METHODS OF USING A TGF-beta KNOCKOUT CELL LINE AND COMPOSITIONS RESULTING THEREFROM

Abstract

This disclosure describes protein compositions including minimal TGFβ1 contamination, methods for making those protein compositions, and methods for using those protein compositions. In one aspect, this disclosure describes protein compositions produced by a TGFβ1 knockout cell line and methods for using those compositions including, for example, in research or as a therapeutic or prophylactic. In another aspect, this disclosure describes a TGFβ1 knockout cell line and methods of making that cell line.

Claims

1. A method for producing a target biologic, wherein the method comprises: culturing a TGFβ1 knockout cell line that overexpresses the target biologic; and purifying the target biologic.

2. The method of claim 1, wherein the method further comprises introducing a vector comprising a gene encoding the target biologic into the TGFβ1 knockout cell line.

3. The method of claim 1, wherein the target biologic comprises a recombinant protein.

4. The method of claim 1, wherein the target biologic comprises a TGFβ superfamily ligand.

5. The method of claim 1, wherein the target biologic comprises an antibody.

6. The method of claim 1, wherein the target biologic comprises a mouse protein or a human protein.

7. The method of claim 1, wherein the cell line comprises a Chinese Hamster Ovary (CHO) cell line.

8. A composition comprising the target biologic produced by the method of claim 1, wherein the composition comprises less than 0.5 ng/mL TGFβ1.

9. A composition comprising the target biologic produced by the method of claim 1, wherein the composition comprises less than 0.5 fg/mL TGFβ1.

10. The composition of claim 8, wherein the composition comprises an undetectable level of TGFβ1.

11. A composition comprising a target biologic, wherein the target biologic comprises a TGFβ superfamily ligand, and wherein the composition comprises less than 0.5 ng/mL TGFβ1.

12. A composition comprising a target biologic, wherein the target biologic comprises a biologic drug, and wherein the composition comprises less than 0.5 ng/mL TGFβ1.

13. The composition of claim 11, wherein the composition comprises less than 0.5 fg/mL TGFβ1.

14. The composition of claim 11, wherein the composition comprises an undetectable level of TGFβ1.

15. The composition of claim 11, wherein the target biologic comprises a recombinant protein.

16. The composition of claim 11, wherein the target biologic comprises a Wnt, a bone morphogenic protein (BMP), an activin, or a growth differentiation factor (GDF), or a combination thereof.

17. The composition of claim 11, wherein the target biologic comprises an antibody.

18. The composition of claim 8, wherein the target biologic comprises a mouse protein or a human protein.

19. The composition of claim 8, wherein the target biologic comprises a biologic drug.

20. The composition of claim 8, wherein the composition comprises a pharmaceutical composition.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0024] FIG. 1 shows the results of an exemplary genomic PCR, demonstrating a genetic mutation in the TGFβ1 genomic sequence of a CHO-S TGFβ1 knockout (KO) PCR product. Sample 1 is a CHO-S wild type (wt) PCR product; Sample 2 is a CHO-S TGFβ1 KO PCR product. The PCR amplicon from the CHO-S TGFβ1 KO DNA (Sample 2) exhibited a band approximately 64-base pairs larger than the PCR amplicon from wild type CHO-S genomic DNA (Sample 1).

[0025] FIG. 2 shows the verification of the introduction of a 64-base pair (bp) insert (indicated by brackets) into the TGFβ1 genomic sequence of the CHO-S TGFβ1 KO cells by DNA sequencing. A stop codon at amino acid 51 (shown by a “-” and highlighted in gray) was introduced by insertions and deletions (INDELs) by the error prone non-homologous end joining (NHEJ) double strand repair mechanism. The positive strand (shown 5′ to 3′) of the DNA sequence is SEQ ID NO:1. The negative strand (shown 3′ to 5′) of the DNA sequence is SEQ ID NO:2. The truncated TGFβ1 amino acid sequence is SEQ ID NO:3. The TGFβ1 amino acid sequence downstream of the stop codon is SEQ ID NO:7.

[0026] FIG. 3 shows exemplary ELISA results indicating that a TGFβ1 KO CHO cell line (CHO-S TGFβ1 KO) produces a significantly reduced amount of TGFβ1, as measured by enzyme-linked immunosorbent assay (ELISA), compared to the wild type CHO cell line (CHO-S wt). Detectable amounts of TGFβ1 were present in the conditioned media of CHO-S wt cells, HEK293 wt, and HEK293 cells expressing recombinant hGDF-15 (HEK wt+hGDF-15). The amount of TGFβ1 present in the conditioned media of CHO-S TFGβ1 knockout samples (CHO-S TGFβ KO) was below the detectable range. TGFβ1 levels were also below the detection limit in stable CHO-S TGFβ1 KO cells expressing recombinant hGDF-15 (CHO-S TGFβ1 KO+hGDF-15). Quantification of the results of FIG. 3 is shown in Table 1. Table 1 also shows that CHO-S TGFβ1 KO cells expressing recombinant hTGFB1 (CHO-S TGFβ1 KO+hTGFβ1) had significantly higher levels of TGFB1 than the parental cell line. In FIG. 3 and Table 1, “*” indicates the measured value was below the detection limit of the assay.

[0027] FIG. 4A shows recombinant hGDF15 protein that includes higher levels of TGFβ1 contamination can cause activation of Smad2 in the DU145 human prostate cancer cell line. The contents of each lane of FIG. 4A are described in Table 2A. FIG. 4B shows recombinant hGDF15 made in CHO-S TGFβ1 knockout cells does not activate phosphorylation of SMAD-2 when used to treat DU145 cells. The contents of each lane of FIG. 4B are described in Table 2B. Recombinant human TGFβ1 protein alone (1 ng/ml, lane 2), recombinant human GDF-15 from HEK293 cells (2 μg/ml, lane 6), and recombinant human GDF-15 from CHO-S wild type cells (2 μg/ml, lane 8) all induce phosphorylation of SMAD-2. The phosphorylation of SMAD-2 can be rescued with Chicken anti-TGFβ 1 antibody treatment demonstrating that this phosphorylation of SMAD-2 is TGFβ1 dependent. Recombinant human GDF-15 produced in TGFβ1 KO CHO-S cell line does not show any SMAD-2 phosphorylation in DU145 cells at a 2 μg/ml dose, demonstrating removal of the TGFβ1 protein contamination (lane 4).

[0028] FIG. 5 shows TGFβ1 negatively affects osteoblast differentiation. MC3T3-E1 preosteoblasts were treated with recombinant mouse Wnt-3a protein from a lot known to have a relatively high level of TGFβ1 contamination (78 picograms (pg) of TGFβ1 protein per microgram (μg) of Wnt-3a protein) (circles). The same dose of mouse Wnt-3a protein was added to MC3T3-E1 cells in the presence of a saturating dose (50 μg/mL) of a chicken anti-TGFβ 1 blocking antibody (triangles). Osteoblast differentiation was measured by quantifying intracellular alkaline phosphatase enzyme activity after three days of treatment.

[0029] FIG. 6 shows TGFβ1 negatively affects Wnt-3a-induced osteoblast differentiation. MC3T3-E1 preosteoblasts treated with recombinant human Wnt-3a protein exhibited a dose-responsive induction of osteoblast differentiation, signified by increasing alkaline phosphatase activity three days after addition of the Wnt-3a protein. The median effective dose (ED50) for this effect was 1.78 ng/mL (circles). When a constant dose of 20 ng/mL of recombinant human Wnt-3a protein was added to MC3T3-E1 cells for three days, a relatively high level of alkaline phosphatase activity resulted (flat line and triangles). If the same steady dose of 20 ng/mL of recombinant human Wnt-3a protein was added to cells and a dose titration of recombinant human TGFβ1 was added to this steady 20 ng/mL dose of Human Wnt-3a protein (squares), inhibition of Wnt-3a-mediated osteoblast differentiation was observed at a neutralizing dose of 50 percent (%) (ND50) of 53.3 femtogram per milliliter (fg/mL) TGFβ1.

