Recombinant TNF ligand family member polypeptides with antibody binding domain and uses therefor

Abstract

The present invention relates in general to the field of TNF ligand family members. In more detail the present invention relates to polypeptides comprising at least three components A, each of which comprises the sequence of a TNF homology domain (THD) of a TNF ligand family member, or a functional derivative thereof, and comprising at least one component B consisting of a V.sub.L region and a V.sub.H region linked directly to each other with a linker sequence L which has a length of <12 amino acids. Furthermore, the present invention also relates to nucleic acids encoding such polypeptides and pharmaceutical compositions thereof.

Claims

1. Polypeptide comprising: a) at least three components A, each of which comprises the sequence of a TNF homology domain (THD) of TNF Related Apoptosis Inducing Ligand (TRAIL) having at least 90% sequence identity to SEQ ID NO: 05, and b) at least one component B consisting of a V.sub.L region and a V.sub.H region linked directly to each other with a linker sequence L which has a length of ≦12 amino acids, wherein the V.sub.L and V.sub.H regions are of an antibody binding to a cell surface molecule selected from the group consisting of: a cytokine receptor, a growth factor receptor, an integrin, a cell adhesion molecule and/or a cell type- or tissue-specific cell surface antigen, cell surface expressed tumor-associated antigens (TAA), and carbohydrates.

2. Polypeptide according to claim 1, wherein the at least three components A are identical.

3. Polypeptide according to claim 1, wherein at least one component A comprises a sequence selected from the group consisting of SEQ ID NOs: 1-5.

4. Polypeptide according to claim 1, wherein all at least three components A comprise the sequence of SEQ ID NO: 1 and/or SEQ ID NO: 5.

5. Polypeptide according to claim 1, wherein the at least 3 components A are directly linked to each other via at least two intervening peptide linkers (peptide linker P).

6. Polypeptide according to claim 5, wherein the at least two peptide linkers P are chosen from SEQ ID NOs: 48, 88 and/or 90.

7. Polypeptide according to claim 1, wherein the V.sub.L region is selected from the group consisting of SEQ ID NOs: 92 and 130-132, wherein the V.sub.H region is selected from the group consisting of SEQ ID NOs: 93 and 133-135, or wherein the V.sub.L and V.sub.H regions are selected from the group consisting of SEQ ID NOs: 92 and 93, SEQ ID NOs: 130 and 133, SEQ ID NOs: 131 and 134 and SEQ ID NOs: 132 and 135.

8. Polypeptide according to claim 1, wherein the V.sub.L and V.sub.H region of a component B are linked with a linker sequence L chosen from SEQ ID NOs: 41-48, 50-51 and 53-75.

9. Polypeptide according to claim 1, wherein component B has the sequence of SEQ ID NO: 94, 136, 137 or 138.

10. Polypeptide according to claim 1, wherein the polypeptide comprises a glycosylation motif and/or is glycosylated and/or comprises an albumin binding domain (ABD).

11. Polypeptide according to claim 1, wherein the polypeptide comprises an albumin binding domain (ABD) located between component A and component B or downstream of component A.

12. Polypeptide according to claim 1, wherein the polypeptide comprises the sequence of SEQ ID NO: 103, SEQ ID NO: 107, SEQ ID NO: 102, SEQ ID NO: 125, SEQ ID NO: 126, SEQ ID NO: 127, SEQ ID NO: 128 or SEQ ID NO: 129.

13. Polypeptide complex comprising the polypeptide according to claim 1.

14. Polypeptide complex according to claim 13, which is a dimeric complex.

15. Pharmaceutical composition comprising at least one polypeptide according to claim 1, optionally further comprising at least one pharmaceutically acceptable carrier, adjuvant, and/or vehicle.

16. A method for treating cancer, comprising administering to a subject in need thereof the polypeptide according to claim 1.

17. Nucleic acid encoding for a polypeptide comprising: a) at least three components A, each of which comprises the sequence of a TNF homology domain (THD) of TNF Related Apoptosis Inducing Ligand (TRAIL) having at least 90% sequence identity to SEQ ID NO: 05, and b) at least one B consisting of a V.sub.L region and a V.sub.H region linked directly to each other with a linker sequence L which has a length of ≦12 amino acids, wherein the V.sub.L and V.sub.H regions are of an antibody binding to a cell surface molecule selected from the group consisting of: a cytokine receptor, a growth factor receptor, an integrin, a cell adhesion molecule and/or a cell type- or tissue-specific cell surface antigen, cell surface expressed tumor-associated antigens (TAA), and carbohydrates.

Description

(1) In the following a brief description of the appended figures will be given. The figures are intended to illustrate the present invention in more detail. However, they are not intended to limit the subject matter of the invention to any extent.

(2) FIG. 1: Schematic illustration of exemplary polypeptides according to the present invention: A) K=V.sub.H leader (e.g. SEQ ID NO: 101); F=tag (i.e. FLAG tag; see for instance SEQ ID NO: 100); L=linker sequence L, e.g. 15 aa long (e.g. (GGGGS).sub.3; SEQ ID NO: 52); X=peptide linker X (see for instance SEQ ID NO: 95); P=peptide linker P (e.g. SEQ ID NO: 48); A=component A (e.g. TRAIL aa residues 95-281, SEQ ID NO: 5). An example for such polypeptide is SEQ ID NO: 96 (FIG. 20). B) K=V.sub.H leader (e.g. SEQ ID NO: 101); F=tag (i.e. FLAG tag; see for instance SEQ ID NO: 100); L=linker sequence L, ≦12 aa long (e.g. (GGGGS); SEQ ID NO:50); X=peptide linker X (see for instance SEQ ID NO: 95); P=peptide linker P (e.g. SEQ ID NO: 48); A=component A (e.g. TRAIL aa residues 95-281; SEQ ID NO: 5). An example for such polypeptide is SEQ ID NO: 102 (FIG. 21). C) K=V.sub.H leader (e.g. SEQ ID NO: 101); F=tag (i.e. FLAG tag; see for instance SEQ ID NO: 100); L=linker sequence L, e.g. 15 aa long (e.g. (GGGGS).sub.3; SEQ ID NO: 52); X=peptide linker X including glycosylation site (see for instance SEQ ID NO: 105); P=peptide linker P (e.g. SEQ ID NO: 48); A=component A (e.g. TRAIL aa residues 95-281; SEQ ID NO: 5). An example for such polypeptide is SEQ ID NO: 97 (FIG. 22). D) K=V.sub.H leader (e.g. SEQ ID NO: 101); F=tag (i.e. FLAG tag; see for instance SEQ ID NO: 100); L=linker sequence L, e.g. 15 aa long (e.g. (GGGGS).sub.3; SEQ ID NO: 52); X=peptide linker X including glycosylation site (see for instance SEQ ID NO: 105); P=peptide linker P including glycosylation site (e.g. P: (SEQ ID NO: 90); P.sub.2: SEQ ID NO: 88); A=component A (e.g. TRAIL aa residues 95-281; SEQ ID NO: 5). An example for such polypeptide is SEQ ID NO: 98 (FIG. 23). E) K=V.sub.H leader (e.g. SEQ ID NO: 101); ABD=Albumin binding domain (see for instance SEQ ID NO: 99); Q=Linker sequence (e.g. GGSGGGGSGG; SEQ ID NO: 71); L=linker sequence L, ≦12 aa long (e.g. (GGGGS); SEQ ID NO: 50); X=peptide linker X including glycosylation site (see for instance SEQ ID NO: 106); P=peptide linker P with (e.g. SEQ ID NO: 88) or without glycosylation site (e.g. SEQ ID NO 48: GGGSGGGS); A=component A (e.g. TRAIL aa residues 95-281, SEQ ID NO: 5). An example for such polypeptide is SEQ ID NO: 103 (FIG. 24). F) K=V.sub.H leader (e.g. SEQ ID NO: 101); F=tag (i.e. FLAG tag; see for instance SEQ ID NO: 100); L=linker sequence L, ≦12 aa long (e.g. (GGGGS); SEQ ID NO: 50); Q=Linker sequence (e.g. GGS; SEQ ID NO: 55); ABD=Albumin binding domain (see for instance SEQ ID NO: 99); X=peptide linker X including glycosylation site (see for instance SEQ ID NO: 105); P=peptide linker P (e.g. GGGSGGGS; SEQ ID NO: 48); A=component A (e.g. TRAIL aa residues 95-281, SEQ ID NO: 5). An example for such polypeptide is SEQ ID NO: 125 (FIG. 25). G) K=V.sub.H leader (e.g. SEQ ID NO: 101); F=tag (i.e. FLAG tag; see for instance SEQ ID NO: 100); L=linker sequence L, <12 aa long (e.g. (GGGGS); SEQ ID NO: 50); X=peptide linker X including glycosylation site (see for instance SEQ ID NO: 105); P.sub.1=peptide linker P, (e.g. GGGSGGGS; SEQ ID NO: 48); P.sub.2=peptide linker P.sub.2 (e.g. GGSGG; SEQ ID NO: 61); A=component A (e.g. TRAIL aa residues 95-281, SEQ ID NO: 5); ABD=Albumin binding domain (see for instance SEQ ID NO: 99). An example for such polypeptide is SEQ ID NO: 126 (FIG. 26).

