Antibodies Directed To Her 3 And Uses Thereof

Abstract

ABSTRACTANTIBODIES DIRECTED TO HER-3 AND USES THEREOFThe present invention relates to binding proteins that bind to HER-3 and polynucleotides encoding the same. Expression vectors and host cells comprising the same for the production of the binding protein of the invention are also provided. In addition, the invention provides compositions and methods for diagnosing and treating diseases associated with HER-3 mediated signal transduction and/or its ligand heregulin.

Information

Applicants

Inventors

Specification

Antibodies directed to HER-3 and uses thereof DESCRIPTION
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present Invention relates to binding proteins including antibodies and binding fragments tliereof that bind to HER-3 and polynucleotides encoding the same. Expression vectors and host cells comprising the same for the production of the binding protein of the invention are also provided. In addition, the invention provides compositions and methods for diagnosing and treating diseases associated with HER-3 mediated signal transduction and/or Its ligand heregulin.
2. Background of the Technology
The human epidermal growth factor receptor 3 (HER-3, also known as ErbB3) is a receptor protein tyrosine kinase and belongs to the epidermal growth factor receptor (EGFR) subfamily of receptor protein tyrosine kinases, which also includes HER-1 (also known as EGFR), HER-2, and HER-4 (Plowman et aL, Proc. Natl. Acad. Sci. U.S.A. 87 (1990), 4905-4909; Kraus etal., Proc. Natl. Acad. Sci. U.S.A. 86 (1989), 9193-9197; and Kraus et al., Proc. Naa. Acad. Sci. U.S.A. 90 (1993). 2900-2904). Like the prototypical epidermal growth factor receptor, the transmembrane receptor HER-3 consists of an extracellular ligand-binding domain (ECD), a dimerlzation domain within the ECD, a transmembrane domain, an intracellular protein tyrosine kinase domain (TKD) and a C-terminal phosphorylation domain.
The ligand Heregulin (HRG) binds to the extracellular domain of HER-3 and activates the receptor-mediated signaling pathway by promoting dimerlzation with other human epidermal growth factor receptor (HER)

family members and transphosphorylation of its intracellular domain. Dlmer formation between HER family members expands the signaling potential of HER-3 and is a means not only for signal diversification but also signal amplification. For example the HER-2/HER-3 heterodimer induces one of the most important mitogenic signals among HER family members.
HER-3 has been found to be overexpressed in several types of cancer such as breast, gastrointestinal and pancreatic cancers. Interestingly a correlation between the expression of HER-2/HER-3 and the progression from a non¬invasive to an invasive stage has been shown (Alimandi et ai. Oncogene 10,1813-1821; deFazio etai, Cancer B7, 487-498; Naidu et al., Br. J. Cancer 78, 1385-1390). Accordingly, agents that interfere with HER-3 mediated signaling are desirable. Murine or chimeric HER-3 antibodies have been reported, such as in US5968511, US5480968 and WO03013602.
A humanized monoclonal antibody against HER-2, Herceptin®, has recently been shown to interfere with HER-2 mediated signaling and is therapeutically effective in humans (Fendly et al., Hybridoma 6, 359-370; Hudziak et al., Mol. Cell. Biol. 9, 1165-1172; Stebbing et al., Cancer Treat. Rev. 26, 287-290). Herceptin* has been shown to act through two different mechanisms, i.e. the engagement of the effector cells of the immune system as well as a direct cytotoxic, apoptosis inducing effect.
However, only patients with highly amplified HER-2 respond significantly to Herceptin* therapy, thus limiting the number of patients suitable for therapy. In addition the development of resistance to drugs or a change in the expression or epitope sequence of HER-2 on tumor cells may render even those approachable patients unreactive with the antibody and therefore abrogating its therapeutic benefits. Therefore more drugs for target based therapies approaching further members of the HER family, such as HER-3, are needed.

BRIEF DESCRIPTION OF THE DRAWING FIGURES
Fig. 1 shows the extent of HER-3 expression In a panel of human cancer cell lines and demonstrates that HER-3 is expressed in a variety of human cancers.
Fig. 2 shows the results of the FACS analysis of HER-3 antibody binding to either Rati cells stably expressing the different members of the HER family or only empty vector.
Fig. 3 shows antibody binding competition bins mapped to HER3 domains.
Fig. 4. show the results of the indirect FACS Scatchard antibody affinity analysis performed with anti-HER-3 antibodies of the invention. The analysis indicates that the anti-HER-3 antibodies of the invention possess high affinities and strong binding constants for HER-3 expressed on the cell surface
Fig. 5 shows the accelerated endocytosis of HER-3 induced by anti-HER-3 antibodies of the invention.
Figs. 6 a-e show the results of a ligand competition assay performed with anti-HER-3 antibodies of the invention. The results demonstrate that the antibodies of the invention specifically reduce binding of f ^l]-a-HRG/f ^l]-p-HRG to cells expressing endogenous HER-3.
Fig. 7a shows the results of a HER-3 phosphotyrosine ELISA performed with anti-HER-3 antibodies of the invention. Antibodies according to the present invention were able to inhibit p-HRG-mediated HER-3 activation as indicated by increased receptor tyrosine phosphorylation. Furthermore fig. 7b shows representative results of this experiment with titrated antibody.
Fig. 8 shows the result of a p42/p44 MAP-Kinase ELISA performed with

anti-HER-3 antibodies of the invention. Antibodies according to the present invention were able to reduce p-HRG-mediated p42/p44 MAP-Klnase activation as indicated by increased IVIAP-Kinase phosphorylation.
Fig. 9 shows the result of a phospho-AKT ELISA perfonned with anti-HER-3 antibodies of the invention. Antibodies according to the present invention were able to reduce p-HRG-mediated AKT activation as indicated by AKT phosphorylation.
Fig. 10 shows the inhibition of MCF7 cell proliferation by human anti-HER-3 antibodies of the invention. Antibodies according to the present invention inhibit HRG-induced cell growth in human cancer cells.
Fig. 11 shows the transmigration of MCF7 cells inhibited by human anti-HER-3 antibodies of the invention.
Figs. 12a-i shows the inhibition of the anchorage independent cell growth by human HER-3 antibodies of the invention.
Fig. 13 shows the inhibition of xenograft growth of T47D human breast cancer cells by a human anti-HER-3 antibody of the invention.
Fig. 14 shows the reduction of BxPC3 human pancreas cancer cells in mice after administration of anti Her3 (U1-59 and U1-53) or anti EGFR (Erbitux) antibodies.
Fig. 15 shows the reduction of xenograft growth of BxPC3 human pancreas cancer cells by a human anti-HER-3 antibody of the invention and in combination with anti EGFR (Erbitux) antibodies.
Fig. 16 demonstrates that antibodies of the invention delay human melanoma (HT144) cell growth in nu/nu mice.

Fig. 17 shows the reduction of xenograft growth of HT-29 human colon carcinoma cells by human HER-3 antibodies of the Invention (U1-53, U1~59 and U1-7).
Fig. 18 shows the reduction of xenograft growth of Calu-3 human lung cancer cells by human anti-HER-3 antibodies of the invention (U1-59, U1-53 and U1-7).
Fig. 19 shows the reduction of xenograft growth of BxPC-3 human pancreas cancer cells by human anti-HER-3 antibodies of the invention (U1-7, U1-59 and U1-53).
Fig. 20 demonstrates that an antibody of the invention (U1-59) causes suppression of HER-3 in BxPC3 human pancreas cancer xenografts.
SUMMARY OF THE INVENTION
A first aspect of the present invention relates to an isolated binding protein that binds to HER-3.

protein or a CHO or other cell type transformed with a nucleic acid molecule encoding a binding protein of the invention.
Another aspect of the present invention relates to a method for producing a binding protein of the invention by preparing said binding protein from a tissue, product or secretion of an animal, plant or fungus transgenic for a nucleic acid molecule or nucleic acid molecules encoding the binding protein of the invention. Preferably, a binding protein of the invention is prepared from the tissue, product or secretion of a transgenic animal such as cow, sheep, rabbit, chicl1 X 106 G418 c.f.u./ml; m.o.l. of 10) for 4-12 h in the presence of Polybrene (4 mg/ml; Aldrich). After changing the medium, selection of Rati cells with G418 was started. Usually, stable clones were picked after selection for 21 days.
EXAMPLE 2: HER-3 expression in human cancer cell lines

Receptor tyrosine kinases, as for example HER-3, play a crucial role In the nitiation and progression of hyperproliferative diseases such as the transition from benign hyperplastic cell growth towards a malignant cararcinoma. Since HER-3 expression varies between tumor cells and normal tissue an analysis of HER-3 expression is a critical factor for identification of patient subgroups that would benefit from treatment with binding proteins of the invention. Thus, HER-3 expression was quantified in a panel of human cancer cell lines to elucidate the role of HER-3 in human cancer formation. Cancer cell lines were grown as recommended by the ATCC. In detail, 105 cells were harvested with 10 mM EDTA in PBS, washed once with FACS buffer (PBS, 3 % FCS, 0.4 % azide) and seeded on a 96-well round bottom plate. The cells were spun for 3 min at 1000 rpm to remove supernatant and then resuspended with a-HER-3 antibody 2D1D12 (WO03013602) (3 μg/ml). Cell suspensions were incubated on ice for 1 hr, washed twice with FACS buffer and resuspended with secondary antibody (100 μl/well) donkey-anti-human-PE (Jackson) diluted 1:50 in FACS buffer. The cell suspensions were incubated on ice and in the dark for 30 min, washed twice with FACS buffer and analyzed (FACS, Beckman Coulter). Fig. 1 shows representative results of the analysis and demonstrates that HER-3 is expressed in a variety of human cancers.
EXAMPLE 3: Immunization and titerino
The HER-3 ECD protein, that was prepared as described in Example 1 and C32 cells (Human melanoma; ATCC #CRL-1585) were used as an antigen. Monoclonal antibodies against HER-3 were developed by sequentially immunizing XenoMouse® mice (XenoMouse® strains: XMG1 and XMG4, Abgenix, Inc. Fremont, CA). XenoMouse® animals were immunized via footpad route for all injections. The total volume of each injection was 50 pi per mouse, 25 pi per footpad.
For cohort #1 (10 XMG1 mice), the initial immunization was with 10 pg of