[0030] FIG. 7 shows that the hWnt3a proteins, with relatively higher levels of TGFβ1 protein (see Table 3), show inhibition of MC3T3-E1 differentiation at the highest doses of hWnt3a protein, while the hWnt3a protein purified from the CHO-S TGFβ1 KO line do not show this inhibition of differentiation at higher hWnt3a doses. Lot #DLGC02 from a CHO-S TGFβ1 KO line (circles) shows a sigmoidal curve with a dose responsive increase in osteoblast differentiation assayed by alkaline phosphatase activity assays. The top three doses of both CHO-S wt-derived lots (RSK51 (squares) and RSK69 (triangles)) show significant reduction in osteoinductive activity in the highest three hWnt3a doses (1.67 ug/ml, 0.556 ug/ml, and 0.185 ug/ml), indicative of TGFβ1 inhibitory activity. This “dip” in activity was not observed when MC3T3-E1 cells were treated at the same doses with lot #DGLC02 (hWnt3a protein derived from the CHO-S TGFβ1 KO line) (circles).

[0031] FIG. 8 shows that hWnt3a proteins with high or low levels of TGFβ1 protein show similar activity in a HEK293 TCF9-Secreted Alkaline Phosphatase (SEAP) Wnt-responsive reporter assay. hWnt3a proteins from CHO-S TGFβ1 KO cells (lot #DLGC02, circles), CHO-S wt (lot #RSK51, squares), and CHO-S wt (lot #RSK69, triangles) all showed comparable activity. The ED50 for these lots: RSK51=145 ng/ml, RSK69=191 ng/ml, and DLGC02=221 ng/ml.

[0032] FIG. 9 shows that hWnt3a protein with high levels of TGFβ1 contamination results in lower osteoinductive activity at high hWnt3a doses in a MC3T3-E1 assay than hWnt3a protein purified from TGFβ1 KO CHO-S cells. The dose-response curve from hWnt3a purified from wild type CHO-S cells (hWnt3a, lot RSK 51, circles) showed a pronounced reduction in activity at the highest four doses of hWnt3a. When the same doses of hWnt3a from the same lot were added to the MC3T3-E1 cells in the presence of a static dose of 10 μg/ml of the TGFβ1 blocking antibody (hWnt3a, lot RSK51+10 μg/ml of TGFβ1 blocking antibody, triangles), the effect of high dose hWnt3a treatment doses was “rescued” to higher levels. hWnt3a protein purified from TGFβ1 KO CHO-S cells (hWnt3a, lot DLGC02, squares) which is known to have a lower level of TGFβ1 contamination (4.97 pg of TGFβ1/μg of hWnt3a, see Table 3) showed a much less pronounced drop in activity at high hWnt3a doses than hWnt3a purified from wild type CHO-S cells (hWnt3a, lot RSK51, diamonds) which has a much higher level of TGFβ1 (114.6 pg of TGFβ1/μg of hWnt3a, see Table 3.) The low level of TGFβ1 levels in purified hWnt3a lot DLG02 (squares) could be further rescued to higher activity levels with addition of 10 μg/ml of TGF1 blocking antibody (triangles) showing that even relatively lower levels of TGFβ1 contamination present in the serum used in the cell culture production media can result in unwanted off-target effects in osteoinductive cellular bioassays.

DETAILED DESCRIPTION OF ILLUSTRATIVE EMBODIMENTS

[0033] This disclosure describes compositions including a target biologic and minimal TGFβ1 contamination, methods for making those compositions, and methods for using those compositions. In some aspects, a method for making the composition includes culturing a TGFβ1 knockout cell line that overexpresses the target biologic and purifying the target biologic.

[0034] On occasion, endogenous proteins produced by host cell lines may be copurified with a target protein produced by the cell line. If the endogenous protein has very potent bioactivity, even a very low level of contamination may negatively affect the performance of the product. Chinese Hamster Ovary (CHO) cells are often used by pharmaceutical and biotech companies for the manufacturing of recombinant antibodies and other proteins.

[0035] One endogenous protein that is produced by CHO cells that may negatively affect the performance of the target protein is TGFβ1 (Beatson et al. Biotechnol Bioeng 108, 2759-2764 (2011)). TGFβ1 can elicit a biological effect at concentrations in a range as low as femtogram per milliliter (fg/mL) in very sensitive assays like a MC3T3-E1 osteoblast differentiation assay. (See, for example, FIG. 5 and FIG. 6.)

[0036] Due to the importance of TGFβ1 to cell proliferation and health, knocking out TGFβ1 in CHO cells was expected to negatively impact the cells (Strutz et al. Kidney Int 59, 579-592 (2001)). For example, it was unpredictable whether the cells would become sick or would exhibit poor growth kinetics compared to wild type cells.

[0037] As further described herein, however, knocking out TGFβ1 in CHO cells does allow for the propagation of CHO cells that do not produce TGFβ1. Although slightly slower growth parameters were observed in the knockout cell line, the cell line has been used to successfully produce hWnt3a, hBMP10, and hGDF15 recombinant proteins.

[0038] Moreover, in contrast to antibody columns—which were previously used to remove TGFβ1 from compositions including recombinant proteins, using cells that do not produce TGFβ1 allows for a more complete elimination of TGFβ1. In addition, using antibodies to remove TGFβ1 is not effective at removing latent TGFβ1, and latent TGFβ1 can be activated to produce biologically active mature TGFβ1 (Shi et al. Nature 474, 343-349 (2011)).

[0039] Furthermore, using a TGFβ1 knockout line to produce a recombinant protein is more cost-efficient than TGFβ1 antibody removal alone because milligrams of an anti-TGFβ1 antibody are otherwise needed to effectively remove TGFβ1.

[0040] Thus, in some embodiments, this disclosure describes methods for minimizing the introduction of TGFβ1 into a protein production process by eliminating a cell's ability to produce TGFβ1. Overall, eliminating a cell's ability to produce TGFβ1 provides a lower-cost alternative compared to removal of TGFβ1 with anti-TGFβ1 antibodies, provides for the ablation of both latent and mature TGFβ1 proteins, and results in more consistent removal of TGFβ1 contamination compared to removal with anti-TGFβ1 antibodies alone.

Compositions

[0041] In one aspect this disclosure describes compositions that include a target biologic and very low levels of TGFβ1. Exemplary biologics may include, for example, a recombinant protein (including, for example, a monoclonal antibody, a glycoprotein, a peptide hormone, or a toxin), a ribonucleoprotein, a non-peptide hormone (including, for example, a steroid hormone or an eicosanoid hormone), a glycoprotein, a glycolipid, a subcellular organelle, a blood component, etc. Combinations of the exemplary biologics are also contemplated including, for example, a recombinant glycoprotein or a subcellular organelle isolated from a blood component, etc. In some embodiments, the target biologic includes a biologic drug, that is, a biologic used for treatment of a disease or condition in a subject.

[0042] In some embodiments, the composition may preferably be produced by a TGFβ1 knockout cell line including, for example, a TGFβ1 knockout cell line described herein. In some embodiments, a composition produced by a TGFβ1 knockout cell line may be further purified to remove TGFβ1 including, for example, with an anti-TGFβ 1 antibody. In an exemplary embodiment, an antibody column including an anti-TGFβ 1 antibody may be used to remove TGFβ1. An anti-TGFβ1 antibody may be used to remove TGFβ1 introduced during cell culture (including, for example, as a component of fetal bovine serum (FBS)) of the TGFβ1 knockout cell line.