(3) FIG. 2: Purified polypeptides of SEQ ID NO: 104 (lane 1), SEQ ID NO: 96 (lane 2), SEQ ID NO: 102 (lane 3) and of glycosylated SEQ ID NO: 97 (lane 4) were analyzed by SDS-PAGE (reducing conditions) followed by silver staining (upper, left, 1 μg protein per lane) or Western blotting (lower, 250 ng protein per lane) using monoclonal anti-TRAIL or anti-FLAG antibodies in combination with alkaline phosphatase-conjugated secondary antibody. The glycosylated polypeptide of SEQ ID NO: 97 was treated with N-glycosidase and analysed by SDS-PAGE and Coomassie staining (upper, right). SEQ ID NO: 104 is a single chain TRAIL polypeptide with three copies of TRAIL 95-281 (SEQ ID NO: 5) linked by two glycine linkers and comprising an N-terminal Flag-tag. However, the polypeptide does not comprise any region B as specified herein.

(4) FIG. 3 Purified polypeptides of SEQ ID NO: 104 (upper left), SEQ ID NO: 96 (lower left), SEQ ID NO: 102 (upper right) and of glycosylated SEQ ID NO: 97 (lower right) were separated by size exclusion chromatography on a BioSuite 250 column. The retention times of the molecular weight standards thyroglobulin (669 kDa), β-amylase (200 kDa), bovine serum albumin (67 kDa) and carbonic anhydrase (29 kDa) are indicated by dotted lines.

(5) FIG. 4 Flow cytometric analysis of expression of EGF receptor and proapoptotic TRAIL receptors DR4 and DR5 in Jurkat, Huh-7 and HepG2 cell lines.

(6) FIG. 5 Flow cytometric analysis: (A) Blocking of the binding of Cetuximab to target-negative (HepG2) and target-positive cells (Huh-7) by an excess of SEQ ID NO: 102; (B) Binding of SEQ ID NO: 102 and SEQ ID NO: 96 to HepG.sub.2 and Huh-7 cells. “scFv” represents an anti-EGFR specific antibody fragment used for competition.

(7) FIG. 6 Target-independent induction of cell death: (A) EGFR low/negative HepG2 cells were sensitized with 500 ng/ml Bortezomib and treated with serial dilutions of SEQ ID NO: 102 (open squares), KillerTRAIL™ (inverted filled triangle), SEQ ID NO: 96 (open circles) and SEQ ID NO:104. Cell viability was determined using crystal violet staining. Results from four independent experiments are shown (mean±S.E.M.). (B) Jurkat cells (1×10.sup.5/well) were used for a similar experiment as described in (A). Jurkat cells were not sensitized. Results from three independent experiments are shown (mean±S.E.M.). Viability of Jurkat cells was determined using the MTT assay.

(8) FIG. 7. EGFR-directed induction of cell death in EGFR+Huh-7 hepatocellular carcinoma cells sensitized with 250 ng/ml Bortezomib and treated in duplicates with serial dilutions of SEQ ID NO: 102 (open squares), SEQ ID NO: 96 (open circles), SEQ ID NO: 97 (open diamond), and SEQ ID NO: 104 as control (left side of panel). For quantification of the targeting effect, cells were additionally preincubated with an excess of EGFR-specific antibody Cetuximab (70 nM) before adding the test polypeptides (graphs of SEQ ID NO: 102 (middle panel) and SEQ ID NO: 96 (right panel). For easy comparison, dose response curves of the respective reagents in absence of cetuximab from the left panel were plotted again in these two panels. Results from four independent experiments are shown (mean±S.E.M.).

(9) FIG. 8: EGFR-directed induction of cell death in Huh-7 hepatocellular carcinoma cells as in FIG. 7 with the exception that constant concentrations of the test proteins were used for preincubation (30 min), followed by addition of serial dilutions of Bortezomib. Control: Bortezomib alone. Results from three independent experiments are shown (mean±S.E.M.).

(10) FIG. 9: EGFR specific TRAIL fusion proteins lack hepatotoxic activity. Groups of 3 CD1 mice were treated intraperitonally with indicated fusion proteins or control reagents: negative control: PBS; positive control: aggregated FasL (CD95L) fusion protein. (A) Plasma samples were prepared after 4 h and 24 h and the activity of alanine aminotransferase (ALT) was assayed using an enzymatic assay (Bioo Scientific, Austin, Tex.). Dashed line indicates upper normal level of ALT (physiologic range in adult human: 35-50 U/L). (B) Mice were sacrificed after 24 h except for positive control (animals treated with an aggregated FasL (CD95L) fusion protein show phenotypic signs of severe systemic toxicity after 2-4 hrs and die after ˜5 hrs, samples were taken after 4 hrs) and liver biopsies were taken for determination of caspase-3 activity using a specific AMC-coupled peptide substrate (Enzo Lifesciences, Lörrach, Germany).

(11) FIG. 10: EGFR-directed induction of cell death by single-chain TRAIL fusions on NCI-H460 cells. NCI-H460 non-small lung cancer cells (3×10.sup.4/well) were seeded in 96-well plates and cultivated for 24 h. Then, cells were sensitized with 2.5 μg/ml cycloheximide and treated in duplicates with serial dilutions of the indicated fusion proteins. After 16 h, cell viability was determined using crystal violet staining. Results from two independent experiments are shown (mean±S.E.M.). EC.sub.50 values were 1.2±0.08×10.sup.−12 M for SEQ ID NO: 96, 3.4±0.28×10.sup.−13 M for SEQ ID NO: 102 and 5.8±1.5×10.sup.−12 M for SEQ ID NO: 98.

(12) FIG. 11: Biochemical characterization of N-glycosylated polypeptide according to the present invention (SEQ ID NO: 98). (A) SEQ ID NO: 98 was treated with N-glycosidase and analysed by SDS-PAGE and Coomassie staining. (B) SEQ ID NO: 98 and SEQ ID NO: 96 were separated by size exclusion chromatography on a BioSuite 450 column (Waters). The retention times of the molecular weight standards apoferritin (443 kDa), β-amylase (200 kDa), bovine serum albumin (67 kDa) and carbonic anhydrase (29 kDa) were indicated by dotted lines.

