FIELD OF THE INVENTION

Generally, the invention relates to the field of biological pharmaceuticals as well as their use in in oncology, e.g. prostate cancer, lung cancer, etc. More specifically, the invention relates to a heterodimeric TGFBR2/TGFBR3 -derived composition that is capable of inhibiting TGF-P cytokine.

BACKGROUND

Transforming growth factor beta (TGF-P) isoforms, TGF-pi, P2, and P3, are small (25 kDa) homodimeric signaling proteins. They are secreted in a latent form and are activated by multiple mechanisms, including integrin binding and proteolysis. They form a complex with the TGF-P type I and type II receptors (TpRI and TpRII) for signal transduction, in which TpRII phosphorylates and activates TpRI. The phospho-TpRI then phosphorylates intracellular Smad2 and Smad3, which form a complex with the common-mediator Smad, Smad4, to regulate gene expression.

TGF-Ps are potent growth inhibitors in normal epithelial cells, including normal prostate epithelial cells, by stimulating apoptosis and inhibiting G1 to S cell cycle progression. The deletion of Smad4 has been furthermore shown to drive the invasion and metastasis of indolent prostate tumors with Pten deletion in a mouse model, demonstrating the tumor suppressive activity of the TGF-p/Smad pathway in the prostate gland. Thus, it is not surprising that carcinoma cells in general and prostate cancer (PCa) cells in particular are resistant to TGF-P-induced growth inhibition and that loss of or reduced expression of the signaling receptors, TpRI, TpRII, or the non-signaling TGF-P type III receptor, also known as betaglycan, is often observed during the progression of human PCa. (see Qin T, et al., Oncotarget, 2016, Vol. 7, (No. 52), pp: 86087- 86102, the entire teachings of which are incorporated by reference herein).

Prostate carcinoma cells, while responding poorly to TGF-P-mediated growth inhibition, often produce much higher levels of TGF-P isoforms than their normal counterparts. Furthermore, latent TGF-P is activated by the protease prostate specific antigen (PSA), which is an androgen receptor (AR) target gene abundantly secreted by advanced androgen-independent PCa cells. Serum TGF- pi levels have been shown to correlate with tumor burden, metastasis, and serum PSA in PCa patients and an increased level of TGF-P 1 is strongly associated with PCa progression and poor clinical outcome. Excessive levels of TGF-P may act on tumor stromal cells in a paracrine fashion to promote disease progression. TGF-P’ s tumor promoting activity may be related to its ability to generate and maintain cancer stem cells, including PCa stem cells, which are AR negative and presumably sensitive to TGF-p. TGF-Ps are also known to stimulate the conversion of CD4+CD25- T cells to CD4+CD25+Foxp3+ regulatory T-cells, which inhibit anti-tumor immunity. Treatments with TGF-P inhibitors, such as soluble betaglycan or a pan-isoform neutralizing antibody, have been shown to have beneficial effects in animal models of PCa, including inhibition of the growth and angiogenesis of tumors formed by AR negative human PCa cells or inhibition of regulatory T-cell production and tumor progression. Thus, there are multiple mechanisms by which TGF-Ps promote the progression of advanced disease and treatment with TGF-P inhibitors appears to be a viable strategy for attenuating disease progression. The TGF-P pathway is known however to be tumor suppressive in normal and some experimental models of early stage adenocarcinomas as mentioned above, and even advanced tumors may contain early and late stages of lesions due to tumor heterogeneity. Thus, the greatest perceived risk of TGF-P antagonists in treating late stage PCa is the potential progression of early-stage tumor cells in which TGF-P pathway is still tumor suppressive.

TGFBR1, TGFBR2 and TGFBR3 proteins are single spanning membrane receptors. They consist of signal peptide, extracellular domain, transmembrane and cytoplasmic domains. TGFBR1 exists in two isoforms, Isoform-1 and Isoform-3, where Isoform-1 is commonly accepted as the canonical form and is four amino acid residues shorter that the non-canonical Isoform 3. Extracellular portion of TGFBR1 consists of 125 and 129 amino acid residues for Isoform-1 and Isoform-2, respectively, while extracellular domain of TGFBR2 contains 144 amino acids. Extracellular domain of TGBFR3 consists of 767 amino acid residues and includes N-terminal BGo and C-terminal BGzp domains (Fig. 1). BGo domain binds TGF- P whereas BGzp binds both TGF- P and activins.

SUMMARY OF THE INVENTION

This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.

In certain aspects, the present invention provides for a heterodimeric recombinant protein composition capable of binding human TGF-p. The recombinant protein composition included a first polypeptide which includes amino acid sequence of SEQ ID NO. 1 or of SEQ ID NO. 2; and a second polypeptide which includes amino acid sequence of SEQ ID NO. 3, or of SEQ ID NO. 4, or of SEQ ID NO. 5.

In certain aspects, the present invention provides for a therapeutic composition which contains a recombinant protein composition capable of binding human TGF-p. The recombinant protein composition included a first polypeptide which includes amino acid sequence of SEQ ID NO. 1 or of SEQ ID NO. 2, a second polypeptide which includes amino acid sequence of SEQ ID NO. 3, or of SEQ ID NO. 4, or of SEQ ID NO. 5, and a pharmaceutically acceptable carrier or excipient. The therapeutic composition may further comprising about 0.02% PS-20, about 3% D- mannitol, about 20 mM NaCl, about 70 mM L-proline, and about 25 mM histidine at pH 6.0 .