[0043] The target biologic may include a protein including, for example, a recombinant protein that has a use in research or therapy. The minimization or elimination of TGFβ1 from a composition including a protein that has a use in research or therapy may be important because TGFβ1 has many functionalities (both desired and unwanted). These functionalities can have important consequences in, for example, immune-oncology and regenerative medicine workflows. See, for example (Glick Cancer Biol Ther 3, 276-283 (2004), Park et al. Cancer Discov 6, 1366-1381 (2016), Tamayo et al. Int J Mot Sci 19, 3928 (2018), Wu et al. Bone Res 4, 16009 (2016)).

[0044] In some embodiments, when the target biologic includes a protein, the protein may include a TGFβ superfamily ligand. Exemplary TGFβ superfamily ligands include Wingless-type MMTV Integration Site Family proteins (Wnts), bone morphogenic proteins (BMPs), activins, and growth differentiation factors (GDFs).

[0045] The target biologic may be from or derived from any suitable species including, for example, mouse, human, rat, etc.

[0046] A Wnt may include, for example, Wnt1, Wnt2, Wnt2b, Wnt3, Wnt3a, Wnt4, Wnt5a, Wnt5b, Wnt6, Wnt7a, Wnt7b, Wnt8a, Wnt8b, Wnt9a, Wnt9b, Wnt10a, Wnt10b, Wnt11, or Wnt16, or a combination thereof. A low level of TGFβ1 contamination in compositions including a Wnt protein may be particularly beneficial because, due to the difficulty of manufacturing recombinant Wnt proteins, the Wnt protein concentration in Wnt protein formulations is generally low. Consequently, users of these low concentration compositions typically have to add a relatively higher volume of the Wnt protein formulation, resulting in the corresponding addition of more contaminating TGFβ1.

[0047] A BMP may include, for example, BMP2, BMP3, BMP4, BMPS, BMP6, BMP7, BMP8a, BMP8b, BMP10, BMP11, or BMP15, or a combination thereof. A low level of TGFβ1 contamination in compositions including a BMP is particularly important because while BMPs promote mesenchymal stem cell to osteoblast differentiation, TGFβ1 inhibits osteoblast differentiation. (See, for example, Wu et al. Bone Res. 4:16009 (2016), & Examples 3 and 4.) BMPs have multiple clinical applications. For example, BMP2 and BMP7 are approved for therapeutic clinical use for non-union fracture healing, spinal fusions, and oral surgery.

[0048] A GDF may include, for example GDF1, GDF3, GDF5, GDF6, GDF8, GDF9, GDF10, GDF11, or GDF15, or a combination thereof. A low level of TGFβ1 contamination in compositions including a GDF is particularly important to accurately determine the effects of the GDF (Olsen et al. PLoS One 12, e0187349 (2017)).

[0049] In some embodiments, the target biologic may include a protein that has a therapeutic use including, for example, an antibody. In some embodiments, the antibody may be a monoclonal antibody. Antibodies may target a protein of an immune checkpoint pathway. For example, exemplary antibodies may target a protein of the CTLA-4 or PD-1 pathway or both. Exemplary antibodies include nivolumab (anti-PD1), pembrolizumab (anti-PD1), atezolizumab (anti-PD-L1), durvalumab (anti-PD-L1), avelumab (anti-PD-L1), tremelimumab (anti-CTLA-4), or ipilimumab (anti-CTLA-4), or combinations thereof, or biosimilars thereof. A low level of TGFβ1 contamination in compositions including an antibody targeting a protein of the PD-1 pathway may be particularly important because TGFβ1-producing cells have been found to upregulate multiple components of the PD-1 signaling pathway, inhibiting antitumor immunity (Park et al. Cancer Discov 6, 1366-1381 (2016)).

[0050] Additional exemplary antibodies may include monoclonal antibodies such as bevacizumab, trastuzumab, adalimumab (HUMIRA), infliximab, rituximab, and biosimilars thereof. Other exemplary non-antibody target biologics with therapeutic effects include, for example, etanercept, epoetin alfa, pegfilgrastim, filgrastim, etc.

[0051] In some embodiments, the composition includes less than 0.5 ng/mL TGFβ1 protein, less than 0.1 ng/mL TGFβ1 protein, less than 0.01 ng/mL TGFβ1, less than 0.05 ng/mL (50 pg/mL) TGFβ1 protein, less than 0.02 ng/mL (20 pg/mL) TGFβ1 protein, less than 0.01 ng/mL (10 pg/mL) TGFβ1 protein, less than 0.005 ng/mL (5 pg/mL) TGFβ1 protein, less than 1 pg/mL TGFβ1 protein, less than 0.5 pg/mL (500 fg/mL) TGFβ1 protein, less than 0.01 pg/mL (100 fg/mL) TGFβ1 protein, less than 0.05 pg/mL (50 fg/mL) TGFβ1 protein, less than 0.005 pg/mL (5 fg/mL) TGFβ1 protein, or less than 0.001 pg/mL (1 fg/mL) TGFβ1 protein.

[0052] In some embodiments, the composition includes an undetectable level of TGFβ1. In some embodiments, the composition includes an undetectable level of TGFβ1 as measured using the Quantikine ELISA Kit (Catalog No. #DB100B, R&D Systems, Minneapolis, Minn.). The Quantikine ELISA Kit uses a quantitative sandwich enzyme immunoassay technique to detect natural and recombinant TGFβ1 and exhibits a minimum detectable concentration of human TGF-β1 in a range of 1.7 pg/mL to 15.4 pg/mL (mean 4.61 pg/mL). The minimum detectable concentration may be determined by adding two standard deviations to the mean optical density (O.D.) value of twenty zero standard replicates and calculating the corresponding concentration.

[0053] In an exemplary embodiment, the composition may include a Wnt or BMP in a range of 0.05 mg/mL to 0.1 mg/mL and TGFβ1 at a concentration of less than 0.02 ng/mL (20 pg/mL). At these levels, the negative effects of TGFβ1 contamination on Wnt's effects in MC3T3-E1 cells are no longer observed in many assays.

[0054] In another exemplary embodiment, the composition may include a Wnt or BMP in a range of 0.05 mg/mL to 0.1 mg/mL and TGFβ1 at a concentration of less than 0.001 pg/mL (1 fg/mL) TGFβ1 protein. At these levels, the negative effects of TGFβ1 contamination on Wnt's effects in MC3T3-E1 cells is expected to no longer be observed in the alkaline phosphatase (ALP) activity assay of FIG. 6. (The ND50 of recombinant TGFβ1 in this assay was 53 fg/mL.)

[0055] In some embodiments, the composition may be lyophilized. Such compositions may include a buffer, for example, bicarbonate, for reconstitution prior to use or administration, or the buffer may be included in the lyophilized composition for reconstitution with, for example, water. The lyophilized composition can be provided in a syringe, optionally packaged in combination with the buffer for reconstitution, such that the reconstituted composition can be immediately administered to a patient.

[0056] In some embodiments, the composition may include at least 40 μg/mL, at least 50 μg/mL, at least 100 μg/mL, at least 150 μg/mL, or at least 200 μg/mL of the target biologic. In some embodiments, the composition may include up to 50 μg/mL, up to 100 μg/mL, up to 150 μg/mL, up to 200 μg/mL, up to 300 μg/mL, or up to 400 μg/mL of the target biologic.

