Rheinland-Pfalz" and was conducted in accordance with the "Declaration of Helsinki." Informed written consent was obtained from all participants. In 3,041 LURIC participants, genotypes, homoarginine levels and all necessary covariates were available.

2. The Cardiovascular Risk in Young Finns Study (YFS)

In brief, the YFS cohort is a Finnish longitudinal population study sample on the evolution of cardiovascular risk factors from childhood to adulthood (Raitakari et al, 2008, Int J Epidemiol. 37: 1220-1226). In the present study, we used the variables measured in 2001. For these subjects, genotyping was performed in 2009 using a custom-built Illumina Human 670k BeadChip at the Welcome Trust Sanger Institute. In 2,074 subjects, genotype, risks factors, and phenotype data were complete, and they formed the present study population. This study was conducted according to the guidelines of the Declaration of Helsinki and the study protocol was approved by local ethics committees. All participants gave their informed consent.

3. Tampere Vascular Study (TVS)

Vascular sample series from femoral arteries, carotid arteries and abdominal aortas were obtained during open vascular procedures between 2005 and 2008. The study has been approved by the Ethics Committee of Tampere University Hospital. All clinical investigations were conducted according to the declaration of Helsinki. TVS whole blood and monocyte collections were performed in 2008, the angio graphically verified patient samples to new TVS study were pre-selected from a larger population based study (Nieminen et al, 2006, BMC cardiovascular disorders 6: 9) - 55 individuals with angiographically verified coronary artery disease and 45 without were selected. The participant pool comprises patients who have undergone an exercise stress test at Tampere University Hospital. Laboratory Procedures

For LURIC patients fasting blood samples were obtained by venipuncture in the early morning. Blood glucose, cholesterol and triglycerides (TG) were measured by standard laboratory procedures as described previously (Winkelmann et al, 2001, Pharmacogenomics 2: Sl-73). HDL cholesterol (HDL-C) was measured after separating lipoproteins with a combined ultracentrifugation-precipitation method (Winkelmann et al, 2001, Pharmacogenomics 2: Sl-73). Genomic DNA was prepared from EDTA anticoagulated peripheral blood by using a common salting-out procedure. Homocysteine was determined by HPLC (Waters millennium chromatography with fluorescence detector 470). Arginine was measured in serum samples with a conventional amino acid analysis technique, involving separation of amino acids by ion exchange chromatography followed by postcolumn continuous reaction with ninhydrin (Moore et al., 1958, Federation proceedings 17: 1107- 1115). Creatine was determined in serum by liquid chromatography tandem mass spectrometry (LC-MS/MS) (Carlings et al, 2008, Ann Clin Biochem. 45 : 575-584). For YFS subjects venous blood samples were drawn after an overnight fast. Standard enzymatic methods were used for serum total cholesterol, TG, and HDL-C.

Homoarginine was measured in all studies in serum stored at -80°C by a reversed phase high- performance liquid chromatography at the Clinical Institute of Medical and Chemical Laboratory Diagnostics, Medical University of Graz (Meinitzer et al, 2007, Clin Chim Acta. 384: 141-148; Teerlmk et al, 2002, Analytical biochemistry 303: 131-137). Intraday coefficients of variation (CVs) at different concentrations (mean levels) were 4.7% (1.21 μιηοΙ/L) and 2.2% (3.53 μιηοΙ/L), and between-day CVs were 7.9% (1.25 μιηοΙ/L) and 6.8% (3.66 umol/L), respectively.

Genotyping and quality control

In LURIC, genotyping was done by using the Affymetrix Human SNP Array 6.0 at the LURIC Study facility while in YFS genotyping was done using a custom-built Illumina Human 670k BeadChip at the Welcome Trust Sanger Institute (Table 4). Genotype imputation in both studies was performed using MACH 1.0 (Li et al, 2010, Genet Epidemiol 34: 816-834; Li et al, 2009, Genomics Hum Genet. 10: 387-406) and HapMap II CEU (release 22, NCBI build 36, dbSNP 126) samples as a reference. After imputation, 2,543,887 SNPs were available. SNPs with a squared correlation of > 0.3 between imputed and true genotypes were considered well imputed. For TVS samples, genotyping was done using the Illumina HumanHap660W-Quad BeadChip (Illumina, Inc., San Diego, CA, USA) according to manufacturer's recommendation.

