WO2006125632A2

WO2006125632A2 - Agonistic antibodies that bind to the tweak receptor fn14 and thereby modulate adiposity-associated phenotypes as well as their use in therapy

Info

Publication number: WO2006125632A2
Application number: PCT/EP2006/004974
Authority: WO
Inventors: Dieter Link
Original assignee: Rechtsanwalt Dr. Martin Prager Als Insolvenzverwalter Über Das Vermögen Der Xantos Biomedicine Ag, Pluta Rechtsanwalts Gmbh
Priority date: 2005-05-24
Filing date: 2006-05-24
Publication date: 2006-11-30
Also published as: WO2006125632A3

Abstract

The present invention relates to agonists of the receptor Fn14 as well as their use for activating the signal pathway of the receptor Fn14 for interfering and/or modulating the differentiation of adipocytes and for use in adiposity/obesity and/or metabolic syndrome. In particular, the invention relates to agonists, preferably antibodies that inhibit the differentiation of pre-adipocytes to mature adipocytes and/or the storage of lipid molecules in differentiating or mature adipocytes by receptor-mediated signal transduction.

Description

Agonistic antibodies that bind to the Tweak receptor Fn14 and thereby modulate adiposity-associated phenotypes as well as their use in therapy

Technical description of the invention

The invention relates to pharmaceutical compositions as well as their use for activating the signal pathway of the receptor Fn 14 (synonyms are also TWEAK receptor, TWEAK- R or TNFRSF12A) for interfering and/or modulating the differentiation of adipocytes and for use in adiposity/obesity and/or metabolic syndrome. In particular, the invention relates to antibodies that inhibit the differentiation of pre-adipocytes to mature adipocytes and/or the storage of lipid molecules into differentiating or mature adipocytes by receptor- mediated signal transduction. Also, the invention relates to defined epitopes of the receptor Fn14 that induce signal transduction of the receptor upon specific antibody binding. In particular, the invention relates to antibodies that are directed against defined activating epitopes on the receptor Fn14 and that act as Fn14- activating agents. Preferably, this activation can take place alone or, also, in combination with other Fn 14- activating agents. Furthermore, the invention relates to pharmaceutical compositions and their use for activating the signal pathway of the receptor Fn14, for interfering and/or modulating the differentiation of adipocytes, in particular, for the prevention and/or treatment of adiposity/obesity.

Background of the invention

Presently, adiposity (obesity) and type Il diabetes are two of the most important metabolic diseases that pose an increasingly serious problem, in particular, in the western world. Furthermore, tendencies for a pandemia have also been observed for type Il diabetes.

Adiposity is a condition that is characterized by the overly accumulation of fat tissue in the body. The adipocytes, i.e. those cells that form the fat tissue of the body, are essential for the formation of fat tissue, and, therefore, for developing an obese phenotype. According to recent understanding adiposity is a chronic health dysfunction that is based on a polygenetic predisposition and also caused by environmental factors and is accompanied by a high accompanying and follow-up morbidity. In this context adiposity is associated with a significant risk, in particular, the development of type Il diabetes. Therefore, a long term treatment and care concept for adiposity is indispensable.

The presence of obesity or adiposity can be classified as overweight or (clinical) adiposity in its closer context. It can be defined by the so-called "body mass index" (BMI). The BMI describes the ratio of the body's weight and the body^'s size. The BMI is calculated according to the following formula: BMI = body weight (kg) divided by body size to the square (m²).

The classification of the BMI is established by the WHO (World Health Organization) in dependency of the mortality rate. The normal BMI (18.5 - 24.4 kg/ m²) lies in the range with the lowest relative mortality risk. The BMI values 25 and 30 kg/ m² are the corresponding risk- oriented limits for adults for establishing overweight or (clinical) adiposity.

According to this nomenclature adiposity is divided into two different levels: adiposity level I (BMI = 30 - 34.9), adiposity level Il (BMI = 35 - 39.9), adiposity level III (BMI of more than 40) as well as morbid adiposity/super adiposity (BMI of more than 50). Overweight and adiposity go along with follow up or accompanying diseases (comorbidity). They are a substantial risk factor for the development of type Il diabetes. Frequently, they also result in social isolation and discrimination. They can assist or cause psychic disturbances and, thereby, reduce life expectancy and life quality. At a BMI of more than 30 kg/ m² the mortality (death rate) increases by 1/3, at 40 kg/ m² by a factor 3, at more than 40 kg/ m² by a factor 3 - 20 according the weight. For example, the risk of cancer is also increased (gynaecological carcinomas (endometrial carcinoma - 4 times; mamma- and cervix carcinoma - 2 times), prostate-, gall bladder-, colon- carcinoma)). However, the BMI as an assessment standard for adiposity does not consider the composition of a body's tissue. An increased muscle mass, water content or bone mass can effect the assessment and may be considered optionally. Also, the body's composition of the compartments - fat and fat-free mass (initially, body liquid is not considered here) may be included. According to this two-compartment model the normal range is given for a fat mass < 20% and a fat-free body mass of > 80%. An increased fat content is described as adiposity. Furthermore, an increase in extracellular mass or extracellular water relative to the fat-free body mass indicates adiposity. Studies such as the bioelectric impedance analysis (BIA) or the infrared spectroscopy (NIRI) are suitable methods and methods used by practitioners for determining the body fat status. Also, the fat distribution is important for the risk assessment. An increased abdominal fat accumulation (central or android distribution type) poses a significantly higher risk than the peripheral or gynoid distribution type. Therefore, the "waist-to-hip" ratio (the ratio of the circumference of the waist (cm) and the circumference of the hip (cm)) can also be included in the assessment. In women (> 0.8) or men (> 0.95) increased values are associated with an increased mortality.

Because adiposity must be regarded as an essential risk factor for the development of type Il diabetes, the diagnosis and therapy of adiposity already in early childhood is the most important measure to avoid, for example, the development of paediatric Il diabetes. In childhood and adolescence the BMI can be used in analogy to adults for assessing overweight and adiposity. Because the BMI in childhood and adolescence is influenced in accordance to the physiological changes of the body's fat mass percentage by distinct age- and gender- specific characteristics, one should take into account the age and gender for ones assessment. Because of the low incidence of adiposity-dependent diseases in childhood and adolescence and due to the lack of sufficient long term studies of the health risk of adiposity in childhood and adolescence, there are no established limits for the health risk of the body fat mass in this age range in contrast to the situation for adults. However, for the definition of overweight or adiposity in childhood and adolescence the BMI percentile can be used (obtained by extrapolation) that converges at the age of 18 to a BMI of 25 kg/ m² (overweight) or 30 kg/ m² (adiposity). For example, the 90th or the 97th age- and gender-specific percentile can be used as limit for defining overweight or adiposity. The BMI values 25 and 30 kg/ m² are the corresponding risk- related limits for adults. For a BMI of 25 kg/m² the relative risk for men for acquiring type Il diabetes is 2.2, for women, it is even 5.5.

A treatment of obesity and adiposity is targeted either at slowing down a further increase in weight, but it is preferred to prevent it completely, or, in addition, it is preferred to reduce the body weight. The same is particularly true for the amount of fat tissue of the individual to be treated. The opposite is true for the treatment of a cachexia or lipodystrophy. A treatment of obesity or a tendency for adiposity is preferably indicated for individuals with a BMI of more than 25 in view of the already increased risk of co- morbidities due to overweight. This is particularly true for the case where co-morbidities have already manifested themselves: in particular, when the development of insulin resistance or type Il diabetes is indicated. Preferably, individuals having a BMI of more than 30 without or with co-morbidities that are characterized as having clinical adiposity are treated. This is especially preferred for individuals with a high level of adiposity.

The development of adiposity can be prevented in individuals with a normal BMI or a BMI that qualifies them as overweight, in particular in the case, when there are reasons for assuming that these individuals will tend to develop an adiposity later (increase of the BMI to more than 30 kg/ m²) with an increased probability, in particular, an adiposity associated with co-morbidities.

For example, these reasons may based on a genetic pre-disposition, a sustained dysfunction of the lipid- or lipoprotein metabolism (dyslipidaemia), a life style that can hardly be changed and that assists the development of adiposity, and/or an adiposity due to other diseases. For this purpose a diagnostic system can be helpful that allows for detecting, for example, genetic pre-dispositions directly or indirectly or modified amounts or activities of the receptor Fn14 or its natural ligand or parameters depending thereon, for example, in blood or in fat tissue.

The methods of treatment for adiposity known in the state of the art are essentially limited to a rigorous diet in combination with increased exercise. Presently, a known therapy is limited to the noradrenaline-serotonine-receptor-uptake-inhibitor "Sibutramin" (BASF, Ludwigshafen) or the lipase inhibitor "Xenical" (Roche, Basel, Switzerland). However, the application range of these medicaments is limited due to the occurrence of side effects/complications. For example, Sibutramin cannot be employed with high blood pressure - a typical complication for adiposity. Furthermore, the disease can only be treated symptomatically.

One of the most frequent co-morbidities of adiposity is diabetes, in particular type Il diabetes.

Diabetes mellitus (generally known as "sugar disease") is a chronic metabolic disease that is characterized by an increased level of sugar in the blood. Different causes of the disease and also different disease characteristics require the discrimination of two types, type I and type Il diabetes. Type I diabetes (formerly: juvenile diabetes) most often begins in adolescence and develops because of an immunological destruction of islet cells of the pancreas. These islet cells produce the hormone insulin that is responsible for the utilization of glucose from food. From the destruction of the islet cells an absolute insulin deficiency results. The glucose in the food cannot be metabolized any more and the blood level of sugar increases. The treatment of type I diabetes is done by the administration of insulin.

The type Il diabetes (formerly: adult or old age diabetes) regularly develops at a later age. It is characterized in that body cells, where the insulin is supposed to function, do not react sufficiently to insulin. Amongst others, this may be ascribed to a resistance of adipocytes and skeletal muscle cells for insulin. Interestingly, mostly the older more mature adipocytes are resistant, whereas younger adipocytes regularly do not have an insulin resistance.

Such an insulin resistance is considered to be the result of prolonged increased blood sugar and insulin levels as they are, for example, observed in overweight people. The therapy of type Il diabetes is done stepwise: at first a diet for lowering the blood sugar level in general is attempted. If these dietary measures are not sufficient for treatment, subsequently medicaments lowering blood sugar are administered, and in advanced stages insulin is administered, too.

However, the administration of insulin to patients with type Il diabetes can result in hyperinsulinaemia (increased insulin level in the plasma, serum) due to the insulin resistance in the target tissues. Then, a therapy with insulin is no longer possible (Curr. Diab. Rep., October 3 (5), page 378 - 85, (2003)).

In the context of a previously conducted project of the applicant for identifying protein factors having relevance for obesity, adiposity or type Il diabetes or cachexia, the applicant discovered the protein TNFSF12 (synonym: TWEAK), preferably in a trimeric configuration and in its soluble form as a new modulator of the differentiation of adipocytes.

The modulation of the differentiation of adipocytes by an activation of the TNFSF12 signal transduction system is pursued as a therapeutic strategy independently of whether the TNFSF12/Fn14 signal transduction system is involved in the disposition, aetiology or associated symptoms of the disease. Aberrant TNFSF12 expression or aberrant Fn14 receptor activity and/or signal transduction are no requirements or obstacles for performing the therapeutic method described herein.

Furthermore, it was surprisingly found that this protein TNFSF12 can be employed for the treatment of adiposity. It was demonstrated that TNFSF12 is capable of reversibly inhibiting the differentiation or the function of murine and also primary human adipocytes and to obtain a reduction of fat deposits in vivo. This allows for the first time the treatment of adiposity on the level of fat cell differentiation.

This protein TNFSF12 unfolds a therapeutic effect preferably by directly affecting certain fat tissues within an individual or also indirectly by other fat tissue compartments or also by the interaction with other tissues that are involved in the regulation of the energy homoeostasis.

The explanation for the found phenotype of the protein TNFSF12 results from the property of TNFSF12 as a receptor ligand that can elicit signal transduction upon binding:

The protein TNFSF12 is a cell surface-associated type Il membrane protein of 249 amino acids in length that was originally described as a member of the tumor necrosis factor (TNF) superfamily (Review in: Wiley et al., Cytokine & Growth Factor Reviews, 14, page: 241 - 249, 2003). A biologically active part is cleaved into the extracellular environment as a soluble protein with a length of about 156 amino acids and this can in turn activate the corresponding receptor. This receptor activation itself can initiate changes in the cell, for example, assist cell division, enhance cellular migration and lead to the secretion of cytokines that are discussed in the context of inflammatory or cytotoxic processes.