[0057] In some embodiments, the composition may be a pharmaceutical composition. Pharmaceutical compositions may be formulated in a variety of forms adapted to the chosen route of administration. The composition will vary depending on mode of administration and dosage unit. For example, for parenteral administration, isotonic saline can be used. For topical administration a cream, including a carrier such as dimethylsulfoxide (DMSO), or other agents typically found in topical creams that do not block or inhibit activity of the peptide, can be used. Other suitable carriers include, but are not limited to alcohol, phosphate-buffered saline, and other balanced salt solutions. The compounds of this invention can be administered in a variety of ways, including, but not limited to, intravenous, topical, oral, subcutaneous, intraperitoneal, and intramuscular delivery. In some aspects, the composition of the present invention may be formulated for controlled or sustained release of the target biologic. In some aspects, a formulation for controlled or sustained release is suitable for subcutaneous implantation. In some aspects, a formulation for controlled or sustained release includes a patch.

[0058] In some embodiments, including when the composition is a pharmaceutical composition, the composition may include the target biologic as an active agent and a pharmaceutically acceptable carrier. The active agent may be formulated in a pharmaceutical composition and then administered to a vertebrate, particularly a mammal, such as a human patient, a companion animal, or a domesticated animal, in a variety of forms adapted to the chosen route of administration.

[0059] A pharmaceutically acceptable carrier can include, for example, an excipient, a diluent, a solvent, an accessory ingredient, a stabilizer, a protein carrier, or a biological compound. Non-limiting examples of a protein carrier includes keyhole limpet hemocyanin (KLH), bovine serum albumin (BSA), ovalbumin, trehalose, or the like. Non-limiting examples of a biological compound which can serve as a carrier include a glycosaminoglycan, a proteoglycan, and albumin. The carrier can be a synthetic compound, such as dimethyl sulfoxide or a synthetic polymer, such as a polyalkyleneglycol. Ovalbumin, human serum albumin, other proteins, polyethylene glycol, or the like can be employed as the carrier. In some embodiments, the pharmaceutically acceptable carrier includes at least one compound that is not naturally occurring or a product of nature.

TGFβ1 Knockout Cell Line

[0060] In one aspect, this disclosure describes a TGFβ1 knockout cell line and methods of making that cell line.

[0061] The cell line may include any suitable cell line. In some embodiments, the cell line may include a CHO cell line. The development and characterization of a TGFβ1 knock out CHO cell line is described in Example 1. Exemplary CHO cell lines include, for example, a CHO-S line, a CHOK1 line, a CHO-DXB11 cell line, and a CHO-DG44 line. In some embodiments, the cell line may include a human embryonic kidney (HEK) cell line including, for example an HEK 293 cell line. In some embodiments, the cell line may include a NS0 cell line. In an exemplary embodiment, the cell line may include a TGFβ1 KO CHO line, as described in Example 1.

[0062] The knockout cell line may be made using any suitable means. For example, as described in Example 1, the cell line may be made using CRISPR-based gene editing. In some embodiments, CRISPR-based gene editing may include CRISPR/Cas9-based gene editing. Additionally or alternatively, CRISPR-based gene editing may include CRISPR and a non-Cas9 CRISPR endonuclease including, for example, Cas-CLOVER, MAD7, Cas12a (also known as Cpf1), xCas9, SpCas9-NG, etc. Other gene editing technologies such as TALENs, meganucleases, zinc-finger nucleases, transposons, and homologous recombination may also be suitable to make the knockout cell line.

Methods of Using a TGFβ1 Knockout Cell Line

[0063] A TGFβ1 knockout cell line, as described herein may be used to produce a composition that includes a target biologic and a minimal amount TGFβ1. Because the cell media in which the TGFβ1 knockout cell line is grown may include some TGFβ1, it may not be possible to produce a composition that includes no TGFβ1; however, in a preferred embodiment, the composition produced by the TGFβ1 knockout cell line does not include sufficient TGFβ1 to materially affect the activity or action of the target biologic.

[0064] In some aspects, a method for making the composition that includes a target biologic includes culturing a TGFβ1 knockout cell line that overexpresses the target biologic and purifying the target biologic.

[0065] The target biologic may be purified by any suitable means. Exemplary methods of purifying recombinant proteins are discussed in, for example, Wingfield, Curr Protoc Protein Sci. 2015; 80: 6.1.1-6.1.35 and Structural Genomics Consortium et al., Nat Methods 5:135-146 (2008). An exemplary method of purifying a monoclonal antibody is described in Corsiero, Mater. Methods 2016; 6:1481.

[0066] In situations where a composition produced by the TGFβ1 knockout cell line includes some TGFβ1—including when TGFβ1 is added to the cell culture (as, for example, a component of fetal bovine serum)—purifying the target biologic may include using an anti-TGFβ1 antibody, if needed. In an exemplary embodiment, an antibody column including an anti-TGFβ1 antibody may be used to remove TGFβ1 in the composition including, for example, TGFβ1 introduced during cell culture of the TGFβ1 knockout cell line.

[0067] In some embodiments, the TGFβ1 knockout cell line is used to provide a composition that includes less than 0.5 ng/mL TGFβ1 protein, less than 0.1 ng/mL TGFβ1 protein, less than 0.01 ng/mL TGFβ1, less than 0.05 ng/mL (50 pg/mL) TGFβ1 protein, less than 0.02 ng/mL (20 pg/mL) TGFβ1 protein, less than 0.01 ng/mL (10 pg/mL) TGFβ1 protein, less than 0.005 ng/mL (5 pg/mL) TGFβ1 protein, less than 1 pg/mL TGFβ1 protein, less than 0.5 pg/mL (500 fg/mL) TGFβ1 protein, less than 0.01 pg/mL (100 fg/mL) TGFβ1 protein, less than 0.05 pg/mL (50 fg/mL) TGFβ1 protein, less than 0.005 pg/mL (5 fg/mL) TGFβ1 protein, or less than 0.001 pg/mL (1 fg/mL) TGFβ1 protein.

[0068] In some embodiments, the composition includes an undetectable level of TGFβ1. In some embodiments, the composition includes an undetectable level of TGFβ1 as measured using the Quantikine ELISA Kit (Catalog No. #DB100B, R&D Systems, Minneapolis, Minn.). The Quantikine ELISA Kit uses a quantitative sandwich enzyme immunoassay technique to detect natural and recombinant TGFβ1 and exhibits a minimum detectable concentration of human TGF-β1 in a range of 1.7 pg/mL to 15.4 pg/mL (mean 4.61 pg/mL). The minimum detectable concentration may be determined by adding two standard deviations to the mean O.D. value of twenty zero standard replicates and calculating the corresponding concentration.

[0069] In some embodiments, the TGFβ1 knockout cell line is used to provide a composition that includes at least 40 μg/mL, at least 50 μg/mL, at least 100 μg/mL, at least 150 μg/mL, or at least 200 μg/mL of the target biologic. In some embodiments, the composition may include up to 50 μg/mL, up to 100 μg/mL, up to 150 μg/mL, up to 100 μg/mL, up to 300 μg/mL, or up to 400 μg/mL of the target biologic. Exemplary ranges of the target biologic include 40 μg/mL to 400 μg/mL; 100 μg/mL to 300 μg/mL; and 40 μg/mL to 100 μg/mL.

[0070] In some embodiments, the TGFβ1 knockout cell line may be used to provide a composition having an additional feature or features described in the Compositions section of this disclosure.

Methods of Using the Compositions

[0071] In another aspect, this disclosure describes methods of using a composition produced by a TGFβ1 knockout cell line including, for example, a composition having a feature or features as described in the Compositions section of this disclosure.