The first polypeptide of the recombinant heterodimeric protein may contain amino acid sequence of SEQ ID NO. 2, while the second polypeptide includes amino acid sequence of SEQ ID NO. 5.

In certain aspects, the present invention provides for an isolated nucleic acid encoding a polypeptide containing amino acid sequence of: SEQ ID NO. 1, or of SEQ ID NO. 2, or of SEQ ID NO. 3, or of SEQ ID NO. 4, or of SEQ ID NO. 5. The nucleic acid may have its codon usage optimized for high expression of the polypeptide in a mammalian cell.

The isolated nucleic acid may contain a DNA sequence of SEQ ID NO. 6, or of SEQ ID NO. 7, or of SEQ ID NO. 8, or of SEQ ID NO. 9.

In certain aspects, the present invention provides for an isolated nucleic acid containing a DNA sequence of SEQ ID NO. 10. The nucleic acid may be further incorporated into a mammalian cell. The mammalian may be a CHO-K1 cell.

In certain aspects, the present invention provides for a composition comprising a first polypeptide comprising amino acid sequence of SEQ ID NO. 2, and a second polypeptide comprising amino acid sequence of SEQ ID NO. 5, for use in the treatment of idiopathic pulmonary fibrosis, systemic scleroderma, pulmonary arterial hypertension, or a fibrotic disease of heart, liver or kidney. In certain aspects, the present invention provides for a method of treating or preventing idiopathic pulmonary fibrosis, systemic scleroderma, pulmonary arterial hypertension, or a fibrotic disease of heart, liver or kidney. The method includes administering to a patient in need for treating or preventing idiopathic pulmonary fibrosis, systemic scleroderma, pulmonary arterial hypertension, or a fibrotic disease of heart, liver or kidney, a therapeutically effective amount of a pharmaceutical composition comprising a first polypeptide comprising amino acid sequence of SEQ ID NO. 2, and a second polypeptide comprising amino acid sequence of SEQ ID NO. 5.

BRIEF DESCRIPTION OF THE DRAWINGS

The following drawings and descriptions are provided to aid in the understanding of the invention:

Fig- 1 is a schematic diagram of TGFBR3 (betaglycan) domain structure, with the N-terminal orphan domain (BGO) and the N- and C-terminal zona pellucida domains, BGZP-N and BGZP-C, respectively, glycosaminoglycan chains attached to two residues in the ZP-N subdomain are shown schematically as beads on a string, disulfide bonds are represented by S-S, while free cysteines are represented by -SH;

Fig. 2 shown SDS-PAGE analysis of TGFBR2/TGFBR3/Fc heterodimer next to molecular marker, 10 pg per lane was analyzed under reducing and non-reducing conditions to identify potential homo- and hetero-dimers, as a loading control, a mouse IgG2a (m-IgG2a) and a human IgG4 (h-IgG4) were also loaded, markers, kDa are indicated on the right;

Fig- 3 shows a titration curve for TGF-β , MDA-MD-231 cells in serum-free medium were treated with various concentrations of TGF-β and 24 hours post-treatment cell supernatant was quantified for IL-11, EC50 value was calculated from variable slope four parameter algorithm;

Fig. 4 shows inhibition of TGF-β -induced IL-11 secretion by TGFBR2/TGFBR1/Fc chimeras, MDA-MB-231 cells were treated with constant concentration of TGF-β , 2 ng/ml, in the presence of varying concentration of TGFBR2/TGFBR1-Isoform-1/Fc and TGFBR2/TGFBR1- Isoform-3/Fc, IC50 values were calculated from variable slope four parameter algorithm;

Fig. 5 shows Inhibition of TGF-β -induced IL-11 secretion by TGFBR2/TGFBR3/Fc chimera. MDA-MB-231 cells were treated with constant concentration of TGF-β , 2 ng/ml, in the presence of varying concentration of TGFBR2/TGFBR3/Fc, the experiment was repeated twice, left and right panes, and the corresponding IC50 values were calculated from variable slope four parameter algorithm;

Fig. 6 shows inhibition of TGF-β -induced IL-11 secretion by TGFBR2/TGFBR3/Fc chimera, MDA-MB-231 cells were treated with two concentrations of TGF-β , 1 ng/ml and 2 ng/ml, in the presence of varying concentration of TGFBR2/TGFBR3/Fc, data for two TGF-β concentrations are shown on left and right panes, respectively and the corresponding IC50 values were calculated from variable slope four parameter algorithm;

Fig. 7 shows schematic diagram of TGFBR2/TGFBR3/Fc capturing approach, anti-human IgG were chemically crosslinked to the chip surface, TGFBR2/TGFBR3/Fc was captured by the immobilized antibodies and various concentrations of TGF-β were applied to the chip, at regeneration step, TGFBR2/TGFBR3/Fc along with TGF-β bound to it was removed from the capturing antibodies using 3 M magnesium chloride;

Fig. 8 shows sensograms of binding of TGF-β to TGFBR2/TGFBR3/Fc captured by immobilized anti-human antibodies in experiment 1, the curves were analyzed using 1 : 1 binding model;

Fig. 9 shows sensograms of binding of TGF-β to TGFBR2/TGFBR3/Fc captured by immobilized anti-human antibodies in experiment 2, the curves were analyzed using 1 : 1 binding model;

Fig. 10 shows sensograms of binding of TGF-β to TGFBR2/TGFBR3/Fc captured by immobilized anti-human antibodies in experiment 3, the curves were analyzed using 1 : 1 binding model;

