WO1991013095A1

WO1991013095A1 - Hypoxia-induced proteins including an activator of coagulation factor x and methods of preparation and use

Info

Publication number: WO1991013095A1
Application number: PCT/US1991/001280
Authority: WO
Inventors: Satoshi Ogawa; David Stern
Original assignee: The Trustees Of Columbia University In The City Of New York
Priority date: 1990-02-28
Filing date: 1991-02-28
Publication date: 1991-09-05
Also published as: AU7489191A

Abstract

This invention provides a purified Factor X activator isolated from endothelial cell membranes, preferably a purified Factor X activator expression of which by a cell is reversibly induced by hypoxia or by sodium azide. An example is a purified Factor X activator comprising a polypeptide characterized by the ability to migrate as a single band and by an apparent molecular weight of between about 50,000 and about 150,000 daltons, preferably about 100,000 daltons, on a non-reduced SDS-polyacrylamide gel, loss of activity from exposure to mercury chloride, and an isoelectric focusing peak at a pH of about 5. Another characteristic of a purified Factor X activator is an optimum Factor X activating effect at a pH range of approximately 6.0-6.8. This invention also provides a purified nucleic acid molecule encoding a Factor X activator, and antibodies directed to Factor X activator.

Description

HYPOXIA-INDUCED PROTEINS INCLUDING AN ACTIVATOR OP COAGULATION FACTOR X AND METH0D8 OF PREPARATION AND USE

The invention described herein was made in the course of work under Grants Nos. HL34625, HL42507, and HL42833 from the Public Health Service, U.S. Department of Health and Human Services. The U. S. Government has certain rights in this invention.

BACKGROUND OF THE INVENTION

Exposure of endothelium to environments with low oxygen tension is a frequent occurrence in various disorders, especially those associated with compromise of the circulation. Two crucial functions of endothelium, maintenance of a permeability barrier and preservation of the fluidity of blood, are adversely affected by levels of hypoxia that can occur in ischemic syndromes (1-8) . Such hypoxia-induced modulation of cellular function, in endothelium and also in other cell types, results from alterations in metabolic/biosynthetic pathways (9-20) . The effect of hypoxia on endothelial cell coagulation mechanisms affords an example of this: expression of the anticoagulant cofactor thrombomodulin is suppressed by hypoxia, while elaboration of fibronectin is maintained, and synthesis of a novel, membrane-associated activator of Factor X, distinct from the intrinsic and extrinsic pathways, is induced (20) . Thus, although total protein synthesis is decreased by =30- 40% after several days of exposure to an environment with p0₂ «14 mm Hg, the effect of hypoxia on individual proteins, subserving some known endothelial functions, reflects specific changes in their synthesis/expression, which may be considerably greater or less than hypoxia-mediated perturbation of overall protein synthesis. This is the common denominator linking many alterations in cellular properties observed in hypoxia (9-20) . Indeed, a distinct set of alternative mechanisms appears to be initiated by the challenge of adapting to a hypoxic microenvironment. For exa ple, previous studies of tumor cells and fibroblasts showed that the cellular response to anoxia included induction of the synthesis of new proteins, termed oxygen regulated proteins (ORPs) (9-13). ORPs could represent post-translational modifications of proteins produced by normoxic cells, or transcription/translation of genes not constitutively expressed in normoxic endothelium. In fibroblasts, anoxia causes induction of the transcription/translation of retrotransposon-like VL30 element gene products (10-11) . In addition, studies of total cell lysates and released products of anoxic fibroblasts using methods similar to ours, led to the identification of a total of nine major proteins not observed in normoxic cells (10) . Surprisingly, none of these proteins appear to be identical to the membrane- associated ORPs observed in hypoxic endothelial cells by two dimensional gel analysis.

We have identified membrane-associated proteins whose synthesis was induced by hypoxia. Our results indicate that hypoxia induces de novo synthesis of membrane-associated proteins which have the potential to modulate cell surface properties of endothelium, as evidenced by the identification and purification of a novel hypoxia-induced activator of coagulation Factor X.

Factor X activator is a new molecule induced on the endothelial cell surface, for instance in blood vessels, by low surface concentrations of oxygen which are known to activate coagulation. Factor X activator acts as a marker for hypoxemic vessel injury _j_ vivo, and antagonists or blocking antibodies to Factor X activator will prevent thrombosis because Factor X activator contributes to the pathogenesis of ischemic vascular injury. A bypass enzyme which activates Factor X has been isolated from tumor tissue, but has not been isolated from normal tissue, and does not have the same physical characteristics as Factor X activator (34) .

Detection of Factor X activator and the other oxygen regulated proteins is a new and unique way to detect hypoxic damage to a blood vessel. No other method of detection is known. Ischemic damage to heart muscle can be detected by thallium scan. However such a method does not suggest the use of Factor X activator, a new molecule, for detection of such damage not only in heart muscle, but in blood vessels and other tissues.

SϋMMARY OF THE INVENTION

This invention provides a purified Factor X activator isolated from endothelial cell membranes, preferably a purified Factor X activator expression of which by a cell is reversibly induced by hypoxia or by sodium azide. An example is a purified Factor X activator comprising a polypeptide characterized by the ability to migrate as a single band and by an apparent molecular weight of between about 50,000 and about 150,000 daltons, preferably about 100,000 daltons, on a non-reduced SDS-polyacrylamide gel, loss of activity on a reduced SDS-polyacrylamide gel, loss of activity on exposure to mercury chloride, and an isoelectric focussing peak at a pH of about 5. Another characteristic of a purified Factor X activator is an optimum Factor X activating effect at a pH range of approximately 6.0 - 6.8.

This invention also provides a purified nucleic acid molecule encoding a Factor X activator.

This invention further provides an antibody directed to the Factor X activator, such as a monoclonal antibody, in particular an antibody directed to an epitope which is an active site of Factor X activator, and an antibody to which a toxin is coupled.

This invention also provides an inhibitor directed to Factor X activator.

This invention provides a pharmaceutical composition comprising an amount of an antibody directed to an epitope which is an active site of Factor X activator effective to inhibit the activity of Factor X activator, or an antibody to which a toxin is coupled and a pharmaceutically acceptable stabilizer. This invention also provides a pharmaceutical composition comprising an amount of an inhibitor directed to Factor X activator effective to inhibit the activity of Factor X activator, and a pharmaceutically acceptable carrier.

This invention provides an oxygen-regulated cell membrane protein which is expressed in endothelial cells after exposure to hypoxic conditions for a period of at least 23 hours.

This invention also provides an isolated nucleic acid molecule which encodes the oxygen-regulated cell membrane protein.

This invention further provides an antibody directed to the oxygen-regulated cell membrane protein.

This invention provides a method of detecting expression of Factor X activator which comprises contacting a cell with a detectably labeled nucleic acid probe derived from a nucleic acid molecule encoding Factor X activator under conditions which permit the probe to hybridize with mRNA encoding Factor X activator, if any such is present, and detecting expression by detecting any hybridization which occurs.