Expression analyses

Fresh tissue samples were soaked in RNALater solution (Ambion Inc., Austin, TX, USA) and isolated with Trizol reagent (Invitrogen, Carlsbad, CA, USA) and the RNAEasy Kit (Qiagen, Valencia, CA, USA). For the whole blood analysis 2.5 ml of peripheral blood was collected to a PaXgene Blood RNA Tubes (BD). RNA was then isolated with PAXgene Blood RNA Kit (Qiagen) with DNase Set according to manufactures instructions. For separate monocyte analysis peripheral mononuclear cells were isolated from the whole blood samples by Ficoll- Paque density-gradient centrifugation (Amersham Pharmacia Biotech UK Limited, Buckinghamshire, England). Total-RNA was extracted using RNeasy® Mini Kit (Qiagen, Valencia, USA) according to the manufacturer ^' s instructions and stored in -80 °C. The quality of the RNA was evaluated spectrophotometrically. The expression levels were analyzed with an Illumina HumanHT-12 v3 Expression BeadChip (IlluminaThe BeadChips were scanned with the Illumina iScan system. Raw intensity data was exported using the Illumina GenomeStudio software. Further data processing was conducted by means of R language and appropriate Bioconductor modules. Robust multi- array averaging (RMA) (Schmid et al, 2010, BMC Genomics 11 : 349) was used to correct negative intensity values after background subtraction. Between arrays normalization was done using robust spline normalization (RSN) (Schmid et al, 2010, BMC Genomics 11 : 349). Quality control was performed using sample clustering and multi-dimensional scaling. Seven outliers were removed due to low expression profiles, 4 from carotid artery group and 3 from left internal thoracic artery (LITA) group. A more detailed description is provided in the supplementary.

Statistical analyses

For the GWAS analysis, residuals were obtained using linear regression analysis in which homoarginine levels were adjusted for sex, age, and body mass index (relevant covariates were identified using stepAIC function in R software), as well as principal components to control for population stratification and inverse normal transformation was used to obtain a normal distribution. SNPs were evaluated for association with inverse normal transformed residuals using linear regression analyses using the software QUICKTEST (http://toby.freeshell.org/software/quicktest.shtml) in LURIC and ProbABEL (Auchenko et al, 2010, BCM bioinformatics 11 : 134) in YFS. QQ and Manhattan plots were drawn for the analysis of the results. Regional plots were drawn using Locuszoom (Pruim et al, 2010, Bioinformatics 26: 2336-2337). The p value for genome-wide significance was set at p < 5 x 10-8, which corresponds to an a of 0.05 with a Bonferroni correction for one million independent tests. Meta-analysis of both studies was done using an fixed-effects, effective sample-weighted Z-score meta-analysis method, as implemented in the software METAL (Wilier et al, 2010, Bioinformatics 26: 2190-2191). Further statistical analyses were performed using the R statistical software v. 2.15.0 (http://www.r-project.org) and IBM SPSS Statistics v. 20.0 (IBM Corportion). Expression QTL analysis with lead SNPs was performed with the Genevar software using a window of ±1 Mb centered on the SNP (Yang et al, 2010, Bioinformatics 26: 2474-2476). The strength of the relationship between alleles and gene expression intensities was estimated using Spearman's rank correlation and reported as nominal P-values. RESULTS

Study description

For the current analysis 3041 samples from LURIC and 2074 samples from YFS have been included, making a total of 5115 samples. Basic characteristics for both studies are shown in Table 1. LURIC participants were on average significantly older than YFS participants with a much higher percentage of males, higher BMI, higher blood pressure, lower HDL-C and higher triglycerides. While in LURIC about 80% of patients had CAD and about 40% had diabetes mellitus, there were no known CAD patients in the YFS and only 0.7 % had diabetes mellitus. Homoarginine levels were higher in LURIC than in the YFS.