Fig. 11 shows the plasmid map of TGFBR2-Fcl fusion protein expression construct for transient transfection in mammalian cells;

Fig. 12 shows the plasmid map of TGFBR1-Fc2, Isoform 1, fusion protein expression construct for transient transfection in mammalian cells;

Fig. 13 shows the plasmid map of TGFBR1-Fc2, Isoform 3, fusion protein expression construct for transient transfection in mammalian cells; and

Fig. 14 shows the plasmid map of TGFBR3-Fc2 fusion protein expression construct for transient transfection in mammalian cells;

Fig. 15 shows the plasmid map of pAG12B-TGFBR2-Fcl for facilitating plasmid generation for expression of TGFBR2/TGFBR3/Fc in mammalian cells; Fig. 16 shows the plasmid map of pAG12B-TGFBR3-Fc2 for facilitating plasmid generation for expression of TGFBR2/TGFBR3/Fc in mammalian cells;

Fig. 17 shows the plasmid map of pAG12B- TGFBR2/TGFBR3/Fc for co-expressing both pAG12B-TGFBR2-Fcl and TGFBR2/TGFBR3/Fc in mammalian cells, each mediated by its own promoter regulatory region and BGH poly A region;

Fig. 18 shows Western blot analysis of selected 12 clones for the first screen, 7.5ul of cell medium from each clone was mixed with 7.5ul of reduced (A. left panel) of non-reduced (B. right panel), analyzed on a 4-12% SDS gel and detected using the HRP-conjugated anti-human IgG, double bands (corresponding to TGFBR2-Fcl and TGFBR3-Fc2) were detected under the reduced conditions and a single band (~125kDa) was detected under non-reduced conditions;

Fig. 19 shows 2nd screen of single cell clones. Clone #31 and clone #36 from the 1st screen were diluted to average 0.5-1 cell/well for 4x 96-well plates, lOul of cell medium from 96 single cell clones was analyzed by dot blot;

Fig. 20 shows validation of 12 single cell clones, 7.5ul of cell medium from each of 12 selected single cell clones was analyzed by Western blot using anti-human IgG-HRP;

Fig. 21 shows documentation of monoclonality, imaging of colony formation from single cell of clone #A6 and #G9. During the 2 ^nd screen, clones A6 and G9 with obvious single healthy cell were traced everyday by micro-photographing to monitor the growth of two clones (A for clone A6 and B for clone G9);

Fig. 22 shows stability test results: five clones (A6, D10, DI 1, F7 and G9) that were selected from the second screen of single cell clones were used for the stability test for a total ten passages, each clone was seeded in a 96-well plate in triplets with the same amount of cells and passaged twice a week (3-4 days), conditioned media were collected from each passage and analyzed by ELISA assay and detected by using HRP-conjugated anti-human IgG;

Fig. 23 shows SDS-PAGE analysis of purified from conditioned medium TGFBR2/TGFBR3/Fc heterodimeric protein: the conditioned media of clones A6, D10 and G9 were collected at day 10 of post-passaging, the heterodimetic protein was purified by passing the conditioned medium through a Protein A sepharose column and analyzed by SDS-PAGE under non-reducing conditions. DETAILED DESCRIPTION OF THE INVENTION

The teachings disclosed herein are based, in part, upon engineering of a heterodimeric protein assemblies that are capable of binding to human TGF-β and attenuating its function. The heterodimeric protein assemblies of the present teachings are two-subunit chimeric proteins, each subunit comprised of a cytoplasmic fragment of human TGFBR1, TGFBR2, or TGFBR3, fused with a Fc fragment of human IgG.

The terms used in this specification generally have their ordinary meanings in the art, within the context of this invention and in the specific context where each term is used. Certain terms are discussed below or elsewhere in the specification, to provide additional guidance to the practitioner in describing the compositions and methods of the invention and how to make and use them. The scope or meaning of any use of a term will be apparent from the specific context in which the term is used. "About" and "approximately" shall generally mean an acceptable degree of error for the quantity measured given the nature or precision of the measurements. Typically, exemplary degrees of error are within 20 percent (%), preferably within 10%, and more preferably within 5% of a given value or range of values. Alternatively, and particularly in biological systems, the terms "about" and "approximately" may mean values that are within an order of magnitude, preferably within 5- fold and more preferably within 2-fold of a given value. Numerical quantities given herein are approximate unless stated otherwise, meaning that the term "about" or "approximately" can be inferred when not expressly stated.

The methods of the invention may include steps of comparing sequences to each other, including wild-type sequence to one or more mutants (sequence variants). Such comparisons typically comprise alignments of polymer sequences, e.g., using sequence alignment programs and/or algorithms that are well known in the art (for example, BLAST, FASTA and MEGALIGN, to name a few). The skilled artisan can readily appreciate that, in such alignments, where a mutation contains a residue insertion or deletion, the sequence alignment will introduce a "gap" (typically represented by a dash, or "A") in the polymer sequence not containing the inserted or deleted residue.

The methods of the invention may include statistical calculations, e.g. determination of IC50 or EC50 values, etc.. The skilled artisan can readily appreciate that such can be performed using a variety of commercially available software, e.g. PRISM (GraphPad Software Inc, La lolla, CA, USA) or similar.

"Homologous," in all its grammatical forms and spelling variations, refers to the relationship between two proteins that possess a "common evolutionary origin," including proteins from superfamilies in the same species of organism, as well as homologous proteins from different species of organism. Such proteins (and their encoding nucleic acids) have sequence homology, as reflected by their sequence similarity, whether in terms of percent identity or by the presence of specific residues or motifs and conserved positions. However, in common usage and in the instant application, the term "homologous," when modified with an adverb such as "highly," may refer to sequence similarity and may or may not relate to a common evolutionary origin.