This invention also provides a method of detecting expression of Factor X activator in a cell which comprises contacting the cell with an antibody directed to Factor X activator labeled with a detectable marker under conditions well known in the art which permit the antibody to bind to Factor X activator on the surface of the cell, if any such is present, and detecting expression by detecting the marker.

This invention provides a method of detecting expression of an oxygen-regulated cell membrane protein which comprises contacting a cell with a labeled nucleic acid probe derived from a nucleic acid molecule which encodes the membrane protein under conditions which permit the probe to hybridize with mRNA encoding the cell membrane protein, if any such is present, and detecting expression by detecting any hybridization which occurs.

This invention also provides a method of detecting expression of an oxygen-regulated cell membrane protein which comprises contacting the cell with an antibody directed to the cell membrane protein labeled with a detectable marker under conditions which permit the antibody to bind to Factor X activator on the surface of the cell, if any such is present, and detecting expression by detecting the marker.

This invention provides a method for diagnosis of a hypoxemic condition in a tissue comprising detection of Factor X activator.

This invention also provides a method for diagnosis and localization of a hypoxemic condition comprising injecting into a subject a radioactively labeled antibody directed to Factor X activator and determining its binding location by means of a radiation detector.

This invention provides another method for diagnosis of a hypoxemic condition in a tissue comprising detection of expression of an oxygen-regulated cell membrane protein.

This invention also provides a method for diagnosis and localization of a hypoxemic condition comprising injecting into a subject a radioactively labeled antibody directed to an oxygen-regulated cell membrane protein and determining its binding location by means of a radiation detector. This invention provides a method of treating a subject having a hypoxemic condition which comprises administering to the subject an amount of a pharmaceutical composition comprising an antibody directed an epitope which is an active site of Factor X activator and a pharmaceutically acceptable stabilizer, for example albumin.

This invention also provides a method of treating a subject having a hypoxemic condition which comprises administering to the subject an amount of a pharmaceutical composition comprising an inhibitor of Factor X activator effective to block Factor X activator, and a pharmaceutically acceptable carrier.

This invention further provides a method of treating a subject having a tumor which comprises administering to the subject an amount of a pharmaceutical composition comprising an antibody directed to Factor X activator coupled to a toxin effective to necrotize the tumor, and a pharmaceutically acceptable stabilizer.

This invention provides a method for preparing Factor X activator which comprises recovering and purifying Factor X activator from endothelial cells maintained in hypoxic condition, and also a method for preparing Factor X activator which comprises the use of recombinant DNA technology methods well known in the art. BRIEF DESCRIPTION OF THE FIGURES

Figure 1. Autoradioσra s from two-dimensional σels of membranes derived from hypoxic and normoxic endothelial cultures.

Confluent endothelial cells grown in normoxia were either placed in hypoxia (p0₂ « 14 mm Hg) for 24 (A) , or maintained in normoxia for 24 (B) or 48 hr (C) . ³⁵S-Methionine was added to cultures 8 hr prior to harvesting samples. At the end of the experiment endothelial membranes were prepared and samples were processed for SDS-PAGE (see text for details of methods) . Arrow heads (B and C) in hypoxic gels denote spots whose appearance was induced in hypoxia.

Figure 2. Two-dimensional gel electrophoresis of hypoxic and normoxic endothelial cultures: early and delayed ORPs. and dependence of expression on oxygen concentration.

A-B. Confluent endothelial cultures grown in normoxia were placed in hypoxia (p0₂ ~ 14 mm Hg) for 24 hr or 48 hr, or maintained in normoxia for the same times. ³⁵S-Methionine was added to cultures 8 hr prior to harvesting samples and samples were processed as described in the text. A difference map, based on PDQuest analysis and visual inspection of 20 gels, comparing gels with samples from normoxic and hypoxic cultures at the two time points, led to the definition of two groups of ORPs, those whose expression peaked by 24 hr (early ORPs, A) and those achieving maximal expression by 48 hr (delayed ORPs, B) . The ORPs with their arbitrarily defined numbers are shown on a schematic representation of a 2-dimensional gel with the x-axis representing pH and the y-axis representing Mr's (molecular weights were assigned based on comparison with a semilogarithmic plot constructed from the migration of standards run simultaneously) . C-D. The expression profile, based on laser densitometry of spot intensity using the PDQest program, of one early ORP (#2718, C) and one delayed ORP (#4234, D) is shown. For these experiments, the incubation time was 8, 16, 24 and 48 hr. E-F. Expression of ORPs was also dependent on the oxygen concentration and a plot of the intensity of ORPs #3009 (E) and #608 (F) after a 48 hr incubation period versus the p0₂ of the atmosphere is shown.

Figure 3. Comparison of hypoxia-induced endothelial membrane-associated proteins with those induced by heat shock and glucose deprivation.

A. Endothelial cells were exposed to heat shock (43*C for 3 hr) in the presence of ³⁵S-methionine, and samples were processed for 2-dimensional SDS-PAGE as described in the text. The inset shows a difference map based computer analysis (PDQuest program) and visual inspection of duplicate gels. B. Endothelial cells were incubated in glucose-free medium for 16 hr, an additional 8 hr in the presence of ³⁵S-methionine, and then samples of membranes were obtained for 2-dimensional gel analysis. The inset shows a difference map. The difference map in each of the insets reflects the results of multiple gels and multiple exposures, not just the gel in the main part of the figure.

Figure 4. Effect of tumor necrosis factor/cachectin (TNF) and metabolic inhibitors on expression of membrane associated proteins in endothelium.

Endothelial cells were incubated either with A: TNF (1 nM) , B:2-deoxyglucose (25 mM) , C:fluoride (1 mM) or D:azide (1 mM) for 16 hr and an additional 8 hr in the presence of ³⁵S- methionine. Samples were then prepared for 2-dimensional SDS-PAGE as described in the text. Gels were scanned by and analyzed by the PDQuest program (28) , and the difference i age was made. In panels A-C, difference maps, made from the difference image, are shown. In panel D, the main figure shows the autoradiogram and the inset shows the difference map.

5 . .

Figure 5. Properties of the hypoxia-induced endothelial

Factor X activator: isoelectric focussing.

Endothelial cells («4xl0⁹) were made hypoxic (p0₂ «14 mm Hg) for 48 hr, extracted with veronal buffer, and the extract

10 was subjected to preparative isoelectric focussing.

Fraction number (with the corresponding pH) is plotted versus OD280 (diamond, dashed line) and Factor X activating ability (square, solid line) of the samples. The mean of duplicate determinations is shown in the latter case. 15

Figure 6. Characteristics of the hvpoxia-induced endothelial Factor X activator: SDS-PAGE.

