Purification of Mullerian inhibiting substance

Abstract

The present invention discloses a method of purifying Mullerian Inhibiting Substance. The method takes advantage of immunoaffinity chromatography to achieve MIS at up to 95% purity. The method provides for a purified MIS product in which contaminating enzymes having MIS proteolytic activity or inhibitors of MIS antiproliferative activity are substantially removed by addition of an alkali metal halide solution or an alkaline earth metal halide solution prior to elution with an acid solution having a pH of between about 2.5 and 4.0.

Description

FIELD OF THE INVENTION

The present invention relates to a process for the purification of biological materials, especially the purification of Mullerian Inhibiting Substance (MIS).

BACKGROUND OF THE INVENTION

The efficient and rapid purification of biological materials such as proteins, nucleic acids or polysaccharides, has been of great interest. Immunoaffinity chromatography is advantageous since, under the proper conditions, purification of 1,000-10,000 fold can be attained. Cf. Harlow, E. et al., Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory (1988).

One protein for which a rapid and efficient purification system is desirable is the biological modifier Mullerian Inhibiting Substance (MIS) produced in vivo by the ovary and testis. MIS is a testicular protein responsible for regression of the anlagen of the female reproductive tract in the fetal urogenital ridge, as well as other differentiative functions such as inhibition of oocyte meiosis and lung surfactant. Previous studies in A431 cells and more recently in fetal lung suggest that MIS is a potent inhibitor of epidermal growth factor receptor autophosphorylation. Cf. Cigarroa et al., Growth Factors 1:79-91 (1989); Coughlin et al., Mol. and Cell. Endocrinol. 49:75-86 (1985); and Catlin et al., Metabolism.

Human Mullerian Inhibiting Substance has been cloned and expressed recombinantly (rhMIS) in animal cells as a 140 kDa homodimer (Cate et al., Cell 45:685 (1986)). By virtue of carboxy-terminal amino acid homology, MIS is a member of a large gene family that includes TGF-.beta. (Derynck et al., Nature 316:701-5 (1985)), inhibin (Mason et al., Nature 318:659-63 (1985)), activin (Ling et al., Nature 321:779-82 (1986)), Vg1 from Xenopus (Weeks et al., Cell 51:861-7 (1987)), the Drosophila decapentaplegia protein (Padgett et al., Nature 325:81-4 (1987)), and the bone morphogenesis factors (Wozney et al., Science 242:1528-34 (1988)).

RhMIS is a 140 kDa glycoprotein composed of two identical subunits which, under disulfide bond reducing conditions, migrates on polyacrylamide gel electrophoresis at an apparent molecular weight of 70 kDa. The protein can be proteolytically cleaved in approximately one hour by exogenous plasmin into two distinct fragments that migrate electrophoretically as 57 kDa and 12.5 kDa moieties with cleavage at residue 427 of the intact 535 amino acid monomer as demonstrated by Pepinsky (Pepinsky et al., J. Biol. Chem. 263(35):18961-4 (1988)). Prolonged exposure to plasmin can result in cleavage of MIS at additional sites. In addition, purification of MIS by known techniques can be contaminated by other proteases that also cleave MIS.

MIS has been proposed as a potential growth inhibitor of epithelial human tumors of Mullerian origin such as endometrial, Fallopian tubal, cervical, and certain ovarian neoplasms. Experimental evidence using purified bovine MIS support this hypothesis (U.S. patent application Ser. No. 06/792,233, filed Oct. 19, 1985, now U.S. Pat. No. 5,011,687; Fuller et al., J. Clin, Endocrinol. Metab. 54:1051-5 (1982); Fuller et al., Gynecol. Oncol. 22:135-48 (1985)). purified rhMIS used in similar in vitro assays, however, demonstrate limited anti-cancer activity (Wallen et al., Cancer Res. 49:2005-11 (1989)),

Mullerian Inhibiting Substance may be obtained by a variety of different methods. U.S. patent application Ser. No. 06/792,233, filed Oct. 19, 1985, now U.S. Pat. No. 5,011,687, and entitled "Purified Mullerian Inhibiting Substance and Process for Treating Human Ovarian Cancer Cells," describes a process for purifying MIS from testes by using aqueous polar dissociative solutions, separation of DNA and RNA, fractionation by gel filtration chromatography, and isolation of the MIS. U.S. Pat. No. 4,404,188, filed Jul. 29, 1981 and entitled "Purified Mullerian Inhibiting Substance and Method of Purification" describes a process for purifying MIS from testes which comprises treatment with a protease inhibitor, chromatography on ion exchange, chromatography on wheat germ lectin, on concanavalin A and/or on a supported triazinyl dye. U.S. Pat. No. 4,487,833, filed on Mar. 1, 1982 and entitled "Method of Preparing Hybridomas and of Purifying Immunogenic Materials" describes a process for separating MIS using immunoaffinity chromatography. MIS has also been obtained from recombinant DNA techniques as disclosed by Cate (Cate, et al., Cell 45:685-698 (1986)). None of the references, however, describe a method of recovering substantially pure MIS which retains an antiproliferative activity or is essentially free of enzymes having proteolytic activity against MIS.

A need, therefore, continues to exist for the development of highly efficient techniques for immunoaffinity chromatography, wherein a biological substance can be isolated from a complex biological mixture without interference from contaminating proteolytic enzymes or other factors that impede MIS antiproliferative activity. In particular, the need exists for an immunoaffinity chromatography method which can isolate and purify MIS so that the isolate will be essentially free from undesired proteolysis or other factors which may alter MIS activity.

SUMMARY OF THE INVENTION

The purification of MIS according to the immunoaffinity purification process of this invention results in a MIS product which is substantially free of enzymes having proteolytic activity or inhibitors of MIS antiproliferative activity. The present invention achieves this goal by providing for a method of purifying MIS comprising

(a) binding the MIS to an antibodychromatography matrix, said antibody being specific to MIS,

(b) substantially removing contaminating enzymes having MIS proteolytic activity or inhibitors of MIS antiproliferative activity by adding to the matrix an effective amount of an alkali metal halide or an alkaline earth metal halide, and

(c) recovering the MIS by eluting with an acid solution having a pH of between about 2.5 and 4.0.

This invention further provides for a composition comprising MIS obtained by

(a) binding the MIS to an antibody-chromatography matrix, said antibody being specific to MIS,

(b) substantially removing contaminating enzymes having MIS proteolytic activity or inhibitors of MIS antiproliferative activity by adding to the matrix an effective amount of an alkali metal halide or an alkaline earth metal halide, and

(c) recovering the MIS by eluting with an acid solution having a pH of between about 2.5 and 4.0.

Also provided by this invention is a composition comprising MIS, wherein said composition is substantially free of enzymes having MIS proteolytic activity or inhibitors of antiproliferative activity of MIS. Such a composition consists essentially of MIS having a molecular weight of 140 kDa or 70 kDa.

