Neutralizing prolactin receptor antibody Mat3 and its therapeutic use

Abstract

The present invention is directed to the neutralizing prolactin receptor antibody Mat3, and antigen binding fragments, pharmaceutical compositions containing them and their use in the treatment or prevention of benign disorders and indications mediated by the prolactin receptor such as endometriosis, adenomyosis, non-hormonal female contraception, benign breast disease and mastalgia, lactation inhibition, benign prostate hyperplasia, fibroids, hyper- and normoprolactinemic hair loss, and cotreatment in combined hormone therapy to inhibit mammary epithelial cell proliferation and for the treatment and prevention of antiestrogen-resistant breast cancer. The antibody of the invention blocks prolactin receptor-mediated signalling.

Claims

1. An isolated nucleic acid molecule comprising: a nucleic acid sequence encoding an antibody Mat3 or an antigen-binding fragment thereof which antagonizes prolactin receptor (PRLR)-mediated signaling, wherein: a variable heavy chain comprises the sequences of CDR1, CDR2, and CDR3 corresponding to SEQ ID NO: 5, 6, and 7, respectively, a variable light chain comprises the sequences of CDR1, CDR2, and CDR3 corresponding to SEQ ID NO: 8, 9, and 10, respectively, the antibody or antigen-binding fragment thereof comprises an antigen-binding region that binds specifically to one or more regions of the extracellular domain of PRLR, and the PRLR is depicted by the amino acid sequence from position 1 to 210 of SEQ ID NO: 11, SEQ ID NO: 12, or SEQ ID NO: 13.

2. An isolated nucleic acid sequence according to claim 1, wherein the nucleic acid molecule comprises the nucleic acid sequence of SEQ ID NO: 3 and the nucleic acid sequence of SEQ ID NO: 4.

3. An expression vector comprising the nucleic acid molecule of claim 1.

4. A host cell comprising the vector of claim 3, whereby the host cell is a eukaryotic cell or a prokaryotic cell.

Description

DESCRIPTION OF THE FIGURES

(1) FIG. 1: Expression of prolactin-mRNA (PRL-mRNA) (analyzed by real-time TaqMan PCR analysis) in human endometrium and lesions (ectopic tissue) from healthy women and women suffering from endometriosis.

(2) FIG. 2: Expression of prolactin receptor-mRNA (PRLR-mRNA) (analyzed by real-time TaqMan PCR analysis) in human endometrium and lesions (ectopic tissue) from healthy women and women suffering from endometriosis.

(3) FIG. 3A: Neutralizing prolactin receptor antibody Mat3 inhibited sidebranching in mammary glands of mice which have been employed in a hyperprolactinemic surrogate model of benign breast disease. The unspecific antibody had no effect. Healthy normoprolactinemic mice (no pituitary) showed reduced sidebranching, whereas pituitary transplantation (hyperprolactinemia) enhanced sidebranching and lobuloalveolar development. The specific antibody Mat3 antagonized the effects of hyperprolactinemia.

(4) FIG. 3B: The neutralizing prolactin receptor antibody Mat3 inhibited the induction of the prolactin target gene elf5 in mammary glands of mice in a hyperprolactinemic surrogate model of benign breast disease. The unspecific antibody had no effect. Healthy, normoprolactinemic mice (no pituitary) showed reduced elf 5 expression in the mammary gland, whereas pituitary transplantation (hyperprolactinemia) strongly stimulated elf 5 gene expression. The specific antibody Mat3 but not the unspecific control antibody antagonized the effects of hyperprolactinemia.

(5) FIG. 4: Kabat Numbering of framework amino acid positions according to Johnson and Wu (Nucleic Acids Res. 2000, 28, 214-218).

(6) FIG. 5A: FACS analysis results with the anti-PRLR antibody HE06642. Binding of the antibody was determined at a fixed concentration on HEK293 cells expressing the human and mouse PRLR in comparison to the parental cell line not expressing PRLR. Y-axis: #, Median Fluorescence Intensity at 0.37 μg/ml HE06642 as IgG1 molecule. *, 1=HEK293 with human PRLR; 2=HEK293 with murine PRLR; 3=HEK293 without PRLR.

(7) FIG. 5B: FACS analysis results with the anti-PRLR antibody Mat3. Binding of the antibody was determined at a series of different concentrations on HEK293 cells expressing the human PRLR and Ba/F cells expressing the rhesus PRLR in comparison to a cell line (HEK293) not expressing PRLR. Maximal signal intensities at highest antibody concentrations depend on the number of PRLRs expressed on the cell surfaces, i.e. HEK293 and Ba/F cells do not carry the same number of PRLRs on their surface. Y-axis: #, Median Fluorescence Intensity; X-axis: °, different concentrations in μg/ml of the antibody Mat3 as IgG2 molecule; *, 1 (circles)=HEK293 with human PRLR; 2 (triangles)=Ba/F with rhesus monkey PRLR; 3 (squares)=HEK293 without PRLR.

(8) FIG. 6A: Alignment of the sequence region of the Mat3-VL domain with the most similar human V segment identified in VBASE2 (Mat3-VL is 90% identical to germline sequence humIGLV056).

(9) FIG. 6B: Alignment of the sequence region of the Mat3-VH domain with the most similar human V segment identified in VBASE2 (Mat3-VH is 90% identical to germline sequence humIGHV313).

(10) FIG. 6C: Alignment of the sequence region of the HE06642-VL domain with the most similar human V segment identified in VBASE2 (HE06642-VL is 80% identical to germline sequence humIGKV083).

(11) FIG. 6D: Alignment of the sequence region of the HE06642-VH domain with the most similar human V segment identified in VBASE2 (HE06642-VH is 89% identical to germline sequence humIGHV313).

(12) FIG. 7: ELISA-based binding tests of a maturated Fab variant: Fab-containing E coli supernatants were tested for binding to the immobilized extracellular domain of the human PRLR. The figure illustrates the binding of the Fab variants as a bar diagram. The signal intensities (extinction) are given on the y-axes (#), the names of the Fab variants (*) on the x-axes. Elevated signal intensity of the maturated Fab variant 005-C04-20-2 compared to the non-maturated Fab of the parental antibody 005-C04 demonstrate better PRLR-binding of 005-C04-20-2 compared to 005-C04. The “variant” pET28 represents a supernatant of an E. coli strain carrying the Fab-expression plasmid pET28a (Novagen, EMD Chemicals Group, Merck, Darmstadt, Germany) which does not express any Fab.