Confluent endothelial onolayers («4xl0⁹ cells) were

20 . incubated for 48 hr in hypoxia (p0₂ «14 mm Hg) and membrane extracts were subjected to SDS-PAGE (10%) . Lanes of the gel were then silver stained (A) or gel eluted and assayed for ability of samples to activate Factor X (B) . The latter data is plotted as gel slice number versus Factor Xa formed

25

(ng/ml) . C. Endothelial cells were incubated for 48 hr in hypoxia (pθ₂ «14 mm Hg) and membrane extracts were subjected to isoelectric focussing. The fractions with peak activity were then run on preparative SDS-PAGE (10%) , slices of the gel were electroeluted, and re-run on analytical SDS-PAGE (10%) . Lanes of the latter gel were either silver stained (C) , or gel eluted and assayed for ability of samples to activate Factor X (D) . The gel eluted material in lane D was subjected to reduced SDS-PAGE (lane E) . Apparent molecular weights, shown by the arrows, were interpolated from semi-logarithmic plots based on the migration of standard proteins run simultaneously.

Figure 7. Properties of Factor X activation by hypoxic endothelial cells: the effect of pH and reversibility.

A. Endothelial monolayers grown in normoxia were incubated in hypoxia (pθ₂ «14 mm Hg) for 48 hr. Cultures were washed and then incubated in buffer with either pH 6.8 (triangles) or 7.4 (circles) in the presence of the indicated concentration of Factor X. Factor Xa formation was assessed as described in the text and is shown as ng/ml/min. Inset: Double reciprocal plots of the data in each panel calculated using nonlinear regression analysis applied to the Michaelis-Menten equation. B. Endothelial monolayers were made hypoxic as described in A, and incubated with Factor X (1 μM) in buffer with the indicated pH for 30 min at 37*C. Factor Xa formation was then assessed as described in the text. Data are reported as per cent of the Factor X activation seen in reaction mixtures performed at pH 7.4 (ng/ml/10⁵ cells) . C. Endothelial monolayers were incubated for 72 hr in hypoxia (p0₂ »14 mm Hg) or normoxia, and then exposed to an ambient air atmosphere for the indicated time intervals. Factor Xa formation (ng/ml/10⁵ cells) was then studied after the addition of Factor X (1 μM) for 30 min at 37*C as described in the text. Bars represent: N72= cultures exposed to normoxia for 72 hr; H72= cultures exposed to hypoxia for 72 hr; R12=cultures exposed to hypoxia for 72 hr followed by normoxia for 12 hr; R24= cultures exposed to hypoxia for 72 hr followed by normoxia for 24 hr; and R48 cultures exposed to hypoxia for 72 hr followed by normoxia for 48 hr. The mean + S.D. is shown. Figure 8. Characterization of the hypoxia-induced Factor X activator of native endothelium: SDS-PAGE and isoelectric focussing.

A. Vessel segments were exposed to hypoxia (p0₂ »14 mm Hg) for 24 hrs, the endothelial layer was removed by scraping, and cells were prepared for SDS-PAGE as described in the text. Electroeluted gel samples, indicated by the corresponding gel slice, were tested for their ability to activate Factor X. The migration of molecular weight standards run simultaneously is indicated by the arrows. B. Hypoxic endothelial cells from vessel segments were prepared as in A, and subjected to isoelectric focussing. Samples were then tested for their ability to activate added Factor X (1 μM) . In A-B above. Factor Xa formation is expressed as shortening of the clotting time observed when reaction mixtures containing samples from aortic segments, previously incubated with Factor X (1 μM) for 30 min. at 37*c, were added to cephalin. Factor VII/X deficient plasma, and calcium chloride, as described in the text. The mean of duplicate determinations is shown.

DETAILED DESCRIPTION OF THE INVENTION

This invention also provides a purified nucleic acid molecule encoding a Factor X activator. Examples of such nucleic acid molecules are a cDNA molecule encoding a Factor X activator, an isolated genomic DNA molecule encoding a Factor X activator, and an RNA molecule encoding a Factor X activator. Such nucleic acid can be readily obtained by one skilled in the art utilizing well known methods, e.g., the preparation of oligonucleotide probes and the use of such probes to obtain the nucleic acids, such probes being based in turn upon the amino acid sequence of Factor X activator, which may be readily obtained using conventional methods of peptide sequencing. For example, the amino acid sequence of Factor X activator is determined by cleaving the protein into its component peptide fragments, then reacting each peptide fragment with known chemicals that remove and identify individual amino acids starting from either the N- terminal or the C-terminal end, and finally determining how the peptide fragments overlap to assemble the complete sequence of the protein. The whole procedure has been automated and may be performed on a machine. When the amino acid sequence is known, a DNA probe encoding any part of this sequence is made on a DNA synthesizer. mRNA purified from cells known to express Factor X activator, for example endothelial cells maintained in hypoxic conditions, is then contacted with the probe, and the mRNA that hybridizes to the probe is isolated. cDNA is made from the mRNA with reverse transcriptase, and can be used as a probe for genomic DNA encoding Factor X activator, and also sequenced.

This invention further provides an antibody directed to the Factor X activator, such as a monoclonal antibody, in particular an antibody directed to an epitope which is an active site of Factor X activator and such an antibody to which a toxin, such as diptheria toxin, is coupled. Such antibodies can be readily obtained by one skilled in the art utilizing well known methods. For example, antibodies to Factor X activator are raised by injecting the protein into an animal, isolating antiserum from the animal, and using well-known immunological techniques such as precipitation reactions to purify antibodies directed to Factor X activator. Monoclonal antibodies are obtained by isolating B-lymphocytes from the injected animal, selecting those which produce antibodies directed to Factor X activator, fusing them with myeloma cells, and thereby producing hybridomas which produce antibodies directed to Factor X activator. Antibodies directed to an epitope which is an active site of Factor X activator are selected by combining Factor X activator with Factor X, measuring the amount of activated Factor X produced, and then determining which antibody blocks the appearance of activated Factor X. Another method is combining fragments of Factor X activator separately with Factor X to determine which fragments of the protein are capable of activating Factor X, and raising antibodies directed to the active fragment. Yet another method is to analyse the sequence of Factor X activator to determine which part of the protein is most likely to interact with Factor X, synthesize peptides which correspond to that part of the protein, and raise antibodies directed to these peptides. Toxins are coupled to antibodies using methods well known in the art, such as the use of intermediate coupling molecules.

This invention also provides a molecule which acts as an inhibitor of Factor X activator. The inhibitor is isolated by in vitro binding assays with purified Factor X activator using methods well known in the art. For example, natural or synthetic molecules are contacted with Factor X activator to determine whether they are capable of binding it. Then, a molecule is tested for inhibitory activity by combining Factor X activator with Factor X, measuring the amount of activated Factor X produced, and then determining whether the molecule blocks the appearance of activated Factor X. Another method is to analyse the sequence of Factor X activator to determine which part of the protein is most likely to interact with Factor X and synthesize molecules which will bind to that part of the protein, thus inactivating Factor X activator. The inhibitor selected can then be tested for resistance to digestive acids and enzymes to determine its suitability for oral ingestion.