BRIEF DESCRIPTION OF THE DRAWING

The present invention will be better understood by reference to the Description of the Preferred Embodiments when taken together with the attached drawing, wherein:

FIGS 1A-1C shows the sequence structure of rhMIS (SEQ. ID. NO: 1) compared to bovine (SEQ. ID. NO: 2) and rat (SEQ. ID. NO: 3) MIS.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

The present invention provides for an improved process for the purification of Mullerian Inhibiting Substance (MIS). This process takes advantage of the specificity of antigen-antibody interactions to recover a product having a substantially pure MIS. Specifically, this invention incorporates the use of immunoaffinity chromatography. The added benefit of this invention is that the immunoaffinity chromatography method is improved such that the recovered MIS product is substantially free of contaminating enzymes having MIS proteolytic activity or inhibitors of antiproliferative activity of MIS.

Immunoaffinity chromatography is a type of chromatography that makes use of a specific affinity between a substance to be isolated, i.e., a ligand, and a molecule that it can specifically bind. The column material is synthesized by covalently coupling a binding molecule to an insoluble matrix. The column material is able to specifically absorb the substance to be isolated from solution. Elution of the product is accomplished by changing the conditions within the column such that the product is released. This procedure either effects the binding site directly or effects the structure of the bound molecule.

If interaction of the ligand to the binder is primarily electrostatic, desorption can be accomplished by a gradient of increasing ionic strength. Ionic strength can be altered by changing pH. Such a change can alter the degree of ionization of charged groups either on the ligand or at the binding site. In desorbing proteins, however, there may be a secondary effect on the protein molecule which results in a conformational change. When binding is due to a hydrophobic interaction, solvents with reduced polarity are effective in desorbing the bound molecule.

In purifying nucleic acids and proteins, desorption has been generally accomplished with the use of chaotropic substances such as ClO.sub.4.sup.-, CF.sub.3 COO.sup.-, .sup.- SCN, and CCl.sub.3 COO.sup.-. These substances are useful because they are able to break down very strong interactions.

As described by Pepinsky et al., supra, MIS has been isolated by an immunoaffinity chromatography method. According to this method, MIS is eluted from the matrix by a standard chaotropic salt, NaSCN, and can be recovered at up to 95% purity. Evaluation of MIS when purified according to previously used immunoaffinity methods indicates that the protein has an apparent mass of 70 kDa with fractions also existing as 57 kDa and 12.5 kDa.

As demonstrated by the Example herein, the presence of the 57 kDa and 12.5 kDa fragments are the result of proteolysis or inhibition of the activity of the MIS molecule. Using previously described immunoaffinity methods, contaminating proteolytic enzymes or inhibitors of antiproliferative activity of MIS can be eluted with the pure MIS product. This occurs even though these methods are able to achieve an MIS product of up to 95% purity. The proteolytic enzymes initially cleave the MIS molecule into 57 kDa and 12.5 kDa fragments to activate the molecule. However, further cleavage of the 57 kDa fragment to a 34 kDa fragment and to a 22 kDa fragment occurs. Substances which cleave MIS in this manner include serine proteases, such as plasmin, and endopeptidases. These enzymes are not to be considered as all inclusive or limiting in any manner since other enzymes can also proteolytically cleave MIS and such enzymes can be readily determined by those of ordinary skill in the art.

In order to achieve a substantially pure MIS which will be free of contaminating proteases or inhibitors of antiproliferative activity of MIS, the present invention improves upon the immunoaffinity chromatography method previously employed. This improvement results in a composition which comprises MIS and is substantially free of enzymes having MIS proteolytic activity or inhibitors of antiproliferative activity of MIS.

The immunoaffinity chromatography method of this invention is improved by removing contaminating enzymes having MIS proteolytic activity or inhibitors of antiproliferative activity of MIS from an immunoaffinity chromatography matrix by eluting with an effective amount of an alkali metal halide or an alkaline earth metal halide. The MIS is then recovered by eluting with an acid solution having a pH of between about 2.5 and 4.0. This elution with halide followed by acid is alternatively referred to as sequential salt/acid elution.

The terms "Mullerian Inhibiting Substance" and "MIS", as alternatively used herein, are intended to include compounds and materials which are structurally similar to MIS. Examples of such included substances and materials are salts, derivatives, and aglycone forms of MIS. Additionally, the present invention is intended to include mutant forms of MIS which have substantially the same biological activity as MIS. Examples of such mutant forms would be MIS molecules carrying at least one deletion, insertion, or alteration in amino acid sequence. MIS may be obtained from any mammalian source or, as indicated above, from non-mammalian sources through the use of recombinant DNA technology, or from the chemical synthesis of the MIS protein. The purification method of this invention is preferably directed toward the recovery of recombinant human MIS.

The term "protein" is meant to include both synthetic and naturally-occurring amino acid sequences derivable from the naturally occurring amino acid sequence of MIS. The protein is said to be "derivable from the naturally-occurring amino acid sequence of MIS" if it can be obtained by fragmenting the naturally-occurring chosen sequence of MIS, or if it can be synthesized based upon a knowledge of the sequence of the naturally occurring amino acid sequence or of the genetic material (DNA or RNA) which encodes this sequence.

The invention further pertains to polypeptides that, in addition to the chosen sequence, may contain or lack one or more amino acids that may not be present in the naturally-occurring sequence, wherein such polypeptides are functionally similar to or possess antagonist activity to the chosen polypeptide. Such polypeptides for the present invention, are termed "functional derivatives," provided that they demonstrate activity which is substantially similar to or antagonistic to that of MIS.

The MIS composition may be in the form of the free amines (on the N-terminus), or acid-addition salts thereof. Common acid solution salts are hydro halic acid salts, i.e., HBr, HI, or more preferably, HCl.

The purified MIS of this invention can be obtained in solution at up to 95% purity or greater. While the percent purity is comparable to other immunoaffinity purification processes, the MIS composition of this invention is substantially free of contaminating proteolytic enzymes or inhibitors of MIS antiproliferative activity.

For purposes of this invention, purified MIS is considered to be a MIS composition which is substantially free of contaminating proteolytic enzymes or inhibitors of MIS antiproliferative activity regardless of percent purity. The composition is considered to be substantially free of proteolytic enzymes if gel electrophoresis of the purified MIS product indicates a protein having a molecular weight of 140 kDa or 70 kDa. Gel electrophoresis of such a product will not show time dependent proteolytic fragments which are degradation products of MIS. For example the 57 kDa, 12.5 kDa, 34 kDa and 22 kDa degradation fragments of MIS further described herein will not be readily discernable by standard gel electrophoresis methods.

The MIS composition will be considered to be substantially free of inhibitors of MIS antiproliferative activity if the MI blocks proliferation of certain tumor cells. Examples of such tumor cells are included herein and in U.S. patent application Ser. No. 07/683,966, now abandoned, and which is fully incorporated herein by reference. These examples include tumors selected from the group consisting of vulvar epidermoid carcinoma, endometrical adenocarcenoma, cervical carcenoma, endometrial adenocacenoma, ovarian adenoracenoma, and other ocular melanoma. The determination of antiproliferation activity of these tumors as well as any other tumor can be achieved by any of the procedures described herein and in U.S. patent application Ser. No. 07/683,966, now abandoned.