Originally the cell surface molecule DR3 was described as a receptor for TNFSF12 (Marsters et al., Cunr. Biol., 8, page: 525 - 528, 1998). Later it was demonstrated that DR3 is not the physiological receptor for TNFSF12 (Schneider et al., Europ. J. Immun., 29, page: 1785 - 1792, 1999; and Kaptein et al., FEBS Lett, 485, page:135 - 141 , 2000). Instead, there is only one receptor known for TNFSF12 in the present state of the art that is bound by TNFSF12 with physiological activity (Wiley et al., Immunity, 15(5), page: 837 - 846, 2001). This receptor, a member of the TNF superfamily, is described as ,,TWEAK- R" or ..fibroblast growth factor inducible 14" (Fn 14) (Meighan-Mantha et al., J. Biol. Chem., 274, page: 33166 - 33176, 1999; Wiley et al., Cytokine & Growth Factor Reviews, 14, page: 241 - 249, 2003). This receptor is the smallest member of the group of the TNF-receptors. At present no other ligands are known that bind Fn14 (Wiley et al., Cytokine & Growth Factor Reviews, 14, page: 241 - 249, 2003). However, there are also other possible receptors being discussed for TNFSF12 (Polek et al., J. Biol. Chem., 278, page 32317-23, 2003). It was also described that TNFSF12 does not bind to other known TNF-receptors (Wiley et al., Cytokine & Growth Factor Reviews, 14, page: 241 - 249, 2003).

The primary sequence of Fn14 comprises 129 amino acids, wherein the amino terminus of the receptor is processed so that the mature protein consists for 102 amino acids only.

Up to know various cellular activities have been described for TNFSF12 in the literature, e.g. the induction of proliferation and migration of endothelial cells that play a central role in angiogenesis.

Also, the induction of the expression of inflammatory cytokines such as, for example, IL- 6, IL-8 and RANTES has also described inter alia in fibroblasts (Chicheportiche, Arthritis Res, 4, page: 126 - 133, 2002).

Under specific conditions (i.e. co-stimulation with IFN-gamma) the stimulation of cell death has been described in some sensitive tumor cells (Chicheportiche et al., J. Biol. Chem. 272, page: 32401 - 32410, 1997; Nakayama et al., J. Immunol., 170(1), page: 341 - 348, 2003). The cytotoxic activity of monocytes in vitro partially depends on their TWEAK expression and their paracrine effect on said tumor cells (Nakayama et al., J. Exp. Med., 192, page: 1373 - 1380, 2000).

The above observations led to the discussion of TNFSF12 in the context of angiogenetic and also inflammatory disease processes.

Under experimental conditions, i.e. the entry of the blood-brain barrier by suitable methods, an in vivo involvement of transgenic soluble TNFSF12 was demonstrated in inflammatory processes of the central nervous system. It is assumed that this is based on the stimulation of the proliferation of astrocytes (Desplat-Jego et al., J. Neuroimmunol., 133, page: 116 - 123, 2002). TWEAK naturally passes the blood-brain barrier when it is expressed there in infiltrating leucocytes. On the other side, a negative regulation of TWEAK-mRNA was observed in vivo in lipopolysaccharide-induced inflammatory, acute and chronic reaction and animal models for autoimmune diseases (Chicheportiche et al., Biochem. Biophys. Res. Comm., 279, page: 162-165, 2000).

Based on data from cell cultures and transgenic mice TWEAK-mediated activities were also described as possible cause for disease processes in the patent application WO

03/086311.

In this case, the transgenic animals either produced a first a cell-bound, full-length

TWEAK protein or a soluble TWEAK protein fragment.

The transgenic expression of full-length TWEAK protein led to thrombosis in the heart chambers and to their dilatation, to hyperplasia of cells of the bile duct and oval cells in the liver as well as to an enhanced vacuole formation and cell death of liver cells. In the kidneys cysts were formed in the glomeruli of the kidneys. In the lung granulomatoses and the penetration of lymphocytes were recognized in the context of an inflammation.

High doses of soluble TWEAK even led to the death of the animals several months after birth. Even low doses led to, for example, changes of the cellular structure of the heart (dilatative cardiomyopathy) as well as a tubular hyperplasia and glomerulopathy in the kidneys.

Consequently, WO 03/086311 pursued antagonistic therapeutic strategies that are aimed at the inhibition of the TWEAK-dependent signal transduction. As examples for this a genetic inflammatory disease in kidneys with signs of Alport syndrome that is inter alia characterized by the thinning and final destruction of the membranes of the glomeruli by fibrotic substitute tissue is provided. Furthermore, an increased Fn 14 expression was shown in WO 03/086311 and a treatment of animals with said kidney changes with TWEAK antagonists (antagonistic antibodies) led to a reduction of the pathological fibrotic changes.

According to the state of the art, in particular, WO 03/086311 , an increased expression of TWEAK protein in vivo seems to correlate with pathological diagnosis in the heart, the liver, the kidney and the lung, which in turn requires a blockage of TWEAK-dependent activities in the therapeutic application. In WO 03/086311 it was also described that in vitro the exposition of murine pre- adipocytes (3T3-L1 cells) to TNFSF12 can interfere negatively with their ability to incorporate lipids intracellular^. The mechanism is unclear because it was not shown in vivo whether Fn 14 is the only receptor for TNFSF12 and whether an expression of Fn 14 was present in 3T3-L1 cells or also in fat tissue was given. This is particularly relevant because Polek et al. (J. Biol. Chem., 278, page 32317 - 23, 2003) still discuss the possibility of several receptors for TWEAK.

It is known that the differentiation of human adipocytes is hard to demonstrate in murine cell culture systems. For example, murine pre-adipocytes typically display a so-called clonal expansion. This is an increase in the cell count due to a short lived re-entry into the cell cycle after induction for differentiation in previously post-mitotic cells. This phenomenon is not shown by differentiating human adipocytes. Factors that inhibit a clonal expansion can block the differentiation in murine adipocytes. It is therefore unclear whether factors inhibiting differentiation that were identified for murine adipocyte cell cultures can also achieve a blockage of differentiation in primary human adipocytes.

Furthermore, for example, expositions of the protein TNF-alpha in differentiating murine adipocytes cell cultures lead to cell death in a dosage-dependent manner whereas primary human adipocytes do not display such a cell death for the same exposition even though they are inhibited in their differentiation.

In particular, functional data obtained from murine cell cultures can not be transferred to human cell cultures.

Moreover, for a therapeutic utility it is mainly the demonstration of the desired therapeutic effect, i.e. the reduction in weight, that is relevant in vivo. This is so because it is known that many phenomena that are observed in cell cultures do not take place in more complex living organisms or are irrelevant.

In the context of the present invention a dose-dependent and reversible inhibition of the differentiation of adipocytes or the inhibition of the lipid storage in human adipocytes was shown as a result of the TNFSF12-mediated Fn14 signal transduction next to the above described known effects of the TNFSF12/Fn14 ligand receptor system. This inhibition of the differentiation of adipocytes was surprisingly shown by the applicant in fresh human fat cells (or fat precursor cells). Moreover, by the work of the applicant it was for the first time possible to demonstrate that the activation of the TNFSF12/Fn14 ligand receptor system in cell culture as well as by recombinantly providing TNFSF 12 in murine animal models in vivo corresponds with a reduction of the weight and the fat content of animals.

These in vivo effects on animal weight and fat were observed without associated massive inflammatory processes or toxicities in these animal experiments. This was very surprising because according to WO 03/086311 the occurrence of inflammatory processes or toxicities had been expected.

Furthermore, the applicant demonstrated for the first time that agonistic antibodies that bind to the receptor Fn 14 and activate it can elicit a direct effect on the differentiation of fat cells and inhibit fat cells differentiation.

Because of the effect of the TNFSF12/Fn14 ligand receptor system on the differentiation of fat cells and the weight and fat content of animals as demonstrated by the applicant, it is advantageous to utilize these effects on phenotypes induced by the Fn14-dependent signal transduction for therapeutic applications. This allows for a therapeutic use of the Fn14 activation for the prevention and/or Treatment of adiposity and/or adiposity- associated diseases.

In analogy to the TNF-R and other TNF-like receptors it is assumed that the activation of the Fn14 signal pathway preferably occurs when many receptors in close proximity are expressed on the cell surface. This process is known as receptor cluster formation. The TNF ligands are multivalent complexes that can bind to more than one receptor simultaneously and aggregate in this way. Receptor cluster formation as a means for receptor activation has been elaborately described in other systems, in particular, for tyrosine kinase (Ullrich and Schlessinger, Cell, 61, page 203 - 212 (1990); Kolanus et al., Cell, 74, page 171 - 183 (1993)). Accordingly, the administration of Fn14 ligands and/or Fn14-activating agents that also comprise, for example, antibodies, antibody derivatives, binding proteins or binding molecules that can induce the cluster formation and/or a downstream signalling on the surface of target cells can be useful for directly stimulating the Fn14 signal pathway in these cells.

The signal pathway for Fn14 is known amongst others for activating reaction pathways that can potentially modulate immunological or inflammatory of cytotoxic processes. These properties of the Fn14 activation as well as the ability to inhibit these effects by antagonistic antibodies are the basis for the use of Fn14-blocking agents, including specific antagonistic Fn 14 antibodies for the treatment of diseases that are associated with immunological or inflammatory or cytotoxic processes as described by Biogen in WO 03/086311. Also, the use of agonistic antibodies is described in WO03/086311 that effect the induction of the Fn14 signal transduction. Whether such an activation is detrimental for normal tissue due to induced immunological or inflammatory or cytotoxic processes still remains unclear in the state of the art.

Ligands such as TN FSF 12 that activate the receptor Fn 14 can directly inhibit the differentiation of fat cells by ligand/receptor interactions and the subsequent receptor activation and signal transduction. Because receptor activation can also be achieved by agonistic antibodies, for example, those that recognize the corresponding receptor Fn 14 and induce signal transduction upon recognition, agonistic antibodies can also be used instead of natural ligands that specifically activate and target the above described receptor Fn 14.

The relationship of the functionality of the Fn14 signal pathway and the differentiation of adipocytes found by the experiments of the applicant can be employed for generating new therapeutic substances for treating adiposity/obesity and adiposity-associated diseases.

As already described above the treatment of adiposity/obesity and adiposity-associated diseases poses an enormous clinical problem. Previous therapeutic approaches have been and are still symptomatic, either by changing the diet, and in the case of adiposity- associated diabetes, for example, by the administration of PPAR-gamma agonists. Ligands for PPAR-gamma activate the PPAR-gamma transcription factor that can effect a differentiation of the adipocytes or an activation of the fat metabolism in mature adipocytes. Moreover, the disadvantages of these symptomatic treatments (inter alia diabetes) are the occurrence of, for example, cardiovascular complications, diabetic nephropathies, neuropathies or retinophathies that need to be treated therapeutically, too. All previously known therapeutic approaches also have the momentous disadvantage that they are not capable of inhibiting the differentiation or the function of adipocytes. Therefore, one object of the present invention is to provide substances for the treatment and/or prevention of adiposity/obesity and also for the prevention of adiposity-associated diabetes (type II). In particular, the substances comprise agonistic-anti-Fn14 antibodies, antibody derivatives, binding proteins or binding molecules. Because it may be deduced from the previous state of the art that the activation of Fn 14 signal transduction can cause inflammatory and toxic effects in organisms, a specific embodiment of the present invention relates to substances (in particular, agonistic anti-Fn14 antibodies, antibody derivatives, binding proteins or binding molecules) for the treatment and/or prevention of adiposity/obesity and thereby also for the prevention of adiposity/obesity-associated diabetes (type II) that are not associated with any controversial toxicities in effective doses for a relevant therapy.

However, the targeted preparation of said agonistic anti-Fn14 antibodies, antibody derivatives, binding proteins or binding molecules that have as a common property the ability to initiate signal transduction by binding to the Fn 14 is problematic. So it is known that in many cases the binding of antibodies, antibody derivatives, binding proteins or binding molecules alone is not sufficient to activate the receptor (Mueller et all, 2005, J. Neuroimmunol., 159:55 - 65). Moreover, the binding of said molecules must also initiate the molecular event that takes place when the natural ligand binds to the receptor. Said events can be, for example, the initiation of changes in the receptor structures and/or multimerisations of receptors. In these cases it is frequently observed that only the specific binding to a few defined locations of the receptor can cause an optimal receptor activation. However, when these positions on the receptor that may lead to activation upon binding are not known, the generation of activating agents is very difficult, in particular, when no reliable biological test system is available.