This invention provides a pharmaceutical composition comprising an amount of an antibody directed to an epitope which is an active site of Factor X activator effective to inhibit the activity of Factor X activator and thereby to prevent Factor X activation by Factor X activator, and a pharmaceutically acceptable stabilizer. The antibody binds to the active site of Factor X activator in such a way that Factor X activator cannot bind and thereby activate Factor X. Factor X activator induces clot formation by activating Factor X, which in its active state catalyzes the conversion of prothro bin to thro bin. Thrombin catalyzes the conversion of fibrinogen to fibrin, the material of a clot. An example of a pharmaceutically acceptable stabilizer is the protein albumin. The pharmaceutical composition is most effective when administered by injection, since antibodies have fairly long half lives in the circulation.

This invention also provides a pharmaceutical composition comprising an antibody directed to Factor X activator to which a toxin such as diptheria toxin, has been coupled, and a pharmaceutically acceptable stabilizer, such as albumin.

This invention further provides a pharmaceutical composition comprising an amount of an inhibitor directed to Factor X activator effective to inhibit the activity of Factor X activator and thereby to prevent Factor X activation by Factor X activator, and a pharmaceutically acceptable carrier. Examples of pharmaceutically acceptable carriers are distilled water, saline, pH and ionically balanced fluids, and solids such as capsules and tablets. The composition is preferably suitable for oral administration.

This invention provides an oxygen-regulated cell membrane protein which is expressed in endothelial cells after exposure to hypoxic conditions, for example a P0₂ of 12-14 mm Hg, for at least 23 hours.

This invention also provides an isolated nucleic acid molecule which encodes the oxygen-regulated cell membrane protein, such as a cDNA molecule, an isolated genomic DNA molecule, and an RNA molecule. Methods of obtaining such nucleic acid molecules are well known in the art, and are described in more detail supra.

This invention further provides an antibody directed to the oxygen-regulated cell membrane protein, such as a monoclonal antibody. Methods of obtaining such an antibody are well known in the art, and are described in more detail supra.

This invention provides an oxygen-regulated cell membrane protein identified in Figure 2(A) as, for example, #311, #2718, #1530, #1333, #2326, #3009, #5417, #5418, #3328, #8327 and in Figure 2(B) as #403, #688, #1605, #1308, #2227, #2708, #4234, #4420, #6220, #7420, and #8415.

This invention provides a method of detecting expression of Factor X activator which comprises contacting a cell with a detectably labeled nucleic acid probe derived from a nucleic acid molecule encoding Factor X activator under conditions which permit the probe to hybridize with mRNA encoding Factor X activator, if any such is present, and detecting expression by detecting any hybridization which occurs. Such procedures are well known in the art. For example, cells are homogenized and mRNA is extracted from the homogenate by a combination of centrifugation, separation, and column purification. The mRNA is then gel electrophoresed, and the labeled probe is contacted with the mRNA, and binds to complementary strands. The resulting mRNA is then detected by the label on the probe. If the cell is expressing Factor X activator, then the probe will hybridize. If not, no hybridization will occur.

This invention also provides a method of detecting expression of Factor X activator in a cell which comprises contacting the cell with an antibody directed to Factor X activator labeled with a detectable marker under conditions well known in the art which permit the antibody to bind to Factor X activator on the surface of the cell, if any such is present, and detecting expression by detecting the marker using methods well known in the art. For example, if the antibody is fluorescence-labeled, then an amount of the antibody sufficient to label cells is added to a suspension of cells in medium on a slide. After incubation, the slide is observed in a fluorescence microscope which activates the fluorescence label. Any cell expressing Factor X activator has labeled antibody bound to its surface, and the label when stimulated to fluorescence is visible to the eye. Presence of visible label indicates the expression of Factor X activator.

This invention provides a method of detecting expression of an oxygen-regulated cell membrane protein which comprises contacting a cell with a labeled nucleic acid probe derived from a nucleic acid molecule which encodes the oxygen- regulated cell membrane protein under conditions which permit the probe to hybridize with mRNA encoding the cell membrane protein, if any such is present, and detecting expression by detecting any hybridization which occurs. These methods are well known in the art, and are discussed supra in more detail.

This invention also provides a method of detecting expression of an oxygen-regulated cell membrane protein which comprises contacting the cell with an antibody directed to the oxygen-regulated cell membrane protein labeled with a detectable marker under conditions well known in the art which permit the antibody to bind to Factor X activator on the surface of the cell, if any such is present, and detecting expression by detecting the marker using methods well known in the art, which are discussed in more detail supra.

This invention provides a method for diagnosis of a hypoxemic condition in a tissue comprising detection of Factor X activator by either of the methods described supra. namely by detecting mRNA encoding Factor X activator, or by using an antibody directed to Factor X activator. Factor X activator is induced by hypoxic conditions, therefore its presence in a tissue, for example in cells taken from the endothelial vascular tissue of a subject, indicates exposure to hypoxic conditions in that tissue. The condition is then treated as described infra.

This invention also provides a method for diagnosis and localization of a hypoxemic condition comprising injecting into a subject a radioactively labeled antibody directed to Factor X activator and determining its binding location by means of a radiation detector. The labeled antibody will bind to Factor X activator which is being expressed in an area of hypoxia, indicating hypoxic conditions in that area. The condition is then treated as described infra. In particular this localization is useful to visualize a tumor which causes hypoxia in surrounding tissues.

This invention provides another method for diagnosis of a hypoxemic condition in a tissue comprising detection of expression of an oxygen-regulated cell membrane protein by either of the methods described supra. namely detection of expression by isolating mRNA or by use of an antibody. Expression of a protein induced by hypoxia indicates hypoxemic conditions in the tissue to which the expressing cells belong. As discussed infra, hypoxemic conditions are dangerous and can lead to the formation of harmful clots and consequent tissue injury. Therefore it is important to detect hypoxia and treat its causes as soon as possible.

This invention also provides a method for diagnosis and localization of a hypoxemic condition comprising injecting into a subject a radioactively labeled antibody directed to an oxygen-regulated cell membrane protein and determining its binding location by means of a radiation detector. The labeled antibody will bind to the protein, which is being expressed in an area of hypoxia, indicating hypoxic conditions in that area. The condition is then treated as described infra.

This invention provides a method of treating a subject having a hypoxemic condition which comprises administering to the subject an amount of a pharmaceutical composition comprising an antibody directed an epitope which is an active site of Factor X activator and a pharmaceutically acceptable stabilizer, for example albumin. Factor X activator is induced in endothelium as a reaction to lack of oxygen and acidity. The antibody binds to the active site of Factor X activator, preventing it from activating Factor X and thereby causing clot formation as discussed infra.

This invention also provides a method of treating a subjec^* having a hypoxemic condition which comprises administer! to the subject an amount of a pharmaceutical composi' comprising an inhibitor of Factor X activator effective block Factor X activator, and a pharmaceutically acceptable carrier.