(13) FIG. 12: Receptor interaction of EGFR-specific TRAIL fusion proteins. (A) Dose response relationship of TRAIL fusion protein binding to EGFR.sup.+ NCI-H460 cells by indirect immunofluorescence flow cytometry to reveal concentration of half maximum binding (EC.sub.50) (mean SEM, n=4). (B) Colo205 and Huh-7 cells were serum-starved overnight and then incubated with 2 nM of SEQ ID NO: 102, Cetuximab, and PBS for control, respectively. After 10 min, 50 ng/ml EGF were added and cells were incubated for additional 20 min followed by cell lysis. EGF receptors were immunoprecipitated using a specific mouse monoclonal antibody and subjected to SDS-PAGE followed by immunoblotting with phosphotyrosine antibody (anti-pTyr). Total amounts of EGFR were determined by reprobing the membrane with EGFR-specific rabbit polyclonal antibody (anti-EGFR).

(14) FIG. 13: Caspase dependence of cell death and impact of the component B of SEQ ID NO: 102. (A) Colo205 cells (left) and Huh-7 cells (right) were sensitized with 25 ng/ml and 250 ng/ml bortezomib, respectively, and treated with different concentrations of SEQ ID NO: 102 with or without the presence of pan-caspase inhibitor zVADfmk or caspase-3 inhibitor zDEVDfmk (both inhibitors: 20 μM for Colo205 and 10 μM for Huh-7). After 16 h, cell viability was determined using MTT staining (Colo 205) or crystal violet staining (Huh-7) and data were normalized using bortezomib-treated cells as control (mean SEM, n=3). (B) 1×10.sup.4 Colo205 cells per well were grown in 96-well plates using medium with 0.1% FCS. Upon stimulation with 50 ng/ml EGF and sensitization with 10 ng/ml bortezomib, cells were incubated with equimolar concentrations of SEQ ID NO: 102 (open squares), SEQ ID NO: 102+anti-TRAIL mAb 2E5 (filled squares) or Cetuximab (circles) for four days and cell number was assayed by the MTT method using bortezomib/EGF-treated cells as control for normalization (mean SEM, n=2). (C) 1×10.sup.4 Huh-7 cells per well were grown in 96-well plates and treated with 20 ng/ml bortezomib or with a combination of bortezomib and 10 μM zVADfmk. Then, cells were incubated with equimolar concentrations of SEQ ID NO: 102 (open squares), SEQ ID NO: 102+zVADfmk (filled squares), Cetuximab (open circles) or Cetuximab+zVADfmk (filled circles) for three days and cell viability was assayed by the MTT method using bortezomib-treated cells and bortezomib/zVADfmk-treated cells, respectively, as control for normalization (mean SEM, n=3).

(15) FIG. 14: In vitro tolerance of TRAIL fusion proteins to primary tissues. (A) Relative caspase activity (fold increase compared to untreated) in primary human hepatocytes (PHH, mean SEM, n=5) or Huh-7 hepatocarcinoma cells (mean SEM, n=7) after incubation with 1.1 nM SEQ ID NO: 102 in presence or without 500 ng/ml bortezomib. Asterisks indicate statistical significance. (B) Cleavage of caspase-3 in PHH (left) and Huh-7 cells (right) after incubation with 500 ng/ml bortezomib, 1.1 nM SEQ ID NO: 102 or both was analyzed by immunoblotting.

(16) FIG. 15: Antitumor activity of TRAIL fusion proteins in a Colo205 xenograft tumor model. (A) Tumor volume as a function of time after i.p. application of PBS (open diamonds), bortezomib (filled triangles), SEQ ID NO: 104 (lacking component B)+bortezomib (open triangles), SEQ ID NO: 97 (L>12 amino acids)+bortezomib (circles), SEQ ID NO: 102 (L<12 amino acids)+bortezomib (filled diamonds) or SEQ ID NO: 102 only (squares). Arrows, protein application; asterisks, bortezomib application; symbols, mean of tumor volumes 95% confidence interval (CI), n=12 tumors/treatment group. (B) Individual tumor volumes at day 14. Bars, mean of tumor volumes 95% Cl.

(17) FIG. 16: Coomassie-stained SDS-PAGE of affinity-purified SEQ ID NO: 102 (lane 1), SEQ ID NO: 125 (lane 2) and SEQ ID NO: 126 (lane 3). 5 μg of the proteins were loaded under reducing conditions.

(18) FIG. 17: Albumin binding of SEQ ID NO: 125 and SEQ ID NO: 126. Both proteins were incubated for 1 h at RT with equimolar concentrations of human serum albumin (HSA) or mouse serum albumin (MSA) and subsequently separated by size exclusion chromatography. Thyroglobulin (669 kDa), apoferritin (443 kDa), -amylase (200 kDa), bovine serum albumin (67 kDa), carbonic anhydrase (29 kDa) and FLAG peptide (1 kDa) were used as standard proteins/peptides.

(19) FIG. 18: Bioactivity of SEQ ID NO: 125 and SEQ ID NO: 126 in vitro. Huh-7 hepatocarcinoma cells were sensitized with bortezomib (250 ng/ml) and treated with serial dilutions of SEQ ID NO: 102, SEQ ID NO: 125 (left panel) and SEQ ID NO: 126 (right panel) in triplicates. After 16 h, cell viability was determined using crystal violet staining. For quantification of the targeting effect, cells were preincubated with an excess of Cetuximab (70 nM) before adding SEQ ID NO: 102, SEQ ID NO: 125 and SEQ ID NO: 126. SEQ ID NO: 125 and SEQ ID NO: 126 were additionally incubated in presence of 100 μg/ml HSA. The values for SEQ ID NO: 102 were plotted in both panels for comparison.

(20) FIG. 19: Pharmacokinetics of SEQ ID NO: 102 (A) and SEQ ID NO: 125 (B). 25 μg of protein were injected i.v. in CD1 mice and serum samples were taken at the depicted time points followed by detection of scTRAIL molecules via ELISA.

(21) FIG. 20: Sequence of SEQ ID NO: 96.

(22) FIG. 21: Sequence of SEQ ID NO: 102.

(23) FIG. 22: Sequence of SEQ ID NO: 97.

(24) FIG. 23: Sequence of SEQ ID NO: 98.

(25) FIG. 24: Sequence of SEQ ID NO: 103.

(26) FIG. 25: Sequence of SEQ ID NO: 125.

(27) FIG. 26: Sequence of SEQ ID NO: 126.

(28) FIG. 27: Sequence of SEQ ID NO: 127.

(29) FIG. 28: Sequence of SEQ ID NO: 128.

(30) FIG. 29: Sequence of SEQ ID NO: 129.

EXAMPLES

(31) In the following, general examples illustrating various embodiments and aspects of the invention are presented. However, the present invention shall not be limited in scope by the specific embodiments described herein. Indeed, various modifications of the invention in addition to those described herein will become readily apparent to those skilled in the art from the foregoing description, accompanying figures and the examples below. All such modifications fall within the scope of the appended claims.

Example 1

Biochemical Analysis

(32) 1. Polypeptide Production

(33) 1.1 Principle

(34) Three human TRAIL domains encompassing aa residues 95-281 (TRAIL) (SEQ ID NO: 5) were fused with (GGGS).sub.2 peptide linkers P (SEQ ID NO: 48) yielding so called single-chain TRAIL (scTRAIL) (SEQ ID NO: 104). EGFR-specific antibody fragments consisting of V.sub.H (SEQ ID NO: 93) and V.sub.L (SEQ ID NO: 92) were fused N-terminally to scTRAIL (SEQ ID NO: 104). (GGGGS).sub.3 (SEQ ID NO: 52) or GGGGS (SEQ ID NO: 50) peptide linkers between V.sub.H (SEQ ID NO: 93) and V.sub.L (SEQ ID NO: 92) were chosen to obtain a polypeptide according to the present invention (SEQ ID NOs: 96 and 102). A V.sub.H leader (K) (SEQ ID NO: 101) and a FLAG tag (F) (SEQ ID NO: 100) were placed in front of the antibody region. For a glycosylated polypeptide according to the present invention, a linker with two N-glycosylation sites (GNGTSNGTS) (SEQ ID NO: 83) was placed between V.sub.L (SEQ ID NO: 92) and scTRAIL (SEQ ID NO: 104).