Therefore, a further object of the present application is to determine the necessary binding sites or epitopes for activation on the receptor and methods and processes based on this previously unavailable knowledge that allow for an efficient and targeted generation of activating antibodies. Based on the gained knowledge about the binding sites or epitopes on the receptor that are necessary for activation it is then a further object of the present application to identify substances, in particular, agonistic anti-Fn14 antibodies, antibody derivatives, binding proteins or binding molecules, that bind to the binding sites or epitopes on the receptor and by doing so activate the receptor signal transduction and to provide means for the treatment and/or prevention of adiposity/obesity and by doing so to also prevent adiposity/obesity-associated diabetes (type II).

Summary of the invention

According to the invention the above problem of treating and/or preventing adiposity/obesity is solved by generating substances that effectively inhibit the differentiation or function of adipocytes. These substances are characterized according to the invention by their ability to induce the Fn14-mediated signal transduction. Examples of agents that induce Fn14-mediated signal transduction are the natural ligands or homologous ligands (or non-species ligands) of Fn14 as well as antibodies that recognize Fn14 at defined positions (epitopes) and activate at least on of these epitopes by binding. The use of antibodies that recognize and activate Fn14 has the advantage that these are easy to prepare and that they can be used in patients in a more controlled manner due to their good pharmacological behaviour. The invention also comprises epitopes on the receptor Fn 14 that also initiate the molecular event of binding that takes place when the natural ligands bind to the receptor and, therefore, leads to receptor activation and subsequent modulation of the differentiation of adipocytes.

The above problem of providing agents for the treatment and/or prevention of adiposity/obesity is solved by the agonists according to the invention, such as antibodies and derivatives thereof, that recognize and activate the receptor Fn 14. These agonistic antibodies and compositions are produced in a target manner by techniques and methods that are based on the knowledge about the necessary positions (epitopes) of Fn 14 for receptor activation. These agonistic antibodies and compositions containing these are employed for the treatment of adiposity/obesity and further metabolic diseases that are associated with increased or aberrant amount of fat cells by stimulating the large Fn 14 signal pathway.

A preferred embodiment of the invention describes the use of at least one antibody directed against Fn14 (anti-Fn14 ab) for modulating the differentiation of fat cells; the use of a monoclonal or recombinant antibody (anti-Fn14 mab or anti-Fn14 recab) being preferred. A further embodiment of the invention describes antibodies or antibody derivatives that are directed against specific epitopes of the receptor that cause the activation of the receptor by antibody binding. A further embodiment of the invention describes binding molecules (for example, affilines, anticalines, aptameres) that are directed against those epitopes of the receptor that cause activation of the receptor upon antibody binding.

A further embodiment of the invention describes specific epitopes of the receptor that cause activation of the receptor upon binding by antibodies, antibodies derivates, binding proteins or binding molecules.

Furthermore, the invention provides a new screening method for selecting Fn 14- activating agents, wherein these agents comprise amongst others anti-Fn14 antibodies, derivates thereof, binding proteins and binding molecules and allow for the activation of the Fn14 by binding to at least one activating epitope of Fn14.

The approach used for said purpose employs, for example, antibodies against epitopes of the receptor that cause the activation of the receptor upon antibody binding and initiate the receptor signal pathway. This can either be measured directly by measuring the signal transduction events or indirectly, for example, by analysing the modulation of cellular phenotypes by, for example, inhibiting the differentiation of fat cells.

The method used for, for example, testing the agonistic properties and utilities of putative agonistic antibodies for Fn 14 for modulating the differentiation of fat cells is described by reference to examples and preferably comprises the following steps:

1) Undifferentiated human pre-adipocytes are seeded in cell culture dishes and cultivated until post confluence.

2) These cells are incubated together with specific differentiation inducers which normally lead to a detectable differentiation into adipocytes with stored fat.

3) Simultaneously, the cells activated for differentiation in this way are incubated with the agents that are to be tested such as anti-Fn14 antibodies, antibody preparations or formulations.

4) The measurement of the differentiation of cells is done, for example, after > 8 days with a test that is suited for the detection of mature fat cells (for example, with a test for detecting lipids, e.g. Nile Red Assay, that is described in Example 1). 5) Agents that modulate the differentiation of fat cells are recognized because in these cells a deviation of the strength of the test signal occurs that is more than a standard deviation (1 x), preferably more than a 2-fold change of the signal, that is provided by the negative control antibodies.

6) The strength of the differentiation-modulating activities of the tested agents that had been tested positive in the previous step is analyzed by serial dilution and subsequent determinations of dose-dependent changes in phenotype. The specificities of the antibodies for activating epitopes are then confirmed by direct binding experiments, for example, an ELISA with immobilized peptides that represent activating epitopes.

This method that is only listed as an example can be modulated and adapted for the purpose of identifying further Fn 14 activating antibodies that recognize epitopes that cause the activation of the receptor upon binding.

For example, for the described method for generating differentiation-modulated agents also other cells or cell lines that express the Fn 14 naturally, in recombinant form as chimeric receptor or corresponding epitope-containing molecules can be used instead of differentiating pre-adipocytes. By using said cells instead of pre-adipocytes the Fn14- activating effect of agents can be determined, for example, by detecting specific signal transduction events in the cell.

The anti-Fn14 antibody (or a combination of antibodies) that inhibits or modulates the differentiation of adipocytes is an Fn14-activating agent in this experiment. These agents were used for preparing a pharmaceutical composition for treating adiposity/obesity as well as further metabolic diseases that are associated with increased or aberrant amounts of fat cells.

As described above, the invention also relates to the use of functional variants of the above listed antibodies for preparing a pharmaceutical composition for the treatment of adiposity/obesity and further metabolic diseases that are associated with an increased or aberrant amount of fat cells.

Furthermore, the invention also relates to the use of substances and preparations that have activating epitopes of Fn14 for the direct immunization of patients for preventing and/or treating adiposity/obesity and/or further adiposity/obesity- associated diseases that are associated with an increased or aberrant amount of fat cells.

Detailed description of the invention

The following description shall have the purpose of describing preferred embodiments of the invention in an exemplary manner.

Definitions

The term "adipocyte differentiation" relates to the ability of a correspondingly determined precursor cell (pre-adipocyte) to terminally differentiate upon a suitable stimulus. By doing so a specific gene expression program is initiated that phenotypically leads amongst others to an intracellular accumulation of lipid. Such a differentiated adipocyte is also denoted a mature adipocyte.

The term "metabolic syndrome" relates the variable combination of clinical parameters and symptoms that are associated inter alia with adiposity/obesity. Amongst others this frequently includes adiposity and/or high blood pressure and/or high cholesterol and/or insulin resistance.

The term "epitope" (or antibody binding site) is defined as the spatial structure of an antigen that is recognized and bound by an antibody.

A monoclonal antibody (mab) recognizes a single epitope because all antibodies originate from one producing B-cell (= an antibody producing cell) and, therefore, have the same structure. Polyclonal antibodies (pab) are mixtures of different antibodies synthesized by different B-cell clones that may recognize different epitopes because they have different antigen binding sites. An epitope typically comprises several amino acids (for example, 5 - 7 amino acids) that can be located in linear sequence in the primary sequence of the antigen (= continuous epitope or linear epitope) or that can consist of different amino acids of the antigen that are not directly bound (= discontinuous epitope, non-linear epitope).

The term "Fc-domain" of an antibody relates to a part of a molecule that comprises the CH2-, CH3- and hinge regions but that lacks the antigen binding sites. The term "Fn14 activating agent" relates to any agent that is capable of imitating ligand binding to Fn14, cluster formation of Fn14 on the cell surface or that allows or enhances the Fn14 signal pathway or that can influence how the cell interprets the Fn 14 signal on its inside. Examples of Fn14 activating agents are TNFSF12, soluble anti-Fn14 antibodies, cross-linked anti-Fn14 antibodies as well as multivalent anti-Fn14 antibodies.

The term "Fn 14 signal pathway" relates to all molecular reactions in the context of the ligand-, antibody- or binding partner-mediated activation of Fn 14 and the resulting molecular reactions.

The term "anti-Fn14 antibody" ("anti-Fn14 ab") relates to all antibodies that recognize at least one epitope of the Fn14 receptor and bind to it.

The term "anti-Fn14 monoclonal antibody" ("anti-Fn14 mab") relates to all monoclonal antibodies that recognize one single epitope on the Fn 14 receptor and bind to it.

The term "anti-Fn14 antibodies, poly- or monoclonal, cross-linking agent" relates to all agents that can bind to anti-Fn14 antibodies in solution either covalently or non- covalently, so that the antibodies bind to the surface of potential target cells and induce receptor signal transduction. This also includes antibodies that can aggregate the receptors, so that the antibody can bind to the surface of potential target cells and multiply receptor cluster formation there.

The term "functional variant of an antibody" according to the invention relates to an antibody and/or fragment that essentially mediates the biological function or functions of the antibody. In the case of the present antibodies this can be the ability to inhibit the lipid storage in suitable cells. The scope of this term also comprises various derivatives of antibodies, in particular, recombinant, chimeric, humanised or otherwise modified antibodies that induce Fn14 signal transduction.

The term "functional variant" according to the invention relates to all "non-antibody" proteins having similar binding properties such as, for example, anticalines, affilines, single domain antibodies and other specific binding proteins. The term "functional variant" of a polypeptide or a nucleic acid preferably relates to polypeptides or nucleic acids having a sequence similarity, in particular, a sequence identity, of at least 25 %, preferably about 40 %, particularly preferred about 60 %, more preferably about 70 %, most preferred about 80 %, in particular about 90 % and very most preferred 98 % to the polypeptide. Said variants are, for example, homologous polypeptides that originate from other organisms. Other examples of variants are polypeptides or fragments that are coded by the different alleles of a gene. Functional variants preferably also include naturally occurring mutations, in particular, mutations that quantitatively change the activity of the peptides that are coded by these sequences. Furthermore, these variants may preferably result from differential splicing of the coded gene. In a particularly preferred embodiment the term "functional variant" includes derivates having single nucleotide polymorphisms (SNP).

The term "sequence identity" relates to the degree of identity (% identity) of two sequences that, in the case polypeptides, may be determined, for example, by BLASTP 2.2.10, and in the case of nucleic acids, for example, by BLASTN 2.2.10, wherein the low complexity filter is turned off and, in the case of BLASTP₁ the matrix is BLOSUM 62 (Altschul et al. 1997, Nucleic Acids Res., 25:3389 - 3402).

The term "sequence similarity" or "sequence homology" relates to the similarity (% positives) of two nucleotide or polypeptide sequences that is determined, for example, by BLASTN 2.2.10 or by BLASTP 2.2.10, wherein the filter is turned off, and in the case of BLASTP the matrix is BLOSUM 62 (Altschul et al. 1997, Nucleic Acids Res., 25:3389 - 3402).

The term "binding protein" or "binding peptide" according to the invention relates to a class of proteins, peptides or fragments that bind to or inhibit the corresponding molecule including without limitation polyclonal or monoclonal antibodies, antibody fragments and protein scaffolds that are directed against these proteins, peptides or fragments.

The term "aptamer" describes nucleic acids that bind to a polypeptide with high affinity. Aptamers can be isolated from a large pool of different single-stranded RNA molecules by selection methods such as SELEX (see, e.g., Jayasena, Clin. Chem., 45, pp. 1628 - 1650, (1999); Klug and Famulok, M. MoI. Biol. Rep., 20, pp. 97 - 107, (1994); US 5,582,981). Aptamers can also be synthesized and selected in their mirror form, for example, as the L-ribonucleotide (Nolte et al., Nat. Biotechnol., 14, pp.1116 - 1119, (1996); Klussmann et al., Nat. Biotechnol., 14, pp. 1112 - 1115, (1996)). Forms isolated in this way have the advantage, that they are not degraded by naturally occurring ribonucleases and, therefore, have a greater stability.