In order to obtain MIS which is substantially free of proteolytic enzymes or inhibitors of MIS antiproliferative activity, the contaminants are separated from the MIS using immunoaffinity chromatography. Separation occurs by eluting the enzymes or inhibitors with an alkali metal halide or an alkaline earth metal halide. Such a compound will generally be in solution and an effective amount of halide will be between about 0.1M and 2.0M. As alkali metals, the ions of lithium, sodium and potassium are preferred with sodium being most preferable. As alkaline earth metals, the ions of magnesium and calcium are preferred. As halides, the ions of fluorine, chlorine, bromine and iodine are preferred with chlorine being most preferable. When eluting with sodium chloride, a solution of between about 0.1 and 2.0M is preferred. The concentration of halide can also be varied as elution progresses if desired. This can be accomplished by increasing molar concentration of halide in a stepwise fashion. It is preferred that each step be altered after about 0.1-2.0 bed volumes of solution have contacted the chromatography matrix, although such steps can be further modified as desired.

The halide can also be accompanied in solution with an effective amount of a chelating agent. These agents are capable of binding metal ions which can inactivate enzymes that require the metal ions for activity. Such agents include the compounds ethylenediamine tetraacetate (EDTA), and ethylenebis(oxyethylenenitrilo)tetraacetic acid (EGTA). Chelants can be effectively added in a range of between 0.1 and 50 mM.

After the contaminating proteolytic enzymes or inhibitors of MIS antiproliferative activity have been separated, the MIS can be recovered by eluting with an acid solution having a pH of between about 2.0 and 4.0. Although dilutions of strong acids such as HCl can be used, organic acids are preferred because of their relatively mild acid strength. For example, the use of acid amines and imines can be employed as well as monocarboxylic, dicarboxylic and tricarboxylic acids. Preferred as monocarboxylic acids are acetic, propionic and butyric acid. Preferred as dicarboxylic acids are succinic, fumaric and malic acid. Preferred as a tricarboxylic acid is citric acid. Preferred among the amines are the acidic amino acids such as aspartic and glutamic acid. The pH of the acid solution can also be incrementally varied as in the application of the halide.

After the purified MIS product has been eluted, it is preferable to neutralize the product to a pH of between 6.8 and 7.6 to guard against acid hydrolysis. This can be accomplished by various hydroxide compounds such as NaOH or NH.sub.4 OH or by various buffers which can quickly achieve neutralization of the product in the desired pH range. These hydroxide compounds are not to be considered as all inclusive as those of ordinary skill in the art will appreciate.

Preparation of an immunoaffinity chromatography column is well known in the art as demonstrated by Harlow et al.. Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, 1988. For purposes of this invention it is preferable that the chromatography column have a matrix which can be coupled to an antibody specific to MIS. The matrix will generally be solid phase support which binds antibody easily and is commercially available. For example, Protein A can be used as the chromatography matrix. This matrix specifically binds to the Fc domain of antibodies and, after the antibody is bound, the interaction is stabilized by cross-linking with a bifunctional coupling reagent. In preparing a protein A bead-antibody column, any bifunctional coupling reagent can be used. The use of methylpimelimidate is generally preferred.

Another method of preparing an immunoaffinity column couples the antibody to an activated bead. The beads are activated chemically to contain reactive groups. Examples of beads which can be used in this method are agarose beads, cross-linked agarose beads, polyacrylamide beads, copolymers of polyacrylamide and agarose, and polyacrylic beads. Compounds which are generally used to activate the beads are carbonyldiimidoxole, cyanogen bromide, glutaraldehyde, hydroxysuccinimide, and tosyl chloride. After the beads are activated, they are then mixed and coupled with purified antibodies.

Yet another method of coupling antibodies to beads is to activate the antibody first. Purified antibodies can be activated by treating with a bifunctional reagent, one group binding to an appropriate group on the antibody and the other remaining free to bind to the matrix. Reagents which can generally be used for indirect coupling are the water-soluble carbodiimides such as dicyclohexylcarbodiimide (DCCD), 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide HCl (EDAC or DECI), or 1-cyclohexyl-3-(2-morpholinoethyl)-carbodiimide-metho-p-toluene-sulfonate (CMIC or CMLI); the condensing agents for peptide synthesis such as N-ethoxycarbonyl- 2-ethoxy-1,2-dihydroquinoline (EEDQ); glutaraldehyde; or periodate. After the antibodies are activated, they are bound to beads.

Having now generally described this invention, the same will be better understood by reference to a certain, specific example which is included herein for purposes of illustration only and is not intended to be limiting of the invention, unless specified.

EXAMPLE

I. Materials and Methods

A. RhMIS Purification

Recombinant human MIS (rhMIS) was purified from the conditioned media of Chinese hamster ovary (CHO) cells, transfected with a linear construct of the human rhMIS gene and the DHFR gene, amplified by 30M Methotrexate selection, grown to confluence in four liter bioreactors on stainless steel coils as described by Epstein (Epstein et al., In Vitro Cell. and Devel. Biol 25(2):213-6 (1989)) or modified roller bottles in alpha-Modified Eagle's Medium (.alpha.-MEM-), supplemented with 5% female fetal calf serum (FFCS), 10 mg/ml Amikacin, 1.3 g/l -glutamine, 2.0 g/l d-glucose, and 0.1 g/l sodium pyruvate, in the absence of nucleosides. The medium was collected every 3-4 days, and stored at -20.degree. C. Media were thawed and filtered through Whatman #4 filter paper to remove debris, concentrated 20x an a Minitan (Millipore) with a 30 kDa exclusion ultrafilter, and stored at -70.degree. C., until purification. A 5 ml immunoaffinity column was constructed using approximately 50 mg of a Protein A-Sepharose (BioRad) purified monoclonal anti-human rhMIS antibody [6E11], as described by Hudson (Hudson et al., J. Clin. Endocrinol. 70:16-22 (1990)), and covalently attached to Affigel-10 agarose resin (BioRad), per the manufacturer's instructions (approximately 80% coupling efficiency). The column was equilibrated with 100 mls of 20 mM HEPES, pH 7.4, and 200 ml of the concentrated media loaded at 1 column volume/hour at 4.degree. C. After loading, the column was washed with 20 mM HEPES, pH 7.4, until the absorbance at 280 nm returned to baseline (60-100 mls).