(35) 1.2 Plasmids and Cell Lines

(36) An pIRESpuro-scTRAIL expression construct for human scTRAIL (SEQ ID NO: 104) was obtained by EcoRI/NotI cloning of a synthesized sequence coding for three TRAIL components (aa residues 95-281) connected by sequences encoding (GGGS).sub.2 linker motifs into a construct described previously (Schneider et al, 2010, Cell Death. Disease. 2010). For the generation of the EGFR-specific V.sub.H-V.sub.L-scTRAIL expression construct, a synthesized coding sequence of humanized V.sub.H and V.sub.L sequences (huC225) was amplified using the oligonucleotides CGAGGTGCAGCTGGTCGAG (SEQ ID NO: 109) and TGCGGCCGCTCTCTTGATTTC (SEQ ID NO: 110). Next, this template was annealed with the oligonucleotide ATATATCTCGAGGCCAGCGACTACAAAGACGATGACGATAAAGGAGCCGAGGTGCAGCTGGTCGAG (SEQ ID NO: 111) to insert an XhoI site and a FLAG tag coding sequence. After strand elongation, the whole sequence was amplified by the oligonucleotides ATATATCTCGAGGCCAGCGAC (SEQ ID NO: 112) and ATATGAATTCTGCGGCCGCTCTCTTGATTTC (SEQ ID NO: 113). The PCR product was then cloned via XhoI/EcoRI into pCR3 (Invitrogen), carrying an V.sub.H leader. The scTRAIL coding sequence of this construct was then inserted via EcoRI/Xbal sites. The EGFR-specific construct for SEQ ID NO: 102 was derived from pCR3-VH-VL-scTRAIL by shortening linker L from (GGGGS).sub.3 to GGGGS. Therefore, two PCR products were generated using the oligonucleotides (1) CCCACAGCCTCGAGGCCAG (SEQ ID NO: 114) and (2) GAGCCGCCACCGCCACTAG (SEQ ID NO: 115) vas well as (3) CTAGTGGCGGTGGCGGCTCTGATATTCAGCTGA CCCAGTCC (SEQ ID NO: 116) and (4) TGAATTCTGCGGCCGCTCTC (SEQ ID NO: 117). After annealing of the products at the underlined regions and strand elongation, the whole sequence was amplified by the oligonucleotides (1) and (4) followed by XhoI/NotI cloning into pCR3-V.sub.H-V.sub.L-scTRAIL. A glycosylated variant of SEQ ID NO: 96 was generated by two PCR amplifications of the huC225 VH-VL coding sequence in pCR3-VH-VL-scTRAIL using the oligonucleotides (1) CCCACAGCCTCGAGGCCAG (SEQ ID NO: 118) and CCCGTTGCTGGTGCCGTTGCCTGCGGCCGCTCTCTTG (SEQ ID NO: 119), respectively (1) and ATATGAATTCGGATGTCCCGTTGCTGGTGCCGTTG (SEQ ID NO: 120), followed by XhoI/EcoRI cloning in pCR3-VH-VL-scTRAIL. The construct for expression of SEQ ID NO: 98 was generated by two sequential PCR amplifications of the TRAIL coding sequence using the oligonucleotides GCACATCCAATGGGACCAGCGGAACCTCCGAAGAGACTATCTC (SEQ ID NO: 121) and CCCGTTGCTGGTTCCATTACCAGATCCGCCCCCTCC (SEQ ID NO: 122), respectively ATATATGGATCCGGCAACGGCACATCCAATGGGACCAG (SEQ ID NO: 123) and ATATATGGATCCGGTCCCGTTGCTGGTTCCATTAC (SEQ ID NO: 124), followed by BamHI cloning into the expression construct for SEQ ID NO: 97.

(37) HEK293, HepG2, NCI-H460, Colo205 and Jurkat, cells were obtained from the American Type Culture Collection (Manassas, Va.). Cells were cultured in RPMI 1640 medium (Invitrogen, Karlsruhe, Germany) supplemented with 5% fetal calf serum (FCS, HyClone), respectively 10% FCS for HepG2. Huh-7D12 liver carcinoma cells were obtained from Heike Bantel, Hannover Medical School, Hannover, Germany and were cultured in DMEM (Invitrogen, Karlsruhe, Germany) supplemented with 10% FCS.

(38) 1.3 Production and Purification of Recombinant Proteins

(39) The TRAIL fusion proteins of SEQ ID NOs: 96, 97, 98, 102, 125, and 126 were produced in HEK293 cells after stable transfection with the corresponding expression plasmids using Lipofectamine 2000 (Invitrogen) and generation of a pool of stably expressing clones. For protein production, stable clones were expanded and grown in RPMI 1640, 5% FCS, to 90% confluency and subsequently cultured in serum-free Optimem (Invitrogen) supplemented with 50 μM ZnCl.sub.2, replacing media two times every 3 days. The supernatants were pooled and recombinant proteins were purified first by IMAC using Ni-NTA-Agarose (Qiagen, Hilden, Germany). After elution with 100 mM imidazol and dialysis against PBS, the proteins were further purified by affinity chromatography using anti-FLAG mAb M2 agarose (Sigma-Aldrich, Steinheim, Germany). The bound proteins were eluted with 100 μg/ml FLAG peptide (peptides&elephants, Potsdam, Germany) and dialysed against PBS. scTRAIL and SEQ ID NOs: 97 and 98 were purified in a single M2 agarose affinity chromatography step. After concentration of purified proteins using Vivaspin centrifugal concentrators with 50 or 10 kDa MWCO (Sartorius Stedim, Aubagne, France), the protein concentration was measured with a spectrophotometer (NanoDrop products, Wilmington, Del.) and aliquots were stored at −80° C.

(40) 1.4 SDS-PAGE and Western Blot Analysis

(41) Purified polypeptides of SEQ ID NO: 96, SEQ ID NO: 102, SEQ ID NO: 104, 125, 126 and of glycosylated SEQ ID NO: 97 were analyzed by SDS-PAGE (reducing conditions) followed by silver staining (1 μg protein per lane), Coomassie staining or Western blotting (250 ng protein per lane) using monoclonal anti-TRAIL (MAB687, R&D Systems, Wiesbaden, Germany) or anti-FLAG antibodies (M2, Sigma-Aldrich) in combination with alkaline phosphatase-conjugated secondary antibody (Sigma-Aldrich). The glycosylated polypeptide of SEQ ID NO: 97 was treated with N-glycosidase and analysed by SDS-PAGE and Coomassie staining. For deglycosylation, protein (5 μg) was denatured in the presence of SDS and DTT prior to addition of Nonidet P-40 and 500 units of PNGaseF (New England Biolabs, Frankfurt a. M., Germany) according to the supplier's instructions. After 1 h incubation at 37° C., samples were subjected to SDS-PAGE. For Western blotting, an anti-TRAIL antibody MAB687 (R&D Systems, Wiesbaden, Germany) and anti-FLAG M2 mAb (Sigma-Aldrich) were used, followed by an anti-mouse alkaline phosphatase-coupled secondary antibody (Sigma-Aldrich) for detection.

(42) The results of the SDS PAGE/Western Blot analysis verified the increase in molecular mass of SEQ ID NOs: 96, 97 and 102 vs. SEQ ID NO: 104. The increase of molecular mass of SEQ ID NOs: 125 and 126 vs. SEQ ID NO: 102 has been verified by SDS page and Coomassie staining. SEQ ID NO: 97 showed reduced migration in SDS PAGE, which is conform with effective glycosylation of the protein. This was confirmed by PNGaseF treatment of the fusion protein, which removes carbohydrate side chains in glycoproteins, resulting in a shift of the specific band towards the mass of the non-glycosylated form. Furthermore, the assays confirmed the presence of TRAIL as well as of the FLAG-tag in all 4 polypeptides.