The term "sequence similarity," in all its grammatical forms, refers to the degree of identity or correspondence between nucleic acid or amino acid sequences that may or may not share a common evolutionary origin.

The terms “protein” and “polypeptide” are used interchangeably. The polypeptides described herein may be comprised of more than one contiguous amino acid chain, thus forming dimers or other oligomeric formations. In general, the polypetides of the present teachings for use in mammals are expressed in mammalian cells that allow for proper post-translational modifications, such as CHO or HEK293 cell lines, although other mammalian expression cell lines are expected to be useful as well. It is therefore anticipated that the polypeptides of the present teachings may be post-translationally modified without substantially effecting its biological function.

In certain aspects, functional variants of the heterodimeric protein assemblies of the present teachings include fusion proteins having at least a biologically active portion of the human TGFBR1, TGFBR2, or TGFBR3 or a functional fragment thereof, and one or more fusion domains. Well known examples of such fusion domains include, but are not limited to, polyhistidine, Glu-Glu, glutathione S transferase (GST), thioredoxin, protein A, protein G, an immunoglobulin heavy chain constant region (e.g., an Fc), maltose binding protein (MBP), or human serum albumin. A fusion domain may be selected so as to confer a desired property. For example, the TGFBR1, TGFBR2, or TGFBR3 polypeptide portions may be fused with a domain that stabilizes the TGFBR1, TGFBR2, or TGFBR3 polypeptides in vivo (a "stabilizer" domain), optionally via a suitable peptide linker. The term "stabilizing" means anything that increases the half life of a polypeptide in systemic circulation, regardless of whether this is because of decreased destruction, decreased clearance, or other pharmacokinetic effect. Fusions with the Fc portion of an immunoglobulin are known to confer desirable pharmacokinetic properties on certain proteins. Likewise, fusions to human serum albumin can confer desirable properties. Other types of fusion domains that may be selected include multimerizing (e.g., dimerizing, tetramerizing) domains and functional domains that confer an additional biological function, e.g. promoting accumulation at the targeted site of action in vivo.

In an exemplary embodiment, the first subunit is a fusion of a cytoplasmic fragment of human TGFBR2 with a Fc-1 portion of human IgGl (GenBank: J00228.1). The IgG-Fc-1 fragment has five amino acid substitutions to improve heterodimer formation and other functions. The second subunit is a fusion protein of an extracellular domain of either TGFBR1 L34-E125, for Isoform-1, or TGFBR1 L34-E129, for Isoform-2, or an extracellular fragment G21-D379 of human TGFBR3, fused with Fc-2 fragment of IgGl. The Fc-2 fragment fused to TGFBR1 or TGFBR3 extracellular domains has three mutations for function modifications.

The extracellular fragment of TGBFR2 extends from T23 to QI 66. However, last 7 amino acid residues of this domain are LLLVIFQ, wherein 6 out of 7 residues are hydrophobic. These residues were therefore removed, and thus the construct ends with residue D159, resulting in a 137 amino acid fragment. Therefore, in one exemplary embodiment TGFBR2-Fc-1 fusion protein may comprise the amino acid sequence of SEQ. ID NO. 1

TGFBR2-Fc-1 fusion polypeptidel (SEQ ID NO. 1) In another exemplary embodiment TGFBR2-Fc-1 fusion protein may comprise the amino acid sequence of SEQ. ID NO. 2

TGFBR2-Fc-1 fusion polypeptide2 (SEQ ID NO. 2)

To generate TGFBRl-Fc-1 fusion protein, both isoforms were used where coding sequences of extracellular domains of TGFBR1 were fused with a modified human IgGl-Fc fusion protein. Native coding sequence of TGFBR1, L34-E125 or L34-E129 fragments (Isoform-1 and Isoform-2, respectively) were used for fusion with modified IgG-Fc-2 that includes three mutations: S237C, T249W and K292A.

In one exemplary embodiment TGFBRl-Fc-2 fusion protein (Isoform-1) may comprise the amino acid sequence of SEQ. ID NO. 3. In another exemplary embodiment TGFBRl-Fc-2 fusion protein (Isoform-3) may comprise the amino acid sequence of SEQ. ID NO. 4.

To generate TGFBR3-Fc-2 fusion molecule, G21-D379 was fused to a modified human IgGl- Fc-2 fusion protein which includes three mutations S237C, T249W and K292A.

In an exemplary embodiment TGFBR3-Fc-2 fusion protein may comprise the amino acid sequence of SEQ. ID NO. 5. Using well established molecular biology techniques, synthetic gene corresponding to the protein designs of the fusion proteins of the present teachings have been codon-optimized for expression in mammalian cells and synthesized. Invitrogen pTT vector was digested with EcoRI and BamHI restriction enzymes and corresponding constructs digested with EcoRI and BamHI were cloned into the vector. All constructs were verified by sequencing of the entire coding region to ensure correct cloning and coding sequence integrity.

In an exemplary embodiment the DNA fragment with an open reading frame corresponding to TGFBR2-Fc-1 fusion protein may comprise the DNA sequence of SEQ. ID NO. 6. In another exemplary embodiment the DNA fragment with an open reading frame corresponding to TGFBR2-Fc-1 fusion protein may comprise the DNA sequence of SEQ. ID NO. 7. In an exemplary embodiment the DNA fragment with an open reading frame corresponding to TGFBR3-Fc-2 fusion protein may comprise the DNA sequence of SEQ. ID NO. 8.