(13) FIG. 8A and FIG. 8B: Barplot representation of Pepscan ELISA results. Each plotted value represents the average value obtained for 54 peptides with S.E.M (standard error of mean). The black bars represent the relative binding strength of antibody HE06642. The white bars represent the relative binding strength of antibody Mat3. The data was normalized to average over the entire 2916 peptide dataset and corrected for background signal.

(14) FIG. 8A shows the ELISA results for a subset of peptides ranging from amino acids 103-PDPPLELAVEVKQPE-117 indicated as ‘117’ on the X-axis to 127-WSPPTLIDLKTGWFT-141 (indicated as ‘141’). I.e. these peptides, which were shifted by three amino acids along the ECD amino acid sequence, cover the region from amino acid position 103 to 141 of the ECD of human PRLR. The strongest differences observed within this dataset are for peptide 109-LAVEVKQPEDRKPYL-123 (indicated as ‘123’) with a significance p-value of 4×10.sup.−12, and for peptide 121-PYLWIKWSPPTLIDL-135 (indicated as ‘135’) with a significance p-value of 7×10.sup.−4.

(15) FIG. 8B shows the ELISA results for a subset of peptides ranging from 139-WFTLLYEIRLKPEKA-153 (indicated as ‘153’ on the X-axis) to 163-QQTEFKILSLHPGQK-177 (indicated as ‘177’). I.e. these peptides, which were shifted by three amino acids along the ECD amino acid sequence, cover the region from amino acid position 139 to 177 of the ECD of human PRLR. The strongest differences observed within this dataset are for peptide 148-LKPEKAAEWEIHFAG-162 (indicated as ‘162’) with a significance p-value of 6×10.sup.−26, and for peptide 160-FAGQQTEFKILSLHP-174 (indicated as ‘174’) with a significance p-value of 8×10.sup.−8.

(16) These data demonstrate that both antibodies bind to the S2 subdomain of the ECD of human PRLR (amino acid 101 to 210) and therefore are non-competitive to the natural ligand PRL which mainly binds to the S1 domain. However, this peptide scan showed that there are differences in binding to the S2 domains between Mat3 and HE06642. This finding indicates why the antibody Mat3 shows a different species-specificity and potency compared to HE06642.

(17) FIG. 9: Inhibition of prolactin-induced proliferation of BaF3 cells (monoclonal cells stably transfected with human prolactin receptor) by neutralizing prolactin receptor antibodies and unspecific control antibodies. The IC.sub.50 values were determined for the following antibodies in IgG1 format: Mat3 (closed circles), IC.sub.50=0.7 nM [100% inhibition at 1 μg/ml (=6.7 nM)]; HE06.642 (open squares), IC.sub.50=4.2 nM [81% inhibition at 1 μg/ml (=6.7 nM)]; unspecific control antibody (open triangles): no inhibitory effect.

(18) FIG. 10: Inhibition of prolactin-induced proliferation of BaF3 cells (monoclonal cells stably transfected with the murine prolactin receptor) by neutralizing prolactin receptor antibodies and unspecific control antibodies. The IC.sub.50 values were determined for the following antibodies in IgG1 format: Mat3 (closed circles), IC.sub.50=3.0 nM [97.4% inhibition at 1 μg/ml (=6.7 nM)]; HE06.642 (open squares), no inhibitory effect; unspecific control antibody (open triangles): no inhibitory effect.

(19) FIGS. 11A and B: Inhibition of prolactin-induced proliferation of BaF3 cells (monoclonal cells stably transfected with the rhesus prolactin receptor) by neutralizing prolactin receptor antibodies and unspecific control antibodies. The IC.sub.50 values were determined for the following antibodies in IgG1 format: Mat3 (closed circles), IC.sub.50=4.6 nM [99.1% inhibition at 1 μg/ml (=6.7 nM)] (see FIG. 11A); HE06.642 (open squares), IC.sub.50=206 nM [92.4% inhibition at 240 μg/ml (=1600 nM)] (see FIGS. 11A and B); unspecific control antibody (open triangles): no inhibitory effect (see FIGS. 11A and B).

(20) FIG. 12: Treatment of adenomyosis uteri (=endometriosis interna) in SHN mice with neutralizing PRLR antibody Mat3. The results are depicted on the y-axis as disease scores (adenomyosis scores) as described in Example 5. The median disease score for each experimental group is indicated as a horizontal bar. The experimental groups are the following ones: group 1, no pituitary isograft (normoprolactinemic mice develop endometriosis interna to some degree, median disease score=1); group 2, with pituitary isograft (hyperprolactinemia due to pituitary isografting enhances the disease score, median disease score=3); group 3, with pituitary isograft treated with unspecific murine IgG2a isotype control antibody once weekly at a dose of 30 mg/kg (median disease score=3); group 4, with pituitary isograft treated with antibody Mat3 in the murine IgG2a format (Mat3-mIgG2a) once weekly at a dose of 30 mg/kg (Mat3 completely cured the animals. The disease score after 30 mg/kg Mat3 given once weekly was even lower (median disease score=0) than the disease score of normoprolactinemic mice (median disease score=1); group 5, with pituitary isograft treated with antibody Mat3-mIgG2a once weekly at a dose of 10 mg/kg (median disease score=2); group 6, with pituitary isograft treated with antibody Mat3-mIgG2a once weekly at a dose of 3 mg/kg (median disease score=2.5); group 7, with pituitary isograft treated with antibody Mat3-mIgG2a once weekly at a dose of 1 mg/kg (median disease score=2.5). Treatment with antibody Mat3 shows a dose-dependent decrease in the median disease score. Mat3 is therefore suitable to treat endometriosis interna (=adenomyosis uteri) and endometriosis externa in women.