Factor X activator is induced by hypoxic conditions, therefore its presence in cells, for example in cells taken from the endothelial vascular tissue of a subject, indicates exposure to hypoxic conditions. Hypoxemia can result from various conditions, all of which can cause serious tissue damage, for example blood vessel wall injury, if not detected and reversed. A tumor can cause hypoxia. Hypoxia can also occur within a superficially healed wound that contains large areas of damaged vasculature. Thrombosis (clot formation) is also evidenced by hypoxia. All these conditions lead to ischemia, or reduced blood flow to an area. Reduced blood flow means a hypoxic condition will develop. Early detection of Factor X activator expression allows the condition to be corrected before damage occurs. One means of correcting the condition is to inactivate Factor X activator. Factor X activator activates Factor X in the abnormal circumstance of hypoxia, which when activated itself activates prothrombin to thrombin, which in turn cleaves fibrinogen to fibrin, producing a clot. The clot can cause damage to tissues such as vascular tissues, for example blood vessel walls. Therefore it is important to block Factor X activator. Although the clot may already exist, as in thrombosis, further clot formation will be prevented by inactivating Factor X activator, and clot dissolution will be initiated. Conditions for clot maintenance are in dynamic balance with each other. Therefore a circumstance which reduces clot formation will shift the balance to favor clot dissolution. This means that inactivating Factor X activator will shift the balance from clot formation to clot dissolution, preventing further tissue damage. Administering the pharmaceutical compositions described supra to a subject in the very early stages of heart attack can avert the hypoxia-induced damage that leads to tissue death.

This invention further provides a method of treating a subject having a tumor which causes hypoxia which comprises administering to the subject an amount of a pharmaceutical composition comprising an antibody directed to Factor X activator coupled to a toxin such as diptheria toxin effective to necrotize the tumor, and a pharmaceutically acceptable stabilizer. The antibody will bind to Factor X activator in the area of the tumor, and the attached toxin will kill the tumor.

This invention provides a method for preparing Factor X activator which comprises inducing endothelial cells to express Factor X activator by maintaining them under hypoxic conditions or exposing them to sodium azide, recovering the protein from the resulting cells, and purifying the protein so recovered. A detailed example of a method for preparing Factor X activator is provided infra in Experimental Details.

This invention also provides a method for preparing Factor X activator by use of recombinant DNA technology which comprises the use of methods well known in the art. For example, isolated nucleic acid encoding Factor X activator is inserted in a suitable vector, such as an expression vector. A suitable host cell, such as a bacterial cell, or a eucaryotic cell such as a yeast cell, is transfected with the vector. Factor X activator is isolated from the culture medium by affinity purification or by chromatography or by other methods well known in the art.

This invention will be better understood by reference to the Experimental Details which follow, but those skilled in the art will readily appreciate that the specific experiments detailed are only illustrative of the invention, and should not be construed to limit in any way the invention as set forth in the claims which follow thereafter.

Expβrimental Details

Exposure of cultured endothelial cells to hypoxia (p0₂ « 14 mm Hg) does not decrease cell viability, but leads to subtle changes in cellular function, including alterations in cell surface properties, and enhanced/induced synthesis of membrane-associated proteins. In two-dimensional gel analysis of membrane fractions from metabolically labelled cultures two groups each of «10 new spots were identified on autoradiograms derived from hypoxic endothelial cells: one group, whose expression was maximal after 24 hr, and a second group whose expression continued to increase up to 48 hr. The pattern of hypoxia-induced protein induction was distinct from that observed in endothelium responding to heat shock, glucose deprivation or exposure to tumor necrosis factor, but could be duplicated, in part, by sodium azide. Hypoxia is associated with a prothrombotic tendency. Included in the hypoxia-induced endothelial surface proteins is a novel activator of Factor X which was purified to homogeneity from membranes of hypoxic endothelial cells. This was shown by sequential isoelectric focussing and preparative SDS-PAGE to be a polypeptide, Mr «100 kDa and pi «5.0, distinct from intrinsic and extrinsic pathway coagulation factors. The Factor X activator was reversibly induced by hypoxia; and its effectiveness for generating Factor Xa increased under acidic conditions. Although heat shock, cytokines, glucose deprivation and other stimuli failed to induce this Factor X activator, sodium azide did, indicating that inhibition of respiration is closely linked to its induction. Taken together, these results indicate that hypoxia elicits a novel biosynthetic response which includes the expression of new endothelial cell surface molecules. These serve as markers of hypoxemic vessel wall ⁱⁿJ^ury- Materials and Methods

Cell culture, labelling of endothelial cells under hypoxic conditions, and two-dimensional gel analysis. Bovine aortic endothelial cells were grown from aortas of newborn calves in minimal essential medium supplemented with penicillin- streptomycin (100 U/ml-100 μg/ml) , HEPES (pH 7.4; 10 mM) , glutamine and fetal calf serum (10%; Hyclone, Logan, Utah), as described (21-22) . Cultures were characterized as endothelial based on morphologic criteria, and immunofluorescence for protein S, von Willebrand factor and thrombomodulin (23-25) . Bovine vascular smooth muscle cells, also obtained from aortas by further scraping of the vessel surface after removal of the endothelium, were grown in the same medium as endothelial cells and characterized by morphologic criteria and absence of endothelial markers.

After growing to confluence under normoxic conditions (7-10 days post-plating; labelling index <1%; 2xl0⁵ cells/cm²), endothelial cultures were washed, and fresh medium which had been equilibrated with a gas mixture containing the desired concentration of oxygen (the balance of the atmosphere was made up by 5% carbon dioxide and nitrogen) was added. Prior to addition of this medium to the cells, dissolved gas analysis was done (Model ABL-2 from Radiometer, Copenhagen, Denmark) . Cultures were incubated under hypoxia in a chamber which maintained a humidified atmosphere with low oxygen concentrations (Coy Laboratory Products, Ann Arbor, MI) (20) . The oxygen content was continuously monitored and regulated using an oxygen analyzer which controlled an automated valve system. Experiments were carried to completion in the hypoxia chamber, including addition of ³⁵S- methionine, without exposing cultures to ambient air. At intervals throughout these experiments, the oxygen content of culture medium bathing the cells was analyzed. Values shown in the figures are the oxygen pressure of the medium. During the course of these experiments, pH of the medium remained constant and the glucose concentration fell from 10 mM at the start to 5 mM at the end of the experiment (48 hr) . This degree of glucose depletion or even glucose-free medium did not induce the Factor X activator or hypoxia- induced endothelial proteins (see below) . Glucose and lactate concentrations of the medium were assayed using a commercially available kit (Sigma, St. Louis, MO) . Endothelial viability was assessed morphologically and by trypan blue exclusion.

Metabolic labelling of endothelium was performed by incubating cultures in methionine-poor minimal essential medium (methionine content about 10% of that in normal medium and a glucose concentration of 10 mM; Gibco, Grand Island, NY) containing 5% dialyzed fetal calf serum (Hyclone) . Radiolabel, ³⁵S-methionine (>800 Ci/mmol; final concentration, 0.2 mCi/ l) , was added for 8 hrs before the end of the experiment.