In another exemplary embodiment the DNA fragment with an open reading frame corresponding to TGFBR3-Fc-2 fusion protein may comprise the DNA sequence of SEQ. ID NO. 9.

TGFBR3-Fc-2 DNA2 (SEQ ID NO. 9)

EXAMPLES

The following Examples illustrate the forgoing aspects and other aspects of the present teachings. These non-limiting Examples are put forth so as to provide those of ordinary skill in the art with illustrative embodiments as to how the compounds, compositions, articles, devices, and/or methods claimed herein are made and evaluated. The Examples are intended to be purely exemplary of the inventions disclosed herein and are not intended to limit the scope of what the inventors regard as their invention. Efforts have been made to ensure accuracy with respect to numbers (e.g., amounts, temperature, etc.) but some errors and deviations should be accounted for.

Example 1: Construction of plasmids for transient expression of polypeptides of the present teachings.

Plasmid maps for expression of TGFBR2-Fc-1, TGFBR11-Fc-l, TGFBR13-Fc-1 and TGFBR3-Fc-2 constructs for transient transfection in mammalian cells are shown illustratively in Fig. 11, 12, 13, and 14.

Example 2: Transient protein expression of polypeptides of the present teachings.

Transient transfection in Expi293F cells was carried out for protein production. Cells were transfected according to manufacturer's protocol. Plasmid DNA was isolated using CsCl density gradient centrifugation and ethidium bromide was removed by ion exchange column with Dowex AG50W-X8 resin. Expi293F cells were propagated in Freestyle Expi293F Expression Medium to a density of 2.00E+06 - 2.50E+06 cell/ml and then diluted to 1.20E+06 cells/ml by Freestyle Expi293F Expression Medium 18-20 hours before transfection.

At the day of transfection, cell density should be at 2.20E+06 - 2.50E+06 cell/ml. A 125- ml batch of Expi293F cells was used as a standard transfection reaction

The cells were transfected with equivalent amounts of constructs containing TGFBR2-Fc-1 DNA1 (SEQ ID NO. 6) and TGFBR3-Fc-2 DNA1 (SEQ ID NO. 8) (Fig. 11 and Fig. 14) in pTT5 vector (62.5 pg + 62.5 pg). DNA for both constructs was diluted in 3 ml of PBS

Linear Polyethyleneimine (LPEI), 3 mg/ml stock solution, was diluted in 3 ml of PBS to a final concentration of 0.21 mg/ml. Total amount of LPEI added to transfection was 625 pg i.e. 5- fold the mass of combined DNA

To 3 ml of DNA solution, 3 ml of diluted LPEI solution was added, vortexed and incubated at room temperature for 10 min

Mixture of DNA:LPEI, 6 ml was added dropwise to 125 ml of cell suspension 24 hours post-transfection, 3.3 ml of 40% Tryptone N1 in PBS was added to the transfected cell suspension to a final concentration of 1%

On Day 5 after transfection, cell density of transfected cells was usually 4.50E+06 - 5.50E+06 cell/ml and viability - 70% - 88%

Example 3: Purification of polypeptides of the present teachings.

Transiently expressed, as described of foregoing Example 2, TGFBR2-Fc-1/ TGFBR3-Fc-2 (also referred to as TGFBR2/TGFBR3/Fc) heterodimer protein was purified using Protein A Sepharose. Cells were harvested on Day 5 by centrifugation at 300 g for 10 min and the resulting supernatant was centrifuged at 35,000 g for 30 min. Cleared supernatant was filtered through 0.2 p filter to remove cell debris and protein aggregates. Protein A Sepharose, 1 ml, was washed with PBS to remove preservatives and was added to cleared and filtered culture medium. The resin was incubated with the culture medium at 4°C on a shaker for overnight. After incubation, Protein A resin was packed into a column, washed with 10 column volumes of PBS. Protein A-bound proteins were eluted with 100 mM Gly-HCl pH 3.0 and the eluate was immediately neutralized by adding 1/1 Oth of the eluate volume of 1 M Tris-HCl pH 8.8. Eluted protein was extensively dialyzed against PBS, filter sterilized through 0.2 p filter and stored at 4°C. Purity of the resulted

TGFBR2/TGFBR3/Fc preparation was analyzed by SDS-PAGE analysis (Fig. 2)

Example 4: Functional characterization of TGFBR2/TGFBR3/Fc heterodimer protein.

Functional properties of purified TGFBR2/TGFBR3/Fc protein were tested a cell-based assay. MDA-MB-231 cells respond to TGF-β treatment by secretion of IL-11.

Cell maintenance:

1. Propagate MDA-MB-231 cells in the DMEM medium containing 10% FBS on T75;

2. Trypsinize cells, re-suspend in DMEM containing 10% FBS;

3. Test cell viability and count cell numbers using CytoSmart cell counter or equivalent. Standard viability should be >90%, if less, do not use this batch of cells;

4. Prepare dilutions of the cells for plating at cell density of 2.00E+05 cells/ml. Plate 1 ml of the cell suspension per well of 24-well plate;

5. Allow cells to attach for 16-24 hours at 37°C, 5% CO2, then remove medium;

6. Wash wells with fresh serum-free DMEM, 1.0 ml per well twice, add 1ml of serum -free DMEM and incubate 20-24 hours at 37°C, 5% CO2;

7. Wash wells with fresh serum-free DMEM, 1.0 ml per well twice

8. Prepare test substances according to experimental design using serum-free DMEM. If both TGF-β and TGFBR2/TGFBR3/Fc are used, incubate the mixture for 10-15 min at room temp prior to adding to the cells.