(21) FIGS. 13A and B: Inhibition of luciferase reporter gene activity in HEK293 cells stably transfected with the human and murine PRLR. In FIG. 13A the human PRLR-dependent activity is plotted against the antibody concentrations, while FIG. 13B shows the murine PRLR-dependent activity. The luciferase activity is given as percentage of the maximal luciferase activity obtained without addition of any antibody. The IC.sub.50 values were determined for the following antibodies in IgG1 format: Mat3 (closed circles), IC.sub.50=0.5 nM (FIG. 13A, hPRLR) and 1.3 nM (FIG. 13B, mPRLR); HE06.642 (open squares), IC.sub.50=4.6 nM (FIG. 13A, hPRLR) and >>1333 nM (=20 μg/ml) (FIG. 13B, mPRLR); unspecific isotype control antibody (open triangles): no inhibitory effect (see FIGS. 13A and B). These data show the improved activity of Mat3 on hPRLR compared to HE06.642 and demonstrate the activity of Mat3 on mPRLR whereas HE06.642 does not inhibit mPRLR-dependent luciferase activity.

(22) FIG. 14: Cell binding of neutralizing PRLR antibodies on cells expressing PRLR from human, mouse and monkey using flow cytometry. The median fluorescent signal intensity is plotted against the antibody concentration. The following IgG1 antibodies were applied: Mat3 (closed circles), HE06.642 (open squares), unspecific isotype control antibody (open triangles). Different cell lines were tested: A) HEK293 cell line stably transfected with human PRLR, B) HEK293 cell line stably transfected with murine PRLR, C) HEK293 cell line not transfected with any PRLR gene (negative control cell line), D) human breast cancer cell line T47D, E) BaF3 cell line stably transfected with rhesus monkey PRLR, F) BaF3 cell line stably transfected with human PRLR, G) BaF3 cell line stably transfected with murine PRLR. The cell line binding potency of the antibodies on the different cell lines have been deduced from the dose-response curves as EC.sub.50 values (see Table 8). The dose-response plots indicate the superior cell binding properties of Mat3 compared to HE06.642.

(23) Seq ID NO:1 represents amino acid sequence of VH, Mat3

(24) Seq ID NO:2 represents amino acid sequence of VL, Mat3

(25) Seq ID NO:3 represents nucleic acid sequence VH, Mat3

(26) Seq ID NO:4 represents nucleic acid sequence VL, Mat3

(27) Seq ID NO:5 represents amino acid sequence of HCDR1, Mat3

(28) Seq ID NO:6, represents nucleic acid sequence HCDR2, Mat3

(29) Seq ID NO:7 represents nucleic acid sequence HCDR3, Mat3

(30) Seq ID NO:8 represents nucleic acid sequence LCDR1, Mat3

(31) Seq ID NO:9 represents nucleic acid sequence LCDR2, Mat3

(32) Seq ID NO:10 represents nucleic acid sequence LCDR3, Mat3

(33) Seq ID NO:11 represents amino acid sequence of extracellular domain of cynomolgus and rhesus monkey PRLR fused to Fc-His

(34) Seq ID NO:12 represents human ECD_PRLR, amino acid position 1-210, S1 domain 1-100 (S1 domain construct 1-102), S2 domain 101-210

(35) Seq ID NO:13 represents murine ECD_PRLR, amino acid position 1-210

(36) Seq ID NO:14 represents amino acid sequence of VH, HE06642, Novartis (WO2008/22295)

(37) Seq ID NO:15 represents amino acid sequence of VL, HE06642, Novartis (WO2008/22295)

(38) Seq ID NO:16 represents nucleic acid sequence VH, HE06642, Novartis (WO2008/22295)

(39) Seq ID NO:17 represents nucleic acid sequence VL, HE06642, Novartis (WO2008/22295)

EXAMPLES

Example 1

(40) Inhibition of Prolactin-Induced Proliferation of BaF3 Cells (Monoclonal Cells Stably Transfected with Human Prolactin Receptor) by Neutralizing Prolactin Receptor Antibodies and Unspecific Control Antibodies.

(41) To analyze the in vitro efficacy of the neutralizing PRLR antibodies, the inhibition of prolactin-activated cellular proliferation of BaF3 cells was used. The cells were stably transfected with human PRLR and were routinely cultured in RPMI containing 2 mM glutamine in the presence of 10% FCS and 10 ng/ml of human prolactin. After six hours of starvation in prolactin-free medium containing 1% FCS, cells were seeded into 96-well plates at a density of 25000 cells per well. Cells were stimulated with 35 ng/ml prolactin and coincubated with increasing doses of neutralizing PRLR antibodies for two days. Cellular proliferation was analyzed using a CellTiter-Glo Luminescent Cell Viability Assay (Promega). Dose-response curves for the inhibition of prolactin-stimulated cellular growth were generated and IC.sub.50 values calculated. As negative control, stimulation with an unspecific control antibody was used. Antibody Mat3 was tested in comparison to antibody HE 06.642 (both were in the IgG1 format) (FIG. 9).

Example 2

(42) Inhibition of Prolactin-Induced Proliferation of BaF3 Cells (Monoclonal Cells Stably Transfected with the Murine Prolactin Receptor) by Neutralizing Prolactin Receptor Antibodies and Unspecific Control Antibodies.

(43) To analyze the in vitro efficacy of the neutralizing PRLR antibodies, the inhibition of prolactin-activated cellular proliferation of Ba/F3 cells was used. The cells were stably transfected with the murine PRLR and were routinely cultured in RPMI containing 2 mM glutamine in the presence of 10% FCS and 10 ng/ml of human prolactin. After six hours of starvation in prolactin-free medium containing 1% FCS, cells were seeded into 96-well plates at a density of 20000 cells per well. Cells were stimulated with 50 ng/ml prolactin and coincubated with increasing doses of neutralizing PRLR antibodies for three days. Cellular proliferation was analyzed using a CellTiter-Glo Luminescent Cell Viability Assay (Promega). Dose-response curves for the inhibition of prolactin-stimulated cellular growth were generated and IC.sub.50 values calculated. As negative control, stimulation with an unspecific control antibody was used. Antibody Mat3 was tested in comparison to antibody HE 06.642 (both were in the IgG1 format) (FIG. 10).

Example 3

(44) Inhibition of Prolactin-Induced Proliferation of BaF3 Cells (Monoclonal Cells Stably Transfected with the Rhesus Prolactin Receptor) by Neutralizing Prolactin Receptor Antibodies and Unspecific Control Antibodies.