Table 1. Description of the study populations

LURIC YFS

Number of participants 3,041 2,074

Age (years) means ± SD 62.7 (10.6) 31.7 (5.0)

Male sex (%) 70.0 45.3

BMI (kg/m ² ) means ± SD 27.5 (4.0) 25.1 (4.4)

Systolic blood pressure (mmHg) means ± SD 141.2 (23.6) 117.7 (13.1)

Diastolic blood pressure (mmHg) means ± SD 80.9 (11.4) 70.8 (10.8)

Hypertension (%) 58.7 1.9

History of CAD (%) 79.1 -

History of diabetes (%) 40.5 0.6

Active Smokers (%) 23.1 24.1

Total cholesterol (mg/dl) means ± SD 192.4 (10.8) 200.3 (37.9)

HDL-C (mg/dl) means ± SD 38.7 (10.8) 50.7 (12.6)

Triglycerides (mg/dl) 146.0 97.9 median (25th and 75th percentile) (109.0 - 201.0) (71.2 - 142.4)

Homoarginine (μιηοΙ/L) 2.4 1.76 median (25th and 75th percentile) (1.8 - 3.1) (1.4 - 2.2)

Association analyses

Manhattan plots of the association of HapMap2 imputed SNPs with homoarginine serum levels for LURIC and YFS are shown in Figures SI and S2, respectively. The analyses revealed 83 and 161 SNPs in LURIC and in the YFS, respectively, with p values below the genome-wide significance threshold of 5 * 10 ^" . These SNPs are located on chromosome 2 at the CPS1 locus (carbamoyl phosphate synthetase I: EC 6.3.4.16), on chromosome 5 at the AGXT2 locus (alanine-glyoxylate aminotransferase: EC 2.6.1.44) and on chromosome 15 at the GATM locus (glycine amidinotransferase: EC 2.1.4.1). Regional plots of the three regions in LURIC and YFS are shown in Figure 13. In the meta-analysis a total of 186 SNPs were genome-wide significantly associated with homoarginine levels and we also identified another region on chromosome 6 at the MED23/ARG1 locus (homo sapiens mediator complex subunit 23 gene/arginase I) which we considered as a suggestive association (p-value of 2.21 x 10-7 for the lead SNP rsl7060430, Figure 14). A manhattan plot of meta-analysis results is shown in Figure 5, regional plots of the three genome-wide significant regions as well as of the putative region on chromosome 6 are shown in Figures 6 to 9. The strongest association with homoarginine was observed for rs4627277 in LURIC (p = 5.20 x 10 ^~24 ) and rs7171577 in YFS (p = 5.32 x 10 ^"36 ); both SNPs are residing in the same LD block on chromosome 15 at the GATM locus with an r ² = 1. This LD block also harbours rsl 153858 which achieved the lowest p value in the joined meta-analysis with 6.98 x 10 ^"56 (Table 2).

Table 2. Lead SNPs in LURIC, YFS and combined meta-analysis

EAF Beta

SNP Chr Position Alleles LURIC YFS LURIC YFS rs7422339 2 211248752 C/A 0.69 0.64 0.203 0.264 rs28305 5 35080055 C/G 0.09 0.10 0.237 0.173 rs37369 5 35072872 T/C 0.92 0.90 -0.220 -0.301 rsl7060430 6 131960430 G/A 0.97 0.94 -0.295 -0.229 rs4627277 15 43440060 C/A 0.72 0.70 -0.288 -0.415 rsl l53858 15 43439995 C/T 0.72 0.70 -0.288 -0.415

P-value

LURIC YFS Combined Candidate gene

9.88E-11 7.72E-14 1.89E-22 CPS1

1.08E-07 9.84E-04 5.85E-10 AGXT2

4.40E-06 1.24E-08 7.72E-13 AGXT2

9.38E-05 6.58E-04 2.21E-07 MED23/ARG1

5.20E-24 8.16E-36 7.16E-56 GAMT

5.21E-24 7.88E-36 6.98E-56 GAMT

1.65E-23 5.32E-36 2.04E-55 GAMT Chr: chromosome; Alleles: effect allele / reference allele; EAF: effect allele frequency;

In LURIC, mean homoarginine concentrations in individuals homozygous for the minor allele of rsl 153858 (T/T, n = 235) were 2.94± 1.24 μιηοΐ/ΐ, as compared to 2.39±0.93 μπιοΐ/l in individuals homozygous for the major allele (C/C, n = 1598) and 2.72=1=1.11 μιηοΐ/ΐ in heterozygotes (n = 1218). This SNP alone explained 2.9 % of the variance of log homoarginine. A regression model combining the lead SNPs at all four loci was able to explain as much as 4.9 % of the variance in homoarginine levels. To assess whether other SNPs might be significantly associated with homoarginine independent of the lead SNPs, we conducted a conditional analysis by additionally adjusting the GWAS analysis in LURIC for the top SNPs of the three significantly associated loci in the meta-analysis (rs7422339, rs37369, rsl 153858). These analyses found no other genome-wide significantly associated SNPs (Figure 15). The observed genomic control factors in both cohorts were near unity (LURIC: 1.019, YFS: 0.999) suggesting that the GWAS results were not strongly confounded by underlying population stratification.

