WO1994016720A1

WO1994016720A1 - COMPOSITIONS CONTAINING TRANSFORMING GROWTH FACTOR β STABILIZED BY GLYCOSAMINOGLYCANS

Info

Publication number: WO1994016720A1
Application number: PCT/US1994/000490
Authority: WO
Inventors: George Ksander; Ramon A. Burns; Veronica S. De Guzman; Edwin S. Kamemoto; Julie Pon; Ranga Nathan; Audrey A. Chow
Original assignee: Celtrix Pharmaceuticals, Inc.
Priority date: 1993-01-26
Filing date: 1994-01-21
Publication date: 1994-08-04
Also published as: AU6028094A

Abstract

This invention provides a method for stabilizing the activity of a TGF-β by adding a sufficient amount of heparin to the TGF-β.

Description

COMPOSITIONS CONTAINING TRANSFORMING GROWTH FACTOR

BETA STABILIZED BY GLYCOSAMINOGLYCANS

Field of the Invention

This invention relates to stabilizing the structure and activity of TGF-βs and related molecules by addition of heparin.

Background of the Invention

It has previously been disclosed that heparin is able to stabilize fibroblast growth factors (FGF)

(U.S. Patent No. 5,100,668).

Fuller et al. showed that heparin, which increases osteoporosis, modifies osteoclast resorption- stimulating activity; however, the heparin-modulating activity was lost if the serum was stored at 4°C for seven days before addition to the osteoclasts. (1991) J.

Cell. Physiol., 147:208-214.

La Marre et al. found that TGF-β2 is less dissociable from α2-macroglobulin (α2-M) than TGF-β1 by heparin. However, the in vivo form of α2-M-plasmin-TGF-β is not dissociable by 100 μg/ml heparin. (1991) Biochim.

Biophys. Acta, 1091:197-204. α2-M is a cellular

structural protein that also binds to basic fibroblast growth factor (bFGF), platelet-derived growth factor (PDGF), nerve growth factor (NGF), interleukin-1β (IL-1β) and interleukin-6 (IL-6).

Mooradian et al. found that heparin does not alter the binding of labeled TGF-01 to fibronectin. The heparin was tested at a concentration range of 0.01 to 100 μg/ml. (1989) J. Cell. Biochem., 41:189-200. Rosengart et al. found that heparin protects heparin binding growth factor-1 (HBGF) from proteolytic inactivation, thus potentiating the activity of HBGF. However, heat denaturation of the HBGF destroyed the ability of heparin to protect against proteolytic

inactivation. (1988) Biochem. Biophys. Res. Comm.,

152:432-440.

Jaye et al. found that heparin binds to

α-endothelial cell growth factors (α-ECGF) and changes their conformation, as measured by change in intrinsic fluorescence. Heparin was also found to potentiate the mitogenic activity of α-ECGF. (1987) J. Biol. Chem., 262:16612-16617.

Gospodarowicz and Cheng found that 50 μg/ml heparin potentiated the activity of acidic FGF (aFGF) and 5 μg/ml heparin potentiated the activity of bFGF.

Heparin was also found to protect the proteins from inactivation at 200 and 20 μg/ml, respectively. (1986) J. Cell. Phys., 128:475-484.

Schreiber et al. found that 50 μg/ml heparin restores activity to denatured ECGF. (1985) Proc. Natl. Acad. Sci. USA, 82:6138-6142.

Experiments have shown that an interaction between TGF-β and heparin may be a physiologic regulator of TGF-β activity and may potentiate the activity of TGF- β . McCaffrey et al. (1989), J. Cell. Biol., 1989:441- 448. This interaction relies on the ability of heparin to dissociate TGF-β from the TGF-β/α2-M complex,

permitting TGF-β to interact with cell surface receptors. McCaffrey et al. found that 100 μg/ml heparin increased binding of ¹²⁵I-TGF-β to cells, increasing the inhibitory action of TGF-β at concentrations of TGF-β near its apparent K_D. It has also been shown that TGF-β is capable of binding to heparin via specific regions of the TGF-β molecule. McCaffrey et al. (1992), J. Cell.

Physiol., 152:430-440.

These studies did not necessarily imply that heparin binding stabilizes TGF-βs.

When bound to α2-M, TGF-β is rendered unable to affect the proliferation of smooth muscle cells.

