PT1516054E - Fusion constructs containing active sections of tnf ligands - Google Patents

Description

DESCRIPTION

FUSION CONSTRUCTIONS CONTAINING ACTIVE SECTIONS OF TNF LIGANDS Several receptors which belong to the class of TNF receptors are known in the prior art and which in each case interact with at least one TNF ligand as a physiological ligand. TNF receptor family receptors are type I membrane proteins (Nagata et al., Science, 267: 1449, 1995). An example of a well-investigated TNF-TNF receptor system is the Fas receptor (Fas, FasR, CD95), which interacts with the natural ligand FasL and thereby induces an intracellular signal. The importance of the FasL / FasR system for cell-specific cell death has been, in particular, extensively investigated. The rat Fas ligand (Suda et al., Cell 75: 1169, 1993; Lynch et al., Immunity 1: 131, 1994) and the human ligand form (Takahashi et al., International Immunology 6: 1567, 1994) were cloned at the cDNA level. As a member of the TNF ligand family, FasL belongs to the category of type II membrane proteins, i.e. FasL has an extracellular carboxy terminal domain and an intracellular amino terminal domain. TNFÎ ± proteins (tumor necrosis factor a) and TNFβ (tumor necrosis factor β, Eck et al., Journal Biological Chemistry 264: 17595, 1989), for example, also belong to the TNF ligand family. Each of the TNF ligands binds to its physiological TNF receptor. Other examples of such prior art ligands are OX40L (binds to OX40R), CD27L (binds to CD27R), CD30L (binds to CD30R), RANKL (binds RANK-R), CD40L (binds to TRAIL-R1, R2, R3 or R4) and TWEAK (binds to Fn14). 1

However, native forms of the ligands belonging to the family of membrane type II proteins are not suitable as such for medical use. As membrane proteins, they also can not be administered as such, due to the hydrophobic transmembrane domain, in particular.

Thus, attempts have been made in the prior art to obtain fragments of TNF ligand which are still able to exhibit the physiological effect, but which do not have the intracellular segments, or the transmembrane domain. For example, in vitro and in vivo experiments were performed using FasL fragments which only exhibit regions of the FasL domains, which are extracellularly arranged in the native state (sFasL, i.e., soluble FasL). However, these protein fragments were only able to fill in the physiological function of the ligands inappropriately, particularly in the case of FasL, sometimes even observing that the soluble extracellular FasL domain exhibited unwanted effects which were the inverse of the FasL transmembrane protein, which is evidently present in the active form as a trimer under physiological conditions. US6316256 discloses fusion proteins having on the one hand a Fas ligand as well as an Fc region. US6046310 also describes Fc: Fas ligand fusion proteins. Bulfone-Paus et al. (Transplantation, 2000, 7: 1386-1391) disclose an IL-2-IgG-Fas-Ligand fusion protein. Fanslow et al. (Seminars in Immunology, 1994, 5: 267-278) disclose a CD40L-Fc fusion protein. Baum et al. (EMBO J. 1994, 13: 3992-4001) disclose a soluble murine fusion protein OX40L-Fc. WO 01/83525 discloses fusion proteins containing an Fc portion and a portion of a ligand of the TNF family. WO 00/49866 discloses fusion proteins containing a portion of a TNF cytokine and a multimerized portion of a protein selected from the group consisting of the Clq protein family and containing Collectin. Holler et al. describe Fas: COMP and CD40: COMP fusion proteins. The object of the present invention is therefore to provide sequences which not only mimic and reproduce the physiological effects of TNF ligands, for example FasL, but also, are soluble and therefore suitable for use as pharmaceutical agents, in particular for producing medicaments . The invention provides fusion constructs, i.e. either nucleotide sequences or protein sequences derived from nucleotide sequences, which also enable, in particular, the elimination of the phenotype from genetically determined diseases. The fusion constructs of the type according to the invention typically have a structure as set forth in claim 1. According to the invention, the immunoglobulin (Ig) segment, which is located at the N-terminus in the fusion construct, i.e. the component (A), does not exhibit according to the invention the variable region which is characteristic of immunoglobulins and which is responsible for antigen recognition, but instead exhibits only domains or domain segments of the immunoglobulin constant region by for example the CH 2, CH 2 and / or CH 3 domain (s). Accordingly, it is possible according to the invention that the segments of these CH domains according to the invention as components (A) are linked to each other, i.e., for example, the CH1 domain and the CH3 domain, being that component (A) should not, in accordance with the invention, lose the ability to bind to physiological Fc receptors when used in the preparation of a medicament for in utero treatment. In the case of postnatal treatment, the situation appears to be such that, where indicated, it will be advantageous if the constructs according to the invention do not bind to Fc receptors. Thus, it may be preferred, in the case of postnatal therapy, that component (A) specifically no longer has the property of functionally binding to Fc receptors. This loss of function can be achieved, for example, by insertion, deletion or substitution of functional Fc sequences. The immunoglobulin sequence which is derived from the Fc region and which is present as component (A) in the fusion construct according to the invention is capable of dimerizing with another fusion construct. It is preferred that component (A) is able to dimerize with a melt construction according to the invention, which is preferably identical or, alternatively, with another melt construction according to the invention, for example a construction containing another component (B). Dimerization may occur through a disulfide bridge, using the hinge region which in a native state is located between the CH1 domain and the CH2 domain, or by an artificially introduced sequence (or, for example, replacement / insertion of a cysteine ) is capable of covalently (disulfide bridging) or non-covalent dimerization (e.g., leucine zipper or other sequence segments which are suitable for dimerization).

In a preferred embodiment, the constant region of the antibody in the fusion protein will be of human origin, for example, from the GI2765420 antibody, and belongs to the family of 4 IgG class immunoglobulins, especially the IgG1, IgG2, IgG3 or IgG4 classes of preference of the IgG2 or IgG4 class. Alternatively, it is also possible to use immunoglobulin constant regions belonging to the IgG class of other mammals, in particular rodents or primates; however, it is also possible according to the invention to use constant regions of the immunoglobulin classes IgD, IgM, IgA or IgE. Typically, the antibody fragments that are present in the construct according to the invention will comprise the CH3 domain of Fc, or parts thereof and at least a segment part of the CH2 domain of Fc. Alternatively, it is also possible to design fusion constructs according to the invention which contain the CH3 domain and the hinge region for dimerization as component (A).