(43) 1.5 Size Exclusion Chromatography and Albumin Binding Assay

(44) Purified polypeptides of SEQ ID NO: 96, SEQ ID NO: 102, SEQ ID NO: 104 and of glycosylated SEQ ID NO: 97 were separated by size exclusion chromatography on a BioSuite 250 HR SEC (300×7.8) column (Waters, Millipore Corp., Milford, Mass.) equilibrated in PBS and eluted at a flow rate of 0.5 ml/min.

(45) Albumin binding to the polypeptides of SEQ ID NOs: 125 and 126 has been verified by incubating the polypeptides with human serum albumin or mouse serum albumin and determining protein-protein interaction by size exclusion chromatography as described above.

(46) 1.6 Immunoprecipitation and Protein Analysis

(47) For immunoprecipitations, cells were lysed on ice in RIPA buffer (50 mM Tris, pH 7.5, 150 mM NaCl, 10 mM sodium fluoride, 20 mM glycerophosphate, 1 mM EDTA, 1% NP40, 1 mM sodium orthovanadate, 0.5 mM phenylmethylsulfonyl fluoride, 0.1% SDS, 0.25% sodium deoxycholate) with Complete protease inhibitor (Roche Diagnostics, Mannheim, Germany) and lysates were clarified by centrifugation (16 000 g, 10 min, 4° C.). 1.5 mg lysate protein was incubated with 1.5 μg mouse anti-EGFR Ab-13 mAb (Neomarkers, Fremont, Calif., USA) under gentle shaking at 4° C. overnight. Immune complexes were captured with protein G sepharose (KPL, Gaithersburg, Md., USA) and washed three times with RIPA buffer. Proteins were analyzed by SDS-PAGE and Western blotting using mouse anti-phosphotyrosine P-Tyr-100 mAb (Cell Signaling Technology, Danvers, Mass., USA) and rabbit anti-EGFR 1005 antibody (Santa Cruz Biotechnology, Santa Cruz, Calif., USA) followed by HRP-conjugated secondary antibodies. ECL (Pierce Biotechnology, Rockford, Ill., USA) was used for visualization.

(48) Caspases were detected by immunoblotting using a rabbit polyclonal antibody against cleaved caspase-3 (Cell Signaling Technology). GAPDH as internal control was detected with a rabbit polyclonal antibody (Cell Signaling Technology). HRP-conjugated secondary antibodies (Zymed Laboratories, San Fransisco, Calif., USA) and ECL were used for visualization.

Example 2

Flow Cytometry, Cell Death Assay, ALT and Caspase Activities

(49) 2.1 Flow Cytometry

(50) 5×10.sup.5 cells were suspended in PBA buffer (PBS, 0.025% BSA, 0.02% sodium azide) and incubated for 1 h at 4° C. with the indicated scTRAIL fusion proteins (2 μg/ml). After washing the cells three times with PBA buffer, bound fusion proteins were detected by anti-human TRAIL mAb MAB687 (2.5 μg/ml, R&D Systems) and fluorescein isothiocyanate-labelled rabbit anti-mouse IgG Ab (1:200, Sigma-Aldrich), followed by three washing steps with PBA each. For blocking of scTRAIL fusion protein binding to EGFRs (see FIG. 5B), a divalent variant of huC225 (huC225Cys, 50 μg/ml, kindly provided by Celonic GmbH, Jülich, Germany) was added 30 min before addition of SEQ ID NO: 96 and SEQ ID NO: 102, respectively. Expression of TRAIL receptors was detected by anti-TRAIL R1 mAb MAB347 and anti-TRAIL R2 mAb MAB6311 (4 μg/ml each, R&D Systems) in conjunction with anti-mouse IgG-FITC. EGFR expression was detected by a phycoerythrin-labelled anti-human EGFR mAb sc-101 (4 μg/ml, Santa Cruz Biotech., Santa Cruz, Calif.) (see FIG. 4). For binding inhibition, purified SEQ ID NO: 102 (50 μg/ml) was added 30 min prior addition of Alexa Fluor 488-coupled mAb Cetuximab (1 μg/ml) (see FIG. 5A).

(51) The assays determined the expression levels of EGF receptor and proapoptotic TRAIL receptors DR4 and DR5 in Huh-7 and Hep2G hepatocellular carcinoma cell lines and the T cell leukemia line Jurkat, revealing Huh7 as EGFR+, DR4+, DR5 low; HepG2 as EGFR low/negative, DR4+, DR5+; and Jurkat as EGFR negative, DR4 negative, DR5 low (see FIG. 4). Furthermore, blocking of the binding of anti EGFR mab Cetuximab to EGFR+ cells (Huh-7) by an excess of EGFR receptor specific TRAIL protein (SEQ ID NO: 102) revealed functional expression of the EGFR specific V.sub.H-V.sub.L domain within the fusion protein of SEQ ID NO: 102 (FIG. 5A, right panel); as expected, the marginal Cetuximab staining of EGFR low/negative HepG2 cells could not be further reduced by an excess of the fusion protein of SEQ ID NO: 102, and likely reflect nonspecific background staining of the reagent. Likewise, binding of the EGFR targeting fusion proteins SEQ ID NO: 102 and SEQ ID NO: 96 to EGFR+ Huh-7 cells was partially blockable by an excess of an anti-EGFR specific antibody fragment, whereby the remaining signal could be attributed to specific binding of the fusion proteins via the TRAIL domain to their cognate TRAIL receptors. In this experimental setting, binding of fusion proteins was clearly discernable for DR4+DR5+HepG2 cells, too, with little blocking of the signal by addition of an anti-EGFR specific antibody fragment due to low expression of EGFRs at or below the detection level in these cells.

(52) 2.2 Cell Death Assays

(53) Huh-7 (3×10.sup.4), HepG2 (3×10.sup.4), Colo205 (5×10.sup.4 per well) or Jurkat cells (1×10.sup.5) were grown in 100 μl culture medium in 96-well plates for 24 h, followed by treatment with the indicated concentrations of SEQ ID NOs: 96, 102, 125, 126 and 104 or ‘KillerTRAIL’ (Axxora Deutschland GmbH, Lörrach, Germany) in triplicates (see FIGS. 6-8 and 18). As a positive control, cells were killed with 0.25% Triton X-100. Cell death assays with Huh-7 and HepG2 cells were performed in the absence (FIG. 6, Jurkat cells) or presence of Bortezomib (FIG. 7, Huh-7: 250 ng/ml, FIG. 6, HepG2: 500 ng/ml, FIG. 18, Huh-7: 250 ng/ml), Selleck Chemicals, Houston, Tex.). Bortezomib was added 30 min prior incubation with the proapoptotic ligands to sensitize cells for the induction of cell death (FIGS. 6, 7). Alternatively, cells were preincubated for 30 min with the indicated concentrations of TRAIL fusion proteins followed by addition of serial dilutions of Bortezomib (FIG. 8). TRAIL only treated cells are shown in each panel for the applied TRAIL concentration (Bortezomib 0 ng/ml) (FIG. 8). After 16 h incubation, cell viability was determined either by crystal violet staining (Huh-7, HepG2) or the MTT method (Jurkat) (Wuest et al., 2002). In the latter case a lysis buffer consisting of 15% SDS in DMF/H.sub.2O (1:1), pH 4.5 (with 80° A) acetic acid) was used. To demonstrate target antigen-dependent induction of cell death, cells were preincubated for 30 min with competing Cetuximab mAb (10 μg/ml, Merck, Darmstadt, Germany) (FIG. 18) or alternatively EGFR specific huC225Cys (10 μg/ml) (FIG. 7).