[0072] In some embodiments, the composition may be used in research. For example, the composition may be used to test the activity of the target biologic. As described in Example 5, a composition produced by a TGFβ1 knockout cell line including Wnt3a as the target biologic may be used in an assay to test the activity of Wnt3a.

[0073] In some embodiments, the composition may be used to treat a subject. As used herein, the term “subject” includes, but is not limited to, humans and non-human vertebrates. In preferred embodiments, a subject is a mammal, particularly a human. A subject may be an “individual,” “patient,” or “host.” Non-human vertebrates include livestock animals, companion animals, and laboratory animals. Non-human subjects also include non-human primates as well as rodents, such as, but not limited to, a rat or a mouse. Non-human subjects also include, without limitation, chickens, horses, cows, pigs, goats, dogs, cats, guinea pigs, hamsters, mink, and rabbits.

[0074] As used herein “treat,” “treating,” or “treatment” can include therapeutic and/or prophylactic treatments. “Treating a disorder,” as used herein, is not intended to be an absolute term. Treatment may lead to an improved prognosis or a reduction in the frequency or severity of symptoms. Desirable effects of treatment include preventing occurrence or recurrence of disease, alleviation of symptoms, diminishment of any direct or indirect pathological consequences of the disease, decreasing the rate of disease progression, amelioration or palliation of the disease state, and remission or improved prognosis. In some circumstances, the frequency and severity of symptoms may be reduced to non-pathological levels. In some circumstances, the symptoms of an individual receiving the compositions of the invention are only 90%, 80%, 70%, 60%, 50%, 40%, 30%, 20%, 10%, 5%, or 1% as frequent or severe as symptoms experienced by an untreated individual with the disorder.

[0075] The precise dosage and duration of treatment may be a function of the disease being treated and may be determined empirically using known testing protocols or by extrapolation from in vivo or in vitro test data. Concentrations and dosage values may also vary with the severity of the condition to be alleviated. For any particular subject, specific dosage regimens may be adjusted over time according to the individual need and the professional judgment of the person administering or supervising the administration of the compositions.

EXEMPLARY METHOD ASPECTS

[0076] A1. A method for producing a target biologic, wherein the method comprises: [0077] culturing a TGFβ1 knockout cell line that overexpresses the target biologic; and [0078] purifying the target biologic.
A2. The method of Aspect A1, wherein the method further comprises introducing a vector comprising a gene encoding the target biologic into the TGFβ1 knockout cell line.
A3. The method of Aspect A1 or A2, wherein the target biologic comprises a recombinant protein.
A4. The method of any one of Aspects A1 to A3, wherein the target biologic comprises a TGFβ superfamily ligand.
A5. The method of any one of Aspects A1 to A4, wherein the target biologic comprises a Wnt, a bone morphogenic protein (BMP), an activin, or a growth differentiation factor (GDF), or a combination thereof.
A6. The method of any one of Aspects A1 to A5, wherein the target biologic comprises Wnt1, Wnt2, Wnt2b, Wnt3, Wnt3a, Wnt4, Wnt5a, Wnt5b, Wnt6, Wnt7a, Wnt7b, Wnt8a, Wnt8b, Wnt9a, Wnt9b, Wnt10a, Wnt10b, Wnt11, or Wnt16, or a combination thereof.
A7. The method of any one of Aspects A1 to A6, wherein the target biologic comprises BMP2, BMP3, BMP4, BMPS, BMP6, BMP7, BMP8a, BMP8b, BMP10, BMP11, or BMP15, or a combination thereof.
A8. The method of any one of Aspects A1 to A7, wherein the target biologic comprises GDF1, GDF3, GDFS, GDF6, GDF8, GDF9, GDF10, GDF11, or GDF15, or a combination thereof.
A9. The method of any one of Aspects A1 to A8, wherein the target biologic is chosen from Wnt3a, Wnt5a, GDF15, BMP4, or BMP2, or a combination thereof.
A10. The method of any one of Aspects A1 to A9, wherein the target biologic comprises an antibody.
A11. The method of Aspect A10, wherein the antibody comprises a monoclonal antibody.
A12. The method of any one of Aspects A1 to A11, wherein the target biologic comprises a biologic drug.
A13. The method of any one of Aspects A1 to A12, wherein the target biologic comprises a mouse protein.
A14. The method of any one of Aspects A1 to A12, wherein the target biologic comprises a human protein.
A15. The method of any one of Aspects A1 to A14, wherein the cell line comprises a Chinese Hamster Ovary (CHO) cell line.
A16. The method of any one of Aspects A1 to A15, wherein the method further comprises exposing the composition to an anti-TGFβ1 antibody.
A17. The method of Aspect A16, wherein an antibody column comprises the anti-TGFβ1 antibody.
A18. A composition comprising the target biologic produced by the method of any one of Aspects A1 to A17.
A19. The composition of Aspect A18, wherein the composition comprises less than 0.5 ng/mL TGFβ1 protein, less than 0.1 ng/mL TGFβ1 protein, less than 0.01 ng/mL TGFβ1, less than 0.05 ng/mL (50 pg/mL) TGFβ1 protein, less than 0.02 ng/mL (20 pg/mL) TGFβ1 protein, less than 0.01 ng/mL (10 pg/mL) TGFβ1 protein, less than 0.005 ng/mL (5 pg/mL) TGFβ1 protein, less than 1 pg/mL TGFβ1 protein, less than 0.5 pg/mL (500 fg/mL) TGFβ1 protein, less than 0.01 pg/mL (100 fg/mL) TGFβ1 protein, less than 0.05 pg/mL (50 fg/mL) TGFβ1 protein, less than 0.005 pg/mL (5 fg/mL) TGFβ1 protein, or less than 0.001 pg/mL (1 fg/mL) TGFβ1 protein.
A20. The composition of Aspect A18 or A19, wherein the composition comprises an undetectable level of TGFβ1.
A21. The composition of Aspect A20, wherein the level of TGFβ1 is measured using the Quantikine ELISA Kit (Catalog No. #DB100B, R&D Systems, Minneapolis, Minn.).
A22. The composition of any one of Aspects A18 to A21, wherein the composition comprises at least 40 μg/mL, at least 50 μg/mL, at least 100 μg/mL, at least 150 μg/mL, or at least 200 μg/mL of the target biologic.
A23. The composition of any one of Aspects A18 to A21, wherein the composition comprises up to 50 μg/mL, up to 100 μg/mL, up to 150 μg/mL, up to 200 μg/mL, up to 300 μg/mL, or up to 400 μg/mL of the target biologic.