However, it is also possible to use derivatives of the immunoglobulin sequences found in the native state, in particular variants containing at least one substitution, deletion and / or insertion (included herein in the term " variant "). Typically, these variants have at least 90%, preferably at least 95%, and most particularly preferably at least 98% sequence identity with the native sequence. Variants which are particularly preferred in this context are substituent variants which typically contain less than 10, preferably less than 5, and most particularly preferably less than 3 substitutions compared to the respective native sequence. Particularly preferred are the following possible substitutions: Trp with Met, Val, Leu, He, Phe, His or Tyr, or vice versa; Ala with Ser, Thr, Gly, Val, Ile or Leu, or vice versa; Glu with Gin, Asp or Asn, or vice versa; Asp with Glu, Gin or Asn, or vice versa; Arg with Lys, or vice versa; Be with Thr, Ala, Vai or Cys, or vice versa; Tyr with His, Phe or Trp, or vice versa; Gly or Pro with one of the other 19 native amino acids, or vice versa. Particular preference is given for use in the preparation of medicaments for in utero treatments, to modifications of the above nature in which the binding to Fc receptors located on the membrane at least is not prevented and is even where appropriate optimized. However, when constructs according to the invention are used for postnatal therapy, preference is given to minimizing or even abolishing the ability to bind to Fc receptors. Since the fusion constructs according to the present invention should contain as component (A) sequences which must retain the ability to bind to respective physiological receptors for the constant segment of immunoglobulins, amino acids, in particular at positions 230 to 240, more preferably at positions 234 to 237, of the CH2 domain on the antibody fragment used as component (A) will not be consequently altered, or will only be provided with sequence variations which do not impair binding behavior towards Fc receptors. On the other hand, particular preference is given to substitutions or deletions, compared to the native sequence, of immunoglobulin constant regions at positions which result in the elimination or insertion of glycosylation sites, insertion or elimination of disulfide bridges, stability damage or solubility, or enhancement of passage through the cell membrane, upon binding to the Fc receptors. Preference is given in particular to variants which exhibit no structural change, or exhibit only negligible structural changes, as compared to the three-dimensional winding. This structural identity (and hence functional homology) can be determined by graphs of corresponding spectra, for example a circular dichroism spectrum of the native sequence, or a given variant. The shape of the spectrum, in particular in a measurement range between 190 and 240 nm wavelength, can always be used to establish the structural identity. In this regard, see in particular the relevant literature from the manufacturers of CD measuring instruments (e.g. Jasco, Japan). C-terminally to the immunoglobulin fragment, a fusion construct according to the invention typically, but not necessarily, contains a transition region between components (A) and (B), which in turn may contain a linker sequence, which is preferably a peptide sequence. This peptide sequence may have a length of from about 1 to 70 amino acids, where appropriate, up to more amino acids, preferably from 10 to 50 amino acids and particularly preferably from 12 to 30 amino acids. Examples of preferred, particularly strong transition sequences are shown in FIGS. lb to lj (correspondingly marked in these figures). The linker region of the transition sequence may be flanked by further short peptide sequences which may, for example, correspond to DNA restriction cleavage sites. In this regard, any restriction cleavage site familiar to the person skilled in the art of molecular biology may be used. Suitable linker sequences are preferably artificial sequences which contain a high number of proline residues (for example at every second position in the linker region) and, in addition, preferably have a hydrophilic overall character. A linker sequence consisting of at least 30% proline residues is preferred. The hydrophilic character may preferably be achieved by means of at least one amino acid with a positive charge, for example lysine or arginine, or negative charge, for example aspartate or glutamate. In general, the linker region also preferably contains a large number of glycine and / or proline residues in order to confer the necessary flexibility and / or stiffness on the binder region.

However, native sequences, for example the ligand fragments belonging to the family of TNF ligands that are arranged extracellularly, but located immediately in, ie in front of the cell membrane, are also suitable for use as linkers, where appropriate after substitution , deletion or insertion of the native segment. These fragments are preferably 50 AAs which follow extracellularly after the transmembrane region, or subfragments of these first 50 AA. However, it is preferred that these segments have at least 85% sequence identity with the corresponding natural human sequences, most particularly preferably at least 95% sequence identity and particularly preferably at least 99% sequence identity, so to limit the immunogenicity of these linker regions in the fusion protein according to the invention and does not trigger any intrinsic humoral defense reaction. In the context of the present invention, the linker region should preferably not have any immunogenicity.

However, as alternatives to peptide sequences that are attached to the antibody fragment and to the TNF ligand, or a fragment of this TNF ligand, by amide bonds, it is also possible to use compounds having a non-peptidic or pseudopeptide nature, or are based on non-covalent bonds. Examples which may be mentioned in this regard are, in particular, N-hydroxysuccinimide esters and heterobifunctional linkers, such as N-succinimidyl-3- (2-pyridyldithio) propionate (SPDP) or similar cross-linking agents.

In a fusion protein according to the invention and in the direction of the carboxyl terminus, the (B) component is followed to the transition region. In the context of the present invention, fragments containing, in particular, the part of the extracellular segment of the TNF-EDA1 ligand that is closest to the carboxyl terminus are preferred. Regarding the native sequences of the extracellular segments of the above proteins reference is made to the SwissProt database (http://www.expasy.ch/sprot/sprot-top.html).

With respect to EDA1, amino acid fragments 140 to 391, or fragments which are more truncated at the N-terminus (eg 157, 160, 181 or 182 to 391), in particular 200 to 391, most particularly 245 to 391, or fragments which are further truncated at the N-terminus (eg 246 to 391). However, all sequence segments which are located in the above ranges as preferred may also be used in the N-terminal region of the sequences which are present as component (B) in the fusion protein according to the invention, wherein the fusion protein according to the invention typically terminates at the C-terminus with the native C-terminus of the TNF-ligand EDA1. However, it is also possible, where appropriate, that at least one amino acid, typically 2 to 10 amino acids, in rare instances still more than 10 amino acids, be deleted in the C-terminal region of the extracellular TNF ligand fragment which is used as component (B) in the fusion protein according to the invention. The TNF-EDA1 ligand which is present in the fusion protein according to the invention may also contain at least one substitution, deletion and / or insertion, as compared to the native sequence, for the purpose of expressing the desired biological effects, such as solubility, stability or immunogenicity. As an alternative to the native TNF ligand sequence it is also possible according to the invention to use as component (B) in the fusion protein according to the invention, a variant which typically exhibits at least 75%, preferably 85%, and particularly preferably at least 95% sequence identity to the native sequence and which has the same biological effect and is therefore functionally homologous in this regard. This structural identity (and hence functional homology) can be determined by graphically representing suitable spectra, for example a circular dichroism spectrum of the native sequence, or the specific variant. The shape of the spectrum, in particular in a measurement range between 190 and 240 nm wavelength, can always be used to establish a structural identity. In this regard, reference may be made in particular to the relevant literature of the manufacturer of CD measuring instruments (e.g. Jasco, Japan).

Variants which are particularly preferred are substitution variants which typically contain less than 10, preferably less than 5, and most particularly preferably less than 3 substitutions, compared to the respective native sequence. It should be noted that the following substitution possibilities are preferred: Trp with Met, Val, Leu, Ile, Phe, His or Tyr, or vice versa; Ala with Ser, Thr, Gly, Val, Ile or Leu, or vice versa; Glu with Gin, Asp or Asn, or vice versa; Asp with Glu, Gin or Asn, or vice versa; 10

Arg with Lys, or vice versa; Be with Thr, Ala, Vai or Cys, or vice versa; Tyr with His, Phe or Trp, or vice versa; Gly or Pro with one of the other 19 native amino acids, or vice versa.

A further object of the present invention is the use of a fusion protein according to the invention for the preparation and formulation of a medicament for the reversal of genetically determined diseases. This use according to the invention can be applied to all placentas, ie vertebrates having placenta, in particular in human and veterinary medicine. After diagnosis of a genetically determined disease in an embryo, for example by chorion biopsy or amniocentesis, or when a genetically determined disease is suspected in an embryo, based on the relational genetic disposition, in particular the parent and / or mother, the The method according to the invention is suitable for treating the embryo prophylactically and reversing its hereditary phenotype. The use for preparing a medicament for treatment comprises a fusion protein according to the invention as described herein, where appropriate, in a corresponding formulation and is ideally administered to the mother or the mother animal as early as possible during pregnancy. This fusion protein according to the invention is advantageously administered parenterally, preferably intravenously or intraarterially.