(54) 2.3 Alanine Aminotransaminase (ALT) and Caspase Activities

(55) Groups of three CD1 mice (Janvier, Le Genest-St-Isle, France) were treated i.p. with 1 nmol of fusion proteins according to SEQ ID NO: 102 and SEQ ID NO: 97, 0.1 nmol FasL fusion protein (positive control) and PBS (negative control), respectively. Blood samples were taken from the tail after 4 h and 24 h and incubated on ice. Clotted blood was centrifuged (10 000 g, 10 min, 4° C.) and serum samples were stored at −80° C. Activity of alanine aminotransaminase was determined by an enzymatic assay (BIOO Scientific, Austin, Tex., USA). To determine caspase-3 activity in the liver tissue, mice were sacrificed after 24 h (positive control after 5 h) and liver biopsies were taken. Homogenates were prepared in lysis buffer (200 mM NaCl, 20 mM Tris, 1% NP-40, pH 7.4). 10 μg of protein were analyzed by conversion of the fluorogenic substrate Ac-DMQD-AMC (Enzo Life Sciences). Caspase activity in PHH and Huh-7 cells was determined as published by Seidel et al. (Hepatology 2005, 42:113-120).

Example 3

Xenograft Mouse Tumor Model

(56) 8-week-old female NMRI nu/nu mice (Janvier) were injected s.c. with 3×10.sup.6 Colo205 cells in 100 μl PBS at left and right dorsal sides. Treatment started 6 days after tumor cell inoculation when tumors reached about 100 mm.sup.3. Mice received 8 daily i.p. injections of 0.45 nmol of the affinity-purified TRAIL fusion proteins according to SEQ ID NOs: 104, 97, and 102, respectively. On day 1, 3, 5 and 7 of treatment, mice received additionally 5 μg bortezomib in 100 μl PBS i.p. three hours before protein injection. The control groups received 100 μl PBS or 5 μg bortezomib at the same time intervals. Tumor growth was monitored as described in Schneider et al. (Cell Death Dis. 2010; 1: e68) and Kim et al. (Bioconjug. Chem. 2011; 22: 1631-1637). The Tukey's test was applied for statistics.

(57) Description of Results Disclosed in Examples 1, 2 and 3

(58) The main aim of the shown examples was the improvement of the proapoptotic activity of scTRAIL fusion proteins under retention of their tumor selectivity, i.e. non-reactivity towards normal, non-malignant tissue. In principle, the same rules apply for other proapoptotic members of the TNF family, as well as all other non-apoptotic, tissue and immune-regulatory TNF ligands that are inactive as soluble ligands or require membrane targeting to restrict their activity to the relevant tissue or cell types. The examples are also not restricted to the specific target antigen used exemplarily (EGFR), but apply, in principle, to all other tissue or cell selective targets, including, for tumor therapeutic purposes, tumor stroma markers such as fibroblast activation protein.

(59) Construction and Preparation of scTRAIL Fusion Proteins

(60) For generation of functionally improved scFv-TRAIL fusion proteins, first, the genetic code of scTRAIL was adapted for higher protein yields in mammalian expression systems. Among the various linker motifs suitable to connect 3 Trail molecules (components A) the (GGGS)1-4 motifs were tested, with the shorter linkers (GGGS).sub.1 (SEQ ID NO: 47) and (GGGS).sub.2 (SEQ ID NO: 48) being superior to longer linkers with respect to protein stability, tendency to aggregate and display identical or better apoptosis inducing activity.

(61) The inventors used EGFR targeting as a model system. Like other members of the erbB family of receptor tyrosine kinases, the EGFR (erbB1) is an established tumor marker, which is overexpressed in several carcinomas, including lung and liver cancer (Olayioye et al, 2000). For generation of the EGFR specific fusion protein of SEQ ID NO: 96, the construct was N-terminally fused with component B, a humanized and codon-optimized antibody fragment derived from the anti-EGFR mAb Cetuximab (C225) (Naramura et al, 1993) (SEQ ID NO: 94). A FLAG tag (F) was placed N-terminal of component B for purification and detection purposes (FIG. 1A). TRAIL bioactivity depends on the oligomerization state, in particular relevant for TRAILR2 (DR5), which is poorly activated by soluble, trimeric forms of TRAIL (Wajant et al, 2001). For SEQ ID NO: 102 the linker L between V.sub.H and V.sub.L was shortened from (GGGGS).sub.3 (SEQ ID NO: 52) to GGGGS (SEQ ID NO: 50). In an independent approach to improve basal protein stability and protection from proteolytic processing during expression culture conditions, two variants of SEQ ID NO: 96 were designed, in which i) the linker X connecting component A and B comprised two N-glycosylation sites, yielding a monomeric glycosylated form (SEQ ID NO:97) and ii) in addition, the two glycin linkers (P) connecting the three TRAIL components were replaced by linkers (P1, P2) each comprising two N-glycosylation sites, too (FIG. 1) (SEQ ID NO: 98).

(62) Following expression in stably transfected HEK293 cells, the purification of SEQ ID NO: 96 and SEQ ID NO: 102 was accomplished both by IMAC due to intrinsic histidine residues of TRAIL and by M2 mAb affinity chromatography. For example, yields of >3 mg highly pure protein per liter cell culture supernatant were achievable for both fusion proteins. SDS-PAGE and Western blot analysis of the purified proteins revealed single protein bands with an approx.molecular mass of 70 kDa and 100 kDa for scTRAIL (SEQ ID NO: 104) and EGFR specific V.sub.H-V.sub.L-scTRAIL fusion proteins SEQ ID NOs: 96 and 102, respectively, matching the expected calculated molecular masses of the single stranded monomers of 68, 93 and 94 kDa (FIG. 2). The apparent molecular mass of SEQ ID NO: 97 was increased compared to SEQ ID NO: 96, in accordance with effective glycosylation of the introduced linker. N-glycosidase treatment of SEQ ID NO: 97 resulted in a protein with a molecular mass essentially identical to that of its non-glycosylated derivative. The introduction of N-glycosylation sites in SEQ ID NO: 96 improved protein stability and protection from degradation during the production process, evident from strong reduction of degradation products present in culture supernatants (not shown). This allows a single-step purification of glycosylated variants such as SEQ ID NO: 97 with M2 agarose to gain a purification grade comparable to that of non-glycosylated fusion proteins after a two-step purification and thus overall higher yields of purified, bioactive protein.

(63) The gel filtration analysis of fusion proteins (SEQ ID NO: 104) and SEQ ID NO: 96 (both, with and without glycosylation) indicated that the majority of protein (>94%) exists as a monomer (FIG. 3), whereas retention times decrease from scTRAIL (SEQ ID NO: 104) to SEQ ID NO: 97, according to the increase in molecular size. Concerning scTRAIL (SEQ ID NO: 104), the molecular mass deduced from SEC was slightly lower compared to that calculated from SDS-PAGE (FIG. 2), which is a characteristic of this molecule and not a hint for degradation (Schneider et al, 2010). Interestingly, SEC of SEQ ID NO: 102 revealed the presence of two peaks. The major peak near 200 kDa could be attributed to a dimer, whereas the minor peak could represent trimeric and/or tetrameric forms of the fusion protein.

(64) TRAIL fusion proteins comprising an albumin binding domain (ABD) either between component A and component B (SEQ ID NO: 125) or at the C-terminus of the fusion protein (SEQ ID NO: 126) have been purified from HEK293 cells as described above. The increase in molecular weight, which is essentially the result of the introduction of the ABD, can be seen on the Coomassie stained SDS-PAGE gel depicted in FIG. 16.

(65) Size exclusion chromatographie experiments showed that the TRAIL fusion proteins comprising an albumin binding domain are indeed capable of binding both, humen and mouse serum albumin (FIG. 17).