Preparation of anti-Fn14 antibodies for modulating the differentiation of adipocytes

The general method for preparing an antibody or antibody fragment is by methods that are known to the expert, for example, by immunizing a mammal, for example, a rabbit, with the corresponding antigen, whereby, if necessary, corresponding adjuvants, for example, Freund's adjuvant and/or aluminium hydroxide gels or other adjuvants may be used (see, for example, Diamond, B.A. et al., The New England Journal of Medicine, pp. 1344 - 1349, (1981)). The polyclonal antibodies that are formed in the animal as the result of an immunological reaction can later on be isolated from blood by using methods known in the state of the art and may then be purified, for example, by column chromatography. For example, monoclonal antibodies can be prepared in accordance with the known methods of Winter & Milstein (Winter, G. & Milstein, C₁ Nature, 349, pp. 293 - 299, (1991)).

Specific polyclonal antibody serums that are directed against the human Fn 14 can be prepared by employing conventional methods by injecting animals, for example, goats, rabbits or mice, subcutaneously, for example, with an Fn14-derived protein or peptide or derivative that presents the activating epitope. Alternatively, DNA vaccination may be employed. The activating natural epitope can be represented by any chemical substance that has a surface structure and/or form and/or charge that is comparable to the natural epitope.

For this purpose intraperitoneal or subcutaneous injections of additional agents that enhance the immune reaction (adjuvants, e.g. Freund's adjuvant) may also be employed. Polyclonal antiserums that contain the desired antibodies that are directed against the activating epitope of Fn 14 can be expanded by repetitive intraperitoneal immunizations of mice with Fn14-derived protein or peptide or derivative in the absence of adjuvants. The immunization of animals with Fn14, derived proteins, peptides or derivatives thereof that represent the activating epitope can also be effected by either intraperitoneal or intravenous injections. For preparing monoclonal antibodies hybridoma cells can be fused according to classical methods and be screened, for example, by an ELISA (Ling et al., J. Interferon and Cytokine Res., 15, pp. 53 - 59 (1995)). Furthermore, hybridoma cells are assayed for their ability to produce antibodies that recognize the Fn14, derived protein, peptide or derivative or the activating epitope and to modulate the differentiation of fat cells. Pure monoclonal antibodies (IgG) were purified from hybridoma cell culture supematants by means of protein A sepharose.

Various forms of anti-Fn14 antibodies can also be prepared by employing standard methods for producing recombinant DNA (Winter and Milstein, Nature, 349, pp. 293 - 299 (1991)). For example, "chimeric" antibodies, wherein the antigen binding site of an animal antibody is coupled to a human constant domain can be prepared (e.g., Cabilly et al., US 4,816,567; Morrison et al., Proc. Natl. Acad. Sci. U.S.A., 81, pp. 6851 - 6855 (1984)). Chimeric antibodies reduce the observed immune response that becomes pronounced in human clinical studies, where animal antibodies are used. According to the present invention the term antibody and antibody fragment is also understood to include antibodies and/or antigen binding parts thereof that were produced recombinantly and, if required, were modified, for example, chimeric antibodies, humanized antibodies, multifunctional antibodies, bispecific or oligospecific antibodies, single-stranded antibodies and F(ab)- or F(ab)₂ fragments (see, for example, EP 368 684 B1 , US 4,816,567, US 4,816,397, WO 88/01649, WO 93/06213 or WO 98/24884).

Furthermore, recombinant "humanized antibodies" that recognize the activating epitope of Fn14 can be synthesized. Humanized antibodies are chimeric antibodies that for the largest part have human IgG sequences into which regions responsible for the specific antigen binding have been inserted (WO 94/04679). Animals are immunized with the desired antigen, the corresponding antibody is isolated and that part of the variable sequence regions that is responsible for the specific antigen binding is removed. The antigen binding sites originating from the animals are then cloned into the corresponding position of the human antibody gene, wherein the human antigen binding sites had been deleted. Humanized antibodies reduce the use of heterologous (inter-species) sequences in human antibodies and pose a lower risk for inducing an immune response in the treated individual. The production of different classes of recombinant anti-Fn14 antibodies that recognize the activating epitope can also be achieved by preparing chimeric or humanized antibodies with anti-Fn14 variable domains and human constant domains (CH1 , CH2, CH3) that had been isolated from different classes of immunoglobulins. For example, anti-Fn14 antibodies that recognize the receptor activating epitope can be recombinantly produced with an increased affinity for antigen binding sites by cloning the antigen binding sites into vectors that contain the corresponding human constant regions (Arulandam et al., J. Exp. Med., 177, pp. 1439 - 1450 (1993); Lane et al., Eur. J. Immunol., 22, pp. 2573 - 2578 (1993); Traunecker et al., Nature, 339, pp. 68 - 70 (1989)).

As an alternative to classical antibodies it is also possible to employ so-called "protein scaffolds", for example, anticalines, that are based on lipocaline (Beste et al., Proc. Natl. Acad. Sci. USA, 96, pp. 1898 - 1903, (1999)). The natural ligand binding sites of lipocalines, for example, of the retinol-binding protein or bilin-binding protein, can be changed, for example, by employing a "combinatorial protein design" approach, and in such a way that they bind selected haptens (Skerra, Biochem. Biophys. Acta, 1482, pp. 337 - 350, (2000)). For other protein scaffolds it is known that they are alternatives for antibodies (Skerra, J. MoI. Recognit, 13, pp. 167 - 287, (2000)). All these protein-derived alternatives for antibodies can be denoted binding proteins. For this reason and according to the present invention the term binding protein is to be understood as also including the herein described binding proteins and binding molecules, e.g. affilines, anticalines and aptameres, that specifically recognize the receptor-activating epitope of the receptor Fn 14, proteins derived from these, peptides or derivatives thereof, that were produced recombinantly, and if required, were modified.

Epitopes of Fn14 that result in receptor activation upon binding the anti-Fn14 antibody

Antibodies bind with a defined specificity and affinity to the corresponding target molecules. This also true for antibodies that are directed against receptors on the cell surface. The binding of an antibody to such a surface receptor, here Fn14, can result in its activation. For doing so the mere binding of the antibody is necessary but not sufficient. Moreover, the receptor activation depends on the type of binding of the antibody. This type of binding of the antibody must simulate the effect that the binding of a natural ligand has on the receptor. In many cases the natural ligand causes a change in conformation or another property of the receptor, for example, its dimerisation or multimerisation or both. An activation results, when these structural changes result in an intrinsic activity of the receptor protein, for example, an enzymatic activity or a protein- protein interaction activity directly or indirectly. Induced changes of the receptor depend on the molecular position and the positions on the receptor, to which the ligand or antibody binds. Molecular positions that are recognized by antibodies are denoted epitopes. Accordingly, antibodies that lead to the activation of receptors bind to those epitopes that simulate the receptor effects of ligand binding (see above) upon binding.

These peptides, epitopes or artificial chemical structures that correspond functionally to the epitopes can also be used for immunizing animals for the generation of, for example, polyclonal or monoclonal antibodies as well as for generating antibody derivatives and binding proteins by means of recombinant technologies. In addition, it is also possible to use these activating epitopes as immunogen for forming activating antibodies directly in humans. This allows a vaccination of patients whereby their adiposity/obesity and their consequential damages may be prevented and/or treated.

The knowledge about the epitopes that lead to receptor activation upon binding can inter alia be employed for generating agonistic antibodies with a targeted rational method. This can amongst others be achieved by immunizing with said epitopes. On the other side, direct binding assays based on activating epitopes that, for example, are based on the ELISA principle (see Example 5, Figure 6) will also be established and used. Said assays can be employed for identifying and characterizing activating antibodies (see Example 5, Figure 6).

Alternatively, it is also possible to obtain antibodies that bind to said epitopes by competition experiments. In this way potential agonistic antibodies can be identified by peptide competitions. Such an activating antibody is, for example, the agonistic antibody mABX1, that is described in Example 2 and Figure 4 as antagonistic anti-Fn14 antibody. Competition experiments, wherein the antibodies to be analyzed are tested, whether they inhibit the binding of the antibody to be tested upon addition of an excess of a defined epitope indicate an epitope-specific binding. Said competition analysis, the practice of which is within the skill of a skilled person by means of standard methods and which are illustrated in Figure 5 in an exemplary manner, can also be used accordingly for analysing the potential binding of new antibodies to an activating epitope of Fn14. Modulation of fat cell differentiation by specific activation of the Fn14-mediated signal transduction

According to a preferred embodiment the anti-Fn14 antibodies, antibody derivatives or binding proteins thereof used according to the invention function by inhibiting the formation of terminally differentiated adipocytes. On one side, the formation and/or function of terminally differentiated adipocytes can be inhibited by inhibiting the differentiation to terminally differentiated adipocytes, in particular, by inhibiting the transcription signal pathway for differentiation. On the other side, the formation of terminally differentiated adipocytes can be inhibited by inhibiting lipogenesis or by enhancing lipolysis. The inhibition of the differentiation to terminally differentiated adipocytes can be done by measuring the lipid (e.g. by staining with Nile Red). With reporter assays such as, e.g., for PPAR-γ or CEBP-β or -δ, the inhibition of the transcription signal pathway can be measured. By staining with lipophilic coloring agents such as, e.g. Nile Red, the lipid content in adipocytes can be detected. The modulation of the differentiation of adipocytes by activating Fn14 is described for the natural receptor ligands TNFSF12 as well as for activating anti-Fn14 antibodies in Example 2. By doing so, the effect of signal transduction mediated by the ligand (TNFSF12) and the antibody (anti-Fn14) is illustrated clearly and also in a dose-dependent manner (see Figures 1, 2 and 4) either in murine cells or also in fresh human pre-adipocytes.

Because the involvement of Fn14-mediated signal transduction in inflammatory and toxic processes and cell death (for example, in tumor cells) is already known, it was also analyzed, if the inhibition of the differentiation of adipocytes is associated with the induction of cell death. The determination of the cell count of living cells is possible, for example, with the "Alamar Blue" (Biosource) or "CellTiterGlo" (Promega) assays.

The possibility of toxicity or the induction of apoptosis by activating Fn 14 was assayed by the applicant in murine and human cells and is described in Example 3. It was possible to demonstrate that under the conditions employed there was no toxicity. In these experiments it was also demonstrated that the inhibiting effect is reversible. This was demonstrated by the removal of the protein, which subsequently allowed the cells another entry into the differentiation process. Obviously, this is only possible in living cells, a fact that was also confirmed by microscopic measurements of the living cells. Therapeutic use of antibodies and other activators of the Fn14-mediated signal transduction for the prevention and/or treatment of adiposity/obesity and associated diseases

The invention relates to a method for the treatment of a patient, wherein an effective amount of one and/or more of the above defined agonistic agents such as anti-Fn14 antibodies is administered to the patient. Preferably, the patient suffers from adiposity/ obesity. For this method according to the invention all of the embodiments illustrated above for the use according to the invention apply, too.

As already described above, the factors TNFSF12 and the anti-Fn14 antibodies found according to the invention allow for the inhibition of the differentiation of pre-adipocytes to adipocytes and/or the inhibition of the function of mature or maturing adipocytes, for example, by inhibiting the lipogenesis or assisting lipolysis.

The observations that the targeted receptor activation of Fn14 by recombinant supplementation of the receptor-activating ligand TNFSF12 can reduce the weight of animals (see Example 4) is proof for the therapeutic concept described herein.

According to the invention, by inhibiting the complete or partial suppression of a biological function is understood. In this specific case the inhibiting function is the differentiation of adipocytes in patients with adiposity/obesity or the prevention of adiposity/obesity. According to the present invention the term "inhibitor" describes an agent, such as an anti-Fn14 antibody, that preferably inhibits the process of fat storage by itself. In this case the activating antibody can be denoted an inhibitor. The inhibition can, for example, be achieved by the induction of the Fn14-dependent signal transduction and subsequent modulation of the expression of regulated specific genes.

According to a preferred embodiment the inhibitor used, e.g., the anti-Fn14 antibody, elicits its effects by inhibiting the formation of terminally differentiated adipocytes. Preferably and in an exemplary manner, this can take place by interfering with the differentiation to terminally differentiated adipocytes, in particular, by modulating the Fn 14 transcription signal pathway for the differentiation in pre-adipocytes. Furthermore, the invention relates to a pharmaceutical composition that contains one or more agents of the present invention. For all the pharmaceutical compositions according to the present invention the following holds true:

For the preparation of the pharmaceutical composition according to the invention the corresponding molecules are typically formulated together with suitable additives or excipients, such as, for example, physiological buffer solution, for example, sodium chloride solution, deionised water, stabilizers, such as protease- or nuclease inhibitors, preferably aprotinin, ε-aminocaproic acid or pepstatin A or chelators, such as EDTA, gel formulations, such as white vaseline, paraffin with low viscosity or yellow wax, depending on the mode of the administration.