After administration of a pharmaceutical agent, a fusion protein according to the invention and which is of the type described above, binds according to the invention, via its Fc portion, as component (A), to Fc receptors in the placenta and thus, after being internalized, reaches the embryo, typically through the placenta vessels that attach the embryo to the maternal blood circulation. The dose depends on the genetic disease itself and the time of administration (i.e., the embryo development stage), and the drug produced is suitable for treatment as early as possible during the development of the embryo. The medicament prepared enables the administration of Fc: ligand constructs according to the invention, eg FcrEDAl or Fc: EDA2, at least once more, more preferably, regularly during the first, second and / or third month of pregnancy, most particularly preferably , for example, every second day for a period of at least 14 days in the case of a human embryo, where appropriate, however, at longer intervals as well as depending on the dose being chosen.

In principle, however, the dose of a fusion construct according to the invention depends on the method of treatment. In the case of a use for the preparation of a medicament for treatment during embryonic development, typical doses of the administered construct of less than one tenth, preferably less than one hundredth and even more preferably less than one thousandth of the native concentration of the ligating TNF in the newborn. In the case of a use for the preparation of a medicament for treating a newborn or child with a fusion protein according to the invention, the doses will be at least 1/100, more preferably at least 1/10 (always with based on TNF ligand activity) of the serum concentration which is typical of a healthy patient of the same age. Particularly preferred is the use of a fusion protein according to the invention for the formulation of a medicament in a said method when the construction according to the invention administered has an Fc portion at the N-terminus of the fusion construct and the as a component (B), a TNF ligand segment EDA1, in particular EDA1 of human origin, in particular an extracellular segment of this protein which extends from at least amino acid 200 to amino acid 391 (EDA1), being very particularly preferred from amino acid 245 to 391 (EDA1). Still further preferred are the fusion protein constructs according to the invention as shown in Figure 1d, for a corresponding use.

In veterinary medicine, these fusion constructs according to the invention can be used to reverse the progeny phenotype of Tabby mice by treating the mice according to the method during pregnancy. Thus, fusion proteins according to the invention may also be used to produce medicaments which are also suitable for the therapeutic treatment of the analogous human disease which is in an inherited disease which is based on the same genetic defect, namely coupled hypohydrotic ectodermal dysplasia to X (XLHED, Christ-Siemens-Touraine syndrome), the most frequent form of ectodermal dysplasia. XLHED is due to a defect in the EDI gene, and this defect corresponds in human medicine to the following clinical picture of disease: hyposomia, variable degrees of dementia, saddle nose, anhidrosis, absence of teeth or abnormal development of teeth, alopecia, or dysfunction or absence of eccrine sweat glands, which is why affected patients are extremely susceptible to hyperthermia. In addition to XLHED, other ectodermal dysplasias which are based on other hereditary causes may also be treated with the medicament according to the invention. The present invention therefore also relates to the use of fusion constructs according to the invention for the preparation of a medicament for treating diseases genetically determined in humans or animals. The fusion constructs according to the invention, in particular constructs containing segments of the TNF ligand EDA1, for example the fusion constructs according to Figure 1d, are particularly suitable, generally, for the preparation of a medicament for treating genetically determined skin abnormalities, in particular skin structure abnormalities or anomalies of ectodermal structures such as abnormalities of the hair, teeth, or glands (anhydroses or dissidroses of various genes). An example which may be mentioned is the use of the fusion proteins according to the invention for the preparation of a medicament for use in the cosmetic area, for example for treating alopecia, whether acquired (for example in the form of alopecia areata, alopecia atroficans, seborrheic alopecia or premature alopecia) or congenital (especially as anhydrosis syndrome with hypotrichosis) or hirsutism. However, the fusion constructs according to the invention can also generally be used to produce a medicament for treating all forms of ectodermal dysplasia. For this reason, the fusion constructs according to the invention, in particular those involving EDA1, are particularly suitable for producing a medicament for the treatment of hair growth stimulus or inhibition of hair growth, for treating disorders in the growth of hair. teeth, or to treat disorders in the function of glands, including that of sweat glands and / or sebaceous glands, or are suitable for promoting the formation of hair follicles or to heal wounds, in particular to promote healing in patients, for the purpose of ensure the growth of adequate hair in the area of affected skin. The use of such fusion constructs for the preparation of a medicament for administration to a patient, preferably postnatally, very particularly preferably in human therapy in the first year of life, still more preferably in the first 6 months of life, still more preferably in the first 2 months of life, even more preferably in the first 2 weeks of life and even more preferably in the first 3 days after birth, or for administration to the pregnant mother (or to the pregnant mother animal) eg by injection in uterus. The fusion constructs according to the invention may therefore also be used for the preparation of a medicament for treating the aforementioned syndromes, genetic disorders or disorders during embryonic development, by administering them to the mother or the mother animal, with the exception of processes in that the genetic identity of the germ line of the human being is modified.

When constructs according to the invention are used to produce a medicament for treating, in particular, alopecia, or, in particular, for promoting wound healing, for example after skin burns or inflammatory skin diseases, eg in the treatment of areas of skin affected by neurodermatitis or atopy, in particular of the scalp, it is also appropriate to administer such constructs where appropriate using other auxiliary substances, carrier substances, or additives as well as other active compounds where appropriate (eg in the treatment of alopecia: Relaxin, substances having an antiandrogenic effect, eg finasteride, SKL-105657, estrogen, cyproterone acetate, spironolactor, flutamide, minoxidil and / or RU58841) to adults, for example by intravenous or intraarterial subcutaneous administration, or subcutaneous administration, after appropriate processing (formulation), orally, or as a cream, lotion or a for topical use.

Another possible area of use of the constructs according to the invention is in the production of ex vivo skin tissue. For example, the material according to the invention, for example in the form of proteins, nucleic acids or expression vectors, can be used in order to (re) generate the natural skin functionality of patients suffering from skin diseases (e.g. eg diabetic ulcers) or burns, with the formation of eccrine glands, hair follicles and hair. In this context, the material according to the invention is added to the cells or tissues in vitro. For example, the material according to the invention may be used in the traditional method of skin grafting; ie, healthy, autologous skin tissue from the patient, allogeneic skin tissue from deceased persons or donors, or animal tissue, may be stimulated prior to being transplanted and by treatment with the fusion constructs according to the invention for forming hair follicles or eccrine glands, for example in the context of a pretreatment or in the context of extracorporeal culture. However, it is also possible to isolate multipotent precursor cells from the patient's own healthy skin tissue (or other tissue). These cells are formed into skin tissue by addition of suitable differentiation factors. These multipotent precursor cells may, by means of suitable transfections, be transfected with sequences or expression vectors according to the invention, or may be added to these fusion constructs during the formation of the skin tissue. 16

Accordingly, it is possible to use the material according to the invention in the so-called " tissue engineering " (tissue culture) before the cultured tissue is re-transplanted to the patient. For this reason, the tissue to be transplanted (precursor cell, stem cell or tissue cell) (in particular epidermal keratinocytes) or non-human embryonic stem cell (autologous or allogeneic) is in principle first isolated and then cultured in a in a matrix (natural, synthetic or xenogeneic) or in a biodegradable carrier, the cells being suitably provided with nutrients, oxygen, growth factors, and constructs according to the invention, the harmful metabolites being eliminated. The present invention therefore also relates to an in vitro method for culturing skin tissue which exhibits eccrine glands and natural hair growth.