HER-3 ECD protein admixed 1:1 (v/v) with TITERMAX GOLD® (Sigma. Oakville, ON) per mouse. The subsequent five boosts were made with 10 μg of HER-3 ECD protein admixed 1:1 (v/v) with 100 μg alum gel (Sigma, Oakville, ON) in pyrogen-free D-PBS. The sixth boost consisted of 10 μg of HER-3 ECD protein admixed 1:1 (v/v) with TITERMAX GOLD*. The seventh injection consisted of 10 μg of HER-3 ECD protein admixed 1:1 v/v with 100 |jg alum gel. A final boost was made with 10 μg HER-3 ECD protein in
pyrogen-free DPBS, without adjuvant. The XenoMouse* mice were immunized on days 0, 4, 7, 11, 15, 20, 24, and 29 for this protocol and fusions were performed on day 33. The two bleeds were made through Retro-Orbital Bleed procedure on day 13 after the fourth boost, on day 19 after the sixth boost. There was no cohort #2.
For Cohort #3 (10 XMG1 mice) and Cohort #4 (10 XMG4 mice), the first injection was with 107 C32 cells in pyrogen-free Dulbecco's PBS (DPBS) admixed 1:1 (v/v) with TITERMAX GOLD* per mouse. The next four boosts were with 107 C32 cells in pyrogen-free DPBS, admixed with 25 pg of Adju-Phos and 10 pg CpG per mouse. The sixth boost was with 107 C32 ceils in pyrogen-free DPBS, admixed 1:1 (v/v) with TITERMAX GOLD® per mouse. The seventh, eighth, ninth boosts were with 107 C32 cells in pyrogen-free DPBS, admixed with 25 pg of Adju-Phos and 10 pg CpG per mouse. From tenth to fourteen boosts were 5 pg of HER-3 ECD protein in pyrogen-free DPBS, admixed with 25 pg of Adju-Phos and 10 pg CpG per mouse. A final boost consisted of 5 pg of HER-3 ECD protein in pyrogen-free DPBS, without adjuvant. Both Cohort #3 and #4, the XenoMouse* mice were immunized on days 0, 3, 7, 11, 14, 17, 21, 24, 28, 33, 35, 38, 42 and 45 for this protocol and fusions were performed on day 49. The three bleeds were made through Retro-Orbital Bleed procedure on day 12 after the fourth boost, on day 19 after the sixth boost and on day 40 after twelfth boost.
Selection of animals for harvest bv titer
For cohort #1, anti-HER-3 antibody titers in the serum from immunized

XenoMouse mice were determined by ELISA against HER-3 ECD protein.
The specific titer of each XenoMouse* animal was determined from the optical density at 650 nm and is shown in Table 1 below. The titer value is the reciprocal of the greatest dilution of sera with an OD reading two-fold that of background. Therefore, the higher the number, the greater was the humoral immune response to HER-3 ECD.

EXAMPLE 4: Recovery of lymphocytes. B-Cell isolations, fusions and generation of hybridomas
Immunized mice were sacrificed and the lymph nodes were haryested and pooled from each cohort. The lymphoid cells were dissociated by grinding in DMEM to release the cells from the tissues, and the cells were suspended in DMEM. The cells were counted, and 0.9 ml DMEM per 100 million lymphocytes was added to the cell pellet to resuspend the cells gently but completely. Using 100 \i\ of CD90+ magnetic beads per 100 million cells, the cells were labeled by Incubating the cells with the magnetic beads at 4 oC for 15 min. The magnetically-labeled cell suspension containing up to 10o positive cells (or up to 2x10° total cells) was loaded onto a LS+ column and the column washed with DMEM. The total effluent was collected as the CD90-negatiye fraction (most of these cells were expected to be B cells).
The fusion was performed by mixing washed enriched B cells from above and nonsecretory myeloma P3X63Ag8.653 cells purchased from ATCC (Cat. No. CRL 1580) (Kearney et al, J. Immunol. 123, 1979, 1548-1550) at a ratio of 1:1. The cell mixture was gently pelleted by centrifugation at 800 g. After complete removal of the supernatant, the cells were treated with 2 to 4 ml of pronase solution (CalBiochem, Cat. No. 53702; 0.5 mg/ml in PBS) for no more than 2 min. Then 3 to 5 ml of FBS was added to stop the enzyme activity and the suspension was adjusted to 40 ml total volume using electro cell fusion solution, ECFS (0.3 M sucrose, Sigma, Cat. No. S7903, 0.1 mM magnesium acetate, Sigma, Cat. No. M2545, 0.1 mM calcium acetate, Sigma, Cat. No. C4705). The supernatant was removed after centrifugation and the cells were resuspended in 40 ml ECFS. This wash step was repeated and the cells again were resuspended in ECFS to a concentration of 2x106 cells/ml.
Electro-cell fusion was performed using a fusion generator, model ECM2001, Genetronic, Inc., San Diego, CA. The fusion chamber size used was 2.0 ml, using the following instrument settings: Alignment condition:

voltage: 50 V, time: 50 sec; membrane breaking at: voltage: 3000 V, time: 30 μsec; post-fusion holding time: 3 sec.
After ECF, the cell suspensions were carefully removed from the fusion chamber under sterile conditions and transferred into a sterile tube containing the same volume of Hybridoma Culture Medium (DMEM (JRH Biosciences), 15 % FBS (Hyclone), supplemented with L-glutamine, pen/strep, OPI (oxaloacetate, pyruvate, bovine insulin) (all from Sigma) and IL-6 (Boehringer Mannheim). The cells were incubated for 15 to 30 min at 37 "C, and then centrifuged at 400 g for five min. The cells were gently resuspended in a small volume of Hybridoma Selection Medium (Hybridoma Culture Medium supplemented with 0.5x HA (Sigma, Cat. No. A9666)), and the volume was adjusted appropriately with more Hybridoma Selection Medium, based on a final plating of 5x10^ B cells total per 96-well plate and 200 pi per well. The cells were mixed gently and pipetted into 96-well plates and allowed to grow. On day 7 or 10, one-half the medium was removed, and the cells were re-fed with Hybridoma Selection Medium.
EXAMPLE 5: Selection of candidate antibodies bv ELISA
After 14 days of culture, primary screening of hybridoma supematants from the cohort #1 (mice in cohort one were split arbitrarily into fusion #1 and #2) for HER-3-specific antibodies was performed by ELISA using purified his-tagged HER-3 ECD and counter-screening against an irrelevant his-tagged protein by ELISA using goat anti-hulgGFc-HRP (Caltag Inc., Cat. No. HI0507, using concentration was 1:2000 dilution) to detect human IgG binding to HER-3 ECD immobilized on ELISA plates. The old culture supernatants from the positive hybridoma cells growth wells based on primary screen were removed and the HER-3 positive hybridoma cells were suspended with fresh hybridoma culture medium and were transferred to 24-well plates. After 2 days in culture, these supernatants were ready for a secondary confirmation screen. In the secondary confirmation screen for HER-3 specific fully human IgGk antibodies, the positives in the first

screening were screened by ELISA with two sets of detective antibodies: goat anti-hulgGFc-HRP (Caltag Inc., Cat. No. HI0507, using concentration was 1:2000 dilution) for human gamma chain detection and goat anti-hig kappa-HRP (Southern Biotechnology, Cat. No. 2060-05) for human kappa light chain detection. There were 91 fully human IgG/kappa HER-3 specific monoclonal antibodies that were generated from cohort #1.
EXAMPLE 6: Selection of candidate antibodies bv FMAT/FACS
After 14 days of culture, hybridoma supernatants from the cohort #3 and #4 (fusion #3 and #4) were screened for HER-3-specific monoclonal antibodies by FMAT. In the primary screen, hybridoma supernatants at 1:10 final dilution were incubated with Rat1-Her3 cells expressing human l-IER-3 and 400 ng/ml CyS-conjugated Goat F(ab')2 anti-human IgG, Fc-specific antibody (Jackson ImmunoResearch, Cat. No. 109-176-098) at room temperature for 6 hr. The binding of antibodies and detection antibodies complex to cells were measured by FMAT (Applied Biosystems). Non¬specific binding of antibodies to the cells was determined by their binding to parental Rat 1 cells. A total of 420 hybridomas producing HER-3-specific antibodies were selected from primary screen of fusion #3. The supernatants from these expanded cultures were tested again using the same FMAT protocol and 262 of them were confirmed to bind to HER-3 expressing cells specifically. A total of 193 hybridomas producing HER-3 specific antibodies were selected from primary screen of fusion #4. The supernatants from these expanded cultures were tested by FACS and 138 of them were confirmed to bind to HER-3 expressing cells specifically. In the FACS confirmation assay, Rat1-Xher3 cells and parental Rati cells (as negative control) were incubated with hybridoma supernatants at 1:2 dilution for 1 hr at 40C in PBS containing 2 % FBS. Following washing with PBS, the binding of antibodies to the cells were detected by 2.5 pg/ml CyS-conjugated Goat F(ab')2 anti-human IgG, Fc-specific antibody (JIR#109-176-098) and 5 pg/ml PE-conjugated Goat F(ab')2 anti-human kappa-specific antibody (SB# 2063-09). After removing the unbound antibodies by washing with PBS, the

cells were fixed by cytofix (BD# 51-2090KZ) at 1:4 dilution and analyzed by FACSCalibur.
EXAMPLE 7: Selection of hvbridomas for cloning
Antibodies from cohorts 1 and 2 were selected for hybridoma cloning based on specificity for HER-3 over HER1 (EGFR). HER-2 and HER-4 in ELISA using purified recombinant extra-cellular domains available from, for example R&D Biosystems, and FACS-based analysis of human tumor cell lines expressing different HER family members, and a > 5-time increase in mean fluorescent intensity in FACS staining for HER-3 positive cells over background. Based on these criteria, a total of 23 hybridoma lines were selected for cloning by limiting dilution cell plating.
Antibodies from cohorts 3 and 4 were selected for hybridoma cloning based on specificity for HER-3 over HER-1 (EGFR), HER-2 and HER-4 plus three other criteria. The first criterion was an ELISA screen for antibodies with epitopes contained within the L2 domain of HER-3 (see Example "Structural Analysis of anti-HER-3 Antibodies in the Invention).
The second criterion was neutralization of binding of biotinylated heregulin-alpha to HER-3 expressing cells in a FACS based assay. SKBR-3 cells were harvested, washed in culture medium, pelleted via centrifugation and resuspended in culture medium. Resuspended cells were aliquoted into 96-well plates. The plates were centrifuged to pellet the cells. Test antibodies in exhaust hybridoma supernatants were added at 25μl/well and incubated for 1 hr on ice to allow antibody binding. Fifty μ1 of a 10 nM heregulin-alpha (R&D Biosystems, Minneapolis, MN) solution was added to each well for a final concentration of 5 nM and incubated on ice for 1.5 hr. Cells were washed in 150 μl PBS, pelleted by centrifugation and the supernatant removed. Cells were resuspended in 50 pi of goat anti-HRG-alpha polyclonal antibody at 10 μg/m\ and incubated for 45 min of ice. Cells were washed in 200 pi PBS, pelleted by centrifugation and the supernatant