Elution of rhMIS bound to this column was achieved using 2.0M sodium thiocyanate (NaSCN); or 1M Acetic acid, 20 mM HEPES, pH 3.0, with and without a pre-elution step containing 0.5M NaCl, 1 mM EDTA, 0.001% NP-40, 20 mM HEPES, pH 7.4. The majority of the rhMIS protein eluted in a single 2 ml fraction, which was immediately desalted by G-25 size exclusion chromatography in 0.02M HEPES, 0.15M NaCl, 10% glucose, pH 7.4, in the case of chaotropic salt elution, or immediately neutralized with either NaOH or NH.sub.4 OH to a pH between 7.0 and 7.4, when eluted with acid. Depending on the initial pH of the fraction and technique of neutralization, dilutional effects ranged from 10-50%. The resultant rhMIS was examined for total protein according to the method of Bradford (Bradford, Anal. Biochem. 72:248-54 (1976)), for rhMIS concentrations by enzyme-linked immunosorbant assay [ELISA] according to the method of Hudson (Hudson et (al., .J. Clin. Endocrinol 70:16-22 (1990)), on polyacrylamide gel electrophoresis [PAGE] according to the method of Weber (Weber et al., J. Biol. Chem. 244(16):4406-12 (1969)), in Western blot analysis using polyclonal (anti-homo, and anti-N- and C-terminal peptide) antibodies to rhMIS according to the method of Towbin (Towbin et al., PNAS 76:4350-4 (1979)), for NH.sub.2 -terminal sequencing, in organ culture bioassay for Mullerian duct regression, and in tumor antiproliferative assays.

B. RhMIS Bioassay

The standard organ culture bioassay for MIS was performed as described by Donahoe (Donahoe et al., Biol. Reprod. 16:238-43 (1977)). Briefly, 141/2 day female fetal rat urogenital ridges were placed on agar coated stainless steel grids above fortified CMRL 1066 media to which test preparations were added at concentrations less than 20% (v/v). After incubation for 72 hours in humidified 5% CO.sub.2 at 37.degree. C., the specimens were aligned, fixed in 15% formalin, embedded in paraffin, cut in 8 mm cross sections from cephalic to caudal, and then stained with hematoxylin and eosin. The sections were then graded from 0 (no regression) to 5 (complete regression). Female fetal calf serum was used to avoid contamination of the assay with bovine MIS. This change required the addition of 10.sup.-9 M testosterone to aid the expression of the Wolffian duct for morphologic comparison according to the method of Ikawa (Ikawa et al., J. Ped. Surg. 17:453 (1982)).

C. RhMIS ELISA

The enzyme linked immunosorbent assay (ELISA), employing anti-rhMIS monoclonal and polyclonal antibodies, was used to measure the rhMIS content in all preparations. This assay, described by Hudson (Hudson et al., J. Clin. Endocrinol 70:16-22 (1990)), detects intact rhMIS with a sensitivity of 1-2 ng/ml, with minimal cross reactivity with N- and C-terminal fragments of rhMIS, and no recognition of Transforming Growth Factor .beta. (another member of the same gene family as MIS), the humor gonadotropin, follicle stimulating hormone, and luteinizing hormone, or proteins contained in conditional medium of wild type CHO cells.

D. Antiproliferative Assays

Immunoaffinity purified rhMIS was assayed for antitumor activity as described by U.S. patent application Ser. No. 07/683,966, now abandoned, using a double layer agarose colony inhibition assay as described by Fuller (Fuller et al., Gynecol. Oncol 22:135-48 (1985)). Human cancer cell lines, including A431, Hep3b, HEC1, OM431, OM467, and RT4, were plated on an underlayer of 0.6% agarose in .alpha.-MEM+ and FFCS. After addition of rhMIS or vehicle control, plates were incubated for 14 days at 37.degree. C., 5% CO.sub.2, and colonies of more than 30 cells were counted. A liquid media colony inhibition assay that requires fewer cells and 5-7 day incubation period was also used. Colonies were counted using a computer aided automated program.

E. RhMIS Antibody Preparations

Rabbit polyclonal antiserum against 140 kDa homo rhMIS (MGH-1), electroeluted from polyacrylamide gels according to the method of Hunkapillar (Hunkapillar et al., Meth. Enzymol. 91.486 (1983)), was prepared as previously described by Hudson (Hudson et al., J. Clin. Endocrinol. 70:16-22 (1990)). Epitope specific rabbit polyclonal antibodies were raised to regions of the rhMIS molecule NH.sub.2 -terminal (MGH-N1) or COOH-terminal (MGH-Cl) to the monobasic consensus cleavage site at position 427. Synthetic peptides corresponding to residues 411-424 and 471-482 in rhMIS, respectively, constructed from the predicted amino acid sequence of human rhMIS (FIGS. 1A-1C) were used as antigens. These sequences were chosen for their conserved homology among human, bovine and rat MIS, their differences from other members of the supergene family, and their antigenicity and surface probability (i.e., the likelihood that a specific region will present itself for antigen recognition), as predicted by the sequence analysis software package of the Genetics Computer Group at the University of Wisconsin version 5]. This program measures hydrophobicity as outlined by Chou and Fasman (Chou et al., Advances in Enzymology 47:45-147 (1978)), and antigenicity and surface probability by the method of Wolf et al. (Wolf et al., Comput. Appl. Biosci. 4(1):187-91 (1988)).

New Zealand white rabbits were injected in the popliteal lymph nodes with 25-50 .mu.g of relevant peptide, conjugated 1:1 with keyhole limpet hemocyanin complete Freund's adjuvant. These animals were boosted by subcutaneous injection in the back, 4-6 weeks later, with 20-30 .mu.g of unconjugated peptide in incomplete Freund's adjuvant. Animals were bled through an ear vein and the serum stored at -20.degree. C. Polyclonal antiserum was purified by 50% (NH.sub.4).sub.2 SO.sub.4 precipitation, followed by Protein A-Sepharose chromatography. The monoclonal antibody 6E11 was raised according to the method of Kohler (Kohler, G, Immunol. Methods 2:285-98 (1981)) and purified by Protein-A Sepharose affinity chromatography. Control antisera were purchased from DAKO.

F. Temperature, pH, and Storage Effects on

rhMIS Preparations

Immunoaffinity purified rhMIS was sterile filtered using a Millex-GV 0.22 .mu.m filter unit (MILLIPORE), divided into 200 .mu.l aliquots and incubated at -80.degree. C., -20.degree. C., +4.degree. C., +25.degree. C., +36.degree. .C, and +70.degree. C. to determine temperature stability of the rhMIS molecule. To observe the effects of storage at various temperatures, aliquots of rhMIS at 3, 7, 11, 14 and 31 days, were subjected to Bradford protein assay (Bradford, M.M., Anal. Biochem 72:248-54 (1976)), ELISA, polyacrylamide gel electrophoresis in sodium dodecyl sulfate (SDS-PAGE), western blot analysis, organ culture bioassay, and selected aminoterminal analysis. Additionally, preparations of rhMIS were stored at pH 3.0 for 1, 2, and 4 hours, and 2, 3, 5, and 8 days at +4.degree. C. and +36.degree. C. For comparison, a non-cleavable mutant of rhMIS, produced by site-directed mutagenesis of Arg.sup.427 to Thr.sup.427, was tested after a days storage at 37.degree. C.