Further suitable additives are, for example, detergents, such as, for example, Triton X-100 or sodium deoxycholate, but also polyols, such as, e.g., polyethylene glycol or glycerin, sugars, such as, for example, sucrose or glucose, amphoteric substances, such as amino acids, such as glycine or, in particular, taurin or betaine or a protein, such as, for example, bovine or human serum albumin. Detergents, polyols and/or amphoteric substances are preferred.

Preferably, the physiological buffer solution has a pH of about 6.0 - 8.0, in particular a pH of about 6.8 - 7.8, in particular a pH of about 7.4, and/or an osmolarity of about 200 - 400 milliosmol/Iiter, preferably of about 290 - 310 milliosmol/liter. The pH of the medicaments is generally adapted by employing a suitable organic or inorganic buffer, such as, for example, phosphate buffers, tris buffers (tris(hydroxy methyl)amino methane), HEPES buffer ([4-(2-hydroxy ethyl)piperazino]ethane sulfonic acid) or MOPS buffer (3- morpholino-1 -propane sulfonic acid). The choice of the adequate organic buffer generally depends on the desired buffer molarity. For example, a phosphate buffer is suitable for injection or infusion solutions.

Solutions for injection are generally employed, when only a small amount of a solution or suspension, for example, about 1 to about 20 ml is to be administered. Solutions for infusions are generally employed, if a larger amount of a solution or suspension, for example, one or several litres, is to be administered. Because contrary to an infusion solution only a few millilitres are administered in the case of solutions for injection, small differences in the pH or the osmotic pressure of the blood or the tissue liquid are not perceived in comparison to a solution for injection or do not play an important role. Therefore, the dilution of the formulation before its use is generally not necessary. However, when relatively large amounts are to be administered, the formulation according to the invention should be diluted shortly prior to the administration, so that an isotonic solution is obtained. An example of an isotonic solution is a 0.9% saline solution.

In principal, the effective dose will depend on the weight and condition of the subject to be treated. It is to be assumed that the skilled person knows how to determine a suitable dose.

The pharmaceutical composition can be administered in various modes and ways, for example, intramuscularly, subcutaneously, intrathecal^, into the fat tissue, percutaneously (dissolved in DMSO), intravenously or intraperitoneal^ or by fusion or gels, that contain the respective medicament. Further, it is possible to apply the medicament topically and locally. Furthermore, the medicament can be administered by a transdermal therapeutic system (TTS) that allows for the time-controlled release of the medicament. For example, TTS are known from EP 944 398, EP 916 366, EP 889 723 or EP 852 493. A slower release of the protein, peptide or antibody is achieved by a combination with polymers, a prolonged half life by the addition of PEG. Suitable polymers also allow for the oral administration of the pharmaceutical composition, for example, by providing a protected passage through the colon and the targeted penetration of the cell-cell contact sites in the area of the colon epithelium. A combination with chemical active agents (e.g. appetite inhibitors, lipase inhibitors) can further increase this effect.

According to a preferred embodiment the pharmaceutical composition according to the invention having an activating agent is provided in a form that allows for the contact of the active agent with a component of the cell surface (does not enter the cell) (Cardiovasc. Pharmacol. Ther., 7 (3) pp, 171 - 80, (July 2002)).

The listing of the above embodiments is intended to describe the present invention more clearly, but not to limit it.

Furthermore, the invention relates to a method for the prevention of adiposity/obesity and/or the treatment of a patient, wherein an effective amount of one or more of the above defined inhibitors is administered to the patient. Preferably, the patient suffers from adiposity/obesity or diabetes type Il or from diabetes type ll-related diseases. Examples

In the following the invention will be illustrated by examples and figures that are by no means intended to limit the subject-matter of the invention but instead represent embodiments.

In particular, it is shown in the examples that the soluble polypeptides and antibodies employed according to the invention are capable of inhibiting the storage of lipids in murine 3T3-L1 cells and in human pre-adipocytes. The inhibition was demonstrated in primary human adipocytes of subcutaneous as well as of omental origin. In each case the detection was done by the Nile Red assay.

Furthermore, it is demonstrated that the receptor Fn 14, the activation of which effects the modulation of the differentiation of fat cells, is present on human and murine precursor fat cells. This fact is an important prerequisite for explaining the observed phenotype by receptor-mediated signal transduction.

Moreover, it is demonstrated that the activation of the receptor Fn14 that effects the modulation of the differentiation of fat cells leads to no toxic effects in the system of murine and human cell cultures that is used in the present case.

In addition, it is shown that expressed, soluble, murine, homologous variants of TNFSF12 or their protein fragments are capable of reducing the gain in weight in mice without causing general toxic effects.

It was also shown that antibodies that cause the fat cell-modulating phenotype by receptor activation bind to defined epitopes of the receptor Fn 14 and, thereby, initiate signal transduction. One such defined epitope of Fn 14 is represented by the peptide MDCASCRARPHSDFC (see SEQ ID No: 4). The knowledge of this epitope is an essential pre-requisite for the rational generation of defined antibodies that activate the signal transduction by the targeted binding to these epitopes of the receptor.

Description of the figures:

Fig. 1 : This illustrates the dose/response curve that shows the lipid content of 3T3-L1 cells in dependency of the added amount of a recombinant, secreted or soluble protein fragment of TNFSF12. The measured results are specified as relative fluorescence units (RFU) on day 5 after the addition of the protein.

Fig. 2:

This illustrates the percentage inhibition of the lipid storage on day 8 in the form of a dose/response curve by recombinant, secreted or soluble protein fragments of TNFSF12 - in primary human adipocyte cultures of subcutaneous origin. For this purpose the primary human cell cultures were incubated with each of the respective recombinant proteins. The percentage data relates to control cultures that were treated with a protein concentration of 0.025 ng/ml that is still ineffective. A low signal strength means a low lipid storage in the differentiating human adipocytes, a strong signal indicates a strong lipid storage.

Fig. 3: Detection of Fn 14 on murine and human pre-adipocytes

This illustrates the detection of the receptors of TNFSF12, Fn14, on pre-adipocytes. A: mouse 3T3-L1 pre-adipocytes stained with anti-mouse-Fn14-specific antibodies and B: primary human adipocytes stained with anti-human-Fn14-specific antibodies and a suitable fluorophor-labelled secondary antibody and the fluorescence-activated cell sorting method (FACS). The measured results are specified as fluorescence intensities (FL-1) relative to the number of events (or "counts"). The fluorescence intensity profile was compared in the presence (D) or absence (■) of the specific receptor antibodies. The shift of the fluorescence intensity to stronger intensities (shift to the right) meant a binding of the specific receptor antibody together with the fluorescence-labelled secondary antibody to the cell surface. This illustrates the receptor^'s exposition on the cell surface.

Fig. 4:

This illustrates the concentration-dependent inhibition of the lipid storage by added antibody mAB-X1 in comparison to TNFSF12 in primary human adipocytes of subcutaneous origin. The measured results are specified in percent versus lipid storage of normal differentiated cells. The data was obtained on day 8 after the first addition of the corresponding protein and the induction for differentiation. A low signal means a low lipid storage in the differentiating human adipocytes, a strong signal indicates a strong lipid storage. Fig. 5:

This illustrates the degree of binding of an Fn14-specific antibody mAB-X1 to the Fn14 receptor on HeLa cells after pre-incubation of the Fn 14 antibodies with the peptide according to the invention with SEQ-ID #4 by FACS analysis. HeLa cells express the receptor Fn 14 naturally, a fact that was again confirmed by the detection of the binding of TNFSF12 in a comparable FACS analysis.

A: This illustrates an analysis of the competition of the binding of the Fn14-specific antibody mAB-X1 by the peptide with SEQ-ID #4. As a positive control a fusion protein containing the complete human Fn 14 extracellular domain was employed and a control peptide as a negative control. The fluorescence intensity distributions for HeLa cells are specified for each competition experiment obtained with the specific antibody (with competitor - full line or without the competitor - closely dashed line) or an isotype control antibody instead of the Fn14-specific antibody (barred dashed line). The fluorescence intensity distribution after the addition of the competitor is also labelled with an arrow at the maximum. A shift of the fluorescence distribution to the left with the competitor (full line) relative to the distribution without competitor (closely dashed line) illustrates its competitive binding to an Fn14-specific antibody.

B: increasing amounts of peptide having SEQ-I D#4 led to a dose-dependent decrease of the fluorescence intensity measured on the cells. The measured results are specified as the measured geometric mean of the fluorescence intensity distribution.

Fig. 6:

This illustrates the detection of the binding of the Fn14-specific antibody mAB-X1 (black bar) and a non-Fn14-specific isotype antibody (grey bar) as a control to the peptide according to SEQ-ID #4 that is immobilized in a well of a suitable 96-well plate and to an immobilized control peptide of the same length as a control. The binding was measured in the presence of various amounts of the peptides in an ELISA in a dose-dependent manner. The results are specified as the absorption at 450 nm. The dose-dependent signal with the peptide SEQ-ID #4 is significantly stronger for the Fn14-specifιc antibody mAB-X1 than for the control antibody and does not occur when a control peptide having a different sequence is used.

Fig. 7: This illustrates the limitation of the gain in body weight in male C57BL/6 mice two weeks after injection of a suitable expression vector that is dependent on the extent of the expression of soluble or secreted protein fragment of murine TNFSF12(A). The extent of the expression of the protein fragments was determined on serum samples by an ELISA. The negative correlation of the protein expression and the gain in body weight is illustrated. (B): This illustrates the measured weight data of selected fat deposits of animals, wherein an expression of TNFSF12 was detectable (n=5) in comparison to those where this was not detectable or only detectable in relatively low amounts (n=3). Both, the epidydimal fat deposits ("pads") of male animals and the perirenal and colon- associated fat tissue were assayed exemplarily. This result shows that a loss of weight correlates with the TNFSF12 expression in selected fat deposits.

Fig. 8:

This illustrates the insulin-dependent glucose uptake in primary human adipocytes of subcutaneous origin. Ineffective (10^"1° M) and effective (10^"7 M) concentrations of insulin were employed. Effective insulin concentrations result in an increased cellular glucose uptake under the selected conditions (control). The effects of the co-incubation with added protein TNFSF12 (100 ng/ml) are also shown. In this context TNF-alpha (50 ng/ml) served as a positive control because its interference with the insulin-dependent glucose uptake has already been described in adipocytes. The measured results are specified as "insulin-dependent uptake of pmol glucose within 20 min".

Fig. 9:

This illustrates the investigations regarding the proliferating activity or cell survival upon exposure of subconfluent 3T3-L1 cell cultures to soluble TNFSF12. The proliferating activity or the cell count was determined by determining the number of living cells using the AlamarBlue assay and the measured results are specified as relative fluorescence units (RFU).

Fig. 10:

Female C57BI/6J mice were immunized with 30 μg antigen per injection (Peptide coupled to KLH-protein; peptide-sequence: n-MDCASCRARPHSDFC-c, by Eurogentec/Belgium) in the following schedule: Immunizations at days = 0, 14, 28 followed by a boost on day 56. Bleedings were taken on day = 0 (preimmune serum) and on days 38, 66 and on day 87 when mice were sacrificed. Data given above are from day 66, but were comparable in all bleedings tested. Example 1:

Cultivation of pre-adipocyte cells, their induction for differentiation and the measurement of intracellular incorporated lipids

For the propagation and analysis of murine 3T3-L1 cells:

3T3-L1 cells are mouse fibroblast cells of embryonic origin that can differentiate terminally to mature fat cells (adipocytes) under suitable conditions. Therefore, the 3T3-L1 kept in proliferation culture are also called pre-adipocytes. They grow in monolayers and have a doubling time of about 24 h. The 3T3-L1 cells used herein originate from ATCC (CL-173). Because the 3T3-L1 cells spontaneously differentiate at confluency and, thereby, change the properties of the culture, they were not cultivated until confluency (max. 80%). The passaging was done at intervals of 2 days without a medium exchange in between. The cells were used from passage 5 on and until passage 18 only. The maintenance culture of the 3T3-L1 cells was kept in T-75 culture flasks. For passaging every 2 days these were seeded with 4.5 x 10⁵ cells/T-75 flask. There is no medium exchange until the next passage. The seeding of the 3T3-L1 cells for the next experiment was conducted in 96-well microtiter plates: The 3T3-L1 cells were seeded at a cell density of 1.5 x 10⁴ cells/well in 200 μl growth medium/well. Then the cells were incubated without medium exchange for 3 days and subsequently induced for differentiation as it is described further below and contacted with recombinant soluble or secreted protein fragment or antibody. The 3T3-L1 growth medium (GM) had the following composition: DMEM, 10% FCS, 1% penicillin-streptomycin, 2% glutamine, 1% Na-pyruvate). The differentiation medium (DM) was principally added as a double concentrate directly after the addition of recombinant protein or antibody. In this form it consists of a 3T3-L1 growth medium supplemented by induction factors (200 nM insulin, 1 mM IBMX und 2 μM dexamethasone).