Another object of the present invention relates to recombinant DNA constructs on which the fusion proteins according to the invention are based. In the context of the present invention the DNA and protein constructs are treated together by the term " fusion constructs ". Thus, a nucleotide construct according to the invention may, for example, code for the IgG CH2 and CH3 domains, as component (A), and EDA1, suitably linked by a transition region containing a ligand. Accordingly, the underlying recombinant nucleotide sequences according to the invention encode all the protein fusion constructs according to the invention which have been previously described. Particularly preferred are the nucleotide sequences of the fusion constructs described in Figure 1. The nucleotide constructs according to the invention may be obtained, as cDNA, genomic DNA or in synthetic form, using cloning methods or by means of chemical synthesis , various methods of the prior art being suitable. For example, with respect to the sequences of the domain of the immunoglobulin Fc portion that may appear in a nucleotide construct according to the invention, reference is made, for example, to the Sequences of Proteins of Immunologic Interest ", Ia Edition, Kabat et al., US Department of Health and Human Services, 1991, the content of which is incorporated herein in its entirety by reference.

Sequences contained in a nucleotide construct according to the invention will typically be present in segments that are assembled by restriction and ligation. A nucleotide sequence according to the invention will typically contain expression control sequences, including promoters, ribosome binding sites, polyadenylation sites and / or transcription termination sites and, where appropriate, enhancers that are operably linked to nucleotide sequence according to the invention. A fusion protein according to the invention may be expressed by transfection of DNA segments containing these regulatory elements, for example, into plasmid vectors, into bacterial cells, yeast cells, plant cells, insect cells and, preferably, mammalian cells, for example using the calcium phosphate method, electroporation or the like techniques. To obtain expression in eukaryotic cells, the promoter and, where appropriate, the enhancer sequence of, for example, immunoglobulin, SV40, retrovirus, cytomegalovirus, elongation factor, and so on are used. Preferred host cell lines include CHO cells, COS cells, Hela cells, NIH3T3 cells and various myeloma or hybridoma cell lines, including P2 / 0 and NS / 0. The present invention relates to a DNA sequence of this type according to the invention, which encodes a recombinant fusion protein according to the invention, to expression vectors containing a DNA sequence of this type according to the invention. invention and to host cells which are transfected with an expression vector of this type according to the invention. The plasmid vector comprising a DNA construct according to the invention will generally contain a selectable marker, such as gpt, neo, hyg or DHFR, and an amp, tet, kan, etc. gene. for expression in E. coli. A large number of plasmid vectors suitable for expression of heterologous proteins are available from the prior art. A fusion protein according to the invention advantageously contains an N-terminal sequence, for example a hemagglutinin sequence and / or signal sequences and / or at least one other sequence " LEADER " at the amino terminus of the fusion protein, so as to facilitate the secretion of the fusion proteins from the cell, in particular the passage through the ER into the extracellular space, or into the medium.

Methods for constructing nucleotide constructs according to the invention that encode protein constructs according to the invention for their binding to expression control sequences and / or their insertion into plasmids for transfection into cells and selection, and if desired amplification of the cell line genes expressing fusion proteins, can be performed using methods which are known from genetic engineering, including restriction enzyme digestion, ligation, oligonucleotide sequences and PCR reaction (Sambrook et al., 19

Molecular Cloning: Laboratory Manual, 3rd Edition, Cold Spring Harbor Laboratory Press, 2001), all content incorporated herein by reference.

A transfected cell line is selected which expresses and secretes a fusion protein according to the invention and may, where appropriate, be cultured in serum free medium (SFM hybridoma), passed through a decreasing serum concentration and may be , finally, be subcloned. A fusion protein according to the invention may be purified (preferably from serum-free medium) after being secreted, the expression cell line having been cultured. Conventional methods for purifying proteins include filtration, precipitation, protein A affinity chromatography, gel filtration, ion exchange chromatography, electrophoretic methods, or the like. Substantially pure fractions of the fusion protein, at least 90 to 95% homogeneity and preferably at least 98 to 99% homogeneity or more are used for pharmaceutical use. The production of a pharmaceutical agent comprising the fusion protein according to the invention will typically include its formulation in a pharmaceutically acceptable carrier material. This carrier material will typically comprise aqueous carrier materials, using water for injection (WFI), or phosphate buffered water, citrate, or acetate, etc., the pH typically being adjusted to 5.0 to 8.0, preferably 6 , 0 to 7.0. The pharmaceutically acceptable carrier will additionally and preferably contain salt constituents, e.g. sodium chloride or potassium chloride, or other components which make the solution isotonic. In addition, the carrier may contain additional components, such as human serum albumin, polysorbate 80, sugars or amino acids. A composition according to the invention of this nature may be combined in the form of an injection, or as a kit, in a suitable combination with the pharmaceutically acceptable carrier or mediator material, such as sterile water or physiological saline, vegetable oils, mineral oils, higher alcohols, higher fatty acids and non-toxic organic solvents and, where appropriate, other additives. Examples of additional additives which are suitable as fillers (for example saccharides (lactose, sucrose, mannitol or sorbitol), calcium phosphate, cellulose or derivatives thereof), binders (for example starch, gelatin and polyvinylpyrrolidone), stabilizers, preservatives, antiperspirants, surface active substances, colorants, flavorings, emulsifiers, buffering agents, isotonic agents, antioxidants, analgesics or the like. Sodium alginate, polyethylene glycol and / or titanium dioxide are, for example, preferably formulated in the formulations. Corresponding formulations of a pharmaceutical agent are described in Remington (1980), which can be consulted. The concentration of the fusion protein according to the invention in formulations of this nature may range in the range of from 0.001 to 10 mg per ml, preferably in the range of 0.5 to 5 mg / ml. The formulation is preferably injected into the patient or pregnant mother parenterally, i.e., for example, intravenously, intraarterially, subcutaneously or intramuscularly. Alternatively, when the treatment is performed using a method according to the invention, it is also possible to introduce the formulation directly into the vascular system of the embryo, placenta or to inject it into the amniotic sac. The present invention is best explained by the following figures: Figure 1 shows the nucleotide and amino acid sequences of Fc: ligand constructs. The Figure shows, by way of example and diagrammatically, the sequence segments of Fc: ligand constructs. A signal peptide is typically located in the N-terminal region of a fusion protein according to the invention, which is encoded by a corresponding nucleotide sequence, this signal sequence being followed by an Fc segment of IgG (component A), which is in turn followed by a (optional) linker region and, closer to the C-terminus, by a protease cleavage site sequence (optional) and closer to the C-terminus (and at the C-terminus of a fusion construct according to the invention) by a sequence which typically constitutes a sequence of part of a member of the TNF ligand family (component (B)). This partial sequence serving as component (B) in turn typically includes the C-terminal portion of the extracellular segment of a TNF family ligand.