In selected experiments, cultures were incubated under normoxic conditions in the presence of either sodium azide (1 mM) , sodium fluoride (1 mM) , 2-deoxyglucose (25 mM) purified recombinant human tumor necrosis factor/cachectin-α (1 nM; «10⁸ U/mg; provided by Dr. Peter Lo edico, Hoffmann- LaRoche, Nutley, NJ) , or glucose-free medium (in the latter case the serum was dialyzed to remove glucose) . Concentrations of the metabolic inhibitors were chosen based on the following observations. ATP levels of endothelial cells maintained under hypoxia (pθ₂ «14 mm Hg) for 48 hr were similar to those in normoxic cells in the presence of 1 mM sodium azide, 1 mM sodium fluoride or 25 mM 2- deoxyglucose (ATP levels were measured using the luciferase assay [26]). Under the latter conditions, endothelial viability was maintained, based on trypan blue viability, and changes in ATP levels were reversible when the inhibitors were removed (manuscript in preparation) . Heat- shock was performed in normoxia by heating cultures to 42*C for 3 hr. For heat shock studies, ³⁵S-methionine (0.4 mCi/ml) was added 2 hr before harvesting the cultures. In all of the metabolic labelling studies, endothelial cultures were incubated for varying periods at 37 'C in the presence of ³⁵S-methionine and samples were obtained at intervals, including 8, 16, 24 and 48 hr.

At the end of an experiment, cells were scraped from culture vessels and endothelial cell membranes were collected as follows: cultures were washed twice in ice cold serum-free medium, cells were scraped from the growth surface and suspended in 1 ml of Tris (10 mM; pH 7.8)/aprotonin (100 U/ml) , and Dounce homogenized ten times (4 *C) . Then, the same amount of buffer containing Tris (10 mM; pH 7.8)/NaCl (0.3M)/KC1 (10 mM)/CaCl₂ (5 mM)/MgCl₂ (2 mM)/aprotonin (100 U/ml) , was added, the homogenate was centrifuged at lOOOxg for 15 min (4'C), and the supernatant pelleted by ultracentrifugation (100,000xg for 2 hr) . The membrane-rich pellet was resuspended in reduced SDS-gel sample buffer (each sample was derived from «10⁶ cells and was solubilized in 0.1 ml of buffer). Two dimensional gel analysis of membrane-associated proteins, isoelectric focussing in the first dimension followed by reduced SDS- PAGE in the second dimension (27) , was carried out by Protein DataBase, Inc. (28) (Huntington Station, NY) .

Characterization of the hypoxia-induced Factor X activator. The ability of hypoxic endothelial cells to activate Factor X was studied using purified bovine Factor X (100 U/mg; Factor X₁ was used in most studies, although results were identical with Factor X₂) (29-30) and coagulant assays to measure Factor Xa formation, as described previously (20, 30) . In brief, intact monolayers, membrane fractions or samples from the purification procedure described below were incubated with Factor X (1 μM or the indicated concentration) for the times described at 37 " C . Then, an aliquot of the reaction mixture was removed, added to Factor VII/X deficient plasma (Sigma, St. Louis, MO) (60 μl) along with cephalin (60 μl) and CaCl₂ (20 mM; 60 μl) , and clot 5 formation was assessed as described previously (20) . The specificity of this assay has been studied using a blocking monoclonal antibody to bovine tissue factor (see below and [20] and neutralizing antibodies to bovine Factors IX and VIII (see below and [20]). Formation of Factor Xa by

-_jQ coagulant assay correlates with cleavage of the added Factor X with formation of the heavy and light chains, as assessed by SDS-PAGE, and amidolytic activity, as assessed by chromogenic substrate assays (20) . The amount of Factor Xa formed was determined by comparison with a standard curve

15 made with purified Factor Xa. Monospecific polyclonal antibodies to bovine coagulation factors (VII, VIII, IX) and tissue factor were obtained as described previously (20) .

Purification of the Factor X activator included isoelectric

20 focussing (Rotofor Cell, Biorad, Richmond, CA) and SDS-PAGE (31) . In each case, hypoxic endothelial cells («4X10⁹) were processed as follows: monolayers were washed three times in balanced salt solution, harvested by scraping the cells in barbital-buffered saline, and resuspended in veronal buffer 5 (20 mM; pH 7.8) containing PMSF (2 mM) . Extraction was performed three times for 1 hr at 4'C in veronal buffer (5ml/extraction) (20) , cells were pelleted by centrifugation (2000xg for 5 min) and the supernatants were used as follows. For isoelectric focussing, «10 mg of protein was _Q diluted in 50 ml of a solution containing 1.5% ampholyte (Bio-Rad, pH range 3-10)/0.1% octyl-β-glucoside and isoelectric focussing was performed at 12 watts for 4 hr until 1200 volts, the limit, was achieved. Fractions were dialyzed against 0.4 M Tris/HCl (pH 7.5)/0.2 M NaCl/0.1% ₅ octyl-β-glucoside, diluted in barbital-buffered saline, and then tested for their ability to activate Factor X. For SDS-PAGE, extracts of similar numbers of endothelial cells were run on non-reduced 10% gels (31) . Proteins in the gel were either visualized by silver staining (Biorad kit) or electroeluted using an ISCO Sample concentrator (Lincoln, Nebraska) in buffer containing 68 nM N-ethyl orpholine (pH 8.6)/0.05% SDS 3 times for 2 hours at 1 watt. Prior to assessing Factor X activity of gel eluted material, SDS was removed from samples by the method of Konigsberg and Henderson (32) .

Factor X activation by hypoxic endothelial cells was studied at different pH's in the reaction mixture by incubating endothelium with the indicated concentration of Factor X, and assessing Factor Xa formation as described previously (20) . The reversibility of Factor X activator expression was studied by exposing endothelium to hypoxia (pθ₂ «14mm Hg) for 72 hr and then replacing cultures in an ambient air atmosphere for the indicated times.

Experiments with aortic vessel segments (six segments, each«7 cm in length) were carried out by incubating the vessel segment for 24 hr in culture medium under hypoxic conditions (pθ₂ «14 mm Hg) , and then removing the endothelial monolayer by gently scraping with a scalpel. Extracts were then prepared and samples were processed as described above for cultured endothelial cells.

Results

Induction of endothelial cell membrane-associated ORPs. Cell viability of endothelial cells grown to confluence in ambient air then exposed to hypoxia was not affected based on trypan blue exclusion or ability to proliferate after further subculturing in normoxia. However, hypoxia induced the appearance of new proteins, termed by others oxygen regulated proteins or ORPs (9-13) . These are discernible in autoradiograms of 2-dimensional gels of membrane-associated proteins from metabolically labelled cultures. Based on the time course of their appearance after initiation of hypoxic conditions, ORPs could be divided into two groups (Fig. 1) . Exposure of endothelial cells to p0₂ «14 mm Hg for 24 hr resulted in the appearance of approximately 10 ORPs (Fig. IB) . After 48 hr in hypoxia, an additional 10 ORPs were detected (Fig. 1C) , which were unapparent or minimal in normoxic controls (Fig. 1A) . From the analysis of several experiments («20 gels) , it was evident that expression of ORPs consistently fell into these two groups and led us to assign the names early ORPs to those proteins whose expression was maximal by 24 hr, and delayed ORPs, to those proteins whose expression continued to increase up to 48 hr (Fig. 2A-2B) . The time course of expression of one representative ORP from each of these groups demonstrated that maximal expression of the early ORP (ORP #2718; Fig. 2C) occurred before the delayed ORP (ORP #4234; Fig. 2D).