removed. Fifty μl of a solution of rabbit Cy5-labelece anti-goat polyclonal antibody at 5 μg/ml plus 7AAD at 10 μg/ml was added and incubated on ice for 15 min. Cells were washed in 200 μl PBS, pelleted by centrlfugation and the supernatant removed. The cells were resuspended in 100 μl of FACS buffer and read in the FACS. Test HER-3 antibodies that reduced binding of heregulin-alpha were those that had lowest fluorescence intensity. As positive controls, 1:5 serial dilutions from 10,000 ng/ml to 16 ng/ml of a mouse HER-3 mAb (105.5) or the human IgGI HER-3 mAb, U1-49 was used. Negative controls were heregulin-alpha alone, cells alone, goat anti-heregulin-alpha polyclonal antibody alone and rabbit Cy5-iabeled anti-goat polyclonal antibody alone.
The third criterion was relative ranking for affinity and/or higher relative mean fluorescence intensity in FACS using HER-3 expressing cell lines. Relative ranking for affinity was performed by normalizing HER-3-specific antibody concentrations and plotting versus data from limiting antigen ELISA as follows.
Normalization of antigen specific antibody concentrations using high antigen EUSA
Using an ELISA method, supematants for concentration of antigen specific antibody were normalized. Using two anti-HER-3 human IgGI antibodies from cohort 1 of known concentration titrated in parallel, a standard curve was generated and the amount of antigen specific antibody in the test hybridoma supematants from cohorts 3 and 4 were compared to the standard. In this way, the concentration of human HER3 IgG antibody in each hybridoma culture was estimated.
Neutravidin plates were made by coating neutravidin @ 8 μg/ml in 1XPBS/0.05% sodium azide on Costar 3368 medium binding plates at 50 μl/well with overnight incubation at 4° C. The next day the plates were blocked with 1XPBS/1% skim milk. Photobiotinylated his-tagged-HER-3 ECD @ 500 ng/ml In 1XPBS/1% skim milk was bound to the neutravidin

plates by incubating for 1 hour at room temperature. Hybridoma supernatant, serially diluted 1:2.5 from a starting dilution of 1:31 to a final dilution of 1:7568 in1XPBS/1% skin milk/0.05% azide, was added at 50 μl/well, and then incubated for 20 hours at room temperature. Serially dilutions were used to ensure obtaining OD readings for each unknown in the linear range of the assay. Next, a secondary detection antibody, goat anti human IgG Fc HRP at 400 ng/ml in 1XPBX/17oskim milk was added at 50 μl/well. After 1 hour at room temperature, the plates were again washed 5 times with water and 50μl of one-component TMB substrate were added to each well. The reaction was stopped after 30 minutes by the addition of 50 μl of 1M hydrochloric acid to each well and the plates were read at wavelength 450nm. A standard curve was generated from the two lgG1 HER-3 mAbs from cohort 1, serially diluted at 1:2 from 1000 ng/ml to 0.06 ng/ml and assessed in ELISA using the above protocol. For each unknown, OD readings in the linear range of the assay were used to estimate the concentration of human HER-3 IgG in each sample.
The limited antigen analysis is a method that affinity ranks the antigen-specific antibodies prepared in B-cell culture supernatants relative to all other antigen-specific antibodies. In the presence of a very low coating of antigen, only the highest affinity antibodies should be able to bind to any detectable level at equilibrium. (See, e.g., PCT Publication WO/03048730A2 entitled "IDENTIFICATION OF HIGH AFFINITY MOLECULES BY LIMITED DILUTION SCREENING" published on June 12, 2003). In this instance, two mAbs from cohort 1, both of known concentration and known KD, were used as benchmarks in the assay.
Neutravidin plates were made by coating neutravidin at 8 μg/ml in 1XPBS/0.05% sodium azide on Costar 3368 medium binding plates at 50 ul/well with overnight incubation at 4° C. The next day the plates were blocked with 1XPBS/1% skim milk. Biotinylated his-tagged-HER-3 ECD (50 μl/well) was bound to 96-well neutravidin plates at five concentrations: 125, 62.5, 31.2, 15.6, and 7.8 ng/ml in 1XPBS/1% skim milk for 1 hour at room

temperature. Each plate was washed 5 times with water. Hybridoma supernatants diluted 1:31 in 1XPBS/1%skin milk/0.05% azide were added at 50 ul/well. After 20 hours incubation at room temperature on a shaker, the plates were again washed 5 times with dHaO. Next, a secondary detection antibody, goat anti human IgG Fc HRP (Horsh Radish Peroxidase) at 400 ng/ml in 1XPBS/1%skim milk was added at 50 Ml/well. After 1 hour at room temperature, the plates were again washed 5 times with 6H2O and 50^L of one-component TMB substrate were added to each well. The reaction was stopped after 30 minutes by the addition of 50μl. of 1M hydrochloric acid to each well and the plates were read at wavelength 450nm. OD readings from an antigen concentration that yielded OD values in the linear range were used in for data analysis.
Plotting the high antigen data, which comparatively estimates specific antibody concentration (see above for details), versus the limited antigen OD illustrated the relatively higher affinity antibodies, e.g., those that bound had higher OD in the limited antigen assay while having lower amounts of IgG HER-3 antibody in the supematant.
Hybridomas from cohorts 3 and 4 for the 33 best performing antibodies in these sets of assays were advanced to cloning by limiting dilution hybridoma plating.
Alternatively, FACS analysis of HER-3 expression of Ratl/pLXSN and Ratl/HER-3 cells showed similar results (no crossreactivity with endogenous rat epitopes) (Fig. 2).
In detail 1x10* cells were harvested with 10 mM EDTA in PBS, washed once with FACS buffer (PBS, 3 % FCS, 0.4 % azide) and seeded on a 96-well round bottom plate. The cells were spun for 3 min at 1000 rpm to remove supematant and then resuspended with the specific HER-family antibodies (3 pg/ml). Cell suspensions were incubated on ice for 45 min, washed twice with FACS buffer and resuspended with secondary antibody (100 μl/well)

donkey-anti-human-PE (Jackson Immunoresearch, PA) diluted 1:50 in FACS buffer. The cell suspensions were incubated on ice and in the dark for 30 min, washed twice with FACS buffer and analyzed (FACS, Beckman Coulter).
EXAMPLE 8: Structural analysis of anti-HER-3 antibodies of the Invention
In the following discussion, structural information related to antibodies prepared in accordance with the invention is provided. In order to analyze structures of antibodies produced in accordance with the present invention, genes encoding the heavy and light chain fragments were amplified out of the particular hybridoma. Sequencing was accomplished as follows:
The VH and VL transcripts were amplified from individual hybridoma clones in 96 well plate using reverse transcriptase polymerase chain reaction (RT-PCR). Poly(A)+-mRNA was isolated from approximately 2x10' hybridoma cells using a Fast-Track kit (Invitrogen). Four PCR reactions were run for each Hybridoma: two for light chain (kappa (K), and two for gamma heavy chain (y). The QIAGEN OneStep room temperature-PCR kit was used for amplification (Qiagen, Catalog No.210212). In the coupled room temperature-PCR reactions, cDNAs were synthesized with blend of room temperature enzymes (Omniscript and Sensiscript) using antisense sequence specific primer corresponded to C-K, or to a consensus of the CHI regions of Cy genes. Reverse transcription was performed at 50 °C for 1 hr followed by PCR amplification of the cDNA by HotStarTaq DNA Polymerase for high specificity and sensitivity. Each PCR reaction used a mixture of 5'-sense primers; primer sequences were based on the leader sequences of VH and VK available at the Vbase website (http://vbase.mrc-cpe.cam.ac.uk/).
PCR reactions were run at 94 °C for 15 min, initial hot start followed by 40 cycles of 94 "C for 30 sec (denaturatjon), 60 "C for 30 sec (annealing) and 72 "C for 1 min (elongation).

PCR products were purified and directly sequenced using forward and reverse PCR primers using the ABI PRISIVI BigDye terminator cycle sequencing ready reaction Kit (Perkin Elmer). Both strands were sequenced using Prism dye-terminator sequencing kits and an ABI 377 sequencing machine.
Sequence analysis
Analyses of human V heavy and V kappa cDNA sequences of the HERS antibodies were accomplished by aligning the HER-3 sequences with human germline V heavy and V kappa sequences using Abgenix in-house software (5AS). The software identified the usage of the V gene, the D gene and the J gene as well as nucleotide insertions at the recombination junctions and somatic mutations. Amino acid sequences were also generated in silico to identify somatic mutations. Similar results could be obtained with commercially available sequence analysis software and publicly available information on the sequence of human V, D, and J genes, e.g., Vbase (http://vbase.mrc-cpe.cam.ac.uk/).
Molecular cloning of mAb U1-59
Total RNA was extracted from the tissue culture well containing multiple hybridomas lineages, including the hybridoma lineage secreting antibody U1-59. A heavy chain variable region was amplified using 5'-leader VH family specific primers, with 3'-C-gamma primer. A major band was amplified using a VH4 primer, no other bands were visible. The VH4-34 gamma fragment was cloned into pCDNA expression vector in frame with a human gamma 1 constant region gene.
An IgM heavy chain variable region was amplified using 5" VH family specific primers with 3' mu constant region primer. A major band was amplified using VH2 primer, no other bands were visible. The VH2-5 mu fragment was cloned into pCDNA expression vector in frame with a human mu constant

region gene. V kappa chains were amplified and sequenced. Four kappa chain RT-PCR products were Identified. The products were sequenced and after sequence analysis via in silico translation, only three of them had open-reading frames. These three functional kappa chains were cloned out of the oligoclonal U1-59 hybridoma well identified based on V kappa gene usage as (1) VK1 A3-JK2, (2) VK1 A20-JK3 and (3) B3-JK1. All V-kappa were cloned into pCDNA expression vector in frame with a human kappa light chain constant region gene.
Transfections:
Each heavy chain was transfected with each of the kappa chains in transient transfections for a total of 6 heavy chain/kappa light chain pairs. The transfection of the gamma chain with the A20 kappa chain gave poor antibody expression, while no antibody was secreted or detected when the A20 kappa chain was co-transfected with the mu chain. A total of three IgG sups and two IgM sups were available for HER-3 binding assay.