For the isolation of adipocytes from fat tissue and the cultivation of primary human pre- adipocytes:

The isolation and differentiation of pre-adipocytes was conducted as described in Hauner et al. [Methods MoI. Biol. 2001; 155: 239 - 47]. The fat tissue obtained during surgical operations is mechanically freed of connective tissue and remaining blood vessels and subsequently minced. Then it is digested with 200 U/ml collagenase NB4 (Serva Electrophoresis GmbH, Heidelberg) for 90 min at 37 ⁰C and 80 rpm in PBS with 2 % BSA (3 ml collagenase solution/g fat tissue). Afterwards, centrifugation followed for 10 min at 200 g. The formed pellet also contained erythrocytes and remaining connective tissue next to the pre-adipocytes and was taken up in erythrocyte lysis buffer (155 mM NH₄CI, 5.7 mM K₂HPO₄, 0.1 mM EDTA, pH 7.3). After a maximum incubation of 10 min at room temperature it was filtered over a 150 μM filter and centrifuged as described above. Because the remaining pellet still contained remaining connective tissue, it was again purified with a 70 μM filter after being taken up in pre-adipocyte medium (DMEM/F12 (Invitrogen GmbH, Karlsruhe) supplemented with 8 mg/l biotin, 4 mg/l pantothenate, 1.79 g/l NaHCO₃ and 55 mg/l pyruvate). The cells were counted, centrifuged again and resuspeπded in DMEM/F12 with 10 % FCS and 50 μg/ml gentamycin. The seeding was done at a density of 40000 to 55000 cells/cm². The cells were incubated at 37 ⁰C and 5 % CO₂.

Induction for differentiation and addition of the supernatant or the protein to be tested onto 3T3-L1 cells:

Three days after seeding the 3T3-L1 cells 140 μl medium of the 3T3-L1 cells were taken and 50 μl growth medium was added, wherein suitable amounts of recombinant protein (0 - 3 μg/ml protein) or antibody (for example, 0 - 10 μg/ml) had been dissolved. In addition, 100 μl 2X differentiation medium was added to the 3T3-L1 cells followed by an incubation of the 3T3-L1 cells in an incubator for 5 days. After this time the intracellular incorporated lipids were determined.

Differentiation of human pre-adipocytes and SGBS cells:

Human pre-adipocytes were seeded at a density of 40000 to 55000 cells/cm². For SGBS cells (described in Wabitsch et al., 2001, Int. J. Obes. Relat. Metab. Disord. 25:8 - 15) the cell density was lowered by a factor of 10. After the seeding the cells of the primary culture remained for 20 h, the SGBS cells remained for 3 days in proliferation medium (pre-adipocyte medium with 10 % FCS and 50 μg/ml gentamycin). For inducing the differentiation the cells were washed twice with PBS and then incubated in serum-free differentiation medium (pre-adipocyte medium with 66 nM insulin, 1 nM T₃, 100 nM hydrocortisone, 10 μg/ml transferrin, 50 μg/ml gentamycin). Also, for the first 3 days 1 μg/ml troglitazone and 0,5 mM IBMX were added. After this time and in the further course the medium was exchanged every three to four days using differentiation medium. The cells were used, when at least 50% of the cells (relative to the cell count) had incorporated fat. Measurement of the intracellular stored lipid:

For quantitatively determining the storage of cellular lipids the Nile Red reagent (Molecular Probes, Leiden, Netherlands; CAS number 7385-67-3) (Nile Red staining solution: 4 μg/ml Nile Red in PBS/40 % DMSO) was used. The fluorescence was measured at an excitation wavelength of 485 nm and an emission wavelength of 590 nm. The required amount of Nile Red was added to the calculated amount of DMSO and mixed. Subsequently, the calculated amount of PBS is added and the solution is mixed. For performing the Nile Red assay 140 μl medium of the 3T3-L1 cell cultures were removed and 200 μl PBS were added on day 5 (3T3-L1 cells) or days 8 - 12 (primary human pre-adipocytes cultures) after induction for differentiation. Then, 150 μl liquid were removed and 50 μl Nile Red staining solution were added. The incubation of the plates was done for 4 hours at 37 ⁰C in a CO₂ incubator. The read out was done in a fluorescence reader at an excitation of 485 nm and an emission of 590 nm (200 msec readout time). The measured results are specified as relative fluorescence units (RFU) or as percentages relative to normal differentiated (non-inhibited) controls.

Example 2:

Inhibition of the lipid storage in cell cultures by activating the receptor with receptor-specific antibodies or natural ligands

The activation of a cellular receptor can be done by the natural ligand as well as by an antibody, when the activation of said receptor is effected, for example, by successful cluster formation of receptor molecules, which is the present case. The antibody functions agonistically when it results in a similar activity after binding to its target structure that is also generated by the natural ligand. When doing this, the affinity may deviate. The epitope of the antibody can, but must not necessarily, correspond to the binding site of the ligand. What is important is the functional effect on the activation status of the receptor. Not any binding antibody will also result in an activation of the receptor, but only those that bind to specific epitopes in the extracellular domains of the receptor that will result in such an activation.

At first, a receptor was activated by its natural ligand. In the present case the human ligand was also capable of activating the corresponding homologous mouse receptor. The inhibition of lipid storage in mouse 3T3-L1 cells (Fig. 1) or in primary human adipocytes (Fig. 2) was determined with a recombinant, secreted or soluble protein fragment of human TNFSF12 (R&D Systems GmbH, Wiesbaden: Cat No.: 1090-TW; amino terminal 6X HIS-tagged extracellular domain of human TWEAK (arg 93 - his 249, see Chicheportiche et al., 1997, J. Biol. Chem. 272:32401 - 32410)).

The results demonstrate that the storage of lipids in murine 3T3-L1 cells cultures (Fig. 1) as well as in primary human adipocytes (Fig. 2) during differentiation can be inhibited by adding the recombinant soluble protein fragment of TNFSF12 in a dose-dependent manner. The lipid storage can serve as a measure for the differentiation of adipocytes or as a measure for the regulation of lipogenesis or lipolysis in these cells. In both cellular systems there was a half maximum effective concentration of about 30 ng/ml for the recombinant soluble protein fragment of TNFSF12. Also, a comparable inhibition of differentiation was achieved for primary fat cells of omental origin and the human SGBS adipocyte cell line.

The employment of agonistic monoclonal mouse antibodies with specificity for the human TWEAK receptor protein (Fn 14) led to the following results:

The inhibition of the storage of lipids in primary human adipocytes was determined with the antibody mAB-X1 (Fig. 4). The results demonstrate that the storage of lipids in primary human adipocytes (Fig. 4) during the differentiation was inhibited in a dose- dependent manner by adding the antibody mAB-X1. The storage of lipids can serve as a measure for the differentiation of the adipocytes or as a measure for the regulation of lipogenesis or lipolysis in these cells. The addition of 1 μl mAB-X1 antibodies in this experiment led to an inhibition of differentiation that was comparable to the case when 100 ng/ml naturally ligand had been employed.

In order to detect the known receptor of TNFSF12 on the murine 3T3-L1 cell cultures as well as on primary human pre-adipocytes these were detected with species-specific anti- Fn14 antibodies through the FACS method. This is illustrated in Figure 3. In the course the receptor of TNFSF12 was detected with the antibody pair of TWEAK receptor- specific mouse- (A) or human (B) antibody. The staining of the cells was done according to the following scheme: Seeding of 200000 cells in a 6-well dish. 24 h later staining and measurement in the FACS detection device: For doing this, cells were trypsinised and pelleted and subsequently incubated in 200 μl buffer solution (PBS + 1 % FCS) or buffer solution with receptor-specific antibody according to the instructions of the manufacturer (5 μg/ml) for 1 h on ice. Next, 1 ml PBS + 1 % FCS was added and pelleted. The uptake was done in 200 μl PBS buffer with 1 % FCS₁ wherein the secondary antibody (2 μg/ml) was dissolved. A further incubation on ice for another 45 min followed. Subsequently, the cells were again taken up in 1 ml PBS buffer with 1 % FCS and pelleted. After uptake in 500 μl buffer the cells were subsequently measured in the FACS device. For doing this the measurement of the fluorescence intensities was done for living cells only. The result demonstrates that the receptor described for TNFSF12 is expressed on the cell surface of 3T3-L1 cells as well as on the surface of primary human pre-adipocytes.

Example 3:

Investigation of possible adverse effects after receptor activation

The mode and manner how the proteins according to the invention accomplish the inhibition of lipid storage in adipocytes was determined in more detail by further analysis of the cell cultures. An inhibition of the differentiation or the lipid storage in differentiating adipocytes can, for example, be effected by protein-mediated activities such as cell death, preventing the exit from the cell cycle, blocking a differentiation- specific gene expression or blocking lipogenesis or by enhancing lipolysis.

Specifically the determination or the exclusion of cytotoxic mechanisms during the observed inhibition of adipocyte differentiation was assayed as follows: Figure 9 illustrates the determination of the cell count of subconfluent 3T3 cell cultures. The cells were seeded at a density of 3000 cells per well in a 96-well cell culture dish. 24 h later the growth medium of the 3T3-L1 cells was exchanged against medium with reduced FCS content (0.5%). Another 48 h later a further medium exchange to 2 % FCS and a further addition of soluble recombinant TNFSF12 was done. In the course a dose- response relationship was investigated over a concentration range of 0.1 to 500 ng/ml protein. 72 h after the addition of the proteins in medium with 2 % FCS 10 μl AlamarBlue reagent (BioSource company) were added to 100 μl culture medium and incubated for 4 h at 37 ⁰C. The AlamarBlue reagent provides a substrate for living cells that is reacted enzymatically and the product can be measured fluorometrically at an excitation wavelength of 530 nm and an emission wavelength of 590 nm. By doing so an indirect cell count determination is provided in the linear measurement range. These results demonstrate that subconfluent 3T3-L1 cell cultures were neither activated for an increased proliferation by the addition of recombinant soluble protein fragment of TNFSF12 (Fig. 9) nor was a cell death detectable. These results were also confirmed for differentiating 3T3-L1 cell cultures.

For determining the acute in vivo toxicity of a receptor activation the following can be assayed: Following the determination of the maximum tolerated dose (MTD) in mice of the strain C57/BI6/J a) the reduction of the gain in weight in normal weight mice (without previous high calorie diet) and b) the net reduction of adipose tissue in mice that were normal weight or obese at the beginning of the test (and initiating or maintaining a high calorie diet) after addition of the proteins used according to the invention were assayed by weighing or determining the BMI ("body mass index"). The determination of the maximum tolerated dose was done with 3 mice (strain C57/BI6/J wild type) by administering 0.002; 0.02; 0.2; 2.0 and 10 mg protein/kg as a one-time intravenous dose. A subsequent fine adjustment of the dose was done with 3 mice each (strain C57/BI6/J wild type) by the one-time addition of the protein. In a dose tolerance test the dose necessary for application is determined by the repetitive administration of a selected dose over several applications based on blood parameters and serum stability.

In particular, it was shown herein that the one-time intravenous administration of up to 5.4 mg/kg body weight in the case of recombinant soluble protein fragment of TNFSF12 into female C57/BI6/J mice led to no clinical symptoms within 48 that were caused by the test substance.