(45) To analyze the in vitro efficacy of the neutralizing PRLR antibodies, the inhibition of prolactin-activated cellular proliferation of Ba/F3 cells was used. The cells were stably transfected with the rhesus PRLR and were routinely cultured in RPMI containing 2 mM glutamine in the presence of 10% FCS and 10 ng/ml of human prolactin. After six hours of starvation in prolactin-free medium containing 1% FCS, cells were seeded into 96-well plates at a density of 25000 cells per well. Cells were stimulated with 100 ng/ml prolactin and coincubated with increasing doses of neutralizing PRLR antibodies for two days. Cellular proliferation was analyzed using a CellTiter-Glo Luminescent Cell Viability Assay (Promega). Dose-response curves for the inhibition of prolactin-stimulated cellular growth were generated and IC.sub.50 values calculated. As negative control, stimulation with an unspecific control antibody was used. Antibody Mat3 was tested in comparison to antibody HE 06.642 (both were in the IgG1 format) (FIG. 11).

Example 4

(46) Quantitative Analysis of Prolactin and Prolactin Receptor Gene Expression by Real-Time TapMan PCR Analysis in Eu- and Ectopic Endometrium and Endometriotic Lesions from Patients and Healthy Controls.

(47) Real-time Taqman PCR analysis was performed using the ABI Prism 7700 Sequence Detector System according to the manufacturer's instructions (PE Applied Biosystems) and as described Endocrinolgy 2008, 149 (8): 3952-3959) and known by the expert in the field. Relative expression levels of PRL and the PRLR were normalized to the expression of cyclophyllin. We analyzed the expression of PRL and the PRLR in the endometrium from healthy women and in endometrium and endometriotic lesions from patients by using quantitative real-time Taqman PCR analysis. The expression of prolactin and its receptor was clearly upregulated in endometriotic lesions compared to healthy endometrium or endometrium derived from patients.

(48) Results are shown in FIGS. 1 and 2.

(49) These findings imply that autocrine prolactin signalling plays a role in the development and maintenance of endometriosis and adenomyosis uteri (endometriosis interna, a form of endometriosis restricted to the uterus.

Example 5

(50) Treatment of Adenomyosis Uteri (=Endometriosis Interna) in SHN Mice with Neutralizing PRLR Antibody Mat3.

(51) To test the efficacy of neutralizing PRLR antibodies in endometriosis, the adenomyosis uteri model in SHN mice relying on systemic hyperprolactinemia was employed (Acta anat. 116:46-54, 1983). Hyperprolactinemia in SHN mice was induced by pituitary isografting under the kidney capsule of 7 weeks old female mice (Acta anat. 116:46-54, 1983). Neutralizing PRLR antibody Mat3 (30 mg/kg; 10 mg/kg, 3 mg/kg, 1 mg/kg) or unspecific antibody (30 mg/kg) were administered subcutaneously starting two weeks after pituitary isografting. Animals were treated once weekly with the antibodies for seven weeks. The infiltration of the uterine muscular layer by glandular tissue was assessed as described previously (Laboratory Animal Science 1998, 48:64-68). At autopsy (day 66 after pituitary transplantation), uteri were fixed overnight in buffered 4% formalin and embedded in paraffin. The degree of adenomyosis (=endometriosis interna) was assessed as follows: Grade 0=no adenomyosis Grade 1=the inner layer of the myometrium looses its concentric orientation Grade 2=endometrial glands invading the inner layer of the myometrium Grade 3=endometrial glands between the inner and outer layer of the uterine myometrium Grade 4=endometrial glands invading the outer layer of the uterine myometrium Grade 5=endometrial glands outside of the outer layer of the uterine myometrium

(52) The experiment comprised the following experimental groups: 1. Animals without pituitary transplantation, i.e. normoprolactinemic mice 2. Animals with pituitary transplantation, i.e. hyperprolactinemic mice 3. Animals with pituitary transplantation, treated with unspecific control antibody once weekly at a dose of 30 mg/kg 4. Animals with pituitary transplantation, treated with the neutralizing prolactin receptor antibody Mat3 in the murine IgG2a format once weekly at a dose of 30 mg/kg 5. Animals with pituitary transplantation, treated with the neutralizing prolactin receptor antibody Mat3 in the murine IgG2a format once weekly at a dose of 10 mg/kg 6. Animals with pituitary transplantation, treated with the neutralizing prolactin receptor antibody Mat3 in the murine IgG2a format once weekly at a dose of 3 mg/kg 7. Animals with pituitary transplantation, treated with the neutralizing prolactin receptor antibody Mat3 in the murine IgG2a format once weekly at a dose of 1 mg/kg

(53) The neutralizing PRLR antibody Mat3 inhibited endometriosis interna (=adenomyosis) (FIG. 12). The neutralising PRLR antibody Mat3 is therefore suitable to treat endometriosis interna (=adenomyosis uteri) and endometriosis externa in women.

Example 6

(54) Neutralizing PRLR Antibody Mat3 is Suitable for the Treatment of Benign Breast Disease.

(55) An activating PRLR mutation or local or systemic hyperprolactinemia can provoke benign breast disease. Therefore, a hyperprolactinemic mouse model with enhanced proliferation in the mammary gland (hallmark of the most severe forms of benign breast disease) was employed. 12 weeks old female Balb/c mice received a pituitary isograft under the kidney capsule or remained unoperated. Pituitary isografted mice remained untreated or were treated subcutaneously once weekly with neutralizing PRLR antibody Mat3 in the IgG2 format (=IgG2 Mat3) or unspecific control antibody in IgG2 format on day 15, 22, 29, and 36 after pituitary transplantation. Antibody doses were 30 mg/kg. Experimental group size was 8 animals.

(56) The experiment comprised the following experimental groups: 1. Animals without pituitary transplantation, i.e. normoprolactinemic mice 2. Animals with pituitary transplantation, i.e. hyperprolactinemic mice 3. Animals with pituitary transplantation, treated with unspecific control antibody once weekly at a dose of 30 mg/kg 4. Animals with pituitary transplantation, treated with the neutralizing prolactin receptor antibody Mat3 once weekly at a dose of 30 mg/kg

(57) On day 38 after pituitary transplantation mice were sacrificed. Two hours before death, animals received an intraperitoneal injection of BrdU to monitor epithelial cell proliferation. The left inguinal mammary gland was fixed in Carnoy's solution and mammary gland whole mounts were prepared and stained with Carmine alaune. Sidebranching was evaluated in the mammary gland whole mounts. Results are depicted in FIG. 3A. Antibody Mat3 inhibited sidebranching in the mammary gland, unspecific control antibody was without effect (FIG. 3A).