EXEMPLARY COMPOSITION ASPECTS

[0079] B1. A composition comprising a target biologic, [0080] wherein the target biologic comprises a TGFβ superfamily ligand, and [0081] wherein the composition comprises less than 0.5 ng/mL TGFβ1 protein, less than 0.1 ng/mL TGFβ1 protein, less than 0.01 ng/mL TGFβ1, less than 0.05 ng/mL (50 pg/mL) TGFβ1 protein, less than 0.02 ng/mL (20 pg/mL) TGFβ1 protein, less than 0.01 ng/mL (10 pg/mL) TGFβ1 protein, less than 0.005 ng/mL (5 pg/mL) TGFβ1 protein, less than 1 pg/mL TGFβ1 protein, less than 0.5 pg/mL (500 fg/mL) TGFβ1 protein, less than 0.01 pg/mL (100 fg/mL) TGFβ1 protein, less than 0.05 pg/mL (50 fg/mL) TGFβ1 protein, less than 0.005 pg/mL (5 fg/mL) TGFβ1 protein, or less than 0.001 pg/mL (1 fg/mL) TGFβ1 protein.
B2. A composition comprising a target biologic, [0082] wherein the target biologic comprises a biologic drug, and [0083] wherein the composition comprises less than 0.5 ng/mL TGFβ1 protein, less than 0.1 ng/mL TGFβ1 protein, less than 0.01 ng/mL TGFβ1, less than 0.05 ng/mL (50 pg/mL) TGFβ1 protein, less than 0.02 ng/mL (20 pg/mL) TGFβ1 protein, less than 0.01 ng/mL (10 pg/mL) TGFβ1 protein, less than 0.005 ng/mL (5 pg/mL) TGFβ1 protein, less than 1 pg/mL TGFβ1 protein, less than 0.5 pg/mL (500 fg/mL) TGFβ1 protein, less than 0.01 pg/mL (100 fg/mL) TGFβ1 protein, less than 0.05 pg/mL (50 fg/mL) TGFβ1 protein, less than 0.005 pg/mL (5 fg/mL) TGFβ1 protein, or less than 0.001 pg/mL (1 fg/mL) TGFβ1 protein.
B3. The composition of Aspect B1 or B2, wherein the composition comprises an undetectable level of TGFβ1.
B4. The composition of Aspect B3, wherein the level of TGFβ1 is measured using the Quantikine ELISA Kit (Catalog No. #DB100B, R&D Systems, Minneapolis, Minn.).
B5. The composition of any one of Aspects B1 to B4, wherein the composition comprises at least 40 μg/mL, at least 50 μg/mL, at least 100 μg/mL, at least 150 μg/mL, or at least 200 μg/mL of the target biologic.
B6. The composition of any one of Aspects B1 to B5, wherein the composition comprises up to 50 μg/mL, up to 100 μg/mL, up to 150 μg/mL, up to 200 μg/mL, up to 300 μg/mL, or up to 400 μg/mL of the target biologic.
B7. The composition of any one of Aspects B1 to B6, wherein the target biologic comprises a recombinant protein.
B8. The composition of any one of Aspects B1 to B7, wherein the target biologic comprises an antibody.
B9. The composition of Aspect B8, wherein the antibody comprises a monoclonal antibody.
B10. The composition of any one of Aspects B1 to B7, wherein the target biologic comprises a Wnt, a bone morphogenic protein (BMP), an activin, or a growth differentiation factor (GDF), or a combination thereof.
B11. The composition of Aspect B10, wherein the target biologic comprises Wnt1, Wnt2, Wnt2b, Wnt3, Wnt3a, Wnt4, Wnt5a, Wnt5b, Wnt6, Wnt7a, Wnt7b, Wnt8a, Wnt8b, Wnt9a, Wnt9b, Wnt10a, Wnt10b, Wnt11, or Wnt16, or a combination thereof.
B12. The composition of Aspect B10, wherein the target biologic comprises BMP2, BMP3, BMP4, BMP5, BMP6, BMP7, BMP8a, BMP8b, BMP10, BMP11, or BMP15, or a combination thereof.
B13. The composition of Aspect B10, wherein the target biologic comprises GDF1, GDF3, GDFS, GDF6, GDF8, GDF9, GDF10, GDF11, or GDF15, or a combination thereof.
B14. The composition of Aspect B10, wherein the target biologic is chosen from Wnt3a, Wnt5a, GDF15, BMP4, or BMP2, or a combination thereof.
B15. The composition of any one of Aspects B1 to B14, wherein the target biologic comprises a mouse protein.
B16. The composition of any one of Aspects B1 to B14, wherein the target biologic comprises a human protein.
B17. The composition of any one of Aspects B1 to B16, wherein the target biologic comprises a biologic drug.
B18. The composition of any one of Aspects B1 to B17, wherein the composition comprises a pharmaceutical composition.

EXEMPLARY TGFβ1 KNOCKOUT CELL LINE & METHODS OF MAKING ASPECTS

[0084] C1. A TGFβ1 knockout cell line, wherein the TGFβ1 knockout cell line comprises a mutation or deletion in the nucleotides encoding TGFβ1.
C2. The TGFβ1 knockout cell line of Aspect C1, wherein the TGFβ1 knockout cell line comprises a Chinese Hamster Ovary (CHO) cell line or a human embryonic kidney (HEK) cell line.
C3. The TGFβ1 knockout cell line of Aspect C1 or C2, wherein the TGFβ1 knockout cell line comprises a stop codon introduced into the nucleotides encoding TGFβ1.
C4. The TGFβ1 knockout cell line of any one of Aspects C1 to C3, wherein the TGFβ1 knockout cell line overexpresses a target biologic.
C5. The TGFβ1 knockout cell line of Aspect C4, wherein the target biologic comprises a TGFβ superfamily ligand, a recombinant protein, a ribonucleoprotein, a non-peptide, a glycoprotein, a glycolipid, a subcellular organelle, a blood component, a biologic drug, or a combination thereof.
C6. A method of making the TGFβ1 knockout cell line of any one of Aspects C1 to C5.
C7. The method of Aspect C6, wherein the method comprises CRISPR-based gene editing.
C8. The method of Aspect C7, wherein the method comprises CRISPR-Cas9-based gene editing.
C9. The method of Aspect C7 or C8, wherein the method comprises CRISPR-based gene editing of the nucleotides encoding TGFβ1.
C10. The method of any one of Aspects C6 to C9, wherein the method comprises introducing a vector comprising a gene encoding a target biologic into the TGFβ1 knockout cell line

[0085] The present invention is illustrated by the following examples. It is to be understood that the particular examples, materials, amounts, and procedures are to be interpreted broadly in accordance with the scope and spirit of the invention as set forth herein.

EXAMPLES

[0086] All reagents, starting materials, and solvents used in the following examples were purchased from commercial suppliers (such as Sigma Aldrich, St. Louis, Mo.) and were used without further purification unless otherwise indicated.

Example 1

[0087] This Example describes the development and characterization of a TGFβ1 knock out cell line.

[0088] CRISPR/Cas9 gene editing technology was used to perform directed gene modifications in a host Chinese Hamster Ovary (CHO) cell line. Specific guide RNAs were selected to direct the nuclease activity of the Cas9 enzyme to target known sequences in the host cell genome, “knocking out” the functionality of the CHO TGFβ1 gene, rendering the cells incompetent of producing this protein.

[0089] As further described below, the inability of the cells to produce TGFβ1 was confirmed by genomic sequence analysis of individual clones following the gene editing process. Further confirmation was obtained by screening for the presence of TGFβ1 with a Quantikine ELISA kit (R&D Systems, Minneapolis, Minn.).

Generation of TGFβ1 Knockout (KO) Cell Lines

[0090] Knockout of TGFβ1 was performed by B-MoGen Biotechnologies, Inc. (Minneapolis, Minn.) using a gRNA (5′-CAAGACCATCGACATGGAGC-3′) from Synthego Corporation (Menlo Park, Calif.). Clones were isolated and tested for TGFβ1 and TGFβ2 with a Quantikine ELISA kit (R&D Systems, Minneapolis, Minn.). Clones with the lowest level of TGFβ1 were re-isolated. The target region of the target clones was then amplified using PCR, and a clone exhibiting a stop codon in the target region (referred to herein as CHO-S TGFβ1 KO) was selected for further testing.

Genomic Sequence Analysis

[0091] Preparations of CHO-S wildtype (wt) and CHO-S TGFβ1 knockout (KO) genomic DNA were made using Quick-DNA Microprep Kit (Catalog No. D3020, Zymo Research, Irvine, Calif.). PCR was performed on genomic DNA samples to amplify region of interest using GoTaq G2 Hot Start (Catalog No. M7422, Promega, Madison, Wis.).