We further sought to replicate the associations in two different cohorts, the 4D study and the GECOH study. In both studies the association of rsl 153858 with HARG could be replicated (both p < 0.001, Table 5), while there was no association of rs37370 at the AGXT2 locus. In 4D, the association of rsl047891 at the CPS1 locus with HARG (p < 0.001, Table 5) is shown. eQTL expression analysis

We examined the association of the lead SNPs in the three identified loci with gene expression in transcriptomic profiles of three different tissues of the Tampere Vascular Study. The major allele of rsl346268 at the GATM locus (r2 = 0.998 with rsl 153858) was significantly associated with increased mRNA levels of GATM and nearby genes in cells from the arterial plaque and whole blood as well as in monocytes (Figure 9). For the CPS1 locus and the AGXT2 locus no association of the lead SNPs with gene expression could be detected (data not shown). Association of genetic variants with mortality

We investigated the identified genetic variants for possible association with all-cause mortality by means of Cox regression analysis. For the SNPs at the ARG1, CPS1 and AGXT2 loci no association with mortality was detected. However, for rsl 153858, located near the GATM gene, there was a trend in the whole LURIC cohort which became significant after excluding older patients. Hazard ratios for different models are shown in Table 3. Patients homozygous for the major allele, which was associated with lower homoarginine, had a HR of 1.71 (1.07-2.74; P = 0.025) compared to patients homozygous for the minor allele in an age and sex adjusted model including only patients < 67 years. This association remained significant after additional adjustment for conventional risk factors but significance was lost after adjusting for either homoarginine or creatine.

Replication cohorts

4D

The 4D study (Die Deutsche Diabetes Dialyse Studie) was a prospective randomized controlled trial including patients with type 2 diabetes mellitus who had been treated by hemodialysis for less than 2 years. Between March 1998 and October 2002, 1255 patients were recruited in 178 dialysis centers in Germany. Patients were randomly assigned to double-blinded treatment with either 20 mg of atorvastatin (n = 619) or placebo (n = 636) once daily and were followed up until the date of death, censoring, or end of the study in March 2004. The primary end point of the 4D study was defined as a composite of death due to cardiac causes, stroke, and myocardial infarction, whichever occurred first. 4D study end points were centrally adjudicated by 3 members of the endpoints committee blinded to study treatment and according to predefined criteria. The study was approved by the medical ethics committees, and written informed consent was obtained from all participants. Table 3. Hazard ratios (HR) for total mortality according to rsl 153858 genotype.

Model 1 Model 2

HR (95% CI) HR (95% CI)

Patients < 67

years

(n = 1898)

T/T preference ^reference

C/T 1.60 (0.99-2.57) 0.056 1.63 (1.01-2.63) 0.045

c/c 1.71 (1.07-2.74) 0.025 1.69 (1.06-2.70) 0.029

All patients

(n= 3061)

T/T preference preference

l.ff l.ff

C/T 1.18 (0.90 - 1.55) 0.240 1.14 (0.86 - 1.50) 0.360

c/c 1.25 (0.96 - 1.64) 0.098 1.20 (0.92 - 1.58) 0.175

Model 3 P Model 4 P HR (95% CI) HR (95% CI)

-J Qreference -J Qreference

1.54 (0.95-2.48) 0.080 1.48 (0.91 - 2.43) 0.117 1.36 (0.85-2.19) 0.201 1.53 (0.94 - 2.49) 0.084

-J Qreference -J Qreference

1.03 (0.78 - 1.35) 0.849 1.09 (0.83 - 1.45) 0.538 0.98 (0.75 - 1.29) 0.908 1.14 (0.86 - 1.50) 0.366 Model 1 : adjusted for age and gender.