(58) Afterwards the mammary gland whole mounts were embedded in paraffin and BrdU immunostainings were performed as described previously (Endocrinology 149 (8): 3952-3959; 2009). Epithelial cell proliferation was analysed in 4 histological mammary gland slices per animal.

(59) The lateral one third of the right inguinal mammary gland without the lymph node was frozen in liquid nitrogen and processed for RNA preparation. After reverse transcription, real-time Taqman PCR analysis was performed using the ABI Prism 7700 Sequence Detector System according to the manufacturer's instructions (PE Applied Biosystems). Expression of the prolactin target gene elf5 was assessed and normalized to cytokeratin18 expression. Relative mRNA levels were calculated by the comparative ΔCT-method. Pituitary isografting, i.e. hyperprolactinemia enhanced expression of the prolactin target gene elf5 (FIG. 3B). Specific antibody Mat3, but not the unspecific control antibody inhibited elf5 gene expression indicating successful blockade of the prolactin receptor (FIG. 3B).

(60) Antibody Mat3 is therefore suitable to treat benign breast disease.

Example 7

(61) Isolation of Target-Specific Antibodies from Human Antibody Phage Display Libraries

(62) To isolate a panel of antibodies able to neutralize the activity of human PRLR, three human antibody phage display libraries, expressing Fab and scFv fragments, were investigated in parallel. The target used for the library panning was the soluble extracellular domain (ECD) of the human and mouse prolactin receptor, respectively, represented by the amino acids 1-210, of SEQ ID NOs. 12 and 13. Alternative targets were the ECD of PRLR C-terminally linked to six histidines or to a human IgG1-Fc domain via the linker with the amino acid sequence “isoleucine-glutamate-glycine-arginine-methionine-aspartate”.

(63) Selection of target-specific antibodies from phage display was carried out according to methods described by Marks et al. (Methods Mol Biol. 248:161-76, 2004). The phage display library was incubated with 50 pmols of the biotinylated ECD at room temperature for 1 hr and the complex formed was then captured using 100 μl of Streptavidin beads suspension (Dynabeads® M-280 Streptavidin, Invitrogen). Non specific phages were removed by washing the beads with wash buffer (PBS+5% Milk). Bound phages were eluted with 0.5 ml of 100 nM Triethylamine (TEA) and immediately neutralized by addition of an equal volume of IM TRIS-Cl pH 7.4. Eluted phage pool was used to infect TG1 E coli cells growing in logarithmic phase, and phagemid was rescued as described (Methods Mol Biol. 248:161-76, 2004). Selection was repeated for a total of three rounds. Single colonies obtained from TG1 cells infected with eluted phage from the third round of panning were screened for binding activity in an ELISA assay. Briefly, single colonies obtained from the TG1 cell infected with eluted phage were used to inoculate media in 96-well plates.

(64) Microcultures were grown to an OD.sub.60O=0.6 at which point expression of soluble antibody fragment was induced by addition of 1 mM IPTG following overnight culture in a shaker incubator at 30° C. Bacteria were spun down and periplasmic extract was prepared and used to detect antibody binding activity to ECD immobilized on 96-well microplates (96-well flat bottom Immunosorb plates, Nunc) following standard ELISA protocol provided by the microplate manufacturer.

(65) The affinities of the anti-Prolactin Receptor (PRLR) antibodies for binding to the recombinant extracellular domain (ECD) were estimated using the Biacore® 2000 and used for affinity ranking of antibodies.

Example 8

(66) Maturation of Antibody Variants:

(67) Antibody affinity maturation is a two step process where saturation mutagenesis and well-based high throughput screening are combined to identify a small number of mutations resulting in affinity increases. In the first round of affinity maturation positional diversification of wild-type antibody is introduced by site-directed mutagenesis using NNK-trinucleotide cassettes (whereby N represents a 25% mix each of adenine, thymine, guanine, and cytosine nucleotides and K represents a 50% mix each of thymine and guanine nucleotides) according to BMC Biotechnology 7: 65, 2007. This way, all 20 amino acids are introduced at an individual amino acid position. This positional randomization is restricted to the six complementarity determining regions (CDRs). In the second round of affinity maturation beneficial substitutions were recombined and screened for further improvements.

(68) Screening of Maturated “005-C04” Fab Variants by ELISA Tests:

(69) 96 well microtiter plates were coated with 1 μg per milliliter of human PRLR. Plates were incubated over night at 4° C. After blocking with PBS buffer containing 3% bovine serum albumin, normalized E. coli-derived supernatants containing the Fab variants were added. Detection of formed complexes occurred via the addition of an anti-flag antibody (Sigma, A8592) labeled with horseradish peroxidase.

(70) Amplex Red as fluorogenic substrate for horseradish peroxidase was added and incubated for 30 minutes at room temperature. Absorption at 570 nm and extinction at 585 nm was measured using the Tecan Infinite F500 reader. The obtained results the screening hit “005-C04-20-2” are shown in FIG. 7. Individual substitutions in the screening hit “005-C04-20-2” (FIG. 7) beneficial for affinity improvement were evaluated regarding their influence on thermostability of the antibody, in order to ensure cooperative unfolding of the Fab domain during denaturation by temperature elevation. The antibody Mat3 was a derivative of “005-C04-20-2”, in which the thermo-destabilizing substitutions have been reversed to the parental antibody “005-C04”.

Example 9

(71) Cross-Reactivity of Antibodies on Mouse and Human PRLR Expressed on Cell Surfaces

(72) a) In order to understand the missing antiproliferative activity of antibody HE06642 on cells carrying the murine PRLR, the binding characteristics of HE06642 on mouse and human PRLR expressed on cells was determined by flow cytometry on HEK293 cells stably expressing the human and murine PRLR, respectively. The cells as well as the parental HEK293 cell line without PRLR were harvested, centrifuged and resuspended at approximately 5×10.sup.6 cells/ml in 1×PBS containing 2% FBS and 0.1% sodium azide (FACS buffer). The antibody HE06642 was diluted to 2-fold final concentration in FACS buffer and added to appropriate sample wells (50 μl/well). For secondary antibody and autofluorescence controls, 50 μl FACS buffer was added to appropriate wells. 50 μl of cell suspension was added to each sample well. Samples were incubated at 4° C. for one hour, washed twice with cold FACS buffer and resuspended in FACS buffer containing PE-conjugated goat anti-human IgG at a 1:100 dilution. Following a 30 min incubation at 4° C., cells were washed twice with cold FACS buffer, resuspended in FACS buffer containing 1 mg/ml propidium iodide (Invitrogen, San Diego, Calif.) and analyzed by flow cytometry. As shown in FIG. 5A, the antibody HE06.642 only binds to the human PRLR and not to the murine PRLR. This observation is consistent with the finding reported in Example 2 and 12 about the missing activity of HE06.642 in the murine PRLR-dependent proliferation and luciferase reporter gene assays.