Binding activity to HER-3+ cell lines was detected in FACS with the lgG1 mAb consisting of the VH4-34 and the 83 kappa chain. No other VHA/k combinations gave fluorescence signal above background in FACS using HER-3+ cell lines.
Binding competition of the anti-HER-3 antibodies
Multiplexed competitive antibody binning was performed as published in Jia et al. J Immunol Methods. 288, 91-98 (2004) to assess clusters of HER-3 antibodies that competed for binding to HER-3. Tested HER-3 antibodies


Epitope characterization of anti-HER-3 antibodies
The epitopes of human anti-HER-3 antibodies of the invention were characterized. First a dot blot analysis of the reduced, denatured HER-3-His tagged purified ECD protein showed absence of binding by the anti-HER-3 antibodies tested (U1-59. U1-61. U1-41, U1-46. U1-53. U1-43, U1-44, U1-47. U1-52, U1-40, U1-49)) demonstrating that all had epitopes sensitive to reduction of disulfide bonds, suggesting that all had discontinuous epitopes. Next, the antibodies were mapped to defined domains in the HER-3 molecule by engineering various human-rat HER-3 chimeric molecules, based on the division of the HER-3 extra-cellular domain into four domains:
1) LI (D1): the minor ligand-binding domain,
2) S1 (D2): the first cysteine-rich domain,
3) L2 (D3): the major ligand-binding domain, and
4) S2 (D4): the sec cysteine-rich domain.
The extra-cellular domain (ECD) of Human HER-3 cDNA was amplified from RATI-HER-3 cells. The rat HER-3 cDNAs was amplified by RT-PCR from rat liver RNA and confirmed by sequencing. The cDNAs expressing the ECD of human and rat Her3 were cloned into mammalian expression vectors as V5-His fusion proteins. Domains from the human HER-3 ECD were swapped into the scaffold provided by the rat HER-3 ECD by using the Mfe1, BstX1 and Drain internal restriction sites. By this means, various chimeric

rat/human HER-3 ECD HIS fusion proteins (amino acids 1-160, 161-358. 359-575, 1-358, 359-604) were constructed and expressed via transient transfection of HEK 293T cells. Expression of the constructs was confirmed using a rat polyclonal antibody against human HER-3. The human monoclonal antibodies were tested in ELISA for binding to the secreted chimeric ECDs.
Two of the human antibodies, including antibody U1-59, cross-reacted with rat HER-3. To assign binding domains, these mAbs were tested against a truncated form of HER-3 consisting of L1-S1-V5his tagged protein purified from the supernatant of HEK 293T cells transfected with a plasmid DNA encoding the expression of the L1-S1 extra-cellular domains of HER3. mAb U1-59 bound to the L1-S1 protein in ELISA, implying that its epitope is in L1-81. mAb 2.5.1 did not bind to the L1-S1 protein, implying that its epitope is in L2-S2. Further mapping of antibody U1-59 was accomplished using SELD1 time of flight mass spectroscopy with on-chip proteolytic digests of mAb-HER-3 ECD complexes.
Mapping U1-59 epitopes usino SELDI
Further mapping of antibody U1-59 was accomplished using a SELDI time of flight mass spectroscopy with on-chip proteolytic digests of mAb-HER-3 ECD complexes. Protein A was covalently bound to a PS20 protein chip array and used to capture mAb U1-59. Then the complex of the PS20 protein chip and the monoclonal antibody was incubated with HER-3-His purified antigen. Next the antibody-antigen complex was digested with high concentration of Asp-N. The chip was washed, resulting in retention of only the HER-3 peptide bound to the antibody on the chip. The epitope was determined by SELDI and identified by mass of the fragment. The identified 6814 D fragment corresponds to two possible expected peptides generated from a partial digest of the HER-3-his ECD. Both overlapping peptides map to the domain SI. By coupling SELDI results with binding to a HER-3 deletion construct, the epitope was mapped to residues 251 to 325.

The location of the binding domains in the extracellular part of HER-3 that are recognized by the human anti-HER-3 mAbs of the invention are summarized in Table 4. The epitope domain mapping results were consistent with results from antibody competition binding competition bins, with antibodies that cross-competed each other for binding to HER-3 also mapping to the same domains on HER-3 (Fig. 3).

XR = cross-reactive
EXAMPLE 9: Determination of canonical classes of antibodies
Chothia, et al. have described antibody structure in terms of "canonical

classes" for the hypervariable regions of each immunoglobulin chain (J. Mol. Biol., 1987 Aug 20,196(4):901-17). The atomic structures of the Fab and VL fragments of a variety of immunoglobulins were analyzed to determine the relationship between their amino acid sequences and the three-dimensional structures of their antigen binding sites. Chothia, et al. found that there were relatively few residues that, through their packing, hydrogen bonding or the ability to assume unusual phi, psi or omega conformations, were primarily responsible for the main-chain conformations of the hypervariable regions. These residues were found to occur at sites within the hypervariable regions and in the conserved β-sheet framework. By examining sequences of immunoglobulins having unknown structure. Chothia, et al. show that many immunoglobulins have hypervariable regions that are similar in size to one of the known structures and additionally contained identical residues at the sites responsible for the observed conformation.
Their discovery implied that these hypervariable regions have conformations close to those in the known structures. For five of the hypervariable regions, the repertoire of conformations appeared to be limited to a relatively small number of discrete structural classes. These commonly occurring main-chain conformations of the hypervariable regions were termed "canonical staictures." Further wori< by Chothia, et al. (Nature, 1989 Dec 21-28, 342 (6252):877-83) and others (Martin, et al. J. Mol. Biol., 1996 Nov 15, 263(5): 800-15) confirmed that there is a small repertoire of main-chain conformations for at least five of the six hypervariable regions of antibodies.
The CDRs of each antibody described above were analyzed to determine their canonical class. As is known, canonical classes have only been assigned for CDR1 and CDR2 of the antibody heavy chain, along with CDR1, CDR2 and CDR3 of the antibody light chain. The tables below summarizes the results of the analysis. The canonical class data is in the form of HCDR1-HCDR2-LCDR1-LCDR2-LCDR3, wherein "HCDR" refers to the heavy chain CDR and "LCDR" refers to the light chain CDR. Thus, for example, a canonical class of 1-3-2-1-5 refers to an antibody that has a

HCDR1 that falls into canonical class 1, a HCDR2 that falls into canonical class 3, a LCDR1 that falls into canonical class 2, a LCDR2 that falls into canonical class 1, and a LCDR3 that falls into canonical class 5.
Assignments were made to a particular canonical class where there was 70 % or greater identity of the amino acids in the antibody with the amino acids defined for each canonical class. The amino acids defined for each antibody can be found, for example, in the articles by Chothia, et al. referred to above. Table 5 and Table 6 report the canonical class data for each of the HER-3 antibodies. Where there was less than 70 % identity, the canonical class assignment Is marked with an asterisk ("*") to Indicate that the best estimate of the proper canonical class was made, based on the length of each CDR and the totality of the data. Where there was no matching canonical class with the same CDR length, the canonical class assignment is marked with a letter s and a number, such as "s18", meaning the CDR is of size 18. Where there was no sequence data available for one of the heavy or light chains, the canonical class is marked with "Z".

EXAMPLE 10: Determination of antibody affinity
Affinity measurements of anti-HER-3 antibodies of the inyention were performed by Indirect FACS Scatchard analysis. Therefore, 105 cells of interest or SK-Br 3 cells were haryested with 10 mM EDTA in PBS, washed once with FACS buffer (PBS, 3 % FCS, 0.4 % azide) and seeded on a 96-well round bottom plate. The cells were spun for 3 min at 1000 rpm to remove supernatant and then resuspended with a-HER-3 antibody (3 μg/ml) or with antibody dilutions (100 μll\Nell) starting with 20 μglml human monoclonal antibody in FACS buffer, diluted in 1:2 dilution steps. Cell suspensions were incubated on ice for 1 hr, washed twice with FACS buffer and resuspended with secondary antibody (100 μl/well) donkey-anti-human-PE (Jackson) diluted 1:50 in FACS buffer. The cell suspensions were Incubated on ice and in the dark for 30 min, washed twice with FACS buffer and analyzed (FACS, Beckman Coulter). According to the FACS Scatchard analysis, the fluorescence mean was calculated for each measurement. Background staining (= without 1st antibody) was subtracted from each fluorescence mean. Scatchard plot with x-yalue = fluorescence mean and y-value = fluorescence mean/concentration of mAb (nM) was generated. The KD was taken as the absolute yalue of 1/m of linear equation. Fig. 4 shows a kinetic analysis using the U1-59 antibody of the invention. In the following table 8 affinity measurements for certain antibodies of the invention selected in this manner are provided.