McCaffrey et al. (1989). TGF-β is inactive when bound to the glycoprotein decorin (a chondroitin-dermatan

sulfate). Yamaguchi et al. (1990), Nature, 346:281-284.

Although TGF-βs are effective in treating a wide variety of disorders, a drawback to their commercial use is their extreme lability, requiring them to be stored in a lyophilized form or, if in solution, at

-70°C. It would therefore be highly desirable to

discover means to effectively stabilize TGF-βs, e.g., in pharmaceutical formulations.

Brief Description of the Invention

This invention provides a method for stabilizing the activity of TGF-β by adding heparin to the TGF-β in an amount sufficient to stabilize the activity.

The invention further provides a composition containing TGF-β and amount of heparin sufficient to stabilize the activity of the TGF-β.

Detailed Description of the Figures

Figure 1 is a graph depicting heparin stabilization of recombinant TGF-β2 in a collagen slurry. Varying concentrations of heparin were tested with either 0.6 μg/ml or 1.2 μg/ml TGF-β2. TGF-β2 activity was tested approximately once a week for forty-two days. The concentrations of heparin/TGF-β2 tested were as follows: 100 μg/ml heparin/1.2 μg/ml TGF-β2 (■); 10 μg/ml

heparin/1.2 μg/ml TGF-β2 (●); 0 μg/ml heparin/1.2 μg/ml TGF-β2 (◊); 100 μg/ml heparin/0.6 μg/ml TGF-β2 (□); 10 μg/ml heparin/0.6 μg/ml TGF-β2 (O); and 0 μg/ml

heparin/0.6 μg/ml TGF-β2 (◊). Figure 1 is discussed in Example 2.

Figure 2 is a graph depicting the effectiveness of heparin in stabilizing the activity of recombinant TGF-β2 in place of human serum albumin (HSA). Varying concentrations of HSA and combined HSA/heparin were tested with TGF-β2 approximately once a week for forty- two days. The concentrations of HSA and heparin tested were as follows: 2% HSA (O) ; 0.05% HSA (□); and 0.05% HSA + 100 μg/ml heparin (■). Figure 2 is discussed in Example 3. Detailed Description of the Invention

Heparin belongs to a heterogenous group of straight-chain anionic mucopolysaccharides called

glycosaminoglycans. Heparin is composed of a mixture of repeating disaccharide units of glycosamino (GlcN) and an uronic acid which can be either L-iduronic acid or D- glucuronic acid.

Heparin is strongly acidic because of its covalently linked sulfate and carboxylie acid groups. In heparin sodium, the acidic protons of the sulfate units are partially replaced by sodium ions.

A number of proteins that exhibit biological activity (biological response modifiers) have been found to bind to heparin. This fact has been exploited to effect separations of proteins on the basis of their differing affinities for heparin. Hjelmeland and Harvey (1988), Birth Defects: Original Article Series, 24: 87- 102; and Hauschka et al. (1986) J. Biol. Chem.,

261:12665-12674.

The family of peptides known as TGF-βs can both regulate cell growth and differentiation. These polypeptides stimulate and inhibit cell proliferation in a manner depending largely on the cell type. TGFs of some type have been found in almost all tissues from all species of animals examined to date.

TGF-β2 is a well-characterized polypeptide. It has a molecular weight of about 25,000 D and is a dimer composed of two identical subunits of 12,500 D, each linked by a disulfide bond. Cheifetz et al. (1987) Cell, 48:408-415; and Ikeda et al. (1987) Biochemistry.

26: 2406-2410. TGFs are synthesized as high molecular weight precursors which must be processed in order to attain biological activity. During purification TGF processing is generally carried out by exposing the precursor proteins to an acidic pH. Molecules remaining unprocessed must be further purified away from the processed molecules.

In vivo, TGF-β2 has been isolated from bovine demineralized bone, porcine platelets and the human prostatic adenocarcinoma cell line, PC-3. Seyedin et al. (1987) J. Biol. Chem., 262:1946-1949; Cheifetz et al.

(1987); and Ikeda et al. (1987). Cloned TGF-β2 has been expressed in a variety of cell types. Madisen et al.