Expression of ORPs was also dependent on the oxygen concentration in the atmosphere. Although several of the ORPs were induced in a stepwise manner (Fig. 2E) as the oxygen content fell, others appeared only at the lowermost oxygen tensions content (Fig. 2F) . Although these results suggest that hypoxia specifically induced the expression of ORPs, it was important to compare the hypoxic response to that observed with other recognized cellular perturbations, such as heat shock, exposure to glucose deprivation or the cytokine tumor necrosis factor/cachectin- . Two-dimensional gel analysis of endothelial membranes following each of the latter three perturbations demonstrated a complex pattern (Fig. 3A-B, 4A) . There was no overlap between the patterns of protein expression in hypoxia and those observed after these three perturbations (Fig. 3A-B, 4A) , based on computer analysis and visual inspection of the gel patterns. To examine metabolic pathways which could be involved in the induction of ORPs, endothelial cells were incubated with either 2-deoxyglucose, an inhibitor of glycolysis, fluoride, an inhibitor of both glycolysis and fatty acid oxidation, or azide, an inhibitor of respiration at the cytochrome a terminus of the electron transport chain (33), and two- dimensional gel analysis was carried out as for hypoxic cultures (Fig. 4B-D) . Each inhibitor induced different patterns of membrane-associated proteins. Fluoride (l M<) (Fig. 4C) induced six visible spots whose pattern was different from that of ORPs, proteins induced by glucose- deprivation and heat shock proteins. 2-deoxyglucose (Fig. 4B) induced six visible spots; #8617 migrated identically to one glucose-deprivation induced protein, and #5442 migrated identically to one fluoride-induced protein. Among these three inhibitors, only azide induced spots identical to oxygen regulated proteins (#311, #608, #7420, #8416) (Fig 4D) , suggesting that inhibition of respiratory chain may trigger the induction of these ORPs.

Previous studies have indicated that endothelial cells exposed to low oxygen tensions elaborate activities which can modulate mitogenesis of vascular smooth muscle cells (16-17) . This led us to compare the expression of ORPs observed in endothelium with that found in vascular smooth muscle cells. After 24 hr in hypoxia (p0₂ =--14 mm Hg) , smooth muscle cultures remained viable, and autoradiograms from two-dimensional gels did not show ORPs similar to those seen in endothelial cultures (data not shown) . After 48 hr in hypoxia, smooth muscle cell viability was reduced by «80- 90%, protein synthesis was similarly suppressed, and it was difficult to obtain samples suitable for high resolution two-dimensional gels (presumably due to cell death and degradation of proteins) . Thus, the endothelial biosynthetic response to hypoxia differs in many respects from that observed with vascular smooth muscle cells under the same conditions.

Properties of the hypoxia-induced endothelial Factor X activator.

Hypoxemia is associated with abnormalities in the coagulation system, and recent studies have shown that hypoxia perturbs endothelial cell surface coagulant activities (20) . Depression of thrombomodulin is associated with the induction of a membrane-associated, novel Factor X activator. Appearance of Factor X activator activity in endothelium required both hypoxia and de novo protein synthesis (20) , prompting us to examine if it could be related to an ORP.

Isoelectric focussing of membrane extracts from hypoxic endothelial cells demonstrated one peak of Factor X activator activity centered about pH «5.0 (Fig. 5) and separated from the bulk of the protein. Non-reduced SDS- PAGE of similar membrane preparations from hypoxic cultures demonstrated the expected complex pattern when stained for protein (Fig. 6, lane A), but gel elution indicated a single peak of Factor X activator activity corresponding to Mr «100 kDa (Fig. 6, lane B) . Employing sequential preparative isoelectric focussing, SDS-PAGE, electroelution and analytical SDS-PAGE, silver staining of non-reduced gels showed a single band with Mr *100 kDa (Fig. 6, lane C) . Factor X activating capacity of gel eluted material co- migrated with Mr «100 kDa (Fig. 6, lane D) . On reduced gels there was no significant change in migration, and one major band was visible with Mr «100 kDa (Fig. 6, lane E) , but no activity could be detected. Similarly, exposure of purified Factor X activator to reducing agents blocked its activity (data not shown) . PMSF had no effect on Factor X activation by Factor X activator, but HgCl₂ inhibited it, suggesting that the active site involves a sulfhydryl group. Factor X activator activates purified Factor X, but does not cleave Factors IX or III or protein S.

Under nonreducing conditions. Factor Xa formation by the purified hypoxia-induced endothelial Factor X activator was not affected by blocking antibodies to tissue factor. Factors VII, VIII, or Factor IX (Table 1). This is consistent with our previous studies (20) , and supports the hypothesis that the hypoxic activator is distinct from the tissue factor or intrinsic pathways.

Since hypoxia often coexists with acidotic conditions in vivo, the pH dependence of the hypoxic endothelial Factor X activator was examined (Fig. 7A-B) . At pH 6.8, Vmax is increased about four-fold (5.1 versus 1.2 ng/ml/min) and Km falls about two-fold (28 versus 13 μg/ml) compared with these parameters at pH 7.4 (Fig. 7A) . Over a broader pH range, it was evident that the optimum pH for Factor X activation was 6.0-6.8, whereas outside these pH's Factor Xa formation fell off (Fig. 7B) . These data indicate that other environmental factors likely to occur during hypoxemia, such as acidosis, can modulate the functional activity of hypoxia-induced proteins. Furthermore, induction of the Factor X activator in hypoxia was reversible when cultures were re-oxygenated (viability of the cultures was maintained) (Fig. 7C) , indicating that they are not only expressed as part of an irreversible process leading to cell death. TABLE 1

The effect of neutralizing antibodies to certain intrinsic and extrinsic coagulation proteins on Factor Xa formation by the purified Factor X activator of hypoxic endothelium*.

Addition Factor Xa formation

(% of maximal)

No addition 100%

Nonimmune IgG 98 ± 13%

Anti-Factor VII IgG 104 ± 2.8%

Anti-Factor IX IgG 95 ± 3.1%

Anti-Factor VIII IgG 91 ± 3.1%

Anti-tissue factor IgG 91 + 2.8%

^♦Endothelial monolayers grown in normoxia were incubated in hypoxia (pθ2 »14 mm Hg) for 48 hr, and the Factor X activator was purified as described. Purified Factor X activator was then incubated (30 min at 37*C) with the indicated antibody to a bovine coagulation factor/cofactor (final concentration 50 μg/ml) . Next, Factor X (1 μM) was added for 30 min (37^*C). Factor Xa formation is reported as a per cent of the Factor Xa formation observed in wells incubated with hypoxic endothelial cells without the addition of any immunoglobulin (maximal Factor Xa formation was 50 ng/ml/10⁵ cells) . The concentration of antibody to Factor IX was sufficient to neutralize and elute all cell- bound Factor IX. The concentration of anti-tissue factor antibody was sufficient to completely neutralize the maximal endothelial tissue factor activity observed in response to tumor necrosis factor/cachectin.