(66) EGFR-specific Binding of scTRAIL Fusion Proteins

(67) The specific antigen binding of various scTRAIL fusion proteins to EGFR-positive cells was analysed by flow cytometry of two HCC cell lines. Whereas EGFRs in HepG2 cells were barely detectable and thus considered target antigen low/negative, EGFR expression was clearly revealed in Huh-7 cells (FIG. 4), although EGFR levels in this HCC line appear moderate compared with EGFR overexpressing A431 cells (data not shown). Consistent with this, the binding of labelled Cetuximab to the EGFR-positive Huh-7 cells can be blocked by preincubation with 0.5 μM of SEQ ID NO: 102 (FIG. 5). Incubation of HepG2 and Huh-7 cells with SEQ ID NO: 96 or SEQ ID NO: 102 resulted in binding of the proteins to both cell lines, but competition of fusion protein binding by preincubation of cells with the anti-EGFR huC225 (2 μM) was only possible on Huh-7 cells (FIG. 2C). The intermediate fluorescence signal observed upon competition of fusion protein binding to Huh-7 cells likely reflects binding of the TRAIL domain (component A of the fusion protein) to TRAIL receptors. This is consistent with the weaker and non-blockable binding of SEQ ID NO: 96 and SEQ ID NO: 102 on EGFR low/neg. HepG2 cells. Binding competition of both fusion proteins to EGFR-positive cells is an indicator for the structural integrity and functionality of the targeting domain (component B).

(68) Quantitative binding studies of TRAIL fusion proteins to EGFR+, DR4+5+NCI-H460 cells revealed significantly different (P=0.003) EC50 values for TRAIL fusion protein according to SEQ ID NO: 97 (3.6±0.3×10.sup.−10 M) and TRAIL fusion protein according to SEQ ID NO: 102 (1.6±0.3×10.sup.−10 M), implicating an avidity effect of the specific molecular composition of the divalent TRAIL fusion protein according to SEQ ID NO: 102 and therefore potentially superior targeting compared to TRAIL fusion protein according to SEQ ID NO: 97 (FIG. 12A). Furthermore, it has been investigated whether the TRAIL fusion protein according to SEQ ID NO: 102 exhibits the functional activity of blocking EGF-induced EGFR autophosphorylation. Cetuximab served as a positive control in this experiment. Functional blocking of EGF-stimulated receptor activation by the divalent TRAIL fusion protein according to SEQ ID NO: 102 could be demonstrated for both Colo205 (FIG. 12B, left panel) and Huh7 (FIG. 12B, right panel) cells.

(69) Target-independent Induction of Cell Death by scTRAIL Fusion Proteins

(70) To investigate the basic bioactivity of scTRAIL fusion proteins without the influence of targeting domains, we first analysed cell death induction on the target-negative cell lines HepG2 (hepatoma) and Jurkat (T cell leukemia) and compared it with a non targeting scTRAIL molecule. On all cell lines analysed, SEQ ID NO: 102 exerted an approximately tenfold increased bioactivity compared to SEQ ID NO: 96 or its glycosylated form (SEQ ID NO 97). As a reference, the bioactivity of a commercially available highly active TRAIL preparation, so-called ‘KillerTRAIL’ (Enzo Lifesciences), was found to be comparable with the activity of SEQ ID NO: 96. Due to the low or even deficient target antigen expression of HepG2 cells, the apoptosis inducing activity of SEQ ID NO: 102 and SEQ ID NO: 96 was not influenced by the presence of an at least 7-fold excess (70 nM) of Cetuximab (not shown). The bioactivity of SEQ ID NO: 96 on target negative cells did not differ from the one of SEQ ID NO: 104 (scTRAIL). On EGFR-negative, DR4.sup.− DR5.sup.weak Jurkat cells, which are known to be sensitive for apoptosis induced by TRAIL complexes but not by soluble TRAIL, we found a higher apoptosis-inducing activity of SEQ ID NO: 102 compared to KillerTRAIL and no reactivity towards the SEQ ID NOs: 96 and 104 (FIG. 6).

(71) EGFR-directed Enhancement of Cell Death by scTRAIL Fusion Proteins

(72) The EGFR-positive liver carcinoma cell line Huh-7 was chosen to demonstrate the enhancement in bioactivity achievable due to the receptor targeting capacity of SEQ ID NO: 96 and SEQ ID NO: 102. Compared to scTRAIL (SEQ ID NO: 104), SEQ ID NO: 96 showed tenfold better apoptosis-inducing activity on Huh-7 cells (FIG. 7). The competition of this bioactivity with an excess of Cetuximab (70 nM) revealed a right shift of EC50 of SEQ ID NO: 96 in the same order of magnitude, pointing to the functionality of EGFR targeting responsible for the improvement of scTRAIL bioactivity. Glycosylated SEQ ID NO: 97 exerted identical bioactivity compared to its non-glycosylated variant, indicating that glycosylation at this site did not impact bioactivity. SEQ ID NO: 102 showed a tenfold enhanced bioactivity in relation to SEQ ID NO: 96. The competition of activity of SEQ ID NO: 102 with the same molar excess of Cetuximab also resulted in a comparable shift in the dose response curve, confirming that targeting further improves the already increased bioactivity of this fusion protein. Furthermore, dimeric TRAIL fusion proteins comprising an albumin binding domain (ABD) (SEQ ID NOs: 125 and 126) exhibited bioactivity comparable to the dimeric TRAIL fusion protein lacking an ABD (SEQ ID NO: 102) indicating that the introduction of an ABD as performed for TRAIL fusion proteins according to SEQ ID NOs: 125 and 126 does not significantly influence the bioactivity of EGFR-directed enhancement of cell death (FIG. 18).

(73) In another experimental setup, Huh-7 cells were pretreated with a fixed dose of the various TRAIL fusion proteins followed by titration of the apoptosis sensitizer Bortezomib (FIG. 8). At a protein concentration of 1 nM and above, the tested SEQ ID NOs: 102, 96 and 104 were nearly equally efficient in induction of complete apoptosis in these cells when sensitized with Bortezomib. In contrast, at protein concentrations of 0.1 nM and below, a strong synergistic effect of Bortezomib sensitization and the EGFR targeting ability of the constructs became visible. At a protein concentration of 0.05 nM, only SEQ ID NOs: 96 and 102 were able to synergize with Bortezomib, whereby SEQ ID NO: 102 showed higher bioactivity compared to SEQ ID NO: 96 at this concentration. Superior activity of SEQ ID NO: 102 was even more apparent at 0.01 nM of fusion proteins (FIG. 8).

(74) A nearly complete block of cell death by either pan-caspase (zVADfmk) or caspase-3 selective (zDEVDfmk) inhibitors (FIG. 13A) and failure of necrostatin-1 to prevent or reduce cell death (data not shown) indicated that Huh-7 and Colo205 undergo predominantly apoptotic cell death upon treatment with TRAIL fusion proteins. Cetuximab blocked EGF-induced autophosphorylation of EGF receptors (FIG. 12B). Further, cetuximab by itself, though blocking EGF-induced autophosphorylation of EGFR in Colo205 and in Huh7 cells (FIG. 12B), did not substantially affect growth of these two cancer cell lines in a 4-day culture (FIG. 13B, C). Likewise, when SEQ ID NO: 102-induced apoptosis was prevented, either by presence of neutralizing anti-TRAIL antibodies in SEQ ID NO: 102 treated Colo205 cell cultures (FIG. 13B) or by treating Huh-7 cell cultures with pan-caspase inhibitors (FIG. 13C), only a marginal growth inhibition was noted during the 4 day observation period. Together, for the cells and the in vitro conditions studied here, the data indicate that i) SEQ ID NO: 102-induced cell death requires TRAIL signaling and ii) blocking EGFR function by the SEQ ID NO: 102 does not contribute to rapid apoptosis induction.