Figures 1b to 1c show specific sequences of constructs according to the invention, wherein these constructs utilize different members of the TNF ligand family, namely FasL (Figures 1b and 1c, with and without a protease cleavage site), EDA1 and EDA2 (Figures 1d and 1d), TNFα (Figure 1f), CD40L (Figure 1g), TRAIL (Figure 1h), BAFF (Figure 1j) and APRIL (Figure 1i). The constituent sequences of the above-mentioned members of the TNF ligand family are all of human origin and in each case have the native ligand sequence segments located extracellularly, near the C-terminus. The Fc domain (component (A)), which is located near the N-terminus, is connected to the TNF ligand segment (component (B)), which is located near the C-terminus in the fusion protein, via a region linker containing the amino acids PQPQPKPQPKPEPE. In addition, the fusion construct sequences according to the invention shown in Figures 1, 2 and 1 contain a protease cleavage site, each having the sequence LEVLFQGP, so that the PreSCISSION protease (Amersham-Pharmacia) can cut fusion constructs which are designed with this protease cleavage site and thereby separating the immunoglobulin domain from the fusion construct of the TNF ligand segment. However, all sequences which are at least 4 AA in length and which can be recognized by any protease, for example a cysteine or an aspartate protease, are suitable for use as a protease cleavage site sequence. However, short sequences i.e. from 2 to 8 amino acids in length may typically be present between the Fc domain and the linker, or between the linker and the TNF ligand domain. The sequences in Figures 1b to 1j contain the amino acids ARG-SER between the Fc segment and the linker and, in Figures 1c, 1d, 1c, 1g and 1h, at least the tetrapeptide GSLQ, where appropriate, elongated by more amino acids near of the C-terminal of Q is located in the direction of the N-terminus of component (B). These fusion constructs also containing a protease cleavage site, i.e., the fusion constructs in Figures 1, 2, and 1, contain the above tetrapeptide, toward the C-terminus of the protease cleavage site. The linker and the cleavage site of the protease are typically also separated from each other by at least two amino acids, according to the Figure, by the dipeptide GS. The entire region between component (A) and component (B) is called the transition region.

At the N-terminus, i.e. toward the N-terminus to the Fc domain, all constructs shown in Figures 1 to 1] contain a pentadecapeptide (MAIIYLILLFTAVRG) which, in the cases shown, corresponds to a hemagglutinin sequence. In all cases shown, this signal sequence is attached to the Fc segment of the fusion protein (component (A)) via the dipeptide LD.

The linker sequences in Figures 1 to 4 which are not in phase typically correspond to restriction enzyme cleavage sites that are used to link the individual sequence components to each other.

All of the fusion constructs shown in the figures were cloned into a mammalian expression vector PCR-3 (Invitrogen). Figure 2 shows a photograph of a Western blot showing the purification of the Fc: FasL fusion construct according to the invention. The purified protein (5æg / well) was analyzed on a gel with 12% SDS-PAGE under reducing (+ DTT) and non-reducing conditions (-DTT) and using the TRAILR2: Fc fusion construct (which is a construct containing an immunoglobulin domain and a TNF receptor) as a control, on the columns indicated by a (+) marker above the Western blot. Molecular weight markers are indicated in kDa on the left side of the Western blot, while the molecular weights (110 kDa and 57 kDa, respectively), which, as expected, vary as a function of the formation of disulfide bridges (only under non-reducing conditions ) between the Fc domains, which are indicated for the Fc: FasL fusion construct according to the invention, are shown by arrows on the right side of the Western blot. Figure 3 shows the result of gel permeation chromatography. To this end, Fc: FasL (300æg in 200 μΐ PBS) was loaded onto a Superdex-200 column, which was equilibrated in PBS, and eluted at a rate of 0.5 ml / min. The UV profile was recorded online at 280 nm and 0.5 ml fractions were collected. The elution position is plotted below Figure 3 and the molecular weight (in kDa), and the calibration standards are represented graphically above the profile. The apparent molecular weights of the two Fc: FasL peaks were deduced from the calibration curve which is shown in the right-hand chart in Figure 3. The result is shown above the two peaks (850 kDa and, respectively, 440 kDa). The numbers of the fractions investigated are presented in the order in which they were eluted, below the profile. Figure 4 shows electron microscopic photographs, ie, in the order of the constituent Figures, of FasL (Figure 4a), ACRPA-FasL (Figure 4b), ACRP-FasL (Figure 4c), and finally Fc-FasL (Figure 4d). The individual constituent figures in each case contain different preparations of the injected amino acid sequences, in alternative aspects, with a diagrammatic interpretation of the figure determined underneath each photograph being determined by electron microscopy. In this context, the electron microscopic photographs shown in Figure 4a result in a dot pattern in the case of FasL (3 ch, which is the expected trimer of FasL multimerization), whereas in the case of ACRPA: FasL (Figure 4b), result in a rod-shaped model terminated at a point, wherein the point marks the component 25

FasL and the rod shape marks the ACRPA collagen domain. Figure 4c results, in the case of ACRP: FasL, in a double cherry structure composed of two attached rods (ACRP30 collagen domain with a bound FasL domain). Finally, the Fc: FasL structure according to the invention is shown in Figure 4d, where the light spots represent FasL and the dark spots represent the Fc portion of IgG1. The white line that unites them shows the connection between the different subunits. Figure 4e is a figure, in the form of a model, of the three-dimensional structure of an Fc: ligand complex. The structure of Linfotoxin A (PDB accession number 1TNR) was selected as ligand, whereas that of IgG 2a (PDB accession number 1IGT) was selected as the Fc component. Figure 5 shows the specificity of binding of Fc ligands to their respective receptors. In this case, binding of the ligands to their receptors (as receptor fusion protein.COMP) was measured by ELISA and using the following steps: (a) murine anti-human IgG coating, (b) addition of Fc ligands at the desired concentration (as cell supernatant), (c) addition of the respective receptor-COMP fusion construct (each of these transporters carries a signal marker and was also added as a cell supernatant), (d) addition of anti-monoclonal antibody (e) addition of streptavidin coupled to horseradish peroxidase (HRP) and (f) development of the assay with OPD reagent and measurement of absorbance at 490 nm.