PCR Primers:

[0092]

TABLE-US-00001 Forward: (SEQ ID NO: 4) 5′-GCTCCCCTATTTAAGAACAC-3′ Reverse: (SEQ ID NO: 5) 5′-GCTCTGCCGGTGGTTTCCTC-3′

PCR Protocol:

[0093] Step 1: 95° C. for 2 minutes

[0094] Step 2: 95° C. for 30 seconds

[0095] Step 3: 57° C. for 30 seconds

[0096] Step 4: 72° C. for 40 seconds

[0097] Repeat Step 2 to Step 4 35 times

[0098] Step 5: 72° C. for 2 minutes

[0099] The PCR product was run on a 1% agarose gel and an image was taken on the transilluminator visualizing ethidium bromide staining. Results are shown in FIG. 1 and indicate the introduction of a genetic mutation in the TGFβ1 genomic sequence of the CHO-S TGFβ1 KO cells.

[0100] Bands from PCR product of the CHO-S TGFβ1 KO were excised from the gel, purified, and sequenced (Sequencing Primer: 5′-TTCAGGGCTCTCTCCTAACC-3′ (SEQ ID NO:6)). Results are shown in FIG. 2. Sequence analysis confirmed a 64-base pair (bp) insert near the beginning of the TFGβ1 gene, resulting in a stop codon at amino acid 51 introduced by INDELs by the error prone non-homologous end joining (NHEJ) double strand repair mechanism.

TGFβ1 ELISA Data

[0101] CHO-S cells, CHO-S TGFβ1 KO cells, CHO-S TGFβ1 KO cells expressing recombinant hGDF-15, and CHO-S TGFβ1 KO cells expressing recombinant hTGFB1 were plated into EX-CELL ACR CHO Medium (Catalog No. C5467, Sigma Aldrich, St. Louis, Mo.) containing 4 mM L-Glutamine, 1% Pen Strep, and 2.5 mM NaBr and placed in a 33° C. incubator.

[0102] HEK293EBNA cells and HEK293EBNA expressing recombinant hGDF-15 were plated into EX-CELL ACR CHO Medium (Catalog No. C5467, Sigma Aldrich, St. Louis, Mo.) containing 4 mM L-Glutamine and 1% Pen Strep placed in a 37° C. incubator.

[0103] Conditioned media was collected on day 3 and analyzed using a Human TGFβ1 Quantikine ELISA Kit (Catalog No. #DB100B, R&D Systems, Minneapolis, Minn.).

[0104] Results are shown in FIG. 3. Detectable amounts of TGFβ1 were present in the conditioned media of wild type CHO-S wt cells, HEK wt cells, and HEK293 expressing recombinant hGDF-15 cells. (FIG. 3 and Table 1.) The amount of TGFβ1 present in the conditioned media of CHO-S TGFβ1 KO samples and CHO-S TGFβ1 KO cells expressing hGDF-15 was below the detectable range. (FIG. 3 and Table 1.) CHO-S TGFβ1 KO cells expressing recombinant hTGFB1 had significantly higher levels of TGFB1 than the parental cell line (Table 1).

TABLE-US-00002 TABLE 1 Sample ng/mL TGFβ1 CHO-S wt 1.4850 HEK293 wt 0.3193 HEK293 wt + hGDF-15 0.2260 CHO-S TGFβ1 KO * CHO-S TGFβ1 KO + hGDF-15 * CHO-S TGFβ1 KO + hTGFβ1 1216.1796 *Below detection limit

Example 2

[0105] This Example describes the effect of TGFβ1 contamination on the activation of Smad2 by GDF15 in the DU145 human prostate cancer cell line.

[0106] DU145 cells were serum starved for 2 hours prior to adding a protein treatment (with or without TGFβ1 blocking antibody) for 1 hour followed by addition of the hGDF15 protein to the cells for 1 additional hour. The cells were then lysed and analyzing by Western Blot analysis. The antibodies used were a rabbit anti-phospho Smad2 antibody (Cell Signaling Technology, Danvers, Mass.) and a mouse anti-hELF-4E antibody (R&D Systems, Minneapolis, Minn.) (used as a loading control).

[0107] Results are shown in FIG. 4A. Smad2 phosphorylation was detected at a TGFβ1 contamination level of 3.6 pg/mL of media. The contents of each lane of FIG. 4A are described in Table 2A.

[0108] Additional results are also shown in FIG. 4B. The contents of each lane of FIG. 4B are described in Table 2B. Smad2 phosphorylation was detected in DU145 cells treated with 2 μg/ml of recombinant human GDF-15 purified from CHO-S wild type cells but not when DU145 cells were treated with 2 μg/ml of recombinant GDF-15 purified from the CHO-S TGFβ1 knockout CHO-S cell line.

TABLE-US-00003 TABLE 2A TGFβ1 Blocking Protein Levels TGFβ1 in Sample Antibody (pg/μg of Added Lane Protein and Dose protein sample) (1 μg/ml) 2 hGDF15 Lot 2 (HEK293) 500 ng/ml Unknown No 3 hGDF15 Lot 2 (HEK293) 250 ng/ml Unknown No 4 hGDF15 Lot 21 (HEK293) 500 ng/ml 14.31 No 5 hGDF15 Lot 21 (HEK293) 250 ng/ml 14.31 No 6 hGDF15 Lot 11 (HEK293) 500 ng/ml 35.00 No 7 hGDF15 Lot 11 (HEK293) 250 ng/ml 35.00 No 8 Untreated Control 0 No 9 Untreated Control 0 No 10 hGDF15 Lot 11 (HEK293) 250 ng/ml 35.00 Yes 11 hGDF15 Lot 11 (HEK293) 500 ng/ml 35.00 Yes 12 hGDF15 Lot 21 (HEK293) 250 ng/ml 14.31 Yes 13 hGDF15 Lot 21 (HEK293) 500 ng/ml 14.31 Yes 14 hGDF15 Lot 2 (HEK293) 250 ng/ml Unknown Yes 15 hGDF15 Lot 2 (HEK293) 500 ng/ml Unknown Yes 16 hTGFβ1 (CHO) 10 ng/ml (10 ng/ml) Yes 17 hTGFβ1 (CHO) 10 ng/ml (10 ng/ml) No

TABLE-US-00004 TABLE 2B Lane Contents 1 Untreated, blank 2 rhTGFβ1 at 1 ng/mL 3 rhTGFβ1 at 1 ng/mL + Ch × hTGFβ1 4 rhGDF-15 (CHO-s TGFβ1 KO) 5 rhGDF-15 (CHO-s TGFβ1 KO) + Ch × hTGFβ1 6 rhGDF-15 (HEK293EBNA) 7 rhGDF-15 (HEK293EBNA) + Ch × hTGFβ1 8 rhGDF-15 (CHO-s) 9 rhGDF-15 (CHO-s) + Ch × hTGFβ1

Example 3

[0109] This Example describes the effect of TGFβ1 contamination on osteoblast differentiation by Wnt-3a.

[0110] Recombinant Mouse Wnt-3a protein from a lot known to have a relatively high level of TGF1 contamination (78 pg of TGFβ1 per microgram (ug) of Wnt-3a protein) resulted in less differentiation of MC3T3-E1 preosteoblasts toward osteoblasts at higher concentrations of Wnt-3a (circles in FIG. 5). When the same dose of mouse Wnt-3a protein was added to MC3T3-E1 preosteoblasts in the presence of a saturating dose (50 ug/mL) of a chicken anti-TGFβ 1 blocking antibody (Catalog No. AF-101-NA, R&D Systems, Minneapolis, Minn.), a rescue of osteoblast differentiation was observed with alkaline phosphatase activity rebounding to the predicted plateau. Results are shown in FIG. 5.