(73) b) In order to demonstrate cell binding of antibody Mat3 on cells carrying the human and the rhesus monkey PRLR, of which the amino acid sequence of the extracellular domain is identical to the one of cynomolgus monkey, the binding characteristics of Mat3 on human and monkey PRLR expressed on cells was determined by flow cytometry on HEK293 cells stably expressing the human and monkey PRLR, respectively. The cells were harvested, centrifuged and resuspended at approximately 2×10.sup.6 cells/ml in 1×PBS containing 3% FBS and 0.05% sodium azide (FACS buffer). The antibody Mat3 was diluted to 2-fold final concentration in FACS buffer and added to appropriate sample wells (50 μl/well). For secondary antibody and autofluorescence controls, 50 μl FACS buffer was added to appropriate wells. 50 μl of cell suspension was added to each sample well. Samples were incubated at 4° C. for one hour, washed twice with cold FACS buffer and resuspended in FACS buffer containing PE-conjugated goat anti-human IgG at a 1:100 dilution. Following a 30 min incubation at 4° C., cells were washed twice with cold FACS buffer, resuspended in FACS buffer containing 1 mg/ml propidium iodide (Invitrogen, San Diego, Calif.) and analyzed by flow cytometry. As shown in FIG. 5B, the antibody Mat3 binds to the human PRLR as well as to the monkey PRLR. Maximal signal intensities at highest antibody concentrations depend on the number of PRLR expressed on the cell surfaces, i.e. HEK293 and Ba/F cells do not carry the same number of PRLRs on their surface. The EC50 values were calculated based on the dose response curves illustrated in FIG. 5B, in order to derive a measurement value for binding strength of Mat3 on cells (Table 2). The cell-based binding potency of Mat3 was 0.53 nM on HEK293 cells with human PRLR and 2.94 nM on Ba/F cells with monkey PRLR. These data support the finding that Mat3 is not only a very potent human-specific agent, but also at reasonable doses active on monkey PRLR in the low nanomolar range.

Example 10

(74) Binding Studies with Purified Extracellular PRLR Domains Using Surface Plasmon Resonance Analysis

(75) Binding affinities of antibody Mat3 were determined by surface plasmon resonance analysis on a Biacore T100 instrument (GE Healthcare Biacore, Inc.). Antibodies were immobilized onto a CM5 sensor chip through an indirect capturing reagent, anti-human IgG Fc. Reagents from the “Human Antibody Capture Kit” (BR-1008-39, GE Healthcare Biacore, Inc.) were used as described by the manufacturer. Approximately 5000 RU monoclonal mouse anti-human IgG (Fc) antibody were immobilized per cell. Antibody Mat3 was injected at a concentration of 5 μg/ml at 10 μl/min for 10 sec to reach a capturing level of approximately 200 to 600 RU. Various concentrations (400 nM, 200 nM, 100 nM, 50 nM, 25 nM, 12.5 nM, 6.25 nM, and 3.12 nM) in HEPES-EP buffer (GE Healthcare Biacore, Inc.) of the ECD of human, monkey or murine PRLR were injected over immobilized Mat3 antibody at a flow rate of 60 μl/min for 3 minutes and the dissociation was allowed for 10 minutes. The ECDs of the human PRLR (SEQ ID NO: 12), of the monkey PRLR (SEQ ID NO: 11, after proteolytic removal of the Fc-His-tag via Factor Xa digestion) and of the murine PRLR (SEQ ID NO: 13) represented monovalent analytes. Sensograms were generated after in-line reference cell correction followed by buffer sample subtraction. The dissociation equilibrium constant (K.sub.D) was calculated based on the ratio of association (kon) and dissociation rated (koff=kd) constants, obtained by fitting sensograms with a first order 1:1 binding model using BiaEvaluation Software. The monovalent dissociation constants (K.sub.D, affinity) and dissociation rates (kd=koff) values are shown in Table 1.

Example 11

(76) Peptide Scan

(77) a) Peptide Synthesis:

(78) To reconstruct discontinuous epitopes of the target molecule SEQ ID NO. 12 from amino acid position 1 to 210, the ECD of the human PRLR, a library of structured peptides was synthesized. This was done using Pepscan's proprietary Chemically Linked Peptides on Scaffolds (CLIPS) technology (Timmerman et al., 2007, J. Mol. Recognit. 20:283-99; Pepscan Therapeutics, Lelystad, Netherlands). CLIPS technology allows to structure peptides into single loops, double-loops, triple loops, sheet-like folds, helix-like folds and combinations thereof. CLIPS templates were coupled to cysteine residues. The side-chains of multiple cysteines in the peptides were coupled to one or two CLIPS templates. For example, a 0.5 mM solution of the T2 CLIPS template 1,3-bis (bromomethyl) benzene was dissolved in ammonium bicarbonate (20 mM, pH 7.9)/acetonitrile (1:1(v/v). This solution was added onto the peptide arrays. The CLIPS template bound to side-chains of two cysteines as present in the solid-phase bound peptides of the peptide-arrays (455 wells plate with 3 ul wells). The peptide arrays were gently shaken in the solution for 30 to 60 minutes while completely covered in solution. Finally, the peptide arrays were washed extensively with excess of H.sub.2O and sonicated in disrupt-buffer containing 1 percent SDS/0.1 percent beta-mercaptoethanol in PBS (pH 7.2) at 70° C. for 30 minutes, followed by sonication in H2O for another 45 minutes. The T3 CLIPS carrying peptides were made in a similar way but now with three cysteines.