Model 2: model 1 additionally adjusted for LDL-C, BMI, type 2 diabetes, body mass index, smoking, hypertension

Model 3 : model 2 additionally adjusted for homoarginine

Model 4: model 2 additionally adjusted for creatine GECOH

The Graz endocrine causes of hypertension (GECOH) study is a single center diagnostic accuracy study. The study population consists of patients who are routinely referred to an outpatient clinic of a tertiary care center (Department of Internal Medicine, Division of Endocrinology and Nuclear Medicine, Medical University of Graz, Austria) for screening for endocrine hypertension. The main inclusion criterion was arterial hypertension defined according to recent guidelines as an average office blood pressure on at least two occasions of systolic > 140 and/or diastolic > 90 mmHg or a mean 24 ambulatory blood pressure of systolic > 125 and/or diastolic > 80, or a home blood pressure of systolic > 130 and/or diastolic > 85 or ongoing antihypertensive treatment that was initiated due to arterial hypertension. Further inclusion criteria were age of > 18 years and written informed consent. The exclusion criteria were a glomerular filtration rate (GFR) according to the MDRD formula < 30 ml/min/1.73 m2, severe hepatic failure (Child-Pugh class B or C), severe heart failure with NYHA class 3 or 4, acute coronary syndrome within the last two weeks, immunosuppressive therapy, glucocorticoid therapy, ongoing chemotherapy, pregnancy and any other disease with an estimated life expectancy below 1 year. Serum homoarginine levels were available for 221 participants.

Genotyping of lead SNPs in 4D and GECOH

The genotypes of 4D and GECOH were analyzed by real time PCR and pre-designed TaqMan assay (rsl 153858-C_1466373_10; rs37370-C_1018750_l; rs2279651 - C_15968296_10; rsl047891 - C 8793384 1) following the manufacturers ^' instructions (Applied Biosystems).

As internal control 95 DNA samples, blinded distributed within the samples, were genotyped twice. The concordance rate between the replicates was 100 %. The results were scored blinded to the patient status.

Expression analysis in TVS

Fresh tissue samples from TVS were soaked in RNALater solution (Ambion Inc., Austin, TX, USA) and isolated with Trizol reagent (Invitrogen, Carlsbad, CA, USA) and the RNAEasy Kit (Qiagen, Valencia, CA, USA). For the whole blood analysis 2.5 ml of peripheral blood was collected to a PaXgene Blood RNA Tubes (BD). The tube was inverted 8-10 times and then stored at room temperature for at least 2 hours. RNA was then isolated with PAXgene Blood RNA Kit (Qiagen) with DNase Set according to manufactures instructions. For separate monocyte analysis peripheral mononuclear cells were isolated from the whole blood samples by Ficoll-Paque density-gradient centrifugation (Amersham Pharmacia Biotech UK Limited, Buckinghamshire, England). Total-RNA was extracted using RNeasy® Mini Kit (Qiagen, Valencia, USA) according to the manufacturer ^' s instructions and stored in -80 °C. The quality of the RNA was evaluated spectrophotometrically.

The concentration and quality of the RNA was evaluated spectrophotometrically (BioPhotometer, Eppendorf, Wesseling-Berzdorf, Germany). The expression levels were analyzed with an Illumina HumanHT-12 v3 Expression BeadChip (Illumina). In brief, 300- 500 ng of RNA was reversely transcribed into cRNA and biotin-UTP labeled using the Illumina TotalPrep RNA Amplification Kit (Ambion), and 1,500 ng of cRNA was then hybridized to the Illumina HumanHT-12 v3 Expression BeadChip. The BeadChips were scanned with the Illumina iScan system. After background subtraction, raw intensity data was exported using the Illumina GenomeStudio software. Further data processing was conducted by means of R language and appropriate Bioconductor modules. Robust multi-array averaging (RMA) was used to correct negative intensity values after background subtraction. Between arrays normalization was done using robust spline normalization (RSN). Quality control was performed using sample clustering and multi-dimensional scaling. Seven outliers were removed due to low expression profiles, 4 from carotid artery group and 3 from left internal thoracic artery (LIT A) group.