It can be seen clearly from the diagrams shown in Figure 5 that the fusion constructs according to the invention have a strong specificity for the respective TNF receptors that are specific for the ligands (or ligand segments) that are contained in the constructs according to the invention, namely FasL, having a strong specificity for Fas, Trail for Trail R2, EDA1 for the EDA receptor, as expected BAFF and APRIL for the BCMA receptor, which is bifunctional in this respect, TNF for TNFR2 and CD40L for CD40R. Accordingly, the preparation as a construct containing an Fc moiety has no effect on the binding behavior of the component (B) of a construct relative to the corresponding receptor. Figure 6 shows the results of binding experiments using FcrBAFF. As can be seen in Figure 6a, BJAB cells (BAFF-R-positive cells and HEK 293 cells (BAFF-R-negative cells) were incubated with 0.05, 0.2, 0.5, 2.5, 20 or 50 μl of FcrBAFF-containing cell supernatants in a final volume of 100 microliters.The bound FcrBAFF was identified using PE-conjugated anti-human Ig antibody and performing a FACS analysis.The result of quantifying the mean fluorescence (MFI) is shown in the diagram to the right in Figure 6. In this regard, it can be clearly seen in the direct graphical representation for BJAB cells and HEK 293 cells analyzed on the secondary graph (rightmost in Figure 6a) that the mean fluorescence, as a measure of cell binding, increases markedly in the case of BJAB cells, whereas BAFF-R-negative HEK-293 cells exhibit no significantly increased mean fluorescence, even when high concentrations of FcrBAFF according to the invention are present. Consequently, as expected, FcrBAFF also does not bind BAFF-R-negative cells, unlike the situation concerning BAFF-R-positive cells. Figure 6b shows the results of immunoprecipitation experiments. For these, BJAB cells (5x10 7 per assay) were incubated for 15 minutes in 2 ml of medium in the presence of 3æg Fc: FasL, or Fc: BAFF. Cells were collected, washed in PDS, lysed in a lysis buffer (containing 1% NP-40) and immunoprecipitated using Protein A Sepharose. IPs and total cell extracts were analyzed by the Western blot technique using a murine anti-hBAFF-R monoclonal antibody. In this context, BAFF-R-positive BJAB cells were found to bind to Fc: BAFF during incubation (right-hand column in Figure 6b) and can therefore be detected with suitable antibodies in the IP assay. Figure 7 shows the cytotoxicity of Fc: FasL when serial dilutions of Fc: FasL according to the invention (as desired, the column-loaded material, the 850 kDa fraction, or the 440 kDa fraction, as shown in Fig. shown in Figure 3). These serial dilutions were added to FasL-sensitive Jurkat cells and incubated at 37øC for 16 hours. Next, cell viability was analyzed using a PMS / MTS assay. In this regard, it has been found that both the total fraction and the individual fraction have a very similar profile with respect to the cells' ability to survive. Figure 8 shows the phenotypic differences between tabby mice that were treated 10 days after Fc: EDAl birth according to the invention (see Figure 1d), or which were not treated. Wild type mice served as controls (right side). For this, pregnant mice were injected intravenously with 400æg of Fc: EDAl on days 11, 28, 13 and 15 of pregnancy. Progeny from treated tabby mice, untreated tabby mice and wild-type mice were analyzed. As can be seen in Figure 8a, this analysis took into account the aspect of the ear region, in particular the growth of hair in the ear region. The growth of hair in treated animals corresponds to that of wild type animals. The tail region in treated mice also exhibits hair as well as the form usually present in wild type mice. The difference from untreated mice is immediately obvious. Figure 8b shows illustrations of histological sections of tissue from the retro-atrial region, body side, tail and paraffin paw, using hematoxylin / eosin staining. A pronounced bald region can be seen under the ears of untreated tabby mice. The treated tabby mice resemble phenotypically those of the wild type. The side of the body in treated tabby mice exhibits a wild type increase in the number of hair follicles compared to untreated tabby mice. In the tail region, treated mice exhibit the hair growth observed in wild-type mice. On the other hand, untreated tabby mice have no hair in the tail region. As is the case in the wild type, sweat glands can be observed in the paw tissue in treated mice (medium), as indicated by the arrows. In contrast, untreated tabby mice do not develop any sweat glands. The following characteristics were compared: Figure 9 shows a comparison of the results that were obtained when tabby in utero mice were treated with the construct according to the invention, shown in Figure 1d with the results which were obtained in corresponding control animals. Figure 9a shows a comparison of treated tabby mice (center) and untreated tabby mice (left) in adulthood (6 weeks post-birth). The treatment occurred as described in implementation example 4. The appearance of healthy mice as a control is recorded on the right. The treated animals have a coating such as that of wild type animals (including in the critical tail region); in treated mice, the tail has no serrated appearance; the coat of the treated mice exhibits the " monotroic " hair type, which is typical of wild type mice, in particular especially monotroic long hair which is totally absent in the untreated tabby mice. Figure 9b compares the mandible and dental structure of treated and untreated tabby mice and wild-type mice. Tabby mice have an abnormal jaw structure and the teeth also exhibit deformities, in particular a few pronounced cusps. Treatment by a method according to the invention during pregnancy (treatment of the mother animal) allowed recovery of the natural shape of the teeth. Figure 9c shows tissue sections of the eyelid and cornea of Fc-EDAl-treated untreated tabby mice according to the invention and wild-type control animals. The untreated tabby mice characteristically do not possess the Moebius glands in the eyelids. However, these glands reappear after treatment according to the invention. Regeneration of the Moebius glands also prevents corneal keratinization occurring, as can be clearly seen in the tissue section shown in Figure 9c, in untreated tabby mice. Figure 9d shows the results of the sweat test, which shows that functional sweat glands that are totally absent in the untreated tabby mice can be observed in the paw in the case of tabby mice that were treated according to the invention. These sweat glands that have been reacquired after treatment with FcrEDAl allow the demonstration of physiological reactions typical of stimuli such as heat and stress. The sweat test itself is described in the implementation example 4. Figure 10 shows a comparison of the results obtained after the post-natal intraperitoneal treatment of tabby mice with a FcrEDAl fusion construct according to the invention, with the situation in tabby mice untreated and wild-type control animals. Figure 10 compares different regions of the body in the three groups of animals referred to above, in relation to their phenotype. Contrary to the wild-type situation, the skin region behind the ears was not covered with fur in postnatally treated, adult animals. However, the postnatal injection of FcrEDAl was able to recover almost entirely the hairs of the tail, as in the wild type, on the one hand, and, on the other hand, the size of the eye opening which is typical for the wild type. The postnatal treatment according to the invention was able to virtually eliminate the " folded " tail shape ". In addition, sections of tail tissue that were stained with hematoxylin and eosin showed the presence of numerous hair follicles and sebaceous glands associated with hair follicles. Figure 10b shows, microscopically and macroscopically, the phenotypes of the three groups in relation to the sweat glands 31 on the legs. A large number of sweat glands were observed in post-natally treated paws of adult tabby animals, these sweat glands being indistinguishable from wild-type sweat glands, as shown in Figure 10b, using hematoxylin and eosin stained tissue sections. These sweat glands in the treated tabby animals have been found to be functional in a suitable sweat test and to react with physiological stimuli such as heat or stress.

In summary, it can be said that when treatment occurs in utero, the reversal of the phenotype observed after birth does not disappear in the adult animal, such as the reversals observed in the young animal in relation to postnatal treatment with substances according to the invention also persist in the adult animal. The present invention is best illustrated by the following implementation examples.

EXAMPLES OF IMPLEMENTATION

Io Implementation example

Preparation of Fc fusion proteins: Ligand

Preferred fusion proteins according to the invention consist of an Fc portion derived from human immunoglobulin (component (A)), wherein the corresponding DNA sequence does not contain the immunoglobulin stop codon, and a C-terminal segment of a fragment of a ligand belonging to the TNF family (component (B)) (containing the stop codon of the corresponding DNA sequence). In this context, the preferred fusion proteins according to the invention have the following structure: an N-terminal signal peptide, then an Fc segment, for example IgG derivative, a transition region containing a linker region and a site protease cleavage, and a C-terminal segment of the TNF-ligand EDA1. A diagram of a preferred fusion protein according to the invention is shown in FIG. 1a.

In order to prepare a fusion protein according to the invention, for example a fusion protein with a preferred structure as shown in Figure 1a, in particular the sequences shown in Figures 1b to 1c, nucleic acid constructs were cloned into vectors of mammalian expression and expressed in embryonic human kidney cell lines (in HEK-293 cells or in CHO (Chinese hamster ovary) cells). The secreted recombinant fusion proteins were isolated from the conditioned medium by means of affinity chromatography on a Protein A Sepharose column. The yields were several mg per liter of the medium used.