EXAMPLE 11: Anti-HER-3 antibodies of the invention induce HER-3 receptor endocvtosis
HER-3 lias been identified as a factor that can influence initiation and progression of hyperproliferative diseases through serving as an important gatekeeper of HER family mediated cell signaling. Thus, if HER-3 is effectively cleared from the cell surface/membrane by receptor internalization, cell signaling and therefore transformation and/or maintenance of cells in malignancy can be ultimately diminished or suppressed.
In order to investigate whether anti-HER-3 antibodies of the invention are

capable of inducing accelerated endocytosis of HER-3, the relative amount of HER-3 molecules on the cell surface after 0.5 and 4 hr incubation of the cells with anti-HER-3 antibodies of the invention were compared. 3x105 cells were seeded in normal growth medium in 24-well dish and left to grow overnight. Cells were preincubated with 10 MQ/ml anti-HER-3 mAbs in normal growth medium for the indicated times at 37 oC. Cells were detached with 10 mM EDTA and Incubated with 10 pg/ml anti-HER-3 mAbs in wash buffer (PBS, 3 % FCS, 0.04 % azide) for 45 min at 4 "C. Cells were washed twice with wash buffer, incubated with donkey-anti-human-PE secondary antibody (Jackson) diluted 1:100 for 45 min at 4 "C, washed twice with wash buffer and analyzed by FACS (BeckmanCoulter, EXPO).
Data shown in Fig. 5 demonstrate that treatment of cells with anti-HER-3 antibodies leads to internalization of the receptor. Data are shown as % internalization and refer to the reduction of the mean fluorescence intensity of anti-HER3 treated samples relative to control-treated samples.
EXAMPLE 12: Inhibition of ligand binding to human cancer cells SKBr3 bv human anti-HER-3 antibodies of the invention
Radioligand competition experiments were performed in order to quantitate the ability of the anti-HER-3 antibodies of the Invention to inhibit ligand binding to HER-3 in a cell based assay. Therefore, the HER-3 receptor binding assay was performed with 4x105 SK-BR-3 cells which were incubated with varying concentrations of antibodies for 30 min on ice. 1.25 nM [125]-a-HRG/125]-p-HRG were added to each well and the Incubation was continued for 2 hr on ice. The plates were washed five times, air-dried and counted in a scintillation counter. Figs. 6a-e show the results of these experiments performed with representative anti-HER-3 antibodies of the invention and demonstrate that the antibodies of the invention are capable of specifically reducing the binding of [125]-a-HRG/[125].μ.HRG ^ ^ells expressing endogenous HER-3.

EXAMPLE 13: Inhibition of lioand-induced HER-3 phosphorylation bv human anti-HER-3 antibodies of the Invention
ELISA experiments were performed in order to investigate whether the antibodies of the invention are able to block ligand μ-HRG-mediated activation of HER-3. Ligand-mediated HER-3 activation was detected by increased receptor tyrosine phosphorylation.
Day 1: 1 X 96 well dish was coated with 20 μg/ml Collagen I In 0,1 M acetic acid for 4 hr at 37 "C. 2.5x10* cells were seeded in normal growth medium
Day 2: Cells were starved in 100 μl serum free medium for 24 hr.
Day 3: Cells were preincubated with 10 pg/ml anti-HER-3 mAbs for 1 hr at 37 °C and then treated with 30 ng/ml p-HRG-EGF domain (R&D Systems) for 10 min. Medium was flicked out and cells were fixed with 4 % formaldehyde solution in PBS for 1 hr at room temperature. Formaldehyde solution was removed and cells were washed with wash buffer (PBS/0.1 % Tween 20). Cells were quenched with 1 % H2O2, 0.1 % NaNs in wash buffer and incubated for 20 min at room temperature, then blocked with NET-Gelantine for 5 hr at 4 "C. Primary antibody phospho-HER-3 (Tyr1289) (polyclonal rabbit; Cell signaling #4791; 1:300) was added overnight at 4 'C.
Day 4: The plate was washed 3x with wash buffer, then incubated with anti-rabbit-POD diluted 1:3000 in PBS - 0.5 % BSA was added to each well and incubated for 1.5 hr at room temperature. The plate was washed 3x with wash buffer and once with PBS. Tetramethylbenzidine (TMB, Calbiochem) was added and monitored at 650 nm. The reaction was stopped by addition of 100 pi 250 nM MCI and the absorbance was read at 450 nm with a reference wavelength of 650 nm using a Vmax plate reader (Thermo Lab Systems).
Fig. 7a shows representative results of this experiment, demonstrating that

anti-HER-3 antibodies of the invention were able to reduce ligand-mediated HER-3 activation as indicated by decreased receptor tyrosine phosphorylation. Data are shown as percent reduction by therapeutic antibodies relative to a control antibody.
To test potency of mAb U1-53 to inhibit ligand induced HER-3 activation, MCF-7 cells were starved for 24 hr, incubated with mAb U1-53 for 1 hr at 37 "C and stimulated with 10 nM HRG-p for 10 min. Lysates were transferred to 1B4 (mouse anti-HER-3 mAb) ELISA plates and phosphorylation of HER-3 was analysed with antibody 4G10. As shown in Fig. 7b phosphorylation of HER-3 was almost completely inhibited in a dose dependent manner with an IC50 of 0.14 nM.
EXAMPLE 14: Inhibition of ligand-induced D42/D44 MAP-Kinase phosphorylation bv human anti-HER-3 antibodies of the invention
Next ELISA experiments were performed in order to investigate whether the antibodies of the invention are able to block ligand β-HRG-mediated activation of p42/p44 MAP-Kinase. Ligand-mediated HER-3 activation was detected by increased protein (Thr202/Tyr204) phosphorylation.
Day 1: 1 X 96 well dish was coated with 20 pg/ml Collagen I in 0,1 M acetic acid for 4 hr at 37 "C. 3x10* cells were seeded in normal growth medium
Day 2: Cells were starved in 100 pi serum free medium for 24 hr.
Day 3: Cells were preincubated with 5 pg/ml anti-HER-3 mAbs for 1 hr at 37 "C and then treated with 20 ng/ml p-HRG-EGF domain (R&D Systems) for 10 min. Medium was flicked out and cells were fixed with 4 % formaldehyde solution in PBS for 1 hr at room temperature. Formaldehyde solution was removed and cells were washed with wash buffer (PBS/0.1 % Tween 20). Cells were quenched with 1 % H2O2. 0.1 % NaNs in wash buffer and incubated for 20 min at room temperature, then blocked with PBS/0.5 %

BSA for 5 hr at 4 oC. Primary antibody phospho-p44/p42 MAP Kinase (Thr202/Tyr204) (polyclonal rabbit; Cell signaling #9101; 1:3000) was added overnight at 4 oC.
Day 4: The plate was washed 3x with wash buffer, then incubated with anti-rabbit-HRP diluted 1:5000 in PBS - 0.5 % BSA was added to each well and incubated for 1.5 hr at room temperature. The plate was washed 3x with wash buffer and once with PBS. Tetramethylbenzidine (TMB, Calbiochem) was added and monitored at 650 nm. The reaction was stopped by addition of 100 pi 250 nM HCI and The absorbance was read at 450 nm with a reference wavelength of 650 nm using a Vmax plate reader (Thermo Lab Systems).
Fig. 8 shows representative results of this experiment. The antibodies of the invention were able to reduce ligand-mediated p42/p44 MAP-Klnase activation as indicated by decreased phosphorylation. Data are shown as percent reduction by therapeutic antibodies relative to a control antibody.
EXAMPLE 15: Inhibition of B-HRG-induced Phosoho-AKT ohosDhorvlation by human anti-HER-3 antibodies of the invention
In the following ELISA experiment we investigated whether the anti-HER-3 antibodies of the invention are able to block ligand p-HRG-mediated activation of AKT-Kinase. Ligand-mediated AKT activation was detected by increased protein (Ser473) phosphorylation.
Day 1: 1 X 96 well dish was coated with 20 pg/ml Collagen I in 0,1 M acetic add for 4 hr at 37 "C. 3x10® cells were seeded in normal growth medium
Day 2: Cells were starved in 100 pi serum free medium for 24 hr.
Day 3: Cells were preincubated with 5 pg/ml anti-HER-3 mAbs for 1 hr at 37 "C and then treated with 20 ng/ml |3-HRG-EGF domain (R&D Systems)

for 10 min. Medium was flicked out and cells were fixed with 4 % formaldehyde solution in PBS for 1 hr at room temperature. Formaldehyde solution was removed and cells were washed with wash buffer (PBS/0.1 % Tween 20). Cells were quenched with 1 % H2O2. 0.1 % NaNs in wash buffer and incubated for 20 min at room temperature, then blocked with PBS/0.5 % BSA for 5 hr at 4 μC. Primary antibody phospho-Akt (Ser473) (polyclonal rabbit; Cell signaling #9217; 1:1000) was added overnight at 4 μC.
Day 4: The plate was washed 3x with wash buffer, then incubated with anti-rabbit-HRP diluted 1:5000 in PBS-0.5 % BSA was added to each well and incubated for 1.5 hr at room temperature. The plate was washed 3x with wash buffer and once with PBS. Tetramethylbenzidine (TMB, Calbiochem) was added and monitored at 650 nm. The reaction was stopped by addition of 100 μl 250 nM HCI and The absorbance was read at 450 nm with a reference wavelength of 650 nm using a Vmax plate reader (Themno Lab Systems).
Fig. 9 shows representative results of this experiment. The anti-HER-3 antibodies of the invention were able to reduce p-HRG-mediated AKT as indicated by decreased phosphorylation. Data are shown as percent reduction by therapeutic antibodies relative to a control antibody.
EXAMPLE 16: Inhibition of a-HRG/B-HRG-mediated MCF7 cell proliferation bv human anti-HER-3 antibodies of the invention
In vitro experiments were conducted in order to determine the ability of the antibodies of the invention to inhibit HRG-stimulated cell proliferation. 2000 MCF7 cells were seeded in FCS-contalning medium on 96-well plates ovemight. Cells were preincubated in quadruplicates with antibody diluted in medium with 0.5 % FCS for 1 hr at 37 μC. Cells were stimulated with 30 ng/ml a- or 20 ng/ml β-HRG (R&D Systems) by adding ligand directly to antibody solution and were then left to grow for 72 hr. AlamarBlue™ (BIOSOURCE) was added and incubated at 37μC in the dark. Absorbance

was measured at 590 nm every 30 min. The data were taken 90 min after addition of alamar blue. The results as indicated in Fig. 10 show that representative antibodies of the invention inhibit HRG-induced cell growth in human cancer cells. Data are shown as percent reduction by therapeutic antibodies relative to a control antibody.
EXAMPLE 17: Inhibition of B-HRG-induced MCF7 cell migration bv human anti-HER-3 antibodies of the invention
Transmigration experiments were performed in order to investigate whether the antibodies of the invention block cell migration. Serum-starved MCF7 cells were preincubated by adding the indicated amount of antibody to the cell suspension and Incubating both for 45 min at 37 °C. 500 μl cell suspension (50,000 cells) was then placed in the top chamber of collagen I-coated transwells (BD Falcon, 8 μm pores). 750 μl medium (MEM, amino acids, Na-pyruvate, Pen.-Strept., 0,1 % BSA, without fetal calf serum) alone or containing the ligands 3-HRG-EGF domain (R&D Systems) were used in the bottom chamber. Cells were left to migrate for 8 hr at 37 "C and were stained with DAPI.
Stained nuclei were counted manually; percent inhibiton was expressed as inhibition relative to a control antibody.
Fig. 11 shows the result of the experiment demonstrating that representative anti-HER-3 antibodies of the invention reduce HRG-induced cell migration.
EXAMPLE 18: Colonv formation assav (soft aoar assav)
Soft agar assays were conducted in order to investigate the ability of the anti-HER-3 antibodies of the invention to inhibit anchorage independent cell growth. The soft agar colony formation assay is a standard in vitro assay to test for transformed cells, as only such transformed cells can grow in soft agar.