Table 4: Genotyping, imputation and GWAS analysis

LURIC YFS

Study population and design:

Ethnicity Caucasian Caucasian

Study area Rhein-Neckar area, Germany Multi-centre, Finland

Study type CAD case-control Population sample

Genotyping:

a custom-built Illumina

Genotyping platform Affymetrix 6.0

Human 670k BeadChip

LURIC Study nonprofit LLC, the Welcome Trust Sanger

Genotyping centre

Heidelberg Institute

Illuminus clustering

Genotyping calling algorithm Birdseed v2

algorithm Exclusions:

Individual call rate < 95% < 95%

SNP call rate < 98% < 95%

Minor allele frequency < 1% < 1%

HWE p value < 10 ^"6

Number of participants after 3061 2442

exclusions:

Number of SNPs after exclusions: 686, 195 546,677

Imputation:

Imputation package: MACH MACH

Reference panel: HapMap II CEU (release 22, HapMap II CEU (release build 36) 22, build 36)

Number of SNPs after imputation: 2,543,887 2,543,887

Statistical analyses:

Analysis method Linear regression Linear regression

Age, sex , BMI, first three Age, sex , BMI, significant

Adjustments

principle components PCs

GWAS statistics package: QUICKTEST ProbABEL

Genomic Inflation factor (λ) 1.019 0.999

Table 5: Replication of lead SNPs in 4D and GECOH

Homoarginine (μιηοΙ/L)

4D GECOH

SNPID Proxy of Gene N mean N mean rsl047891 rs7422339 CPS1 554 1.23 1.00E-03

C/C 539 1.13

C/A 141 1.13

A/A

rs37370 rs37369 AGXT2 0.393 0.411 C/C 991 1.17 167 2.52

C/T 232 1.16 43 2.73 T/T 14 1.3 1 1.4

rsl 153858 - GATM 5.36E-11 1.32E-04

C/C 691 1.11 98 2.33

C/T 458 1.24 103 2.74

T/T 86 1.34 10 2.91

* Linear regression of SNP on inverse normal transformed residuals of a regression on age, sex and BMI

DISCUSSION In our GWAS performed in two large cohorts of Caucasian patients we identified sequence variants at three genomic loci as significant predictors of serum homoarginine levels. The strongest signal was observed on chromosome 15 at the GATM locus. Glycine amidinotransferase (GATM), also called L-arginine: glycine amidinotransferase (AGAT; EC 2.1.4.1), plays a central role in energy metabolism by catalyzing the conversion of arginine and glycine to ornithine and guanidinoacetate (GAA) (da Silva et al, 2009, Endocrinology and metabolism 296: E256-261). GAA is subsequently converted to creatine, which serves as a buffer and as an "energy shuttle" for phosphate-bound energy, particularly for ATP. Our results suggest that GATM is significantly involved in the metabolism of homoarginine. When arginine is replaced by homoarginine as a substrate, the GATM reaction yields GAA and lysine instead of ornithine. Lysine in turn can be converted to homoarginine again via homocitrulline and homoargininosuccinate.

Allelic variants in the GATM gene have been implicated with Arginine: glycine amidinotransferase deficiency (OMIM #612718), an autosomal recessive disorder characterized by developmental delay, mental retardation, and severe depletion of creatine/phosphocreatine in the brain (Schulze A., 2003, Mol Cell Biol 244: 143-150) GAMT mR A expression is highest in the kidney which is believed to be the major site of homoarginine synthesis (Cullen et al, 2006, Circulation 114: 116-20). In addition, GAMT is also expressed in other tissues like liver and brain, skeletal muscle, heart, lung and salivary glands. The lead SNP identified in our GWAS, rsl 153858, is located about 600 bp downstream of the GATM gene. It is in high LD (r2 = 1) with a non synonymous coding variant, rsl288775 (P = 8.52 x 10-56 in meta-analysis), which codes the exchange of glutamine against histidine at position 110. However, both Poly-Phen2 (Adzhubei et al., 2010, Nature Methods 4: 1073-1081) and SIFT (Kumar et al, 2009, Nature Protocols 4: 1073-1081) predict this exchange to be benign. eQTL expression analysis identified another SNP in linkage to rsl 153858 (rsl346268, P = 1.3 x 10 ^"55 in meta-analysis) that is associated with GATM mRNA levels in cells taken from artery and whole blood as well as in monocytes. The minor allele of rsl346268 was associated with lower levels of mRNA while the minor allele of rsl 153858 was associated with higher serum homoarginine.

A lookup in the ENCODE data using HaploReg showed that rsl 153858 changes a potential binding motif for the transcription factor PAX-1 while rsl 346268 changes an AIRE 2 motif. Several other potential binding motifs are reported in the same haploblock. In line with the role of GATM in the synthesis of creatine we observed a significant association of rsl 153858 with log transformed creatine (P = 0.031) in LURIC. Furthermore and most remarkably, we observed a significant association of rsl 153858 with mortality in a subset of LURIC patients under 67 years of age (but not in the entire cohort) which is lost after adjusting for either homoarginine or creatine. Patients under 67 years of age carrying the major allele of rsl 153858, which is associated with higher mRNA levels and lower homoarginine levels, were at increased risk to die. We believe that the mortality association is diluted in older patients by a higher prevalence of CAD and risk factors such as diabetes or hypertension so that subtle genetic effects might not be discernible any more in these patients.