G. Enzyme Treatments

RhMIS was treated with plasmin for 2 hours, as described by Pepinsky et al., supra, for 1, 3, and 7 days, and with 0.1% trypsin for 1 hr at 4.degree. C. The reactions were quenched with 10% fetal bovine serum. The effect of protease inhibitors was assessed with phenylmethyl sulfonyl fluoride, 0.1 mM; soybean trypsin inhibitor, 1 mM; leupeptin, 1 mM; pepstatin, 1 mM; all used together as a mixture, or tosyllysine chloromethyl ketone, 1 mM, alone. Additionally, rhMIS was assayed for intrinsic enzyme activities. Protease activity was determined using azocasein (2 mg/ml in 0.05M Tris-Cl, pH 8.5) as a substrate (Worthington manual 1970). 0.9 ml was brought to 37.degree. C. in a water bath prior to addition of 0.1 ml of enzyme (1-20 mg/ml Subtilisin) or rhMIS. The reaction was allowed to proceed for 15 minutes at 37.degree. C., and 0.25 ml 70% perchloric acid was added. The samples were placed on ice for 15 minutes, then centrifuged at 4.degree. C. to pellet the precipitate. Absorbance of the supernatant was measured spectrophotometrically at 405 nm. Acid and alkaline phosphatase activities were examined with Sigma chemical kit #104, which measures the ability to hydrolyze p-nitrophenyl phosphate.

H. Sequencing

Acid eluted, immunoaffinity purified rhMIS was submitted as "homo" rhMIS for NH.sub.2 -terminal sequencing. A similar preparation was heated to 37.degree. C. for 8 days to allow cleavage of 70 kDa intact rhMIS to 57 kDa, 34 kDa, 22 kDa, and 12.5 kDa as determined by SDS-PAGE. These bands were then electroeluted from the gel, according to the method of Hunkapillar (Hunkapillar et al.. Meth. Enzymol. 91:486 (1983)), and analyzed for N-terminal sequence. The sequential salt/acid eluted rhMIS was similarly submitted for sequence analysis except the cleavage was done by exogenous plasmin. The 34 kDa fragment, which required concentration by lyophilization prior to PAGE, was electrophoretically transferred to an Immobilon PVDF membrane (Millipore), stained, and sequenced. Edman degradation was performed on a model 470A Applied Biosystems gas phase sequencer; PTH-amino acids were analyzed on an on-line microbore HPLC (ABI model 120A).

I. Electrophoresis

Polyacrylamide gel electrophoresis (PAGE) was carried out by the procedure of Laemmli (Laemmli, Nature 227:680-5 (1970)), as follows: 10 or 15% homogeneous gels were prepared in Hoefer gel casting stands to produce 160.times.160.times.1.5 mm slabs, and gels run at 150V, 30 mA constant current. Proteins in the gels were stained with 0.1% Coomassie Brilliant Blue R250 (Sigma) in 50% methanol, 10% acetic acid for 1 hour before destaining in 50% methanol, 10% acetic acid. As appropriate, samples were reduced using 0.75 M 2-mercaptoethanol with heating to boiling for 1o min. Molecular weight standards were obtained from BioRad.

For western analysis, gels were incubated in transfer buffer (3 g Tris, 14.4 g/l glycine) prior to electrophoretic transfer to nitrocellulose or Immobilon-PVDF (Milligen Corp.) sheets according to the BioRad technical bulletin no. 83-0050. After the transfer, the sheets were used directly for microsequencing, or blocked by incubation with 30 mg/ml bovine serum albumin for 30 minutes at room temperature, with shaking. Thereafter, the blots were incubated with 1:500 to 1:1000 fold dilutions of rabbit polyclonal anti-rhMIS or anti-rhMIS peptide antisera for 2 hrs, then washed with 0.05M Tris-Cl, 0.15M NaCl prior to addition of a 1:1000 dilution of goal anti-rabbit horseradish peroxidase conjugate (BioRad). Antibody complexes were visualized by the addition of BioRad color reagent for 1-4 minutes, prior to quenching the reaction with water.

For microsequencing, proteins were transferred according to the BioRad technical bulletin No. 83-0050 into Immobilon PVDF sheets and stained with Coomassie blue as described above for gels except that acetic acid was omitted.

J. Labelling of RhMIS

RhMIS, prepared by either sequential salt/acid or acid only elution, was iodinated by the chloramine T method of Hunter (Hunter, Proc. Soc. Exp. Biol. Med. 133(3):989-92 (1970)). Briefly, to 0.5 ml (0.3 mg/ml) of rhMIS, 10 .mu.l of Na.sup.125 I (Amersham) and 10 .mu.l of chloramine T (2.5 mg/ml; Sigma) were added. After 90 seconds, the reaction was quenched with 25 .mu.l of NaHSO.sub.3 (7.6 mg/ml). The mixture was desalted on a 10 ml Sephadex G-25 column, equilibrated in 20 mM HEPES, pH 7.4.

Biotinylation (Pierce Chemical Co. Bulletin No.21335) was carried out using water soluble N-hydroxysuccinimide-long chain alkyl-biolin. 0.5 ml of rhMIS (0.3 mg/ml) was dialyzed overnight at 4.degree. C. vs PBS. 5 .mu.l of 4.5 mg/ml NHS-LC-Biotin was added and the mixture incubated at 4.degree. C. with gentle agitation for 4 hours, then dialyzed overnight at 4.degree. C. vs PBS.

Fluorescent rhMIS was prepared with fluorescein isothiocyanate [FITC] according to the method of Johnson (Johnson et al., J. Biol. Chem. 257(10):5632-6 (1982)), as follows. Approximately 200 .mu.g of rhMIS was dialyzed, overnight at 4.degree. C., vs 2.times.4 l 0.05M TRIS, pH 8.8. One tenth volume of fluorescein isothiocyanate (FITC) in TRIS 8.8 was added (50.ANG. FITC/500.ANG. MIS), and the reaction allowed to proceed for 30 minutes at 4.degree. C. The reaction mixture was desalted with a 10 ml Sephadex G-25 column equilibrated in PBS. Protein was monitored at A.sub.280 and fluorescence at A.sub.493. The protein and fluorescence peaks co-eluted with the void volume.

II. RESULTS

Immunoaffinity purified rhMIS migrates on SIS-PAGE under reducing conditions, at an apparent molecular weight of 70 kDa regardless of the method of elution. A 57 kDa minor band was also seen. These two bands, which represented approximately 90% of the Coomassie Brilliant Blue stained protein on reducing gels, as well as the 140 kDa dimeric species in nonreducing gels, bind polyclonal antibody against rhMIS dimer. RhHIS subjected to PAGE without SDS, or under conditions of reverse polarity, failed to enter the gel. The acid elution of rhMIS from the immunoaffinity column resulted in higher yields and greater purity than did chaotropic elution (Table 1). Sequential salt/acid elution resulted in rhMIS being substantially free from enzymes having MIS proteolytic activity.