9. Aspirate medium from the wells and add pre-incubated samples and appropriate controls, 0.5 ml per well

10. Incubate the cells after treatment for 24 hours then harvest the supernatants;

11. Centrifuge the supernatants at 300 x g for 10 min, collect cleared supernatants.

Optimization of assay conditions:

Quantitation of IL- 11 secretion by MDA-MB-231 cells was carried out using anti -human IL- 11 ELISA kit. All steps of the procedure were performed according to manufacturer’s recommendations. To identify suitable conditions for IL-11 production assay, dose-response curve for TGF-β was generated. Starting from 5 ng/ml and using 2-fold serial dilutions, TGF-β was titrated down to 5 pg/ml. Variable slope (four parameter) curve fitting algorithm was used for calculation of EC50 value that was 0.219 ng/ml (Fig. 3). Due to the nature of the ELISA kit used for these series of experiments, absorbance at EC50 resulted only in 0.25-0.30 AU. To study TGFBR2/TGFBR3/Fc inhibition of TGF-β activity, higher concentration of TGF-β were used that are specified for each experiment.

To investigate inhibitory properties of other TGFBR2/TGFBR1/Fc chimeras, MDA-MB-231 cells were treated with TGF-β in the presence of varying concentrations of TGFBR2/TGFBR1- Isoform-l/Fc and TGFBR2/TGFBRl-Isoform-3/Fc. For this experiment, TGF-β was added at a fixed concentration of 2 ng/ml and TGFBR2/TGFBR1/Fc chimeras were titrated from 1,000 ng/ml to 1 ng/ml. Both chimeras demonstrated concentration-dependent inhibition of TGF-β -induced IL- 11 secretion. Calculated IC50 values for TGFBR2/TGFBR1-Isoform-1/Fc and TGFBR2/TGFBRl-Isoform-3/Fc chimeras were 23.1 ng/ml and 12.3 ng/ml, respectively (Fig. 4).

To investigate inhibitory properties of TGFBR2/TGFBR3/Fc fusion protein, MDA-MB-231 cells were treated with TGF-β in the presence of varying concentrations of TGFBR2/TGFBR3/Fc. For this titration experiment, TGF-β was added at a constant concentration of 2 ng/ml and TGFBR2/TGFBR3/Fc fusion protein was titrated from 1,000 ng/ml to 1 ng/ml. TGFBR2/TGFBR3/Fc demonstrated monotonous concentration-dependent inhibition of TGF-fL induced IL-11 secretion. The experiment was repeated twice and calculated IC50 values were 4.78 ng/ml and 2.35 ng/ml, respectively (Fig. 5).

Inhibition of TGFBR2/TGFBR3/Fc was studied at two different concentrations of TGF-β , 1 ng/ml and 2 ng/ml. Maximal concentration of TGFBR2/TGFBR3/Fc was reduced from 1,000 ng/ml used in previous experiments to 100 ng/ml. TGFBR2/TGFBR3/Fc displayed monotonous inhibition of IL-11 secretion for both TGF-β concentrations. TGFBR2/TGFBR3/Fc IC50 values for 1 ng/ml and 2 ng/ml were 2.9 ng/ml and 5.3 ng/ml, respectively (Fig. 6).

Molecular mass of TGF-β is 24 kDa and estimate molecular mass of TGFBR2/TGFBR3/Fc is about 120 - 130 kDa. In terms of molar ratios between TGF-β and TGFBR2/TGFBR3/Fc, for TGF-P concentration of 1 ng/ml or 42 pM, IC50 for TGFBR2/TGFBR3/Fc is 2.9 ng/ml or 24 pM. For 2 ng/ml or 84 pM TGF-fl, IC50 value is 5.2 ng/ml or 43 pM. For both concentrations, TGF-fl is present in apparent 2-fold molar excess over TGFBR2/TGFBR3/Fc. If TGFBR2/TGFBR3/Fc is capable of acting like a dimer and to bind two dimeric TGF-β molecules, then at both TGF-fJ concentrations TGFBR2/TGFBR3/Fc:TGF-β molar ratios are roughly 2: 1. Given that the TGF- P:TGFBR2/TGFBR3/Fc ratio remains the same, it could be suggested that lowering concentrations of TGF-β and TGFBR2/TGFBR3/Fc further would result in apparently lower IC50 values. Due to the nature of this particular assay, it is not practical to use TGF-β concentration below 1 ng/ml because IL-11 secretion response is diminished, and this affects data reliability. Therefore, for this particular assay under tested concentrations of TGF-fJ, IC50 value for TGFBR2/TGFBR3/Fc could be as low as 2.9 ng/ml (24 pM).

Example 5: Characterization of TGFBR2/TGFBR3/Fc - TGF-β interaction in surface plasmon resonance.

To study interaction between TGF-β to TGFBR2/TGFBR3/Fc using surface plasmon resonance, the nature of TGFBR2/TGFBR3/Fc of being human IgGl-Fc fusion protein has been utilized. One of the standard approaches to capture IgG-containing molecules is immobilization of species-specific capturing antibodies on a chip with subsequent loading of the test fusion proteins followed by ligand application. This approach was used for measuring binding parameters for TGFBR2/TGFBR3/Fc and TGF-β as schematically shown in Fig. 7.