The second locus associated with homoarginine levels was CPS1 on chromosome 2. Carbamoyl phosphate synthetase I (CPS1, EC 6.3.4.16) is the rate-limiting enzyme of the hepatic urea cycle catalyzing the production of carbamoyl phosphate from ammonia, bicarbonate, consuming 2 molecules of ATP (Haberle et al, 2011 , Human mutation. 32: 579- 589). Deficiency of this enzyme results in hyperammonemia which may be neonatally lethal or become manifest through environmental factors in adulthood. More than 200 mutations in the CPS1 gene have been reported so far (Haberle et al, 2011, Human mutation. 32: 579- 589). The top SNP in our GWAS, rs7422339, encodes the substitution of asparagine for threonine (T1406N) in a region that is critical for N-acetyl-glutamate binding and results in 20-30% higher enzymatic activity. The same variation has been shown to influence nitric oxide metabolite concentrations and vasodilation, the creatinine serum concentration, homocysteine, and it has also been associated with the risk of pulmonary hypertension in newborns. In line with previous publications, the A allele of rs7422339 was associated with lower arginine levels in LURIC (although not significantly), beyond the association with lower homoarginine. Association with homocysteine was of borderline significance in LURIC with heterozygous individuals having higher levels while there was no association with creatinine (data not shown).

The third locus influencing homoarginine was at 5 l3.2 with the nearest gene being AGXT2. The AGXT2 gene encodes the mitochondrial alanine-glyoxylate aminotransferase, which catalyzes the conversion of glyoxylate to glycine using L-alanine as the amino donor, but it also catalyzes the transamination of asymmetric dimethylarginine (AD MA) to a-keto-5-(N,N- dimethylguanidino) valeric acid (DMGV) (Ogawa et al, 1989, J Biol Chem. 264: 10205- 10209) and overexpression of AGXT2 has been shown to protect from ADM A- induced impairment in NO production in endothelial cells. Furthermore, a recent genome-wide association study of metabolic traits in human urine identified rs37369 in the coding sequence of AGXT2 to be strongly associated with β-aminoisobutyrate (BAIB). This polymorphism encodes a nonsynonymous valine-to-iso leucine (VI 401) substitution and is a likely candidate to be the causative SNP of hyper-P-aminoisobutyric aciduria. Another recent GWAS confirmed AGXT2 to be an important regulator of methylarginines and suggested that this gene represents a novel mechanism for the regulation of blood pressure through the kidney with the T allele at rs37369 causing a modest increase in diastolic blood pressure. The same SNP was identified in our GWAS to be associated with serum homoarginine. We did not observe a significant association with blood pressure although there was a tendency towards higher systolic and diastolic blood pressure in carriers of the T allele (data not shown).

Furthermore, we identified another chromosomal region with a suggestive association to homoarginine serum levels at the MED23/ARG1 locus on chromosome 6. Arginase (ARGl, EC 3.5.3.1) catalyzes the last step of the urea cycle. The isoform encoded by ARGl contributes 98 % of the arginase activity in liver but is also present in red cells. The polymorphisms with suggestive association to homoarginine identified in our study reside in intronic regions of homo sapiens mediator complex subunit 23 gene (MED23) or ARGl gene with both transcripts located on different DNA strands and showing a high degree of overlap. There was no linkage to several pathogenic SNPs in the ARGl gene causing arginase deficiency that have been described previously. In order to attempt to replicate the association of SNPs identified in our GWAS with serum homoarginine we genotyped the lead SNPs in two additional cohorts, the 4D study and the GECOH study. We did not observe any significant association of SNPs at the CPS1, ARG1, AGXT2 or GATM locus with homoarginine in 4D but this might be explained by the high morbidity of these patients all being diabetics with severe kidney disease. Under these circumstances the effect of genetic polymorphisms might be too small to be detectable. The GECOH study includes only 230 patients which limits the power to find significant effects especially for polymorphisms with low minor allele frequency. Nevertheless we were able to replicate the association of rsl 153858 at the GATM locus with homoarginine while there was no significant association of SNPs at the AGXT2 and CPS1 loci.