2nd Fractional Characterization Biochemical characterization of Fc: FasL FasL was biochemically characterized by SDS-PAGE analysis, with an apparent molecular weight of 57 kDa, approximately 110 kDa determined under reducing and non-reducing conditions, respectively. This demonstrates the presence of dimers linked by disulfide bridges (2 ch), as was expected for Fc-containing fusion proteins having a hinge region (see Figure 2 also). 33

A corresponding investigation performed by gel permeation chromatography showed that they eluted two well-defined Fc: FasL peaks, with apparent molecular weights of approximately 440 kDa (main peak) and, respectively, approximately 850 kDa (secondary peak) (see Figure 3). Assuming that FcrFasL has a globular shape, the multimer structure calculated for FasL was 7.7 (7.7 ch) for the major species in the elution profile. However, if the protein is not typically globularly ideal, this calculation is most likely based on an overestimate of the molecular weight, which means that a dimeric ligand (6 ch) structure would be quite possible. The Fc: FasL corresponding to the 440 kDa peak was analyzed electron microscopically using FasL (3 ch), ACRP30: oligomeric FasL (6+ ch) and a deletion mutant of ACRP30: FasL, namely ACRPA: FasL (3 ch) for comparison. The FasL (3c) has a ball-like shape (see Figure 4a), while the ACRPh: FasL (3c) has a ball-like shape with a " (see Figure 4b) and ACRP30: FasL has the appearance of a " double cherry ", which conforms to a dimeric FasL (6 ch) structure (see Figure 4c). Fc: FasL generally assumes a ring structure having 5 " balls " or an open structure having 5 " balls " (see Figure 4d). This in turn corresponds to a dimeric FasL (6 ch), two of the FasL beads being the remaining 3 beads the Fc components of IgG (dimers of 3 polypeptides: 6 ch). A model representation of an Fc ligand was developed using the known crystal structure of lymphotoxin A (a member of the TNF family) and the Fc component of IgG2a (PDB accession numbers 1TNR and 1IGT), the linker region being constructed as a line (see Figure 4e). This representation unambiguously shows that the ligands 34 and the Fc moieties are approximately the same size, in accordance with the observed structure of 5 Fc: FasL beads. Overall, these results indicate that the 440 kDa peak of Fc: FasL is a dimeric FasL (6 ch).

In the present disclosure, the designation " ch " designates " chain " and therefore reflects the number of chains of the TNF ligand (eg APRIL, FasL, BAFF, EDA or Tweak) in the molecule (regardless of whether or not the fusion construct is). If a TNF ligand is present in trimeric form, it has three chains (3 ch), while a dimer of a ligand (for example as a result of the Fc construct according to the invention) has 6 ligand (6 ch) chains.

The other fusion protein sequences according to the invention shown in Figures 1b to 1j were characterized in an analogous manner. 3rd Implementation Example

Biological activity of the Fc constructs: in vitro ligand:

According to the invention, an ELISA-based assay was developed in order to investigate the biological properties of Fc: ligand constructs, namely their ability to bind to their corresponding receptors. In this assay, the Fc: ligand constructs were initially captured by a murine human anti-human IgG monoclonal antibody. In a second step, the soluble receptors were fused to the oligomerization domain of the cartilage oligomeric matrix protein (COMP) (see DE 19963859 and DE 10122140, all of which is incorporated herein by the disclosure), and a signal sequence . Receptors having ligated ligands were identified by means of their respective signal markers (see Figure 5). Using this modified ELISA assay approach, it was found that the Fc: ligand constructs according to the invention did indeed bind to their respective native receptors, i.e. they do not lose their ability to bind as a result of the construct structure and binding also occurs with a high degree of selectivity.

It has also been shown that according to the invention, the Fc-ligand constructs are also capable of binding to endogenous surface expressed receptors. It has been experimentally shown that although the FcrBAFF constructs according to the invention mark BJAB BAFF-R positive cell lines in a dose-dependent manner, they do not label the negative Jurkat BAFF-R cell line (see Figure 6a). In addition, FcrBAFF has been found to be capable of specifically immunoprecipitating the endogenous BAFF receptor in BJAB cells (see Figure 6b).

In addition, the cytotoxicity of FcrFasL in a FasL-sensitive Jurkat cell line was investigated. Fc: FasL was found to be highly cytotoxic in Jurkat cells, with an IC50 of 1 ng / ml (see Figure 7). In addition, according to the invention, there were no significant differences in relation to the specific activity between the complete Fc: FasL preparation according to the invention and the dimeric form of Fc: FasL (6 ch) which was isolated at the peak of 440 kDa and / or the oligomeric form of FcrFasL (12 μl) that was identified at the 850 kDa peak. These results demonstrate that dimeric FasL (6 ch) is already a fully active molecule and that larger complexes no longer increase activity in this respect. 36 4th Implementation example

Biological activity of Fc constructs: ligand in vivo:

Fc: FasL and Fc: EDA fusion proteins were tested in vivo. Injection of Fc: FasL into mice showed that this fusion protein exhibits cytotoxicity in vivo.

Fc fusion construct: EDA: EDA is a member of the TNF ligand family which is physiologically responsible for development and function and also in particular for the morphogenesis of ectodermal structures, most particularly for sweat glands, hair and teeth as epithelial structures of the ectoderm. In contrast, it is known that EDA malfunction causes HED (XL) syndrome in humans, whereas in mice, EDA malfunction leads to what is termed Tabby phenotype, which is essentially characterized by (i) the coating (iii) loss of hair on the tail and skin region behind the ears, (iv) absence of sweat glands on the paws, (v) abnormal formation of teeth and eruption of teeth (vi) characteristic distortion at the tip of the tail, (vii) absence of Moebius glands (glands in the eyelids that secrete a thin film of fat to protect the cornea from drying, thus occurring keratinization of the cornea) and other glands, (viii) reduction in eye opening size and (ix) lower weight gain in the young mouse.

According to the invention, 400æg of Fc: EDAl in utero were injected into female tabby mice at the 11th, 13th and 15th days of pregnancy. This fusion protein was extremely well tolerated and no counter-indicative effects were observed in the treated mice. Although Fc: EDAl treatment did not give any recognizable effects on the treated, female adult mice themselves, a definite and potent reversal of the tabby phenotype was observed in the progeny of treated pregnant mother animals. This reversal of the phenotype was evident on the 10th day after birth. The progeny size of the tabby mice that were treated during pregnancy was substantially larger than those of control mice, with their coatings being soft, black and shiny (see Figures 8/9). In contrast to the situation in the tabby mice, it was also found that the retro-auricular region in the treated tabby mice was hairy and that these mice have normal hairy tails having a large number of hair follicles and the corresponding sebaceous glands, which it was possible to observe on the skin of the tail region (see Figure 8b). In addition, it was possible to discern eccrine sweat glands, which were indistinguishable from the corresponding wild-type sweat glands, in tissue sections of the paws of treated but not untreated tabby mice (see Figure 8b). In tabby mice that were treated in the embryonic state in utero, these sweat glands were still functional in adulthood. A sweat test was performed to determine the functional capacity of the glands: a 3.5% (w / v) solution of I2 in ethanol was rubbed onto the legs of the investigated animals and allowed to dry . Thereafter, a solution of starch in 50% (w / v) mineral oil was applied to the legs. A photograph was taken after applying moderate heat using a hair dryer (5 s). The animals were sacrificed and sections of eyelid and corneal tissue stained with hematoxylin and eosin were prepared. The reversed phenotype was stable, as confirmed by analysis of the adult treated tabby mice (see Figure 9). These mice were significantly larger than tabby mice and their coatings particularly have the longer, monotroic wild type hair variety which is completely absent in the tabby mouse. These treated mice did not have hairless flaps in the retro auricular region and the legs had eccrine sweat glands that were indistinguishable from the wild type. In addition, these sweat glands were functional (Figure 9a), as demonstrated in the sweat test described above. After treatment, the Moebius glands that are always absent in the tabby mouse were once again present and allowed the formation of normal, nonkeratinized corneas, thus providing evidence of the functional competence of the Moebius glands (Figure 9c). The teeth of treated tabby mice were of normal size and exhibited a wild type pattern of tooth cusps. The third molar is small in wild-type mice and is absent in tabby mice. This third molar was also absent in treated tabby mice. The results demonstrate that when administered to pregnant mice, the FcrEDAl according to the invention, as shown in Figure 1d, is able to access the developing embryo via placental Fc receptors and induce induction at EDA receptors , of the formation of native epithelial structures. In this regard, dimerization of EDA1 (6 æl) is most likely of central importance since genetic and biochemical investigations indicate that oligomerization of EDA is essential for its activity. It has also been found in accordance with the invention that after induction and establishment of epithelial structures in the developing embryo, functional EDA no longer needs to be supplied to the adult animal in order to maintain the native phenotype reversed. 5th Implementation example