(75) Binding Affinity of SEQ ID NO: 96 and SEQ ID NO: 102 to Cells

(76) The specific molecular composition of SEQ ID NO: 102 implies avidity effects and thus potential superior targeting functions as compared to SEQ ID NO: 96. Therefore, we determined dissociation constants of both fusion proteins on EGFR-positive, DR4+5+NCI-H460 cells under equilibrium binding conditions at 4° C. by indirect immunofluorescence flow cytometry with an anti-TRAIL antibody. The K.sub.D of the interaction between the scTRAIL fusion proteins and NCI-H460 cells at 4° C. was determined by Lineweaver-Burk kinetic analysis (Lineweaver and Burk, 1934; Benedict et al, 1997) and was found to be 4-fold lower for the fusion protein of SEQ ID NO: 102 (6.5±0.9×10.sup.−8 M for SEQ ID NO: 96 and 1.7±1.2×10.sup.−8 M for SEQ ID NO: 102, respectively). In principle, the measured K.sub.D values reflect cell surface interactions of both functional domains in the fusion protein, the EGFR targeting domain (Component B) and the TRAIL domain, (components A), with their respective receptors. However, because SEQ ID NO: 102 binding to TRAILR+, EGFR negative cells such as Jurkat resulted in only very weak signals in this assay (data not shown), we reason that the signals revealed for NCI-H460 are largely due to binding of the fusion protein via its V.sub.H-V.sub.L domain to EGFRs. In fact, under the assay conditions (4° C.) applied, dynamic clustering of TRAILR that could account for stable receptor ligand interactions and thus apparent enhanced affinity is prevented. Therefore, we attribute this increased affinity largely to an avidity effect of the bivalent targeting domain (component B) of this particular fusion protein (SEQ ID NO: 102).

(77) Lack of systemic toxicity of SEQ ID NO: 102 and pharmacokinetics

(78) To assess whether the strongly increased bioactivity of SEQ ID NO: 102 in vitro diminishes the advantageous tumor selectivity of TRAIL, we studied systemic tolerance and effects of in vivo application on the reportedly most sensitive organ concerning untolerable TRAIL side effects, the liver. Groups of 3 CD1 mice were treated intraperitonally with indicated reagents (FIG. 9): neg. control: PBS; pos. control: aggregated FasL fusion protein; SEQ ID NO: 102 and SEQ ID NO: 96 (A) Plasma samples were prepared after 4 h and 24 h and the activity of alanine aminotransferase (ALT) was assayed using an enzymatic assay. Dashed line indicates upper normal level of ALT (35-50 U/L). (B) Mice were sacrificed after 24 h except for pos. control (animals treated with an aggregated FasL fusion protein show phenotypic signs of severe systemic toxicity after 2-4 hrs and die after ˜5 hrs, samples were taken after 4 hrs) and liver biopsies were taken for determination of caspase-3 activity using a specific AMC-coupled peptide substrate. The data clearly show that the SEQ ID NO: 102, despite its strongly increased apoptotic activity in vitro on target positive tumor cell lines (compared to non-targeted scTRAIL and the most active commercially available TRAIL) remains systemically well tolerated at doses up to 3 mg/kg in mouse models. Moreover, biochemical parameters of organ (liver) specific pathology confirm the phenotypic tolerance to this TRAIL fusion proteins, with only a transient marginal increase in ALT values to the upper normal limit 4 hrs after application. Lack of caspase activation and baseline ALT after 24 hrs, when bioactive fusion protein is still detectable in the blood (Tα.sub.1/2=2 h, Tβ.sub.1/23 h plasma conc at t=24 h: 100 ng/ml with an applied iv dose of 100 μg/animal; proof that SEQ ID NO: 102 maintain tumor selectivity and can be safely applied in vivo.

(79) Furthermore, a comparison of Huh-7 hepatoma cells and primary human hepatocytes (PHH) for caspase-3 activation by SEQ ID NO: 102 in presence of bortezomib showed a strong, bortezomib-dependent caspase-3 activation in the tumor cells, whereas normal liver cells were neither affected by the scTRAIL fusion protein alone nor in combination with bortezomib (FIG. 14A). These results were confirmed by immunoblot analysis of cleaved caspase-3 in Huh-7 and PHH, with no detectable caspase activation in combination treated PHH, whereas robust caspase processing was detectable in sensitive Huh7 carcinoma cells (FIG. 14B).

(80) Antitumoral Activity of SEQ ID NO: 102 in a Xenograft Tumor Model

(81) Given the in vitro data, showing superior bioactivity of the divalent TRAIL fusion protein according to SEQ ID NO: 102 compared with the monovalent TRAIL fusion protein according to SEQ ID NO: 97 or scTRAIL according to SEQ ID NO: 104 in particular at low protein concentrations, eight doses of 0.45 nmol protein were injected i.p. in a daily regimen in combination with bortezomib cotreatment every second day. The systemic treatment started after establishment of solid, vascularized tumors and tumor growth was monitored for 22 days. Bortezomib treatment by itself did not interfere with progressive tumor growth, whereas scTRAIL according to SEQ ID NO: 104 and the monovalent TRAIL fusion protein according to SEQ ID NO: 97 both delayed tumor growth, but at the low dosage applied, did not induce regression of tumors. In contrast, upon treatment with the divalent TRAIL fusion protein according to SEQ ID NO: 102, a strong reduction of tumor size and prolonged survival in all animals, with macroscopically undetectable tumors in 11/12 (+bortezomib) and 9/12 (w/o bortezomib) cases was recorded (FIG. 15). Interestingly, under the treatment conditions applied there was only a slight, but statistically not significant benefit of cotreatment with bortezomib, although at termination of treatment the combination group presented with slower regrowth of tumors (FIG. 15A).

(82) Introduction of an Albumin Binding Domain Increases the In Vivo Half-Life of TRAIL Fusion Proteins

(83) The pharmacokinetics, in particular, the in vivo half-lives for the dimeric TRAIL fusion protein according to SEQ ID NO: 102 lacking an albumin binding domain (ABD) and the dimeric fusion protein comprising an ABD between component A and component B of the TRAIL fusion protein (SEQ ID NO: 125) have been compared. To this end, 25 μg of fusion proteins were injected i.v. in CD1 mice and serum samples were analyzed at certain time points after injection by ELISA assay (FIG. 19). It has been demonstrated that the in vivo serum half-life increases from about 3 hours for the construct without ABD (SEQ ID NO: 102) to about 20 hours for the construct comprising an ABD (SEQ ID NO: 125) (FIG. 19). As indicated above, the constructs comprising an ABD exert a similar bioactivity compared to the constructs lacking an ABD (FIG. 18) indicating that the ABD does not negatively influence bioactivity, but exerts advantageous properties regarding pharmacokinetic properties, such as in vivo serum half-life. Thus, the TRAIL fusion proteins, preferably dimeric TRAIL fusion proteins, as described above comprising an ABD, such as the dimeric TRAIL fusion proteins according to SEQ ID NO: 125 and 126, are particularly preferred embodiments of the polypeptide according to the present invention.

(84) The inventors of the present invention have thus provided evidence for an improved concept in targeted cancer therapy. It may be that polypeptides according to the present invention form oligomers while polypeptides such as SEQ ID NO: 96 remain strictly monomeric. If so, it is surprising that the potential oligomeric structure of the polypeptides according to the present invention does not result in an increased systemic toxicity. The presence of higher-order aggregates in preparations of recombinant TRAIL constructs (e.g. His-TRAIL, crosslinked FLAG-TRAIL) has been reported previously to be responsible for an increased toxicity towards some non-malignant tissue cells (reviewed by Koschny et al, 2007). Thus, the inventors provide new formats of highly active and tumor selective TRAIL molecules with improved in vivo stability and pharmacokinetic properties, thus reaching an unprecedented potential as tumor therapeutic.