TABLE 1______________________________________ PROTEIN CONCEN- MIS % FOLDSAMPLE TRATION (ELISA) MIS PURITY YIELD______________________________________MEDIA 1.75 3.43 +/- 0.196 1 100 mg/ml 1.47 .mu.g/mlIAP (S/A) 0.388 297 .mu.g/ml 77 392 4.4n = 8 mg/mlIAP 1.06 +/- 964 +/- 91 464 13.8(Acid) 0.6 mg/ml 660 .mu.g/mln = 14IAP (SCN) 0.87 +/- 620 +/- 71 362 11.3n = 13 0.3 mg/ml 480 .mu.g/ml______________________________________

RhMIS prepared by single step elution with either chaotropic salt or acid alone undergoes a time and temperature dependent cleavage that is blocked by proteolytic inhibitors. RhMIS prepared by sequential salt and acid elution does not undergo endogenous proteolytic processing. The major species after cleavage are monomers of 57 kDa and 12.5 kDa, consistent with cleavage at residue 427. Amino acid sequence of freshly purified rhMIS yields the known amino terminal sequence LeuArgAlaGluGluProAlaValGlyThr (SEQ. ID. NO: 4). After cleavage, a second sequence AlaAlaGlyAlaThrAlaAlaAspGlyPro (SEQ. ID. NO: 5) is found which, except for the first amino acid, matches the carboxy-terminal sequence of rhMIS beginning at residue 428.

The 34 kDa moiety, seen with prolonged incubation of acid only eluted rhMIS at +36.degree. C., was sequenced successfully after lyophilization and electrophoretic transfer to PVDF membrane. This species yielded the NH.sub.2 -terminal sequence of rhMIS (i.e., LeuArgAlaGluGluProAlaValGlyThr, SEQ. ID. NO: 4), indicative of a second processing event N-terminal to that at residue 427. Sequential salt/acid eluted material showed this same band and the 57 and 12.5 kDa bands after treatment with exogenous plasmin. Occasionally, an additional fragment with apparent molecular weight of 22 kDa was seen, but this fragment was not sequenced. Furthermore, the polyclonal antibodies raised to peptides constructed from the predicted amino acid sequence, both C-terminal or N-terminal to the monobasic protease cleavage site (residue 427), recognized 140 kDa "homo" rhMIS on western blots. Antibody MGH-N1, raised to the peptide corresponding to residues 411-424, recognized only 70 kDa and 57 kDa species. C-terminal anti-peptide (471-482) antibody MGH-C1 recognized the 12.5 kDa species seen in 36.degree. C. treated samples, as well as uncleaved, reduced rhMIS (70 kDa). Additionally, MGH-C1 weakly recognized cleaved N-terminal 57 kDa rhMIS, indicating possible crossreactivity, due to the polyclonal nature of the antibody which was raised to KLH conjugated peptide. A polyclonal antibody, termed MGH-1, which was raised to intact immunoaffinity purified rhMIS, recognized the 140 kDa, 70 kDa, and 57 kDa species, as well as the 34 kDa and 22 kDa fragments. These antibodies recognized similar bands on western blots of a less homogeneous, dye affinity purified rhMIS.

Although protein levels measured by Bradford analysis do not change with the time of incubation, the ELISA values decrease as cleavage increases. In spite of the proteolytic cleavage, however, rhMIS remains biologically active even if the protein is stored at neutral pH for 1 month at temperatures up to 36.degree. C. RhMIS activity is lost, however, by heating to 70.degree. C. for 4 days, by treatment with 0.1% Trypsin, and by storage for greater than 24 hours in acid (1M HAc), at 36.degree. C. Acid storage at 4.degree. C. loses activity more slowly (30-50% at 8 days).

Iodinated, biotinylated, and fluoresceinated rhMIS were all active in the organ culture bioassay at approximately 50% of their prelabelling levels. FITC labelled rhMIS was immunoreactive in the ELISA at only 20% of its original level, while iodinated or biotinylated material retained virtually all of its immunoreactivity. All labelled preparations showed the characteristic rhMIS bands on PAGE.

To determine if rhMIS acts as an enzyme, i.e., an auto protease, it was assayed in a standard subtilisin azocasein assay. RhMIS eluted with acid alone showed subtilisin-like activity of 1.3 .mu.g/mg rhMIS, in the azocasein assay. Sequential salt/acid eluted material had no activity in this assay, indicating that the copurifying protease had been effectively separated from rhMIS. By effectively separated, it is to be understood that the protease had been actually removed or inactivated such that it was no longer effective in cleaving the MIS molecule.

Sequential salt/acid eluted rhMIS, treated with plasmin for 2 and 24 hours, yielded the predicted cleavage at Arg.sup.427, to generate the 57 kDa and 22.5 kDa moieties; these were also seen after prolonged processing of acid-only eluted rhMIS at 36.degree. C. Reaction of sequential salt/acid eluted rhMIS with plasmin for longer periods (3-7 days) shows that after the initial cleavage, the 57 kDa fragment is further cleaved to the 34 kDa species. This species was also observed after prolonged storage at 36.degree. C. of rhMIS purified without the salt pre-elution step. The 34 kDa moiety, generated from the sequential salt/acid eluted rhMIS by prolonged plasmin treatment, was sequenced to yield the NH.sub.2 -terminal sequence of MIS, thus indicating a secondary cleavage event to yield the 34kDa fragment.

To address the frustratingly variable or absent response of rhMIS purified without the salt elution step in antiproliferative assays reported by Wallen (Wallen et al., Cancer Res. 49:2005-11 (1989)), the further refined rhMIS preparations (sequential salt/acid eluted MIS) were tested in similar assays. The rhMIS produced following the salt pre-elution step retains all previous activities, and further is consistently inhibitory in colony inhibition, cell cycle, and subrenal capsule protocols. Examination of the proteins eluted by the 0.5M salt wash shows 3 low molecular weight bands on reducing gels, which were not recognized by either MGH-1 (polyclonal anti-"homo"rhMIS) or MGH-C1 (polyclonal anti-C-terminal peptide).

Analysis of the sequence structure of rhMIS (FIGS. 1A-1C, SEQ. ID. NO: 1) revealed a string of nine leucine repeats beginning at residue 350, followed by a basic region immediately preceding the Arg.sup.427 cleavage site. This string of repeats strongly resembles a leucine zipper as reported by Vinson (Vinson et al., Science 246(4932):911-6 (1989)). Additionally, there are several other leucine rich regions upstream of this zipper, probably representing the hydrophobic core of the molecule. One of the stretches, beginning at Leu.sup.266, shows similarities to the S4 domain of the gated ion channels (Tempel et al., Nature 332:837 (1988)). There was also present at positions 194-8, the sequence Arg Gly Glu Asp Ser (SEQ. ID. NO. 6), strikingly similar to the Arg Gly Asp Ser (SEQ. ID. NO. 7) site for fibronectin binding, which may account for the recognition of a basement membrane binding site for rhMIS by antibody to NH.sub.2 -terminal rhMIS peptide.