Protein conjugation to the chip

Anti-human antibodies were covalently attached to the chip the following protocol.

1. SR7000 Gold Sensor Slide was placed into the instrument and primed with Running Buffer, (lx PBS supplemented with 0.005% Tween 20), for 5 min at 250 pl/min, then allowed to stabilize at 25 pl/min. All steps were carried out at 25 °C; . Protein preparations were diluted using Immobilization Buffer (10 mM Na-acetate pH 5.0) to a final concentration of 25 pg/ml; . Reagents for immobilization procedure were prepared as follows. EDC/NHE activation agent consisting of EDC (l-ethyl-3-(3-dimethylaminopropyl)-carbodiimide) at 40 mg/ml and NHS (N-hydroxy succinimide) at 10 mg/ml in water. 1 M Ethanolamine-HCl pH 8.5 in water; . Activation: EDC/NHS activation agent was injected into the chip at 10 pl/min for 8 min followed by 5 min wash with Running Buffer; . Immobilization: Anti -Human IgG or Fc receptors at a final concentration of 25 pg/ml were injected into the chip at 10 pl/min for 9 min; . Deactivation: Unreacted active groups on the chip surface were blocked by injection of 1 M Ethanolamine-HCl at 10 pl/min for 8 min; . After antibody or Fc receptors conjugation, the chip was washed with Running Buffer for 15 min at 25 pl/min.

Determination of binding constants for TGF-13 and TGFBR2/TGFBR3/Fc interaction

To study interaction of TGF-β and TGFBR2/TGFBR3/Fc, antibody-capturing approach was used. Ant-human IgG were covalently attached to the chip and used for capturing TGFBR2/TGFBR3/Fc (Fig. 7).

1. Chip with immobilized anti-human IgG was equilibrated with Running Buffer at a flow rate of 25 pl/min for 15 min;

2. TGFBR2/TGFBR3 at a final concentration of 25 pg/ml, was loaded at 25 pl/min for 2 min, then the chip was washed for 3 min with Running Buffer to remove unbound material;

3. A series of 2-fold dilutions of TGF-β were prepared using Running Buffer starting from 100 nM concentration. Seven concentrations were used: 100, 50, 25, 12.5, 6.25, 3,125 and 1.56 nM. The ligand was loaded at 25 pl/min for 2 min;

4. After ligand loading, dissociation phase of the experiment was carried out using Running Buffer at 25 pl/min flow rate for 4 min;

5. Regeneration step, dissociation of protein complexes bound by immobilized antihuman IgG was carried out by 3 M MgC1 ₂ running though the chip at 25 pl/min for 30 sec;

6. Series of sensograms for series of TGFBR2/TGFBR3/Fc at different concentrations of TGF-P were generated and used for analysis; 7. A kinetic evaluation of 1 : 1 binding model was used for the analysis of TGF-fl interaction with TGFBR2/TGFBR3/Fc.

8. Dissociation constant, KD, (M), was calculated from association, ka, (l/(M*s) and dissociation, kd, (1/s) constants generated from analysis of series of the concentration curves. KD (M) is a quotient of division of kd, (1/s) by ka, (l/(M*s).

After capturing TGFBR2/TGFBR3/Fc by anti-human antibodies immobilized on chip (Fig. 7), varying concentrations of TGF-β were loaded and binding sensograms were generated (Fig. 7 and Fig. 8). The series of the sensograms were analyzed by kinetic evaluation using 1 : 1 binding model. SPR binding experiment was repeated three times to demonstrate reproducibility. Dissociation constants were calculated from association constant Ka (1/M*s) and dissociation constant Kd (1/s) (Tables 1, 2 and 3). KD, (M) is equal Ka divided by Kd.

According to SDS-PAGE analysis of TGFBR2/TGFBR3/Fc under non-reduced conditions (Figure 6), the preparation used for this study contained more than single band indicating a presence of various homo- and/or hetero-dimers. As the result, association phase of sensograms was not monotonous and thus more difficult to analyze. However, dissociation phase of all sensograms demonstrates remarkable lack of dissociation thus indicating high affinity interaction.

The results are summarized in Tables 1, 2 and 3 and the corresponding series of sensograms are shown on Fig. 8, 9 and 10. Summary table (Table 4) show mean KD, (M) values, and standard deviation (SD) for each experiment. Median value for each experiment and corresponding average is shown as well.

Table 1: Experiment 1 : Binding parameters for binding of TGF-β to captured TGFBR2/TGFBR3/Fc. Binding constant, KD (M), was averaged for all 7 curves and also median was calculated. Table !: Experiment !: Binding parameters for binding of TGF-β to captured

TGFBR2/TGFBR3/Fc. Binding constant, KD (M), was averaged for all 7 curves and also median was calculated.

Table 3: Experiment 3: Binding parameters for binding of TGF-β to captured TGFBR2/TGFBR3/Fc. Binding constant, KD (M), was averaged for all 7 curves and also median was calculated.

Table 4: Summary of binding constant, KD (M), for binding of TGF-β to TGFBR2/TGFBR3/Fc calculated from 3 independent experiments.

Example 6: Generation of stable cell line for overexpression of TGFBR2/TGFBR3/Fc heterodimer protein.