(1989) DNA, 8: 205-212. One of the most commonly used recombinant expression systems is the Chinese hamster ovary (CHO) cell line in which the gene encoding TGF-β2 has been amplified, leading to overexpression of the gene and the secretion of large amounts of precursor TGF-β2. Madisen et al. (1990) Growth Factors, 3: 129-138.

-70°C.

It has now been found that heparin is capable of preserving activity of TGFβs for more than a month when stored in solution at 4°C. Without heparin, the TGFβs would be inactive with 36 hours under these conditions. The addition of sodium heparin to

preparations containing a TGF-β2 serves to maintain the structure and activity of TGF-β2 as assayed by ELISA and cell proliferation assays.

The concentration of heparin suitable to retain TGF-β activity is in the range of about 0.1 to 170 μg per μg of TGF-β. More preferably the concentration of heparin suitable to retain TGF-β activity is in the range of about 10 to 170 μg per μg of TGF-β. Most preferably, the concentration of heparin suitable to retain TGF-β activity is in the range of about 80 to 170 μg per μg of TGF-β. The upper range of 170 μg TGF-β is not limiting as no inhibiting effects of excess heparin have been observed. Therefore the heparin concentration may be raised without a detrimental affect on TGF-β activity, although the highest heparin concentration which may be employed will be determined by other factors, e.g., the pharmacological activity of heparin in vivo.

Heparin belongs to a family of sulfated

mucopolysaccharides called glycosaminoglycans. These include heparin, heparan, heparin sulfate, hyaluronic acid, hexuronylhexoseaminoglycan sulfate, chondroitin sulfate, dermatan sulfate, dextran sulfate, keratan sulfate, and cyclodextrin tetradecasulfate. Other glycosaminoglycans or synthetic polyanionic polymers or fragments of the natural polysaccharides will also be suitable stabilizing TGFβs. The use of the term

"heparin" herein includes other sulfated

mucopolysaccharides. The use of the term "TGF-β" herein includes other forms of TGF-β (e.g., TGF-β-1 to -5 and a heterodimers of the various TGF-βs or of TGF-β1 and TGF- β2). Also included are other members of the TGF-β superfamily, including bone morphogenetic proteins (BMPs), activins, inhibins, the decapentaplegic gene product of Drosophila, and other related proteins.

All patents, patent publications and any publications cited herein, whether supra or infra , are hereby incorporated by reference in their entirety.

Example 1

Effect of Heparin on TGF-β2 Stability in Collagen Slurry TGF-β2 was formulated in a collagen slurry at a concentration of 6.6 μg/ml. Collagen fibrils were precipitated from an acidic solution (0.01 N HCl) containing soluble collagen (J.T. Baker Chemical Co.) by the addition of 0.2M sodium phosphate buffer. The fibrils were isolated by centrifugation, homogenized and diluted with sodium heparin (purified from porcine intestinal mucosa) buffer. The collagen/heparin slurry mixture was combined with a recombinant TGF-β2 working solution containing human serum albumin (HSA) to produce a formulation having a final volume of 60 ml containing: collagen slurry at 7.5 μg/ml, sodium heparin at the concentration indicated, HSA at 0.05% and TGF-β2 at 6.2 μg/ml. Sodium heparin (USP grade, Hepar Indust.) was added to the slurry at concentrations of 0, 3, 10, 30 and 100 μg/ml. TGF-β2 activity was assayed by ELISA after 2 days or 35 days of storage at 4°C.

In general, an ELISA is an immunochemical assay that takes advantage of the specific binding of an antibody to its antigen, as well as the technology for covalently coupling enzymes to antibodies to provide convenient markers. There are two known causes of TGF-β2 instability in formulations, adsorption to surfaces and aggregation (resulting in insolubility). ELISAs are useful for assessing both TGF-β2 instability due to both causes: the ELISA does not detect TGF-β2 which adsorbed to a surface or aggregated, effectively reducing the amount of TGF-β2 detected by the assay.