In this example, an investigation was conducted to determine the extent to which a late, post-natal Fc: EDAl administration according to the invention was capable of alleviating the symptoms of the tabby phenotype in the mice. EDF was therefore given to the young mouse on the 2nd (40 pg), 4th (40 pg), 6th (60 pg), 8th (100 pg) and 10 (100 pg) days (or, alternative experimental test: no 5o (40 pg), no 7o (60 pg), no 9o (60 pg), no 11 o and no 13 o (in each case 100 pg) days after birth); a control group of tabby mice was left untreated.

Appropriately treated mice, or mice of the control group, were analyzed for the following phenotypic characteristics (i) " folded " (ii) formation of hair on the tail and / or behind the ear region and (iii) shape of the eye. In addition, a sweat test was performed on adult mice and sections of paw and tail skin sections of 25-day-old mice were analyzed.

In the case of this post-natal administration of FcrEDAl, it was also possible to obtain a large number of symptoms of the tabby phenotype, in particular the recovery of the tail hair and sweat glands. It was possible to observe the tail hair on the 15th day (just on the 2nd day after birth) in the case of the tabby mouse treated with FcrEDAl (instead of on day 7 to 9 day after birth in the case of wild-type mice). In the case of mice that were first treated on the 5th day after birth, hair formation was observed at 18 ° 40 days. In the case of the treated tabby mouse, the hair is initially formed on the ventral side of the tail. In all treated tabby mice, regardless of the time the treatment started, the structure and density of the tail coat did not differ from that of the tail coat in wild type animals. This late hair formation in the treated tabby mouse can be easily explained by the late induction of hair follicle formation triggered by the Fc: EDA1 according to the invention which has been administered. The experimental approaches described in this example implementation were not able to alleviate the absence of hair (alopecia) behind the ears.

Although postnatal therapy was not able to completely reverse the folded tail shape, it was able to do so to a great extent. Although the eye aperture of the treated tabby mice was significantly increased compared to that of untreated tabby mice, it was not exactly as large as in the wild-type mouse (Figure 10a). Functional sweat glands were developed on the paws of the treated tabby mice, as demonstrated in the tissue section collected from treated 25-day-old tabby mice (Figure 10b). In sweat tests performed on treated adult tabby animals, it was possible to observe the functional capacity of sweat glands that develop in adult animals as a result of the treatment.

In summary, it may be said that the experiments according to the invention are capable of triggering a near complete reversal of a phenotype that will be genetically attributed to the absence of the TNF-EDA ligand. The best curative successes were observed in the case of tabby mice which had already been treated with fusion protein according to the invention in utero. However, postnatal administration of fusion proteins according to the invention, eg Fc: EDAl, to genotypically affected animals also results in considerable curative success, ie a (partial) reversal of the phenotype, in particular in the development of glands sweat (especially on the legs) and tail hair and in the increase of the opening of the eye. 10-02-2010 42

Claims (18)

Recombinant fusion protein comprising an amino acid sequence comprising: (a) in the N-terminal section of the fusion protein, the Fc section of an immunoglobulin, or part of a Fc section, which retains the ability to dimerization, as component (A); (b) in the section located closest to the N-terminus of the fusion protein, the extracellular portion of a TNF ligand or a partial sequence of the extracellular part of a TNF ligand as component (B), and optionally (c) a region (A) and component (B) containing a linker, characterized in that the component (B) is the extracellular portion of the TNF ligand EDAl, or is a partial sequence of the extracellular portion of the TNF ligand EDAl, which has at least 246 to 391 amino acids of the native EDAl sequence. Recombinant fusion protein according to claim 1, characterized in that component (B) has the amino acid sequence 140-391, 157-391, 160-391, 181-391, 182-391, 200-391 or 245 -391 from native EDAl sequence. Recombinant fusion protein according to claim 1 or 2, characterized in that component (A) has the hinge region, the CH2 and CH3 domain of the Fc section of an IgG class immunoglobulin, in particular human IgG . 4. recombinant fusion protein according to any one of the preceding claims, characterized in that component (A) has the amino acid sequence 1 KTHTCPPCPA pellggpsvf lfppkpkdtl misrtpevtg vwdvshedp evkfnwyvdg VEVHNAKTKP REEQYNSTYR WSVLTVLHQ DWLNGKEYKC KVSNKALPAP 1EKTISKMC QPREPQVYTL PPSRDELTKN QVSLTCLVKG FYPSDIAVEW ESNGQPENNY KTTPPVLDSD GSFFLYSKLT VDKSRWQQGN VFSCSVMHEA LHNHYTQKSL SLSPGK Recombinant fusion protein according to any one of the preceding claims, characterized in that the transition region between component (A) and component (B) has the sequence PQPQPKPQPKPEPE. Recombinant fusion protein according to any one of the preceding claims, characterized in that the transition region has a protease restriction site. Recombinant fusion protein according to any one of the preceding claims, characterized in that the N-terminus of the recombinant fusion protein has a signal sequence, for example a secretion signal sequence and / or a signal marker, for example a marker-flag or his-marker. Recombinant fusion protein according to any one of the preceding claims, characterized in that the N-terminus has the sequence MAIIYLILLFTAVRG. Recombinant fusion protein according to any one of the preceding claims, characterized in that components (B) are functionally linked to one another by the recombinant fusion protein sequence. A fusion protein complex according to any one of claims 1 to 9, characterized in that the complex has 6 chains of component (A). 2 A DNA sequence, characterized in that the DNA sequence encodes a recombinant fusion protein according to any one of claims 1 to 9. An expression vector, wherein the expression vector contains a DNA sequence according to claim 11. A host cell, characterized by being transfected with an expression vector according to claim 12. A compound which contains a recombinant fusion protein according to any one of claims 1 to 9, a complex according to claim 10, a DNA sequence according to claim 11, an expression vector according to claim 12, or a host cell according to claim 13. Use of a recombinant fusion protein according to any one of claims 1 to 9, a complex according to claim 10, a DNA sequence according to claim 11, an expression vector according to claim 12 , or a host cell according to claim 14, for the preparation of a drug for treating genetic diseases, in particular ectodermal dysplasia, or other diseases, disorders or anomalies of ectodermal structures, such as, for example, hair anomalies, teeth or glands, in particular to treat alopecia, or to heal wounds. Use according to claim 15 for preparing a drug for treating ectodermal dysplasia, in particular hypohydrotic ectodermal dysplasia coupled to X. Use according to claim 15 for preparing a drug for treating alopecia, hirsutism, for treating wounds or malfunctions of sweat or sebaceous glands, or deficiency of sweat glands or sebaceous glands. Use according to any one of claims 15 to 17 for the preparation of a drug for treating the mother / mother animal during pregnancy / pregnancy wherein the available drug is administered in a suitable dosage, parenterally. 10-02-2010 4