750 to 2000 cells (depending on the cell line) were preincubated with indicated antibodies at 10 pg/ml in IMDM medium (GIbco) for 30 min and resuspended in 0.4 % Difco noble agar. The cell suspension was plated on 0.75 % agarose underlayer containing 20 % FCS in quadruplicate in a 96-well plate. Colonies were allowed to form for 14 days and were then stained with 50 μl MTT (0.5 mg/ml in PBS) overnight. Figs. 12 a-l show the results of these experiments performed with three representative antibodies of the invention. These results demonstrate that anti-HER-3 antibodies of the invention reduce anchorage independent cell growth of MDA-MB361 and NCI-ADR breast cancer cells (Fig. 12a.b), MKN-28 gastric cancer (Fig. 12c), HT144 melanoma cells (Fig. 12d), Skov3 ovary carcinoma cells (Fig. 12e), PPC-1 prostate cancer cells (Fig. 12f),BX-PC3 pancreas cancer cells (Fig. 12g), A431 epidenmoid carcinoma cells (Fig. 12h) and lung carcinoma cells (Fig. 121). Colonies were counted with a Scanalyzer HTS camera system (Lemnatec, Wuerselen).
EXAMPLE 19: Human anti-HER-3 antibodies inhibit human breast carcinoma growth in nude mice
The anti-tumor efficacy of therapeutic antibodies is often evaluated in human xenograft tumor studies. In these studies, human tumors grow as xenografts in immunocompromised mice and therapeutic efficacy is measured by the degree of tumor growth inhibition. In order to determine, if the anti-HER-3 antibodies of the invention Interfere with tumor growth of human breast cancer cells in nude mice, 5x10° T47D cells were implanted in female NMRI nude/nude mice. Tumors were subcutaneous, grown on the back of the animal. Treatments began when tumors reached a mean volume of 20 mm'; eight days post implantation. Prior to first treatment, mice were randomized and statistical tests perfomned to assure uniformity in starting tumor volumes (mean, median and standard deviation) across treatment groups. Treatment started with a loading dose of 50 mg/kg followed by 25 mg/kg injections once a week by intraperitoneal injection. A control arm received doxorubicin (pharmaceutical grade). All animals were supplemented with 0.5

mg/kg/week oestrogen injected i.p.
Details of the treatment groups are given below.

Data for median tumor volume (Fig. 13) demonstrated that administration of an anti-HER-3 antibody of the invention resulted in reduction of tumor growth.
EXAMPLE 20: Human anti-HER-3 antibodies inhibit human pancreatic tumor growth in SCID mice
To test the therapeutic potential of anti-HER3 antibodies in other solid tumor types the anti-HER-3 antibodies, U1-53 and U1-59, were tested in mice with established tumors derived from the human pancreatic tumor cell line BxPC3. As controls sets of mice treated with either the vehicle control, PBS, or the established therapeutic antibody, Erbitux, were included. 5x108 BxPC3 cells were inoculated subcutaneously without Matrigel into CB17 SCiD mice. Mice bearing established tumors with a mean volume of 140mm2 received 50mg/kg of U1-53, U1-59, Erbitux or the equivalent volume of PBS via intraperitoneal injection. Thereafter the mice received once weekly 25mg/kg injections for the duration of the study.
The results for this experiment are shown in Fig. 14. U1-53 and U1-59 reduced the growth of the human pancreatic tumors in a cytostatic fashion. Notably, in this experiment, U1-53 and U1-59 were more effective than the EGFR-targeting antibody Erbitux at delaying tumor growth. These data demonstrated the therapeutic efficacy of anti-HER-3 antibodies of the

invention in comparison to a benchmark therapeutic agent.
EXAMPLE 21: Comibinina the human antl-HER-3 antibodies with anti-EGFR antibodies increases anti-tumor activity
The monotherapy of hyperproliferative diseases with targeted antibodies is often hampered by problems such as, on the one hand, the development of resistance to drugs, and on the other hand, a change in the antigenicity. For example, loss of antigenicity after prolonged treatment may render tumor cells insensitive to therapeutic antibodies, since those tumor cells that do not express or have lost the targeted antigen have a selective growth advantage. These problems might be evaded by using the antibodies of the invention in combination with a therapeutic antibody that targets a different receptor on the tumor cells, or another antineoplastic agent. Intervening in multiple signaling pathways or even related pathways but at multiple intervention steps might also provide therapeutic benefit. These combined treatment modalities are likely to be more efficacious, because they combine two anti-cancer agents, each operating via a different mechanism of action.
In order to demonstrate the feasibility of the anti-HER-3 antibodies of the invention, U1-53 and U1-59, as suitable combination agents, we compared monotherapeutic administrations of U1-53 or U1-59 with those in which either U1-53 or U1-59 was combined with the anti-EGR specific antibody, Erbitux. 5x106 BxPC3 cells were inoculated subcutaneously with Matrigel into CB17 SCID mice. After tumor volumes had reached 200 mm3 mice were randomized into individual treatment groups. Weekly intraperitoneal administrations of U1-53, U1-59 and Erbitux as single agents or combinations of either anti-HER3 antibodies with Erbitux or as a cocktail of two anti HER-3 antibodies were performed. All antibodies were dosed at a single loading doses of 50 mg/kg/week, followed by weekly injections of 25 mg/kg for six weeks. Control arms received bi-weekly administrations of Gemcitabine (120 mg/kg), weekly pooled human IgG or weekly vehicle (PBS) injections. The regimens are detailed below.


The results for this experiment are shown in Fig. 15. Antibodies U1-53 and U1-59, when administered as single agents, delayed the growth of the human pancreatic tumors to the same degree as Gemcitabine, which is often used as a standard anti-pancreatic cancer chemotherapy. Co¬administration of Erbitux with U1-53 or U1-59 resulted in a significantly greater reduction of tumor growth than observed with either single agent administration of U1-53, U1-59 or Erbitux. Thus, a beneficial therapeutic response can be achieved by combining the anti-HER-3 antibodies of the invention with suitable antibodies that target separate tumor antigens.
In summary, the anti-HER-3 antibodies of the invention have potent therapeutic efficacy against human tumors in vivo. They can be effectively combined with other anti-neoplastic therapeutics for increased anti-tumor activity.

EXAMPLE 22: Human anti-HER-3 antibodies inhibit human melanoma tumor growth In nu/nu mice
IVIembers of the erbB family of receptors, including Her3, are abnormally expressed in a large variety of epithelial cancers and they are known to play important roles in the growth and survival of many these solid tumors. These tumors include melanomas, head and neck squamous cell cancers, non-small cell lung cancers and prostate, glioma, gastric, breast, colorectal, pancreatic, ovarian cancers. In order to verify, that the anti-IHer3 antibodies of the invention are not restricted in their anti-cancer activity to individual tumor types, e.g. pancreatic cancers (see Example 21), but can be used as therapeutics against many HER-3-dependent tumors, we tested U1-53 and U1-59 in additional xenograft studies. One example is shown in Fig. 16. 5 x 105 human melanoma cells, HT144, were injected subcutaneously into CB17 SCID mice, followed by immediate subsequent intraperitoneal injection of 50mg/kg of U1-53 and U1-59, the equivalent volume of PBS or Dacarbacin (DITC) at 200mg/kg. Thereafter, mice received 25mg/kg of U1-53 or U1-59 once weekly, whereas DITC was given once every two weeks at 200mg/kg.
The median tumor volumes from each treatment group are shown in Figure 16. Administration of the antibodies of the invention resulted in growth reduction of the human melanomas when compared to tumors that had been treated with the vehicle control. These results demonstrate that the antibodies of the invention are not restricted in their therapeutic potential and target a wide variety of HER-3 expressing cancers.
EXAMPLE 23: Human anti-HER-3 antibodies inhibit growth of colon carcinoma xenografts in mice
HT-29 human colon carcinoma cells were suspended in medium with 2:1 ratio of Matrigel to a final concentration of 10 x 106 cells/ml. 0.2 ml of cell suspension were injected s.c. into the right flank of 4-5-week-old GDI nu/nu mice. A total of 95 mice were used.

The mice were randomly assigned to control and treatment groups. The treatment started on the same day. Duration of treatment was 29 days. Upon completion of the study, three tumours per group were collected 3 hours after administration of treatment. The tumours were fast-frozen and kept at -80 "C.
The following treatment protocol was carried out:
Control group: non-specific human IgG 25 mg/kg, twice weekly, intraperitoneal
Treatment group: antibody U1-53, 25 mg/kg, twice weekly, intraperitoneal
Treatment group: antibody U1-7, 25 mg/kg, twice weekly, intraperitoneal
Treatment group: antibody U1-59, 25 mg/kg, twice weekly, intraperitoneal
Treatment group 5-FU: 5-fluorouracil, 50 mg/kg, 9d x 5, intraperitoneal
The median tumor volumes from each group are shown in Fig. 17. Administration of the antibodies of the invention resulted in growth reduction of the HT-29 colon carcinoma tumors when compared to tumors that had been treated with non-specific human IgGl
EXAMPLE 24: Human anti-HER-3 antibodies inhibit lung cancer growth in mice
Calu-3 human non-small cell lung cancer cells were suspended in medium with 1:1 ratio of Matrigel to a final concentration of 5 x 106 cells/ml. 0.05 ml of cell suspension were injected s.c. into the right flank of 9-week-oid female CB17 scid mice. A total of 60 mice were used.