The TGF-β2 ELISA is performed as described by J. Dasch et al. (1990) Ann. N.Y. Acad. Sci., 593:303-305. with the following modifications. A microtiter plate was first coated with an anti-TGFβ2 monoclonal antibody

(1D11.16, Celtrix Pharmaceuticals). The test sample was then added and the binding of the TGF-β2 to the

monoclonal resulted in immobilization of TGF-β2 on the plate. The plate was then rinsed to remove unbound TGF- β2 , and a second anti-TGF-β2 monoclonal antibody (3C7.14, Celtrix Laboratories) was added, followed by addition of an antibody-peroxidase conjugate (rat anti-mouse IgG2b, Zymed Labs). The resulting immobilized complex was assayed spectrophotometrically following addition of ABTS substrate (Kirkegaard and Perry Laboratories, Inc.). The level of sensitivity is 0.5 ng/ml TGF-β2. A standard curve was generated over a TGF-β2 concentration range of 0.5-100 ng/ml. The concentration of TGF-β2 in test samples, appropriately diluted, was determined by

interpolation from the standard curve.

The sample size for performing the ELISA assay was as follows: 33.3 μl of the final slurry formulation was combined with 966.7 μl of the ELISA sample buffer. The final concentration of TGF-β2 in the assay system was 200 ng/ml.

The TGF-β2 activity/recovery was assayed by ELISA, as described above. The sample for the day 35 assay was 114 μl TGF-β2 (assuming initial concentration) and was diluted to 50 ng/ml. The results of this assay are summarized in Table 1. Table 1

Heparin Stabilizes rTGF-β2 in a Collagen Slurry

These data indicate that at both time points little or no TGF-β2 activity was detected in the collagen slurry without heparin, while essentially complete (95%) recovery was obtained when 100 μg/ml heparin was added. Intermediate levels of activity were measured in the presence of heparin at concentrations between 0 and 100 μg/ml. In addition, the rate of loss of rTGF-β2 activity was slower in proportion to the amount of heparin present.

Example 2

Linear Relationship of Heparin/TGF-β2 Concentration The results obtained in Example 1 were confirmed by a second experiment in which heparin was present at concentrations of 0, 10 or 100 μg/ml, rTGF-β2 was present at concentrations of 6 μg/ml or 1.2 μg/ml, and the preparations were stored at 4°C for up to 42 days. Activity was measured and determined by ELISA assay as described in Example 1. The sample size and concentration are the same as described in Example 1. The final collagen slurry concentration was 7.5 mg/ml. The results obtained are depicted in Figure 1. Under these conditions, heparin at 100 μg/ml preserved the activity of the two concentrations of rTGF-β2 up to 42 days. Heparin at the lower

concentration of 10 μg/ml stabilized 0.6 μg/ml rTGF-β2 but not 1.2 μg/ml rTGF-β2, demonstrating that higher concentrations of heparin are required to preserve the activity of higher concentrations of rTGF-β.

Example 3

Comparison of Stabilizing Activity of Heparin

Compared to that of Human Serum Albumin

Human serum albumin (HSA) is frequently used to stabilize proteins. Heparin was tested for its ability to replace HSA as a stabilizer for TGF-β2. The results obtained in the following example show that heparin is not only able to replace HSA in stabilizing TGF-β2 but results in greater stability for a longer period of time. The results are depicted in Figure 2. A first collagen slurry was prepared as described in Example 1, containing 6.2 μg/ml TGF-β2 and 2% or 0.05% human serum albumin (HSA). In addition a second collagen slurry was prepared as described in Example 1 with 0.05% HSA and 100 μg/ml heparin.

The results obtained show that after storage at

4°C for 2 or 5 days, no TGF-β2 activity could be detected in the slurry containing only 0.05% HSA, while full activity was present in the slurry containing 2% HSA. Thus, HSA stabilizes TGF-β2 activity in a collagen slurry for a limited time. However, when 100 μg/ml heparin was present along with 0.05% HSA, full TGF-β2 activity was recovered. Thus, heparin is able to substitute for HSA as a stabilizer of TGF-β2. Further, the addition of heparin resulted in increased stability of TGF-β2 for at least 6 weeks, whereas even a large amount, 2%, of HSA was not effective in preserving stability longer than 10 days.

Example 4

Effect of Heparin on TGF-β2 Stability

in Aqueous Formulations

In Example 1, the TGF-β2 solution was incubated with 20 mM NaPO₄, pH 7, for 2 hours at room temperature. Under these conditions TGF-β2 lost all activity in the absence of heparin, as shown in Table 2, but retained full activity in the presence of heparin.