CONCLUSION

In our GWAS we identified three chromosomal regions significantly associated with serum homoarginine and another region with suggestive association. All of these regions harbor genes that are located in the mitochondrium and have been linked to arginine metabolism.

The strongest association, which was replicated in the GECOH study, was found for GATM which catabolizes arginine/homoarginine whereas CPS1 and ARG1 are involved in the urea cycle and the synthesis of arginine/homoarginine. For rsl 153858 there was an association with mortality in LURIC patients younger than 67 years, compatible with the idea that homoarginine deficiency and mortality are causally linked. This idea, however, needs to be proven in mechanistic studies.

The invention further relates to the following items:

1 A method of prevention or treatment of myocardial failure, wherein homoarginine or a precursor thereof is administered to a subject.

2 The method of claim 1, wherein the myocardial failure relates to ventricular dysfunction, preferably as a consequence of coronary heart disease, arterial hypertension, aortic stenosis, myocarditis, cardiomyopathy, and/or inherited abnormalities of the myocardium, more preferably to a systolic or diastolic dysfunction. The method claims 1 or 2, wherein the myocardial failure is characterized by cardiac remodelling, preferably by ventricular remodelling, more preferably by ventricular hypertrophy. The method of claims 1 to 3, wherein the method results in a change of at least one biological parameter indicating myocardial performance. The method of claim 4, wherein the biological parameter indicating myocardial performance is defined by the concentration of at least one endogenously expressed biomarker, by the steady state level or progression of fibrosis and/or by the patient ^' s ventricular ejection fraction. The method of claims 4 or 5, wherein the method results in a down-regulation of concentration of at least one endogenously expressed biomarker, in a reduction of fibrosis and/or in an increase of the patient's ventricular ejection fraction. The method of claim 6, wherein the down-regulation of concentration of the at least one endogenously expressed biomarker is characterized by a decrease in concentration of at least 30 %, 40 % , 50 %, 60 %, 70 %, 80 %, or 90 %. The method of claims 5 to 7, wherein the at least one endogenously expressed biomarker is selected from the group consisting of natriuretic peptides and beta- myosin heavy chain (β-MHC), preferably from the group consisting of atrial natriuretic factor (ANF), brain natriuretic peptide (BNP) and N-terminal pro-B-type natriuretic peptide (NT-proBNP). The method of claim 6, wherein the patient's ventricular ejection fraction is increased to at least 35 % of volume per volume, preferably to at least 40 % of volume per volume, more preferably to 45-54 % of volume per volume, and most preferably to at least 55 % of volume per volume. The method of any of the preceding claims, wherein the homoarginine or the metabolic precursor thereof is administered by means of oral or parenteral administration, preferably by intramuscular, subcutaneous or intravenous administration.

The method of any of the preceding claims, wherein the homoarginine or the metabolic precursor thereof is administered for a period of several weeks, months or years.

The method of any of the preceding claims, wherein the homoarginine or the metabolic precursor thereof is administered in conjunction with at least one additional treatment of myocardial failure, wherein the additional treatment of myocardial function is preferably in form of one or more inhibitor(s) of the angiotensin converting enzyme (ACE), one or more inhibitor(s) of the neutral endopeptidase (NEP), one or more inhibitors of matrix metalloproteases (MMPs), one or more beta blocker(s), one or more ATi receptor blocker(s), one or more diuretic agent(s) and/or one ore more aldosterone antagonist(s).

The method of any of the preceding claims, wherein the metabolic precursor is a medicine, an endogenous compound or an enzyme involved in the homoarginine metabolic pathway, preferably involved in the up-regulation of endogenous serum or tissue levels of homoarginine.

The method of claim 13, wherein the metabolic precursor is a gene product or a regulator of a gene product associated with the CPSl locus on chromosome 2, the AGXT2 locus on chromosome 5, the GATM locus on chromosome 15 and/or the MED23/ARG1 locus on chromosome 6.

The method of claims 13 or 14, wherein the metabolic precursor is selected from the group of enzymes consisting of L-arginine: glycine amidinotransferase (AGAT), carbamoyl phosphatase synthetase I (CPSl), mitochondrial alanine-glyoxlate aminotransferase and arginase (ARG1).