The mice were randomly selected to control and treatment groups. Treatment started on the same day. The duration of treatment was 32 days.
The following treatment protocol was carried out:
PBS vehicle group
hG control group: non-specific human IgG: 25 mg/kg. twice weekly, intraperitoneal
Treatment group antibody U1-53,25 mg/kg, twice weekly, intraperitoneal
Treatment group antibody U1-7, 25 mg/kg, twice weekly, intraperitoneal
Treatment group antibody U1-59, 25 mg/kg, twice weekly, intraperitoneal
The median tumor volumes from each control and treatment group are shown in Fig. 18. Administration of the antibodies of the invention resulted in growth reduction of the human non-small lung cancer xenografts when compared to tumors that had been treated with the PBS vehicle control or non-specific human IgG.
EXAMPLE 25: Human anti-HER-3 antibodies inhibit human pancreatic tumor growth in Balb/C-mice
Human pancreatic BxPC3 tumor cells were suspended in medium with a 2:1 ratio of Matrigel to a final concentration of 5 x 106 cells per ml. 0.2 ml of cell suspension were injected s.c. into the right flank of 5-7- week-old female BalbC nu/nu mice. A total of 100 mice were used.
The mice were randomly distributed into control and treatment groups. The treatment started on the same day. The treatment duration was 27 days.

The following treatment protocol was carried out:
hIgG control group: non-specific human lgG2, 25 mg/kg, twice weekly, Intraperitoneal
Treatment group antibody U1-53, 25 mg/kg, twice weekly, intraperitoneal
Treatment group antibody U1-7, 25 mg/kg, twice weekly, intraperitoneal
Treatment group antibody U1-59,25 mg/kg, weekly, intraperitoneal
Gemzar treatment group, gemcitabine, 80 mg/kg, weekly, intraperitoneal
The median tumor volumes from each control and treatment group are shown in Fig. 19. Administration of the antibodies of the invention resulted in growth reduction of the human pancreatic tumors when compared to tumors that had been treated with non-specific human IgG or with Gemzar.
The inhibition of HER-3 in the human pancreatic tumors could also be shown in a pharmacodynamic experiment. The BxPC3 tumor xenografts were grown as described above. 3 mice were treated with 500 μg of an lgG1 control antibody and 3 mice were treated with 500 μg of the anti-HER-3 antibody U1-59. The mice were treated on day 1 and day 4 and then sacrificed on day 5 to measure the antibody-dependent inhibition of HER-3 phosphorylation (pHER-3).
The tumors were homogenized in a standard RIPA buffer with protease inhibitors. 50 pg clear lysate was separated on a 4-20 % Tris-glycine gel, transferred onto a nitrocellulose membrane and blocked in 3 % bovine serum albumin (BSA). Immunoblotting was performed using an anti-pHER-3 antibody (antibody 21D3, Cell Signaling technology). An anti-actin antibody (AB a-2066, Sigma) was used as a control.

The expression was detected by enhanced chemiluminescence (Annersham Biosciences, PIscataway, NJ). The images were captured with the Versadoc 5000 Imaging System (BioRad, Hercules, CA).
The results are shown in Fig. 20. After administration of the human anti-HER-3-antibody U1-59 phosphorylation of HER-3 was no longer detectable. Thus, the antibodies of the invention are capable of significantly reducing HER-3 activation in human pancreatic tumor cells.
EXAMPLE 26: Use of antl-HER-3 antibodies of the invention as a diagnostic agent
Anti-HER-3 mAb can be used in the diagnostic of malignant diseases. HER-3 is expressed on tumor cells in a very distinct way compared to normal tissue and, therefore, an expression analysis of HER-3 would assist in the primary diagnosis of solid tumors, staging and grading of solid tumors, assessment of prognostic criteria for proliferative diseases and neoplasias and risk management in patients with HER-3 positive tumors.
A. Detection of HER-3 antigen in a sample
An Enzyme-Linked Immunosorbent Assay (ELISA) for the detection of HER-3 antigen in a sample is developed. In the assay, wells of a microtiter plate, such as a 96-well microtiter plate or a 384-well microtiter plate, are adsorbed for several hr with a first fully human monoclonal antibody directed against the HER-3 antigen. The immobilized antibody serves as a capture antibody for any of the HER-3 antigen that may be present in a test sample. The wells are rinsed and treated with a blocking agent such as milk protein or albumin to prevent nonspecific adsorption of the analyte.
Subsequently the wells are treated with a test sample suspected of containing the HER-3 antigen, or with a solution containing a standard amount of the HER-3 antigen. Such a sample is, for example, a serum

sample from a subject suspected of having levels of circulating HER-3 antigen considered to be diagnostic of a pathology. After rinsing away the test sample or standard, the wells are treated with a second fully human monoclonal anti-HER-3 antibody of the invention that is labelled by conjugation with biotin. The labeled anti-HER-3 antibody serves as a detecting antibody. After rinsing away excess secondary antibody, the wells are treated with avidin-conjugated horseradish peroxidase (HRP) and a suitable chromogenic substrate. The concentration of the HER-3 antigen in the test samples is determined by comparison with a standard curve developed from the standard samples.
B. Detection of HER3-antiQen in Immunohistochemistrv (IHC^
In order to determine HER3-antigen in tissue sections by IHC. Paraffin-embedded tissues are first deparaffinized in xylene for 2 x 5 min and then hydrated with 100% Ethanol 2x3 min, 95% Ethanol 1 min and rinsed in distilled water. Antigenic epitopes masked by formalin-fixation and paraffin-embedding are exposed by epitope unmasking, enzymatic digestion or saponin. For epitope unmasking paraffin sections are heated in a steamer, water bath or microwave oven for 20-40 min in a epitope retrieval solution as for example 2N HCI solution (pH 1.0). In the case of an enzyme digestion, tissue sections are incubated at 37''C for 10-30 minutes in different enzyme solutions such as protienase K, trypsin, pronase, pepsin etc.
After rinsing away the epitope retrieval solution or excess enzyme, tissue sections are treated with a blocking buffer to prevent unspecific interactions. The primary antibody is incubated at appropriate dilutions in dilution buffer for 1 hour at room temperature or overnight. Excess primary antibody is rinsed away and sections are incubated in peroxidase blocking solution for 10 min at room temperature. After another washing step, tissue sections are incubated with a secondary antibody labelled with a group that might serve as an anchor for an enzyme. Examples therefore are biotin labelled secondary antibodies that are recognized by streptavidin coupled

horseradish peroxidase. Detection of said antibody/enzyme complex is achieved by incubating with a suitable chromogenic substrate.
C. Determination of HER-3 antigen concentration in serum of patients
A sandwich ELISA is developed to quantify HER-3 levels in human serum. The two fully human monoclonal antl-HER-3 antibodies used in the sandwich ELISA, recognized different domains on the HER-3 molecule and do not compete for binding, for example, see Example 8. The ELISA is performed as follows: 50 \i\ of capture anti-HER-3 antibody in coating buffer (0.1 M NaHCOa. pH 9.6) at a concentration of 2 pg/ml were coated on ELISA plates (Fisher). After incubation at 4 °C overnight, the plates are treated with 200 pi of blocking buffer (0.5 % BSA, 0.1 % Tween 20, 0.01 % Thimerosal in PBS) for 1 hr at 25 °C. The plates were washed (3x) using 0.05 % Tween 20 in PBS (washing buffer, WB). Normal or patient sera (Clinomics, Bioreclaimation) are diluted in blocking buffer containing 50 % human serum. The plates are incubated with serum samples overnight at 4 "C, washed with WB, and then incubated with 100 pl/well of biotinylated detection anti-HER-3 antibody for 1 hr at 25 "C. After washing, the plates are incubated with HRP-Streptavidin for 15 min, washed as before, and then treated with 100 pl/well of o-phenylenediamine in H2O2 (Sigma developing solution) for color generation. The reaction is stopped with 50 pl/well of H2S04 (2 M) and analyzed using an ELISA plate reader at 492 nm. The concentration of HER-3 antigen in serum samples is calculated by comparison to dilutions of purified HER-3 antigen using a four parameter curve fitting program.
Staging of cancer in a patient
Based on the results set forth and discussed under items A, B and C, through use of the present invention, it is possible to stage a cancer in a subject based on expression levels of the HER-3 antigen. For a given type of cancer, samples of blood are taken from subjects diagnosed as being at

various stages in the progression of the disease, and/or at various points in the therapeutic treatment of the cancer. The concentration of the HER-3 antigen present in the blood samples is determined using a method that specifically determines the amount of the antigen that is present. Such a method includes an ELISA method, such as the method described under items A. and B. Using a population of samples that provides statistically significant results for each stage of progression or therapy, a range of concentrations of the HER-3 antigen that may be considered characteristic of each stage is designated.
In order to stage the progression of the cancer in a subject under study, or to characterize the response of the subject to a course of therapy, a sample of blood is taken from the subject and the concentration of the HER-3 antigen present in the sample is determined. The concentration so obtained is used to identify in which range of concentrations the value falls. The range so identified con-elates with a stage of progression or a stage of therapy identified in the various populations of diagnosed subjects, thereby providing a stage in the subject under study.
EXAMPLE 27: Uses of anti-HER-3 antibodies and antibody conjugates of the invention for treatment or prevention of hvperoroliferative diseases
Many solid tumors are driven by HER family mediated signalling and it has been demonstrated that HER-3 is a crucial partner through complex formation between HER-1, HER-2 and HER-4. Therefore, a reduction or elimination of HER-3 mediated signaling would impact all other HER family members and impair cell signaling leading to a wide window of therapeutic interventions and potential in combination therapy with other targeted agents, biologies and cytotoxic agents. Thus, anti-HER-3 antibodies of the invention can be used for treatment of certain hyperproliferative or HER-3 assodated disorders, that are based on a number of Actors as for example HER-3 expression. Tumor types as breast cancer, gastrointestinal cancer, pancreas cancer, prostate cancer, ovarian cancer, stomach cancer.