TGF-β2 was incubated with 20 mM NaPO₄, 0.25% HSA, pH 7, for 3 hours at 37°C then for 3 days at 4°C. The results are shown in Table 2. Under these conditions TGF-β2 also lost all activity in the absence of heparin but retained full activity in the presence of effective concentrations of heparin. Thus, the ability of heparin to stabilize TGF-β in non-collagenous solutions occurs over a wide range of TGF-β concentrations and in the absence or presence of excipients such as HSA.

These results indicate that heparin is also able to stabilize the activity of TGF-β in aqueous solutions not containing collagen.

Table 2

Heparin Stabilized rTGF-β in

Buffered Aqueous Solutions

Example 5

Effect of Heparin on Biological Activity of TGF-β2 The following example demonstrates that heparin is able to stabilize the biological activity of TGF-β in addition to increasing detectability in ELISA assays.

55 μg of rTGF-β2 were incubated in the absence or presence of 100 μg/ml heparin for 3 hours at room temperature at pH 7 in phosphate-buffered saline (PBS, 20 mM NaPO₄, 130 mM NaCl). At the end of the incubation, 10 μl samples were assayed for TGF-β2 activity using the mink lung cell (MvlLu) antiproliferation cell culture assay according to the method described by Ogawa et al. (1991) Meth. Enzymol. 198: 317-327. In this assay, the ability of TGF-β2 to inhibit cell proliferation is determined by measuring acid phosphatase activity. A small ED₅₀ reflects high activity. TGF-β2 stock solution of 100 μg/ml was also assayed without incubation as a control. Both standard and samples were diluted to 9.9 μg/ml, added to the assay plate and subject to further dilution in the assay plate. The concentration range is 0.0015-3.3 ng/ml. The results are summarized in Table 3. Note that in a separate experiment under identical conditions, ELISA data showed 100% recovery of TGF-β2 activity in the presence of heparin and < 1% activity in the absence of heparin.

Table 3

Heparin Stabilizes Biological Activity of rTGF-β2

These data show that heparin preserved the biological activity of rTGF-β2 in conditions under which TGF-β is not otherwise active.

Claims

1. A method of stabilizing the activity of a transforming growth factor β comprising adding a

glycosaminoglycan to a solution of transforming growth factor β in an amount sufficient to stabilize activity of the transforming growth factor β .

2. The method according to claim 1 wherein the glycosaminoglycan is selected from the group

consisting of neparin, chondroitin sulfate, dermatan sulfate and keratan sulfate.

3. The method according to claim 2 wherein the glycosaminoglycan is heparin and is present in an amount of at least 0.1 μg heparin per μg of the transforming growth factor β .

4. The method according to claim 3 wherein the heparin is present at a concentration of about 0.1 -

170 μg per μg of the transforming growth factor β .

5. The method according to claim 3 wherein the heparin is present at a concentration of about 10 - 170 μg per μg of the transforming growth factor β .

6. The method according to claim 3 wherein the heparin is present at a concentration of about 80 - 170 μg per μg of the transforming growth factor β .

7. The method according to claim 1 wherein the transforming growth factor is selected from the group consisting of TGF-β1, TGF-β2, TGF-β3 and heterodimers thereof.

8. The method according to claim 7 wherein the transforming growth factor β is TGF-β2.

9. A composition comprising an amount of a transforming growth factor β effective to exert

biological activity and an amount of glycosaminoglycan sufficient to stabilize the activity of the transforming growth factor β .

10. The composition according to claim 9 wherein the glycosaminoglycan is selected from the group consisting of heparin, chondroitin sulfate, dermatan sulfate and keratan sulfate.

11. The composition according to claim 10 wherein the glycosaminoglycan is heparin and is present in an amount of at least 0.1 μg heparin per μg of the transforming growth factor β .

12. The method according to claim 11 wherein the heparin is present at a concentration of about 0.1 - 170 μg per μg of the transforming growth factor β .

13. The method according to claim 11 wherein the heparin is present at a concentration of about 10 -

170 μg per μg of the transforming growth factor β .

14. The method according to claim 11 wherein the heparin is present at a concentration of about 80 - 170 μg per μg of the transforming growth factor β .

15. The composition according to claim 9 wherein the transforming growth factor β is selected from the group consisting of TGF-β1, TGF-β2, TGF-β3 and heterodimers thereof.

16. The composition according to claim 15 wherein the transforming growth factor β is TGF-β2.