Because the differentiation of adipocytes is inter alia an insulin-dependent process, it was to be determined, whether the receptor activation can lead to a negative modulation of the insulin signalling in adipocytes. In these experiments that are shown in Figure 8 assays relating to the interference of TNFSF12 with the insulin-dependent glucose uptake in 3T3-L1 cells were conducted. 500000 primary adipocytes (50000 SGBS cells in the case of FGF-16) on day 12 after the induction of differentiation were kept in a 3 cm dish and treated as follows for determining the insulin-dependent glucose uptake: Experimental set up: Controls: 1 cell culture for background (0¹ insulin stimulation). An empty well/dish for determining the background. The stimulation was done with 10^'7, without or 10^"10 insulin, because 10^'10 insulin is not stimulating in this context. In each 3 cm dish there were 800 μl medium DMEM/Ham F12 with 5 mM glucose. Differentiated cells were washed four times with 2 ml warm PBS. Then 800 μl DMEM/F12 with 5 mM glucose and each of the proteins according to the invention were added. Subsequently, there was an incubation for 1 h at 37 ⁰C. Subsequent addition of insulin. Plates were incubated for 15 min at 37⁰C at the lowest agitation frequency.

2-Deoxy-D-[1-H³]-glucose (AmershamTRK 383) was diluted 1 : 10 in DMEM/F12. During the incubation period tubes were provided for the scintillation counter: one for each measuring point and an additional one for the empty well and one for the complete amount of used 2-deoxy-D-[1-H³]-glucose. After 15 min incubation 8 μl of the diluted 2-deoxy-D-[1-H³]-glucose were added. The plates were again incubated with light agitation for 20 min at 37 ⁰C. The plates were placed on ice after 20 min. The medium was suctioned off by a pump. Cells were washed 3 times with 1 ml ice cold PBS each. 500 μl IGEPAL (Sigma I 3021) were added per well followed by 20 min lysis on ice. The complete volume of each well was pipetted into the scintillation cups. In addition, 500 μl IGEPAL plus 8 μl of the diluted 2-deoxy-D-[1 -H^-glucose into one cup. In each cup 3.5 ml of the scintillation liquid (Rotiszint eco plus, Roth, 0016.2) were pipetted. Measurement in the scintillography device over one minute per sample followed. The readout was done with the help of the program Mikro Beta Windows. For the presentation of the results the basal glucose uptake is subtracted, so that only the insulin-dependent part of the glucose uptake is shown.

The receptor activation by TNFSF12 in this experimental set up demonstrated no negative influence of the insulin-dependent glucose uptake.

Example 4:

Reduction of weight or gain in weight in the mouse model

Preferred animal models are those that correspond to the human situation of adiposity/obesity or type Il diabetes in a specific manner. Preferably, rodents such as, for example, mice or rats, as well as dogs, pigs and primates are used. A particularly preferred experimental embodiment is a diet-induced adiposity, optionally associated with an insulin resistance or type Il diabetes in mice or rats induced by high calorie diet. The high calorie diet, that is, for example, a 20 - 45 kcal% fat diet, can be administered before or simultaneously to the administration of the proteins or fragments or the pharmaceutical composition of the invention. Accordingly, the animals can be of normal weight or already overweight at the time point of administering the protein or protein fragment. The goal is a reduction or prevention of a gain in weight and/or a weight reduction; optionally, also an enhancement of the insulin-sensitivity or the glucose tolerance. This can either be done during the initiation of or continued high calorie diet or during a change to normal or hypocaloric diet - in the case of already overweight animals. The change of the diets is initiated at the time of the first administration of the protein or protein fragment or the pharmaceutical composition.

After the determination of the maximum tolerated dose (MTD) in mice of the strain C57/BI6/J (MTD) a) the reduction of the gain in weight in normal weight mice (without previous high calorie diet) and b) the net reduction of the adipose tissue in mice that were normal weight or obese at the beginning of the test (at the beginning or when maintaining the high calorie diet) after the addition of the proteins was determined by weighing or determining the BMI ("body mass index").

The determination of the maximum tolerated dose is done with 3 mice (strain C57/BI6/J wild type) by administering 0.002, 0.02, 0.2, 2.0 and 10 mg protein/kg as a one-time intravenous dose. A subsequent fine adjustment of the doses is done with 3 mice each (strain C57/BI6/J wild type) by the one-time administration of the protein. In a dose tolerance test the dose necessary for application is determined by repetitive administration of a selected dose over several applications based on blood parameters and serum stability.

The administered antibodies are, in principle, detectable by ELISA in blood serum. A reduction of the amount of detectable protein fragment in the blood serum over time is assumed, because this is expected for a protein distributing itself in an organism. The inhibitory activity in the mouse serums can be measured by the addition into the 3T3-L1 cell culture assay by Nile Red lipid determination. In this way it can be demonstrated that the antibodies can be administered intravenously and without causing clinical toxicity.

For further experiments the above-determined maximum dose is reduced to the lowest dose that still has a therapeutic effect. In the following test for determining the change in weight, said determined dose is administered parenterally (intravenously, intraperitoneal^, subcutaneously or intramuscularly). The intraperitoneal, intravenous or subcutaneous application is preferred. For determining the reduction of the gain in weight in normal weight mice normal weight mice are kept on a normal calorie diet (at most 20 kcal% fat diet) or on a high calorie diet (45 kcal%) either without (untreated) or with the addition of the proteins according to the invention (treated) for 10 to 14 weeks. During and after these experiments the gain in weight is determined by weighing and/or determining the BMI.

For treated mice that are kept on a normal calorie diet a reduction in weight or reduced gain in weight is observed contrary to untreated mice. For treated mice that are kept on a high calorie diet a lesser gain in weight or a weight reduction is observed in comparison to untreated mice. This reduction in weight can also be attributed to a reduced fat tissue mass.

For determining the decrease of the gain in weight in obese mice, obese mice (that had been kept on a high calorie diet (45 kcal% fat diet)) were kept on a normal calorie diet (at most 20 kcal% fat diet) and/or on a continued high calorie diet either without (untreated) and with the addition of the proteins according to the invention (treated) for 10 to 14 weeks. During or after these experiments the gain in weight was determined by weighing and/or determining the BMI.

For treated mice that were kept on a normal calorie diet or continued high calorie diet a stronger reduction in weight was shown in comparison to untreated mice. For untreated mice on a high calorie diet a further increase in weight is observed. However, for treated mice a weight reduction or a lesser gain in weight is noted.

In an animal experiment the following was demonstrated: 8 five week old male C57BL6/J mice were each injected with an expression construct for recombinant secreted and soluble murine protein fragment of TNFSF12. It was the purpose of this experiment to observe an influence of the protein expression on the gain in body weight. The animals were kept on a normal calorie diet for the complete observation period.

A TNFSF12 expression construct was used that contained a fragment of the mouse homologue that is described further below. This fragment coded for a recombinant soluble, secreted murine form of TNFSF12.

The expression constructs were prepared by performing an RT-PCR with a suitable murine RNA source (TNFSF12: RNA from 3T3-L1 cells) and with suitable primers (comprising the sequences coding for: mouse-TNFSF12: Arg105 - His249). In addition, 6 histidine residues at the 5'-end and 2x TGA stop codons and sequences for the restriction cleavage sites Xhol and BgI Il at the 3'end were introduced.

Used primers: m12-F-His:

5 '-CATCACCATCACCATCACCGAGCTATTGCAGCCCATTATG-S ^'

(compare to SEQ-ID: #1) m12-R: δ'-AGATCTCGAGTCATCAGTGAACTTGAAAGA-S^'

(compare to SEQ-ID: #2)

The reaction product was inserted directly into the vector pSECTag/FRTΛ/5-His-TOPO (Invitrogen, Karlsruhe, Germany) according to the instructions of the manufacturer for the cloning of PCR products. In this way the secretion signal of the protein for the Ig-kappa chain precedes the protein fragment of TNFSF12 aminoterminally in the reading frame followed by 6 histidine residues. Therefore, the open reading frame generated in this manner coded for the following secreted proteins:

Mouse-TNFSF12 reading frame (compare to SEQ-ID: #3):

METDTLLLWVLLLWVPGSTGDAAQPARRARRTKLALHHHHHHRAIAAHYEVHPR PGQDGAQAGVDGTVSGWEETKINSSSPLRYDRQIGEFTVIRAGLYYLYCQVHFD EGKA VYLKLDLLVNGVLALRCLEEFSATAASSPGPQLRLCQVSGLLPLRPGSSLR IRTLPWAHLKAAPFLTYFGLFQVH

Subsequently, the reading frame was cloned into the target vector pBS-HCRHPI-A (Miao et al, 2003, Human Gene Therapy 14:1297 - 1305) that was intended for the in vivo gene expression in animals: The excision from the pSECTag/FRTΛ/5-His-TOPO vectors was effected with the assistance of the restriction cleavage sites Nhel and BgIII and filling the overlapping ends - introduction into pBS-HCRHPI-A was done by opening with the restriction enzyme EcoRV and ligating the blunt ends with each of the introduced fragments. The resulting plasmid was designated pXAPOI (mTNFSF12).

The transfection of the plasmids in C57BL6/J mice coding for recombinant secreted and soluble murine protein fragment of TNFSF12 was done according to a method that was originally described by Zhang et al. (Gene Therapy 2000; 7:1344). The entry of the plasmid DNA, preferably into liver tissue, is achieved by hydrodynamic pressure that is caused by the fast injection of the plasmid-containing solution into the tail vein. There, a gene expression can be realized for weeks up to months. The method in brief: 50 μg endotoxin-free prepared plasmid were administered in PBS solution in an amount of solution corresponding to 10 % of the body weight of the mouse to be injected with a velocity 1 ml/5 sec. Then, the animals were weighed up to 8 weeks after the injection on a weekly basis. The content of the expressed secreted soluble murine protein fragment of TNFSF12 was determined with an ELISA in obtained blood sera. ELISA antibody pairs for murine TNFSF12: For coating: anti-mTWEAK antibody (R&D Systems, Wiesbaden, Germany, cat. no. # AF1237), detection antibody: biotin-conjugated anti-mTWEAK antibody (e-bioscience, clone MTW-1 , cat. no. # 13-9913). The correlation of protein expression and body weight was investigated.

The experiments described in Fig. 7 have demonstrated that the degree of expression of murine, recombinant, secreted, soluble protein fragment of TNFSF12 clearly correlates negatively with the observed gain in weight during the period of investigation. Based on the present results a correlation coefficient of -0.72 resulted for TNFSF12 (Fig. 7A). The correlation coefficient indicates how strict a correlation is. Values between -1 and +1 are possible. A value of 0 means that there is no correlation. +1 or -1 indicate the presence of an absolutely strict (for minus: negative) correlation. These correlations were observed over the complete observation period of 8 weeks.

The trend indicated by the correlation was also observed (Fig. 7) by weighing selected fat tissue deposits (..epidydimal fat pads" - defined well-measurable fat tissue at the gonads of male animals, perirenal and colon-associated fat tissue). The reduction of the body weight corresponded to the reduction of the weight of selected fat deposits.

It is therefore obvious that further experimental test series, in particular, with larger animals numbers, will confirm the weight-reducing effect and that this can also be attributed to a limiting of the extent and weight of fat tissue deposits.

More detailed experiments are preferred, wherein at the time point of administering the inhibitor 12 week old C57BI6 mice were kept on a normal calorie diet or a high calorie diet and subsequently on a high calorie diet for further 8 weeks until the time of injection. Under these conditions of a diet-induced adiposity/obesity it is to be expected that the presence of the protein fragments of TNFSF12 described herein leads to a significantly reduced gain in weight or an absolute weight reduction that is preferably caused by a significantly reduced weight of the fat tissue deposits.

More detailed experiments are preferred, wherein at the time point of administering the inhibitor 12 week old C57BI6 mice were kept on a normal calorie diet or a high calorie diet and subsequently on a high calorie diet for another 8 weeks until the time of injection. Under these conditions of a diet-induced adiposity/obesity it is to be expected that the presence of the protein fragments of TNFSF12 described herein will lead to a significantly reduced gain in weight or an absolute reduction in weight that is preferably caused by a significantly reduced weight of the fat tissue deposits.