(79) b) Pepscan ELISA:

(80) The binding of antibody Mat3 and of antibody HE06642 to each peptide was tested in a PEPSCAN-based ELISA (Slootstra et al., 1996, Molecular Diversity 1: 87-96). The peptide arrays were pre-incubated with 5% to 100%-binding buffer (1 hr, 20° C.). The binding buffer was composed of 1% Tween-80, 4% horse-serum, 5% Ovalbumin (w/v) and was diluted with PBS. After washing the peptide arrays were incubated with primary antibody solution (1 to 5 μg/ml) in PBS containing 1% Tween-80 (overnight at 4° C.). After washing, the peptide arrays were incubated with a 1/1000 dilution in 100% binding buffer of an antibody peroxidase conjugate for one hour at 25° C. (anti-human, humpo). After washing, the peroxidase substrate 2,2′-azino-di-3-ethylbenzthiazoline sulfonate (ABTS) and 2 microliters/milliliter of 3 percent H.sub.2O.sub.2 were added. After one hour, the color development was measured. The color development was quantified with a charge coupled device (CCD)-camera and an image processing system.

(81) c) Data Processing:

(82) The raw data were optical values obtained by a CCD-camera. The values ranged from 0 to 3000 mAU, similar to a standard 96-well plate ELISA-reader. The results were quantified and stored into the Peplab database. The binding values were extracted for analysis. Occasionally a well contained an air-bubble resulting in a false-positive value, the cards were manually inspected and any values caused by an air-bubble were scored as 0.

(83) d) Data Analysis and Representation:

(84) A heat map is a graphical representation of data where the empirical values from experimental data are organized in a two-dimensional map, and are then represented as colors (Brinton, 1914, Graphic Methods for Presenting Facts, New York: The Engineering Magazine Company; Gower and Digby, 1981), “Expressing complex relationships in two dimensions” in Interpreting Multivariate Data, ed. Barnett, V., Chichester, UK: John Wiley & Sons, pp. 83-118.). For double-looped CLIPS peptides, such a two-dimensional map could be derived from the independent sequences of the first loop and the second loop.

(85) For the target protein (ECD of human PRLR), the 2916 CLIPS peptides had sequences that were effectively permutations of 54 unique sub-sequences, combined in two sequential CLIPS loops. Thus, the observed Pepscan ELISA data could be plotted in a 54×54 matrix, where each X coordinate was the amino acid sequence of the first loop, and each Y coordinate was the amino acid sequence of the second loop. For each XY coordinate in the matrix, the Pepscan ELISA value was placed that is derived from the peptide with sequence X+Y. To further facilitate the visualization, Pepscan ELISA values were replaced with colors from a continuous gradient. In this case, extremely low values were colored green, extremely high values were colored red, and average values were colored black. When this color map was applied to the described data matrix, a color heat map resulted. The peptides revealing the most striking differences in the ELISA results between the antibodies Mat3 and HE06642 were selected based on these heat map representation. For these peptides, each ELISA signal value was processed for representation in a bar plot (FIGS. 8A and 8B). Each plotted value represented the average value obtained for 54 peptides with S.E.M (standard error of mean). The data was normalized to average over the entire 2916 peptide dataset and corrected for background signal.

(86) FIG. 8A shows the ELISA results for a subset of peptides ranging from amino acids 103-PDPPLELAVEVKQPE-117 indicated as ‘117’ on the X-axis to 127-WSPPTLIDLKTGWFT-141 (indicated as ‘141’). I.e. these peptides, which were shifted by three amino acids along the ECD amino acid sequence, cover the region from amino acid position 103 to 141 of the ECD of human PRLR. The strongest differences observed within this dataset are for peptide 109-LAVEVKQPEDRKPYL-123 (indicated as ‘123’) with a significance p-value of 4×10.sup.−12, and for peptide 121-PYLWIKWSPPTLIDL-135 (indicated as ‘135’) with a significance p-value of 7×10.sup.−40. FIG. 8B shows the ELISA results for a subset of peptides ranging from 139-WFTLLYEIRLKPEKA-153 (indicated as ‘153’ on the X-axis) to 163-QQTEFKILSLHPGQK-177 (indicated as ‘177’). I.e. these peptides, which were shifted by three amino acids along the ECD amino acid sequence, cover the region from amino acid position 139 to 177 of the ECD of human PRLR. The strongest differences observed within this dataset are for peptide 148-LKPEKAAEWEIHFAG-162 (indicated as ‘162’) with a significance p-value of 6×10.sup.−26, and for peptide 160-FAGQQTEFKILSLHP-174 (indicated as ‘174’) with a significance p-value of 8×10.sup.−8. These data demonstrate that both antibodies bind to the S2 subdomain of the ECD of human PRLR (amino acid 101 to 210) and therefore are non-competitive to the natural ligand PRL which mainly binds to the S1 domain. However, this peptide scan showed that there are differences in binding to the S2 domains between Mat3 and HE06642. This finding indicates why the antibody Mat3 shows a different species-specificity and potency compared to HE06642.

Example 12

(87) Inhibition of Luciferase Reporter Gene Activity in HEK293 Cells Stably Transfected with the Human and Murine PRLR

(88) To further analyze the in vitro activity of the neutralizing PRLR antibody Mat3 on the human and the murine PRLR, a reporter gene assay was used. HEK293 cells stably transfected with the murine PRLR were transiently transfected with a luciferase reporter gene under the control of LHREs (lactogenic hormone response elements). Afterwards, cells were seeded at a density of 20000 cells per well (80 μl) on a 96-well plate in DMEM High Glucose medium with 4.5 g/L glucose, 2 mM Glutamax, 0.5% FCS, and 1% Penicillin/Streptomycin. The next day 200 ng/ml human prolactin (10 μl) with and without increasing doses of test antibodies (10 μl) were added. Twenty-four hours later, luciferase activity was determined. For comparison, the antibody HE06642 and a nonspecific antibody targeting choleratoxin called CTX were also tested. All antibodies were tested as IgG1 molecules (FIGS. 13A and 13B).