[0111] These results demonstrate that neutralizing TGFβ1 in purified mouse Wnt-3a protein removes a negative regulator of osteoblast differentiation.

Example 4

[0112] This Example describes the effect of TGFβ1 on a Wnt-3a-mediated increase in alkaline phosphatase (ALP) expression.

[0113] MC3T3/E1 osteoblast cells were seeded overnight at 1×10.sup.4 cells/well. The next day, media from was removed (leaving the MC3T3/E1 osteoblast cells on the plate), and new media including Wnt-3a or Wnt-3a and TGFβ1 was added. The cells were incubated for 3 additional days in the presence of Wnt-3a or Wnt-3a and TGFβ1, and alkaline phosphatase (ALP) activity was measured (Pacifici et al., Exp. Cell Res. 1991; 195, 38-46). Results are shown in FIG. 6.

[0114] MC3T3-E1 preosteoblasts treated with recombinant human Wnt-3a protein alone exhibited dose-responsive induction of osteoblast differentiation, signified by increasing ALP activity three days after addition of the Wnt-3a protein. (FIG. 6, circles).

[0115] When a constant dose of 20 ng/mL of recombinant human Wnt-3a protein was added to MC3T3-E1 cells for three days, a relatively high level of ALP activity resulted. (FIG. 6, flat line and triangles.) If the same steady dose of 20 ng/mL of recombinant human Wnt-3a protein was added to cells and a dose titration of recombinant human TGFβ1 was added (FIG. 6, squares), inhibition of Wnt-3a-mediated osteoblast differentiation was observed at a neutralizing dose of 50% (ND50) of 53.3 fg/mL TGFβ1.

Example 5

[0116] This Example describes the effect of TGFβ1 contamination of human Wnt3a (hWnt3a) purified from CHO wt and CHO TGFβ1 KO cell lines in assays designed to measure the effects of hWnt3a.

[0117] A CHO TGFβ1 KO cell line was prepared as described in Example 1. All CHO-S or CHO-S TGF1 KO lines expressing hWnt3a were made by transfection of a DNA expression plasmid expressing hWnt3a and selected with puromycin to select for clonal stably integrated cell lines overexpressing hWnt3a. Recombinant human Wnt3a purified from CHO-S wt cells stably expressing hWnt3a shows relatively high levels of TGFβ1 protein compared to reduced levels of TGFβ1 protein observed in hWnt3a isolated from a TGFβ1 knockout CHO-S cell line that stably overexpresses hWnt3a (CHO-S TGFβ1 KO+Wnt3a). Results are shown in Table 3. Lot #RSK69 had 165 pg TGFβ1 protein per μg of hWnt3a protein and lot #RSK51 had 114.6 pg TGFβ1 protein per μg of hWnt3a protein. In contrast, hWnt3a protein purified from the CHO-S TGFβ1 KO line (lot #DLGC02) had 4.97 pg TGFβ1 protein per μg of hWnt3a protein. The low level of TGFβ1 protein in the hWnt3a protein purified from the TGFβ1 KO line is due to the use of 2% fetal bovine serum (FBS) included in the cell culture media; including FBS is necessary to culture the cells which produce active recombinant Wnt proteins. The source of this TGFβ1 protein is not cell-derived but from serum used during cell culture.

TABLE-US-00005 TABLE 3 TGFβ1 Levels (per microgram Cell Line Protein Lot # hWant3a) CHO-S wt + hWnt3a hWnt3a RSK69 165 pg CHO-S wt + hWnt3a hWnt3a RSK51 114.6 pg CHO-S TGFβ1 KO + hWnt3a hWnt3a DLGC02 4.97 pg

[0118] The hWnt3a protein from lots purified from wild type CHO-S cells showed inhibitory effects at the highest levels of hWnt3a treatment in a MC3T3-E1 osteoblast differentiation assay (FIG. 7 & FIG. 9). Reduced osteoinductive activity was observed at higher doses of hWnt3a (1.66 μg/ml, 0.556 μg/ml, and 0.185 μg/ml) purified from wild type CHO-S cells but was not observed when treating the MC3T3-E1 cells with hWnt3a derived from CHO-S TGFβ1 KO cells (FIG. 7 & FIG. 9). The “dip” in activity at higher doses of Wnt3a protein is due to TGFβ1 contamination and can be reversed with a TGFβ1 function blocking antibody (see, e.g., FIG. 5 & FIG. 9). Moreover, as shown in FIG. 9, hWnt3a protein with high levels of TGFβ1 contamination results in lower osteo-inductive activity at high hWnt3a doses in a MC3T3-E1 assay than hWnt3a protein purified from TGFβ1 KO CHO-S cells or hWnt3a from CHO-S TGFβ1 KO cells.

[0119] The recombinant hWnt3a purified from CHO-S TGFβ1 KO cells showed substantially lower levels of TGFβ1 compared to hWnt3a purified from CHO-S wt cells (see Table 3). This low level of TGFβ1 contamination in recombinant hWnt3a purified from CHO-S TGFβ1 KO cells arises from the 2% serum that is necessary during cell culture to produce active Wnt proteins. Addition of the TGFβ1 blocking antibody rescued the inhibitory effects of the lower level TGFβ1 contamination arising from the serum in CHO-S TGFβ1 KO cells expressing hWnt3a (FIG. 9). These data demonstrate that hWnt3a purified from CHO-S TGFβ1 KO cells contains significantly less TGFβ1 contamination compared to hWnt3a purified from CHO-S wt cells expressing hWnt3a. In addition, these data further demonstrate that the inhibitory effects on MC3T3-E1 osteoblast differentiation correlates with relative amounts of TGFβ1 contamination in purified hWnt3a proteins.

[0120] Interestingly, TGFβ1 contamination did not affect all assay systems. For example, no major functional differences between hWnt3a proteins purified from CHO-S wt and CHO-S TGFβ1 KO cells were detected in in a HEK293 Wnt reporter assay, as further described below.

[0121] A stable cell line was generated to incorporate nine T cell factor (TCF) binding elements upstream of a SEAP reporter gene (Korinek et al. Science 275, 1784-1787 (1997)). Wnt proteins bind to Wnt receptors on the cell surface of this HEK293 cell line resulting in a cascade of intracellular events culminating in β-catenin binding to TCF transcription factors resulting in production of the reporter gene SEAP. SEAP is secreted into the media and SEAP activity assays provide an indication of Wnt pathway activity. Recombinant TGFβ1 protein has been tested in this Wnt reporter assay previously with no activation of this Wnt specific assay.

[0122] Results are shown in FIG. 8. These results were not surprising; titrations of TGFβ1 protein in this HEK293 Wnt reporter have not been shown to have any effect in this Wnt reporter system. These data demonstrate that TGFβ1 contamination is cell- and assay-context dependent, and not all biological activity assays will show sensitivity to TGFβ1 contamination.

[0123] The complete disclosure of all patents, patent applications, and publications, and electronically available material (including, for instance, nucleotide sequence submissions in, e.g., GenBank and RefSeq, and amino acid sequence submissions in, e.g., SwissProt, PIR, PRF, PDB, and translations from annotated coding regions in GenBank and RefSeq) cited herein are incorporated by reference. In the event that any inconsistency exists between the disclosure of the present application and the disclosure(s) of any document incorporated herein by reference, the disclosure of the present application shall govern. The foregoing detailed description and examples have been given for clarity of understanding only. No unnecessary limitations are to be understood therefrom. The invention is not limited to the exact details shown and described, for variations obvious to one skilled in the art will be included within the invention defined by the claims.