In this Example, the amino acid sequences of TGFRB2-Fcl fusion polypeptide2 (SEQ ID NO. 2) and TGFRB3-Fc2 fusion polypeptide (SEQ ID NO. 5) for the composition of TGFBR2/TGFBR3/Fc fusion protein were codon optimized for mammalian cell expression. DNAs encoding for codon optimized TGFRB2-Fcl DNA2 (SEQ ID NO. 7) with an additional PacI site at the 5’ end and a BGH poly A region plus an Asci site at the 3’ end and TGFRB3-Fc2 DNA2 (SEQ ID NO. 9) were then chemically synthesized. The DNA fragment that contains the coding region of TGFRB2-Fcl was cloned into an intermediate vector pAG12B between Nhel and Xhol sites to generate pAG12B-TGFRB2-Fcl vector (Fig 15). The DNA fragment that contains the coding region of TGFRB3-Fc2 was cloned into pAG12B vector between the Nhel and Xhol sites to generate pAG12B-TGFRB3-Fc2 vector (Fig 16). To generate vector pAG12B- TGFBR2/TGFBR3/Fc (Fig 17), the Mlul-Xhol fragment from the vector pAG12B-TGFRB3-Fc2 that contains the entire promoter region and TGFRB3-Fc2 was cloned into vector pAG12BTGFRB2-Fcl between Asci and Xhol sites. The isolated positive clones from each stage were validated by DNA sequencing analysis using a variety of primers. The full sequence of the pAG12B- TGFBR2/TGFBR3/Fc comprises SEQ. ID NO. 10. pAG12B- TGFBR2/TGFBR3/Fc DNA (SEQ ID NO. 10)

CHO-K1 cells (ATCC CCL-61) were maintained in DMEM/F12 media supplemented with 10% FBS at 37oC with 5% CO2. Transfection of CHO-K1 cells with pAG12B- TGFBR2/TGFBR3/Fc vector in the presence of transfection reagent “SuperFect” was performed according to the manufacturer’s protocol (Qiagen #301305). In brief, 2ug of DNA, 8ul of Superfect and 200ul of DMEM medium (no FBS) were mixed and incubated at room temperature for 15min. CHO-K1 cells were seeded in a 6-well plate at least 2h prior transfection and the transfection mix was added drop-wise to the surface of each well of freshly seeded cells. Transfected cells were then passaged 48h post-transfection. Stably transfected cells were 1 : 10 diluted and selected by adding hygromycin B to final 300ug/ml in complete medium. The hygromycin B-containing medium was changed twice a week until individual colonies became visible (3-4 weeks). For the first screen, 96 individual clones derived from single cells (most likely) were picked and amplified in a 96-well plate. After incubated at 37oC/5% CO2 for 10-14 days, conditioned media from each well were collected and analyzed by dot blot (or Western blot) using HRP-conjugated goat antihuman IgG (Pierce) (Fig 18). Two highly expressed clones were picked for the second single cell clone screen. To make sure the final developed TGFBR2/TGFBR3/Fc/CHO cell line compliance to the clonality requirement, limiting dilution cloning (LDC) method was employed. Two clones (#31 and #36) from the first screen were passaged twice with >98% cell viability. The procedure described below using one 50ml conical tube is sufficient to seed 2 x 96-well plate for each of the two clones isolated from the first screen. a) Treat passaged cells with trypsin-EDTA and dilute the cells with the complete medium to a final concentration of 1000 viable cells/ml; b) After carefully counting, further dilute cells with 50ml of complete medium to 0.5-1 viable cell/0.2ml; c) Using the multiple channel pipettor to transfer 0.2ml of the diluted cells into each well of the 96-well plate (2 plates for #31 and 2 plates for #36); d) After 12h of plating, examine the plates and label those wells that contain a single viable cell; e) Monitor two wells with single healthy cell by daily microimaging (Fig 19 and 20); f) Incubate the plates at 37oC and 5% CO2 for 12-14 days g) Transfer 5ul of conditioned medium from each of the wells that were labeled from single cells to a new 96-well dot blot apparatus; h) Examine the expression of TGFBR2/TGFBR3/Fc from single cell clones (Fig. 21)

Cells from clones #A6 and #G9 were amplified in the complete medium and the productivity of the two clones were assessed by ELISA assay. Comparing to the purified TGFBR2/TGFBR3/Fc, an estimate production for clone A6 and G9 is 50-100mg/L.

For the stability test, five clones from the second screen (A6, DIO, DI 1, F7 and G9) were selected and tested for their stability. 10,000 cells/well was seeded in a 96-well plate with triplicates for each clone. Cultured media were collected from each passage (twice a week) and the expression of TGFBR2/TGFBR3/Fc was monitored for 10 passages by ELISA assay using the primary antibody against human IgG (Fig. 22).

TGFBR2/TGFBR3/Fc protein heterodimer of this example was purified from conditioned media essentially as described in the forgoing Example 3. Purity of purified TGFBR2/TGFBR3/Fc heterodimer protein was assessed by SDS-PAGE under non-reducing conditions (Fig. 23). Purified protein was equilibrated against a buffer containing 0.02% PS-20, 3% D-mannitol, 20 mM NaCl, 70 mM L-proline, 25 mM histidine, pH 6.0, and aliquoted for storage at 4°C.

All publications and patents mentioned herein are hereby incorporated by reference in their entirety as if each individual publication or patent was specifically and individually indicated to be incorporated by reference.

While specific embodiments of the subject matter have been discussed, the above specification is illustrative and not restrictive. Many variations will become apparent to those skilled in the art upon review of this specification and the claims below. The full scope of the invention should be determined by reference to the claims, along with their full scope of equivalents, and the specification, along with such variations.