Having now fully described this invention, it will be appreciated by those skilled in the art that the same can be performed within a wide range of equivalent parameters of composition and conditions without departing from the spirit or scope of the invention or of any embodiment thereof.

__________________________________________________________________________SEQUENCE LISTING(1) GENERAL INFORMATION:(iii) NUMBER OF SEQUENCES: 7(2) INFORMATION FOR SEQ ID NO:1:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 560 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(ii) MOLECULE TYPE: protein(xi) SEQUENCE DESCRIPTION: SEQ ID NO:1:MetArgAspLeuPro LeuThrSerLeuAlaLeuValLeuSerAlaLeu151015GlyAlaLeuLeuGlyThrGluAlaLeuArgAlaGluGluProAlaVal20 2530GlyThrSerGlyLeuIlePheArgGluAspLeuAspTrpProProGly354045IleProGlnGluProLeu CysLeuValAlaLeuGlyGlyAspSerAsn505560GlySerSerSerProLeuArgValValGlyAlaLeuSerAlaTyrGlu6570 7580GlnAlaPheLeuGlyAlaValGlnArgAlaArgTrpGlyProArgAsp859095LeuAlaThrPheGlyVal CysAsnThrGlyAspArgGlnAlaAlaLeu100105110ProSerLeuArgArgLeuGlyAlaTrpLeuArgAspProGlyGlyGln115 120125ArgLeuValValLeuHisLeuGluGluValThrTrpGluProThrPro130135140SerLeuArgPheGlnGluProProProG lyGlyAlaGlyProProGlu145150155160LeuAlaLeuLeuValLeuTyrProGlyProGlyProGluValThrVal165 170175ThrArgAlaGlyLeuProGlyAlaGlnSerLeuCysProSerArgAsp180185190ThrArgTyrLeuValLeuAl aValAspArgProAlaGlyAlaTrpArg195200205GlySerGlyLeuAlaLeuThrLeuGlnProArgGlyGluAspSerArg210215 220LeuSerThrAlaArgLeuGlnAlaLeuLeuPheGlyAspAspHisArg225230235240CysPheThrArgMetThrProAla LeuLeuLeuLeuProArgSerGlu245250255ProAlaProLeuProAlaHisGlyGlnLeuAspThrValProPhePro260 265270ProProArgProSerAlaGluLeuGluGluSerProProSerAlaAsp275280285ProPheLeuGluThrLeuThrArg LeuValArgAlaLeuArgValPro290295300ProAlaArgAlaSerAlaProArgLeuAlaLeuAspProAspAlaLeu305310 315320AlaGlyPheProGlnGlyLeuValAsnLeuSerAspProAlaAlaLeu325330335GluArgLeuLeuAspGlyGluG luProLeuLeuLeuLeuLeuArgPro340345350ThrAlaAlaThrThrGlyAspProAlaProLeuHisAspProThrSer355 360365AlaProTrpAlaThrAlaLeuAlaArgArgValAlaAlaGluLeuGln370375380AlaAlaAlaAlaGluLeuArgSerLeuProGl yLeuProProAlaThr385390395400AlaProLeuLeuAlaArgLeuLeuAlaLeuCysProGlyGlyProGly405 410415GlyLeuGlyAspProLeuArgAlaLeuLeuLeuLeuLysAlaLeuGln420425430GlyLeuArgValGluTrpArgGly ArgAspProArgGlyProGlyArg435440445AlaGlnArgSerAlaGlyAlaThrAlaAlaAspGlyProCysAlaLeu450455 460ArgGluLeuSerValAspLeuArgAlaGluArgSerValLeuIlePro465470475480GluThrTyrGlnAlaAsnAsnCysGln GlyValCysGlyTrpProGln485490495SerAspArgAsnProArgTyrGlyAsnHisValValLeuLeuLeuLys500 505510MetGlnAlaArgGlyAlaAlaLeuAlaArgProProCysCysValPro515520525ThrAlaTyrAlaGlyLysLeuLeuIleS erLeuSerGluGluArgIle530535540SerAlaHisHisValProAsnMetValAlaThrGluCysGlyCysArg54555055 5560(2) INFORMATION FOR SEQ ID NO:2:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 575 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(ii) MOLECULE TYPE: protein(xi) SEQUENCE DESCRIPTION: SEQ ID NO:2:MetProGlyProSerLeuSerLeuAlaLeuValLeuSerAlaMetGl y151015AlaLeuLeuArgProGlyThrProArgGluGluValPheSerThrSer202530 AlaLeuProArgGluGlnAlaThrGlySerGlyAlaLeuIlePheGln354045GlnAlaTrpAspTrpProLeuSerSerLeuTrpLeuProGlySerPro 505560LeuAspProLeuCysLeuValThrLeuHisGlySerGlyAsnGlySer65707580Ar gAlaProLeuArgValValGlyValLeuSerSerTyrGluGlnAla859095PheLeuGluAlaValArgArgThrHisTrpGlyLeuSerAspLeuThr 100105110ThrPheAlaValCysProAlaGlyAsnGlyGlnProValLeuProHis115120125Leu GlnArgLeuGlnAlaTrpLeuGlyGluProGlyGlyArgTrpLeu130135140ValValLeuHisLeuGluGluValThrTrpGluProThrProLeuLeu145 150155160ArgPheGlnGluProProProGlyGlyAlaSerProProGluLeuAla165170175L euLeuValValTyrProGlyProGlyLeuGluValThrValThrGly180185190AlaGlyLeuProGlyThrGlnSerLeuCysLeuThrAlaAspSerAsp 195200205PheLeuAlaLeuValValAspHisProGluGlyAlaTrpArgArgPro210215220GlyLeuAlaL euThrLeuArgArgArgGlyAsnGlyAlaLeuLeuSer225230235240ThrAlaGlnLeuGlnAlaLeuLeuPheGlyAlaAspSerArgCysPhe 245250255ThrArgLysThrProAlaLeuLeuLeuLeuLeuProAlaArgSerSer260265270Al aProMetProAlaHisGlyArgLeuAspLeuValProPheProGln275280285ProArgAlaSerProGluProGluGluAlaProProSerAlaAspPro 290295300PheLeuGluThrLeuThrArgLeuValArgAlaLeuAlaGlyProPro305310315320AlaArg AlaSerProProArgLeuAlaLeuAspProGlyAlaLeuAla325330335GlyPheProGlnGlyGlnValAsnLeuSerAspProAlaAlaLeuGlu 340345350ArgLeuLeuAspGlyGluGluProLeuLeuLeuLeuLeuProProThr355360365AlaAla ThrThrGlyValProAlaThrProGlnGlyProLysSerPro370375380LeuTrpAlaAlaGlyLeuAlaArgArgValAlaAlaGluLeuGlnAla385 390395400ValAlaAlaGluLeuArgAlaLeuProGlyLeuProProAlaAlaPro405410415ProL euLeuAlaArgLeuLeuAlaLeuCysProGlyAsnProAspSer420425430ProGlyGlyProLeuArgAlaLeuLeuLeuLeuLysAlaLeuGlnGly 435440445LeuArgAlaGluTrpArgGlyArgGluArgSerGlySerAlaArgAla450455460GlnArgSerAlaGl yAlaAlaAlaAlaAspGlyProCysAlaLeuArg465470475480GluLeuSerValAspLeuArgAlaGluArgSerValLeuIleProGlu 485490495ThrTyrGlnAlaAsnAsnCysGlnGlyAlaCysGlyTrpProGlnSer500505510AspArg