endometrial cancer, salivary gland cancer, lung cancer, kidney cancer, colon cancer, colorectal cancer, thyroid cancer, bladder cancer, glioma, melanoma, other HER-3 expressing or overexpressing cancers, appear to present prefen-ed indications, but indications are not limited to those on the preceding list In addition the following groups of patients will benefit from anti-HER-3 directed mAb treatment:
• Patients with resistance to anti-HER-2 mAb treatment
• Patients not eligible for the treatment with anti-HER-2 mAb
• Patients with resistance to anti-HER-1 mAb or small molecule anti-EGFR inhibitor
• Patients with non-small cell lung cancer resistant to eriotinib or gefitinib.
Anti-HER-3 antibodies of the invention would be used either as a monotherapy or in combination with one or more agent in a so called "combination therapy". Said combination therapy may include, but is not limited to, agents that were specified previously in the invention. Combination therapy with anti-HER3 antibodies and other agents may extend patient survival, time to tumor progression or quality of patient life. Protocol and administration design will address therapeutic efficacy as well as the ability to reduce the usual doses of standard therapies, as for example chemo- or radiation therapy.
Treatment of humans with anti-HER-3 antibodies of the invention
To determine the in vivo effects of anti-HER-3 antibody treatment in human patients with tumors, such human patients are injected over a certain amount of time with an effective amount of anti-HER-3 antibody of the invention. At periodic times during the treatment, the human patients are monitored to determine whether their tumors progress, in particular, whether the tumors grow and metastasize.
A tumor patient treated with the anti-HER-3 antibodies of the invention has a lower level of tumor growth and/or metastasis compared to the level of tumor growth and metastasis in tumor patients treated with the current standard of

care therapeutics.
Treatment with anti-HER-3 antibody conjugates of the invention
To determine the in vivo effects of anti-HER-3 antibody conjugates of the invention, human patients or animals exhibiting tumors are Injected over a certain amount of time with an effective amount of anti-HER-3 antibody conjugate of the invention. For example, the anti-HER-3 antibody conjugate administered is DM1-anti-HER-3 antibody conjugate, an auristatin-anti-HER-3 antibody conjugate or radioisotope-anti-HER-3 antibody conjugate. At periodic times during the treatment, the human patients or animals are monitored to determine whether their tumors progress, in particular, whether the tumors grow and metastasize.
A human patient or animal exhibiting tumors and undergoing treatment with, for example, DM1-anti-HER-3 antibody or radioisotope-anti-HER-3 antibody conjugates has a lower level of tumor growth and metastasis when compared to a control patient or animal exhibiting tumors and undergoing treatment with an alternate therapy. Control DM1-antibodies that may be used in animals include conjugates comprising DM1 linked to antibodies of the same isotype of the anti-HER-3 antibodies of the invention, but more specifically, not having the ability to bind to HER-3 tumor antigen. Control radioisotope-antibodies that may be used in animal tests include conjugates comprising radioisotope linked to antibodies of the same isotype of the anti-HER-3 antibodies of the invention, but more specifically, not having the ability to bind to HER-3 tumor antigen. Note: the control conjugates would not be administered to humans.
GENERAL REMARKS
The foregoing written speciHcation Is considered to be sufficient to enable one skilled in the art to practice the invention. The present invention is not to be limited in scope by the construct deposited, since the deposited

embodiment is intended as a single illustration of certain objects of the invention and any constructs that are functionally equivalent are within the scope of this invention. The deposit of material herein does not constitute an admission that the written description herein contained is inadequate to enable the practice of any object of the invention, including the best mode thereof, nor is it to be construed as limiting the scope of the claims to the specific illustrations that it represents.
The foregoing description and Examples detail certain preferred embodiments of the invention and describes the best mode contemplated by the inventors. It will be appreciated, however, that no matter how detailed the foregoing may appear in text, the invention may be practiced in many ways and the invention should be construed in accordance with the appended claims and any equivalents thereof.
Furthermore, unless othenA/ise defined, scientific and technical terms used in connection with the present invention shall have the meanings that are commonly understood by those of ordinary skill in the art. Moreover, unless otherwise required by context, singular terms shall include pluralities and plural terms shall include the singular. Generally, nomenclatures utilized in connection with, and techniques of, cell and tissue culture, molecular biology, and protein and oligo- or polynucleotide chemistry and hybridization described herein are those well known and commonly used in the art. Standard techniques are used for recombinant DNA, oligonucleotide synthesis, and tissue culture and transformation (e.g. electroporation, lipofection). Enzymatic reactions and purification techniques are performed according to manufacturer's specifications or as commonly accomplished In the art or as described herein. The foregoing techniques and procedures are generally performed according to conventional methods well known in the art and as described in various general and more specific references that are cited and discussed throughout the present specification. See e.g. Sambrook et al. Molecular Cloning: A Laboratory Manual (3rd ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. (2001)), which is

incorporated herein by reference. The nomenclatures utilized in connection with, and the laboratory procedures and techniques of, analytical chemistry, synthetic organic chemistry, and medicinal and pharmaceutical chemistry described herein are those well known and commonly used in the art. Standard techniques are used for chemical syntheses, chemical analyses, pharmaceutical preparation, formulation, and delivery, and treatment of patients.
INCORPORATION BY REFERENCE
All references cited herein, including patents, patent applications, papers, text books, and the like. Including the references cited therein, are hereby incorporated herein by reference in their entirety.

The isolated binding protein of any one of Claims 1-15, wherein the binding of the binding protein to HER-3 reduces HER-3 chosphorylation.
18. The isolated binding protein of any one of Claims 1-15, wherein the binding of the binding protein to HER-3 reduces cell proliferation.
19. The isolated binding protein of any one of Claims 1-15, wherein the binding of the binding protein to HER-3 reduces cell migration.
20. The Isolated binding protein of any one of Claims 1-15, wherein the binding of the binding protein to HER-3 increases the downregulation of HER-3.
21. The isolated binding protein of any one of Claims 1-20 which is an antibody.
22. The isolated binding protein of Claims 21, wherein the antibody is a monoclonal antibody, a polyclonal antibody, a recombinant antibody, a humanized antibody, a human antibody, a chimeric antibody, a multispecific antibody, or an antibody fragment thereof.
23. The isolated binding protein of Claim 22, wherein the antibody fragment is a Fab fragment, a Fab' fragment, a F(ab')2 fragment, a Fv fragment, a diabody, or a single chain antibody molecule.
24. The isolated binding protein of any one of Claims 1-23, wherein said isolated binding protein is of the IgGI-, lgG2-, lgG3- or lgG4-type.
25. The isolated binding protein of any one of Claims 1-

24, wherein the binding protein Is coupled to a labelling group.
26. The isolated binding protein of Claim 25, wherein the labelling group is a radioisotope or radionuclide, a fluorescent group, an enzymatic group, a chemiluminescent group, a biotinyl group, or a predetermined polypeptide epitope.
27. The isolated binding protein of any one of Claims 1-24, wherein the binding protein Is coupled to an effector group.
28. The Isolated binding protein of Claim 27, wherein the effector group is a radioisotope or radionuclide, a toxin, or a therapeutic or chemotherapeutic group.
29. The isolated binding protein of Claim 28, wherein the therapeutic or chemotherapeutic group is selected from the group consisting of calicheamicin, auristatln-PE, geldanamycin, maytansine and derivatives thereof.
30. An isolated nucleic acid molecule encoding an binding protein of any one of Claims 1-29.
31. The isolated nucleic acid molecule of Claim 30, wherein the nucleic acid molecule is operably linked to a control sequence.
32. A vector comprising the nucleic acid molecule of Claim 30.
33. A vector comprising the nucleic acid molecule of Claim 31.
34. A host cell transformed with the vector of Claim 32.

35. A host cell transformed with the vector of Claim 33.
36. A process for preparing the isolated binding protein of any one of Claims 1-29, comprising the step of Isolating said binding protein from a host cell
37. The process of Claim 36, wherein the host cell is a mammalian cell, a plant cell, a fungal cell, or a prokaryotic cell.
38. A pharmaceutical composition comprising as an active agent at least one isolated binding protein of any one of Claims 1-29, and a pharmaceutically acceptable carrier, diluent or adjuvant.
39. The composition of Claim 37 or 38 for therapeutic use.
40. The composition of Claim 37 or 38 for diagnostic use.
41. A method for treating or preventing a disease associated with HER-3 in a patient, comprising administering a phamnaceutically effective amount of a pharmaceutical composition of Claim 38 or 39 to a subject in need thereof.
42. The method of Claim 41, wherein the disease is a hyperproliferative disease.
43. The method of Claim 42, wherein said hyperproliferative disease is selected from the group consisting of breast cancer, gastrointestinal cancer, pancreas cancer, prostate cancer, ovarian cancer, stomach cancer, endometrial cancer, salivary gland cancer, lung cancer, kidney cancer, colon cancer, colorectal cancer, thyroid cancer, bladder cancer, glioma, melanoma, testis cancer, soft tissue sarcoma, head and neck cancer, other HER-3 expressing or overexpressing cancers, and formation of tumor metastases.

44. The method of Claim 42 or 43, wherein said hyperprollferative disease is associated with increased HER-3 phosphorylation, increased HER-2/HER-3 heterodimerization or an increased activity of Pl3-kinase, c-jun-terminal kinase. AkT, ERK2 and/or PYK2.
45. A method for diagnosing a disease associated with HER-3. comprising:
(a) contacting a sample with the binding protein of any
one of Claims 1-26, under conditions suitable to allow binding of said binding protein to HER-3; and
(c) identifying binding of said binding protein to HER-3.
46. The method of Claim 45, wherein the disease is a hyperprollferative disease.
47. The method of Claim 46, wherein said hyperprollferative disease is selected from the group consisting of breast cancer, gastrointestinal cancer, pancreas cancer, prostate cancer, ovarian cancer, stomach cancer, endometrial cancer, salivary gland cancer, lung cancer, kidney cancer, colon cancer, colorectal cancer, thyroid cancer, bladder cancer, glioma, melanoma, other HER-3 expressing or overexpressing cancers, testis cancer, soft tissue sarcoma, head and neck cancer, and formation of tumor metastases.
48. The method of Claim 46 or 47, wherein said hyperprollferative disease is associated with increased HER-3 phosphorylation, increased HER-2/HER-3 heterodimerization or an increased activity of PI3-kinase, c-jun-terminal kinase, AKT, ERK2 and/or PYK2.
49. A kit comprising the isolated binding protein of any one of Claims 1-28.

50. The kit of Claim 49, comprising a further therapeutic
agent.
51. The kit of Claim 50 wherein the further therapeutic
agent is an antineoplastic agent.
52. The kit of Claim 51, wherein the anti-neoplastic agent
is an anti-tumor antibody or a chemotherapeutic agent.

Documents

Orders