Example 13

(89) Binding Studies with Purified Extracellular PRLR Domains Using Surface Plasmon Resonance Analysis

(90) Binding affinities of antibodies Mat3 and HE06642 were determined by surface plasmon resonance analysis on a Biacore T100 instrument (GE Healthcare Biacore, Inc.) in parallel. Antibodies were immobilized onto a CM5 sensor chip through an indirect capturing reagent, anti-human IgG Fc. Reagents from the “Human Antibody Capture Kit” (BR-1008-39, GE Healthcare Biacore, Inc.) were used as described by the manufacturer. Approximately 5000 RU monoclonal mouse anti-human IgG (Fc) antibody were immobilized per cell. Each test antibody was injected at a concentration of 5 μg/ml at 10 μl/min for 10 sec to reach a capturing level of approximately 200 to 600 RU. Various concentrations (400 nM, 200 nM, 100 nM, 50 nM, 25 nM, 12.5 nM, 6.25 nM, and 3.12 nM) in HEPES-EP buffer (GE Healthcare Biacore, Inc.) of the ECD of human, monkey or murine PRLR were injected over the immobilized test antibody at a flow rate of 60 μl/min for 3 minutes and the dissociation was allowed for 10 minutes. The ECDs of the human PRLR (SEQ ID NO: 12), of the monkey PRLR (SEQ ID NO: 11, after proteolytic removal of the Fc-His-tag via Factor Xa digestion) and of the murine PRLR (SEQ ID NO: 13) represented monovalent analytes. Sensograms were generated after in-line reference cell correction followed by buffer sample subtraction. The dissociation equilibrium constant (KD) was calculated based on the ratio of association (kon) and dissociation rated (koff=kd) constants, obtained by fitting sensograms with a first order 1:1 binding model using BiaEvaluation Software (see Table 7).

(91) TABLE-US-00007 TABLE 7 Monovalent dissociation constants and dissociation rates of purified extracellular domains of human, monkey and murine PRLR (expressed in HEK293 cells) determined for Mat3 and HE06.642 by surface plasmon resonance Human PRLR Monkey PRLR Murine PRLR K.sub.D [M] Kd [1/s] K.sub.D [M] Kd [1/s] K.sub.D [M] Kd [1/s] Mat3 1.3 × 10.sup.−9 4.14 × 10.sup.−4 9.6 × 10.sup.−9 1.02 × 10.sup.−3  0.4 × 10.sup.−9 1.64 × 10.sup.−4 HE06.642# 0.8 × 10.sup.−9 5.33 × 10.sup.−4  17 × 10.sup.−9 5.79 × 10.sup.−3 15.0 × 10.sup.−9 9.10 × 10.sup.−3 #The affinities disclosed for HE06642 in WO2008/022295 are K.sub.D (human PRLR) = 2.6 × 10.sup.−9 M, K.sub.D (monkey PRLR) = 38.9 × 10.sup.−9 M, and K.sub.D (murine PRLR) = 2.7 × 10.sup.−9 M.

(92) As shown in Table 7, Mat3 exhibits improved affinities to monkey and murine PRLR compared to HE06.642. The improved cell binding and antiproliferative activity of Mat3 on human PRLR is not reflected by improved affinity to hPRLR compared to HE06.642. Despite binding to the purified ECD of murine PRLR, HE06.642 does not bind cells or inhibits proliferation of cells expressing the murine PRLR (FIGS. 10 and 14, Table 8). In contrast, Mat3 blocks cell proliferation in nanomolar to subnanomolar scale in all three species.

(93) Additionally, the soluble ECD-PRLR (SEQ ID NO: 12) was captured on the surface via the immobilized test antibodies Mat3 and HE06642, therefore, the epitope of the capture antibody was blocked for all bound ECD-PRLR molecules. Human PRL was immediately passed over the surface to bind to the immobilized ECD-PRLR. This way, it could be measured whether PRL bound the ECD-PRLR, although the ECD is captured by the test antibody.

(94) It could be observed that PRL binds the ECD-PRLR independently from binding to Mat3 and HE06.642. Thus, both antibodies do not compete with the natural ligand PRL on ECD-PRLR.

Example 14

(95) Cell-Binding Studies of Antibodies on Various Cell Lines Expressing Human, Mouse and Monkey PRLR

(96) A cell binding study was performed with antibody Mat3 and HE06642 as well as a choleratoxin-specific isotype control, all being in IgG1 format. The tested cell lines were HEK293 cells stably expressing human and mouse PRLR, respectively, the human breast cancer cell line T47D as well as the Ba/F cell lines expressing human, mouse and rhesus monkey PRLR. The cell binding was determined by flow cytometry on the above mentioned cells. The cells were harvested, centrifuged and resuspended at approximately 2×10.sup.6 cells/ml in 1×PBS containing 3% FBS and 0.05% sodium azide (FACS buffer). Each test antibody was diluted to 2-fold final concentration in FACS buffer and added to appropriate sample wells (50 μl/well). For secondary antibody and autofluorescence controls, 50 μl FACS buffer was added to appropriate wells. 50 μl of cell suspension was added to each sample well. Samples were incubated at 4° C. for one hour, washed twice with cold FACS buffer and resuspended in FACS buffer containing PE-conjugated goat anti-human IgG at a 1:100 dilution. Following a 30 min incubation at 4° C., cells were washed twice with cold FACS buffer, resuspended in FACS buffer containing 1 mg/ml propidium iodide (Invitrogen, San Diego, Calif.) and analyzed by flow cytometry.

(97) The obtained data are shown as dose-response curves in FIG. 14. From these curves EC50 values were deduced indicating the cell binding potencies (Table 8). In conclusion, the data indicate the superior cell binding properties of Mat3 compared to HE06.642 across the different PRLR-expressing cell lines.

(98) TABLE-US-00008 TABLE 8 Cell binding potency of neutralizing PRLR antibodies Mat3 and He06.642 in IgG1 format on cells expressing PRLR from human, mouse and monkey deduced from flow cytometry Cell binding (EC.sub.50 [M]) # Cell line Mat3 HE06.642 A HEK293-human PRLR 0.8 × 10.sup.−9 1.6 × 10.sup.−9 B HEK293-murine PRLR 0.1 × 10.sup.−9 —* C HEK293 w/o PRLR —* —* D Human T47D 0.4 × 10.sup.−9 0.8 × 10.sup.−9 E BaF3-rhesus PRLR 1.2 × 10.sup.−9 3.2 × 10.sup.−9 F BaF3-human PRLR 0.1 × 10.sup.−9 0.2 × 10.sup.−9 G BaF3-mouse PRLR 0.2 × 10.sup.−9 —* *no significant binding; # dose-response diagrams in FIG. 14