AsnProArgTyrGlyAsnHisValValLeuLeuLeuLysMet515520525GlnAlaArgGlyAlaThrLeuAlaArgProProCysCysValProThr530 535540AlaTyrThrGlyLysLeuLeuIleSerLeuSerGluGluArgIleSer545550555560AlaHisHis ValProAsnMetValAlaThrGluCysGlyCysArg565570575(2) INFORMATION FOR SEQ ID NO:3:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 553 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear( ii) MOLECULE TYPE: protein(xi) SEQUENCE DESCRIPTION: SEQ ID NO:3:MetGlnGlyProHisLeuSerLeuLeuLeuLeuLeuLeuAlaThrMet151015GlyAlaValLeuGlnAlaAspThrVa lGluGluLeuThrAsnThrArg202530GlyLeuIlePheLeuGluAspGlyValTrpProProSerSerProPro3540 45GluProLeuCysLeuValAlaValArgGlyGluGlyAspThrSerLys505560AlaSerLeuThrValValGlyGlyLeuHisSerTyrGl uGlnAlaPhe65707580LeuGluAlaValGlnGluSerArgTrpGlyProGlnAspLeuAlaThr8590 95PheGlyValCysSerThrAspSerGlnThrThrLeuProAlaLeuGln100105110ArgLeuGlyAlaTrpLeuGlyGluThrGlyG luGlnGlnLeuLeuVal115120125LeuHisLeuAlaGluValIleTrpGluProGlnLeuLeuLeuLysPhe130135 140GlnGluProProProGlyGlyAlaSerArgTrpGluGlnAlaLeuLeu145150155160ValLeuTyrProGlyProGlyProGlnValThrVa lThrGlyAlaGly165170175LeuGlnGlyThrGlnSerLeuCysProThrArgAspThrArgTyrLeu180185 190ValLeuThrValHisPheProAlaGlyAlaTrpSerGlySerGlyLeu195200205AlaLeuThrLeuGlnProSerLysGluGlyAlaThr LeuThrIleAla210215220GlnLeuGlnAlaPheLeuPheGlySerAspSerArgCysPheThrArg225230235 240LysThrProThrLeuValLeuLeuProProThrGlyProThrProGln245250255ProAlaHisGlyGlnLeuAspThrValProPhe ProGlnProGlyLeu260265270SerLeuGluProGluAspLeuProHisSerAlaAspProPheLeuGlu275280 285ThrLeuThrArgLeuValArgAlaLeuArgGlyProLeuThrArgAla290295300SerAsnThrArgLeuAlaLeuAspProGlyAlaLeuAlaSerP hePro305310315320GlnGlyLeuValAsnLeuSerAspProValAlaLeuGlyArgLeuLeu325330 335AspGlyGluGluProLeuLeuLeuLeuLeuSerProAlaAlaAlaThr340345350ValGlyGluProMetArgLeuHisSerProThrS erAlaProTrpAla355360365AlaGlyLeuAlaArgArgValAlaValGluLeuGlnAlaAlaAlaSer37037538 0GluLeuArgAspLeuProGlyLeuProProThrAlaProProLeuLeu385390395400SerArgLeuLeuAlaLeuCysProAsnAspSerArgSe rAlaGlyAsp405410415ProLeuArgAlaLeuLeuLeuLeuLysAlaLeuGlnGlyLeuArgAla420425 430GluTrpArgGlyArgGluGlyArgGlyArgAlaGlyArgSerLysGly435440445ThrGlyThrAspGlyLeuCysAlaLeuArgGluLeuSer ValAspLeu450455460ArgAlaGluArgSerValLeuIleProGluThrTyrGlnAlaAsnAsn465470475 480CysGlnGlyAlaCysGlyTrpProTrpSerAspArgAsnProArgTyr485490495GlyAsnHisValValLeuLeuLeuLysMetGlnAla ArgGlyAlaAla500505510LeuGlyArgLeuProCysCysValProThrAlaTyrThrGlyLysLeu515520 525LeuIleSerLeuSerGluGluHisIleSerAlaHisHisValProAsn530535540MetValAlaThrGluCysGlyCysArg545 550(2) INFORMATION FOR SEQ ID NO:4:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 10 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(ii) MOLECULE TYPE: protein(xi) SEQUENCE DESCRIPTION: SEQ ID NO:4:LeuArgAlaGluGluProAlaValGlyThr15 10(2) INFORMATION FOR SEQ ID NO:5:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 10 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(ii) MOLECULE TYPE: protein(xi) SEQUENCE DESCRIPTION: SEQ ID NO:5:AlaAlaGlyAlaThrAlaAlaAspGlyPro15 10(2) INFORMATION FOR SEQ ID NO:6:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 5 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(ii) MOLECULE TYPE: protein(xi) SEQUENCE DESCRIPTION: SEQ ID NO:6:ArgGlyGluAspSer15(2) INFORMATION FOR SEQ ID NO:7:(i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 4 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(ii) MOLECULE TYPE: protein(xi) SEQUENCE DESCRIPTION: SEQ ID NO:7:ArgGlyAspSer

Claims

1. A method of purifying recombinant MIS from host cells capable of expressing recombinant MIS comprising

(a) binding the recombinant MIS to an antibody-chromatography matrix, said antibody being specific to MIS,

(b) substantially removing contaminating enzymes having MIS proteolytic activity or inhibitors of MIS antiproliferative activity by adding to the matrix an effective amount of an alkali metal halide solution wherein said solution contains an effective amount of chelating agent, and

(c) recovering the recombinant MIS by eluting with an acid solution having a pH of between about 2.5 and 4.0.

2. The method of claim 1, further comprising neutralizing the recovered MIS to a pH of between about 6.8 and 7.6.

3. The method of claim 1, wherein the alkali metal halide solution is between about 0.1M and 2.0 M.

4. The method of claim 1, wherein the alkali metal halide is sodium chloride.

5. The method of claim 1, wherein the acid is acetic acid.

6. The method of claim 3, wherein the alkali metal halide solution is between about 0.5M and 2.0M.

7. The method of claim 6, wherein the alkali metal halide solution is about 0.5M.

8. The method of claim 1, wherein said host cells are Chinese hamster ovary cells.

9. The method of claim 1, wherein said chelating agent is EDTA.