Example 5:

Epitopes of Fn14 that lead to receptor activation upon binding by anti-Fn14 antibodies

Antibodies that are directed against receptors on the cell surface bind with defined specificity and affinity. The binding of an antibody to Fn 14 can cause its activation. For doing so the mere binding of the antibody is necessary, however, not sufficient, because receptor activation depends on the type of binding of the antibody. This type of binding by the antibody must simulate the effect that a natural ligand has on the receptor upon binding. In many cases the natural ligand causes a change in conformation or another property of the receptor such as, for example, its dimerisation or multimerisation or both. An activation takes places when these structural changes leads to an instrinsic activity of the receptor protein, for example, an enzyme activity or protein-protein interaction activity directly or indirectly. Because the induced changes of the receptor depend on the epitopes on the receptor to which the ligand or antibody binds, antibodies that lead to the activation of receptors bind to those epitopes (here denoted "activating epitope") that simulate upon binding the receptor effects of the ligand binding. The peptide of the sequence MDCASCRARPHSDFC is such an activating epitope of the receptor Fn14. Antibodies binding to said peptide are capable of activating Fn14.

Direct binding experiments were conducted in order to identify antibodies that bind to the peptide of the sequence MDCASCRARPHSDFC (SEQ-ID #4) that represents the activating epitope of Fn14. These experiments are based on the known principle of a state of the art ELISA. For this ELISA test the binding of potential activating antibodies to the peptide with the sequence MDCASCRARPHSDFC is measured by immobilizing the peptide on a solid surface (e.g. microtiter plate) and, thereafter, the ability of the antibodies to bind the immobilized peptide is analyzed. As a control in these tests it is necessary to determine the unspecific binding of antibodies to a control peptide which does not represent the activating epitope. Antibodies that demonstrate a markedly enhanced binding to the activating epitope in comparison to the binding to the control peptide (i.e. to the peptide of the sequence MDCASCRARPHSDFC) are defined as "binders" of the activating epitope. A further control in these experiments is the analysis of the specificity of the binding to the activating epitope. In order to ensure this, the binding of a further irrelevant antibody to the activating epitope as well as to a control peptide is analyzed. These antibodies must bind to the activating epitope much weaker in comparison to antibodies that recognize the activating epitope.

Figure 6 illustrates an example for the identification of an antibody that binds specifically to the activating epitope of Fn14 that is represented by the peptide of the sequence MDCASCRARPHSDFC. For doing this, 0.02, 0.2, 1 mg/ml of the peptide MDCASCRARPHSDFC (activating epitope) or a control peptide are placed on microtiter plates (MaxiSorb Plate) by incubating for 15 h at 4 ⁰C in carbonate buffer (0.1 M carbonate buffer, 8.4 g NaHCO₃ and 3.56 g Na₂CO₃ per I; pH 9.5). After treating the plate with blocking buffer (PBS, pH 7.4, 0.5 % BSA, 5 % skimmed milk, 10 % FCS, 0.1 % Tween20; 2 h at RT) and washing buffer (4 x 300 μl PBS + 0.05 % Tween20) for reducing unspecific antibody binding 5 μg/ml of a control antibody (mouse anti-GFP antibody; clone JL-8, Co. Clontech) or one of the anti-Fn14 antibodies (mouse anti- human Fn14; clone mAB-X1) to be tested was added in sample buffer (PBS + 10 % FCS, 0.1 % Tween20) and incubated for 2 h at RT. After further wash steps for removing unbound antibody (4 x 300 μl PBS + 0.05 % Tween20) enzyme-coupled secondary antibody (160 ng/l, goat-anti-mouse antibody coupled to HRPO (horse radish peroxidase), Co. Jackson Laboratories) in sample buffer (PBS + 10 % FCS, 0.1 % Tween20) was added for detecting the specific antibody binding. This was incubated for 1.5 h at RT. After another 4 wash steps for removing unbound secondary antibody- enzyme conjugate (4 x 300 μl PBS + 0.05 % Tween20) the specifically bound test- or control antibody was determined by the addition of the substrate TMB (tetra methyl benzidine, Sigma) and the subsequent incubation for 30 min at RT. After ending the enzymatic reaction with 50 μl 1 M HCI the strength of the enzyme-dependent staining reaction was determined in an ELISA reader by determining the absorption at 450 nm. The results of such an ELISA analysis are shown in Figure 6. It is clearly recognizable that the antibody mAB-X1 has a markedly enhanced binding to the activating epitope in comparison to the binding to the control peptide (i.e. to the peptide of the sequence MDCASCRARPHSDFC). Accordingly, mAB X1 is clearly identified as a binder of the activating epitope. Furthermore, the analysis of the binding of a control antibody shows that this binding and the used test are specific. The control antibody shows a low binding to the control peptide and to the peptide that represents the activating epitope. Antibodies according to the invention are preferably identified by this ELISA method.

In order to confirm that the peptide with the sequence MDCASCRARPHSDFC that represents the activating epitope of Fn 14 is recognized by an Fn14-specific antibody further competition experiments can be performed. These experiments are based on the reduction or prevention of the specific binding of activating anti-Fn14 antibodies to cell surfaces by adding peptides that represent the activating epitope. On the other side, control peptides or peptides that represent other epitopes, cannot influence this binding.

Figure 5 shows the results of these competition experiments. The activating epitope in this experiment is again represented by the peptide with the sequence MDCASCRARPHSDFC. For performing the experiment the activating antibody (1 μg/ml) was added to the test peptide in concentrations of 0.25, 0.5, 0.75, 1 mg/ml peptide (in 10 % DMSO) in 50 μl 1 x PBS + 0.5 % BSA, 0.02 % Na-azide (= PBA buffer) and incubated for 15 h at 4⁰C. Subsequently, 1.3 x 10⁵ HeLa cells that carry the receptor on their surface received 200 μl PBA with 50 μl of the antibody peptide (protein) mixture, incubation for 30 min on ice to allow for antibody binding, then pelleting, the supernatant is removed and the cell pellet is washed with 200 μl PBA to remove unbound antibody. Then, AlexaFluor488 conjugated goat-anti-mouse antibody (Co. Molecular Probes) was added in a concentration of 5 μg/ml, incubation for 30 min on ice, then the cells were washed twice with 200 μl PBA and then the number of antibody-bound cells as well as the strength of the binding to these cells was determined by FACS analysis. For doing this, the cells were measured in a volume of 200 μl in the FACScalibur (Becton Dickinson) in the FL1-H canal without using "gates".

The results of these FACS analysis are shown in Figure 5. In Figure 5 A it is clearly recognizable that the antibody mAB-X1 is inhibited by the peptide with the sequence MDCASCRARPHSDFC that represents the activating epitope of Fn14 in its binding to the cells. The fact that this inhibition takes place in a dose-dependent manner (Figure 5B) is a further indication for the specificity of the antibody binding and the specific competition of the peptide with the sequence MDCASCRARPHSDFC that represents the activating epitope.

Control peptides with other sequences of Fn 14 or completely different proteins demonstrated in these experiments as expected no significant shift of the FACS signal to the left - in the direction of reduced labelling of HeLa cells by the antibody - whereas the addition of the complete extracellular domain of Fn 14 in a fusion with Fc protein (Co. R&D Systems; employed amount 10 μg/ml) also led to a pronounced shift of the signal to the left (Figure 5A).

The results of an immunization experiment in mice is shown in Figure 10. Mice were immunized with a peptide with the sequence MDCASCRARPHSDFC or with a control peptide comprised of a scrambled version of the above sequence. Immunizations were performed at days 0, 14 and 28. Bleedings were taken at days 0 (pre-immune serum) and on days 38, 66 and final bleeding on day 87. As can be seen in Figure 10, the serum from the immunized mice is able to inhibit accumulation of lipid and differentiation of the 3T3-L1 pre-adipocytes to mature adipocytes, while both the no-serum control and the serum from the same mice prior to immunization with the peptide comprised of the epitope sequence MDCASCRARPHSDFC show no effect on fat cell differentiation and lipid accumulation. First this shows that the epitope sequence claimed in this invention is able to produce agonistic antibodies directed against the Fn 14 receptor and that the serum containing these antibodies is able to inhibit adipocyte differentiation. Second, this approach outlines the feasibility of using peptides containing this epitope sequence for vaccination studies and therapies for the treatment of obesity.

Claims

TRANSLATED CLAIMS 1 to 25

1. Use of an Fn14 activating agent, preferably an anti Fn14 antibody (anti-Fn14-Ab) for the preparation of a pharmaceutical composition for the prevention and/or treatment of obesity and/or obesity associated diabetes (type II) and/or metabolic syndrome and/or other obesity associated disorders.

2. Use of an Fn14 activating agent according to claim 1, wherein the agent causes an activation of the Fn14 signal pathway.

3. Use of an Fn 14 activating agent according to any of claims 1 or 2, wherein the activation takes place alone or in combination with other lymphotoxin β-receptor activating agents.

4. Use of an Fn 14 activating agent according to any of claims 1 to 3, wherein the agent inhibits the differentiation of pre-adipocytes to mature adipocytes and/or the storage of lipid molecules in differentiating and/or already mature adipocytes and/or promotes lypolysis.

5. Use of an Fn 14 activating agent according to any of claims 1 to 4, wherein the agent comprises an agonistic anti-LT-β-R antibody.

6. Use of an Fn 14 activating agent according to any of claims 1 to 5, wherein the agent is directed to a least one activating epitope of Fn 14, which has the sequence according to SEQ ID NO: 4.

7. Use of an Fn 14 activating agent according to any of claims 1 to 6, wherein the agent comprises polyclonal and/or monoclonal anti-Fn14 antibodies.

8. Use of an Fn 14 activating agent according to any of claims 1 to 7, wherein the agent comprises functional variants and/or derivatives of anti-Fn14 antibodies, in particular molecules with similar binding properties such as chimeric, recombinant, humanized or otherwise modified antibodies, antibody fragments, single domain antibodies and other specific binding proteins.

9. Use of an Fn 14 activating agent according to any of claims 1 to 8, wherein the agent comprises an agonist, in particular a protein with similar binding properties as antikalin, affilin or a non-protein such as aptamer.

10. A pharmaceutical composition comprising a β-receptor activating agent as defined in one of claims 1 to 9.

11. A peptide, characterized in that it has at least one of the activating epitopes of Fn14 selected from the group of sequences according to SEQ ID NO:4 or a peptide which comprises a sub-sequence of at least five amino acids of SEQ ID NO:4 with the proviso that the peptide is not the complete sequence of Fn 14.

12. An antibody characterized in that it binds a peptide of SEQ ID NO:4.

13. An antibody according to claim 12, characterized in that it activates Fn14.

14. Use of a peptide according to claim 11 for the preparation of an activating anti- Fn14 antibody.

15. A pharmaceutical composition comprising a peptide according to claim 11.

16. Use of a peptide according to claim 11 for the preparation of a vaccine.

17. Use of a peptide according to claim 11 for the preparation of a vaccine for the prevention and/or treatment of obesity and/or obesity associated diabetes (type II) and /or metabolic syndrome and/or other obesity associated disorders.

18. A method for the identification of an Fn 14 activating agent, comprising the step of contacting of the agent(s), which has (have) to be tested, with at least one substance comprising an activating epitope of Fn14, preferably a peptide according to claim 11 and detecting of an interaction of the agent with the epitope.

19. A method according to claim 18, characterized in that the substance is expressed from a cell and the phenotype of the cell is determined in the presence or absence of said agent.

20. A method for identifying an Fn 14 activating agent, wherein the activation of Fn 14 is effected by the binding of the agent to at least one activating epitope of Fn14, preferably a peptide according to claim 11 and the method comprising the following steps:

a) sewing undifferentiated pre-adipocytes and cultivating until post confluence,

b) inducing the cells and simultaneously incubating the cells activated for differentiation with agents which are to be tested,

c) determination of the level of differentiation of the cells, wherein the determination comprises the detection of intracellular lipid levels,

d) identifying of an active agent using the deviation of the test signal strength compared to a negative control,

e) testing of the positively identified agent of step d) for specific binding to an activating epitope of Fn14 or a peptide according to claim 11.

21. A method according to any of claims 18 to 20, characterized in that the cell is a mammalian cell, preferably of human origin.

22. A method according to any of the claims 17 to 20, characterized in that the cell is a primary pre-adipocyte or a primary adipocyte or stems from primary pre- adipocytes or primary adipocytes.

23. A method according to any of claims 17 to 22, characterized in that the change of the biological activity of pre-adipocytes or mature adipocytes, preferably 3T3-L1- cell or human pre-adipocytes or adipocytes, is determined.

24. A method according to any of claims 17 to 23, characterized in that the change of the biological activity consists of a inhibition or decrease of the fat storage in cells, preferably pre-adipocytes, differentiating adipocytes or mature adipocytes.

25. A method according to any of claims 17 to 24, characterized in that the fat storage is measured via determination of intracellular stored lipids.