Studies with metabolic inhibitors in normoxic endothelial cultures were carried out to determine whether Factor X activator expression was under the influence of cellular metabolic pathways (Table 2) . Neither 2-deoxyglucose nor fluoride led to expression of Factor X activating capacity of endothelium. In contrast, the respiratory inhibitor sodium azide induced cell surface expression of a direct activator of Factor X. Expression of Factor X activator was not observed in endothelial cells subjected to heat-shock, TNFα, or glucose deprivation.

To begin to extrapolate these results with cultured endothelium to the in vivo setting, induction of the Factor X activator was studied on the native endothelium of aortic segments (Fig. 8) . Vessel segments exposed to hypoxia for 24 hr demonstrated the capacity to activate Factor X, as reported previously (20) . The endothelial monolayer of the segments was removed by gentle scraping and subjected to SDS-PAGE and isoelectric focussing by the same protocol used for samples of the cultured cells. The Factor X activator induced in hypoxic native endothelium also has Mr *=100 kDa on nonreduced SDS-PAGE (Fig. 8A) and pi «5.0 (Fig. 8B) .

TABLE 2

Effect of metabolic inhibitors on the ability of endothelial cells to activate Factor X*.

Condition Factor Xa formation

(% of maximal)

Normoxia 5.3 + 1.5%

Hypoxia 100%

Fluoride 6 + 1.5%

Deoxyglucose 5 + 1.2% Azide 59 + 7.9%

^♦Endothelial monolayers grown in normoxia and then treated in one of the fallowing ways: 48 hr of further incubation in normoxia (normoxia) , 48 hr of incubation in hypoxia (p0₂ «14 mm Hg) (hypoxia), addition of sodium fluoride (1 mM) for 48 hr (fluoride), addition of 2-deoxyglucose (25 mM) for 48 hr (deoxyglucose) , or sodium azide (1 mM) for 48 hr (azide) . Then, cultures were washed and Factor X (1 μM) was added for 30 min at 37^*C. Factor Xa formation was assessed as described in the text. Data is reported as Factor Xa formed as a per cent of that observed in cultures exposed to hypoxia (maximal Factor Xa formation was 40 ng/ml/10⁵ cells) . The mean + SD is shown.

Discussion

Work by other investigators indicating that anoxia can induce the synthesis of proteins (ORPs) by tumor cells and fibroblasts (9-13) , led us to define the pattern of protein biosynthesis in endothelium under hypoxic conditions. Endothelial cell viability is well-maintained in hypoxia, and protein synthesis is only moderately suppressed (20) , even after prolonged exposure to p0₂*s of 10-20 mm Hg level which have been measured in vivo (7-8) . Our study has focussed on endothelial proteins associated with membrane fractions of hypoxic cell cultures, since our initial emphasis was to identify molecules which could be responsible for the altered coagulant properties of hypoxic endothelium, and to search for markers of this state in vivo. The studies reported here demonstrate that hypoxia induces synthesis of at least twenty membrane-associated proteins, visualized by two-dimensional gel analysis. Although expression of some membrane proteins is enhanced by hypoxia (ORP #1330, #1512, #3009, #5418), others appear to be induced de novo. Further, some of the hypoxia-induced proteins, such as #608 (Fig. 2E) , are selectively expressed only at lower oxygen tensions, suggesting that they indeed represent novel proteins. Pilot studies, in which endothelial cell surface proteins were radioiodinated using the lactoperoxidase method, have confirmed that many of the ORPs are accessible on the cell surface. In addition, the hypoxia-induced endothelial activator of Factor X appears to be accessible to the substrate on the cell surface, since the Factor Xa formation occurs on intact endothelial cells.

ORPs could represent post-translational modifications of proteins produced by normoxic cells, or transcription/translation of genes not constitutively expressed in normoxic endothelium. For example, other investigators have observed production of proteins not expressed in normoxic fibroblasts or tumor cells, by their anoxic counterparts (9-13) . In fibroblasts, anoxia causes induction of the transcription/translation of retrotransposon-like VL30 element gene products (10-11) . in addition, studies of total cell lysates and released products of anoxic fibroblasts using methods similar to ours, led to the identification of a total of nine major proteins not observed in normoxic cells (10) . Surprisingly, none of these proteins appear to be identical to the membrane-associated ORPs observed in hypoxic endothelial cells by two dimensional gel analysis. Further, we note 1) the absence of any similarity between the pattern of membrane-associated proteins induced in endothelium subjected to heat-shock, glucose deprivation or the cytokine tumor necrosis factor/cachectin and those observed in hypoxia; and 2) hypoxia of vascular smooth muscle cells did not induce proteins which co igrated with the endothelial membrane-associated ORPs. These results suggest a unique response by endothelial cells to hypoxia.

The capacity of ORPs to modulate functional properties of the endothelial cell surface is emphasized by the reversible induction of a novel Factor X activator in hypoxic cultures. This activator of Factor X, which is also expressed by the native endothelium of hypoxic aortic segments, is a membrane-associated protein, pi «5.0 and Mr corresponding to «100 kDa. Comparing these results with our two-dimensional gel study of hypoxic endothelial cultures, there is one spot (#608), induced only in the presence of low concentrations of oxygen and with a time course of a delayed ORP (the latter characteristics closely resemble the pattern of induction for the Factor X activator in hypoxic endothelium (20) , which could represent the hypoxia-induced Factor X activator. Production of antibodies to the purified Factor X activator will be required to determine definitively if it is identical to ORP #608. Based on the results presented here and our previous observations (20) , the hypoxia-induced Factor X activator appears to be distinct from tissue factor/Factor Vila and the Factor IXa/VIIIa complex. Furthermore, the migration of the activator on nonreduced SDS-PAGE is distinct from that of the tumor procoagulant (34) (the latter has Mr «68 kDa) , an activator of Factor X present in malignant tissue which bears certain similarities to the hypoxia-induced Factor X activator described previously, including inhibition by mercury chloride and ability to be extracted from membranes with low ionic strength veronal buffer (34) . Our results are consistent with the expression of a novel activator of Factor X by hypoxic endothelium. The functional effectiveness of this activator for the formation of Factor Xa at pH «7.4, however, is low as compared with the classical intrinsic and extrinsic systems (35-36) . As the pH of the reaction mixture is decreased. Factor Xa formation increases. Since acidotic conditions are likely to prevail in hypoxemic tissues, this suggests that the activator may have functional activity under pathophysiologic conditions. In addition, other elements of the hypoxemic milieu may further enhance its function.

Mechanisms underlying the induction of ORPs and the Factor X activator are unclear at this time. Studies with metabolic inhibitors indicated that only azide, an inhibitor of cytochromes a+a₃ (33) , could mimic hypoxia in inducing an endothelial Factor X activator. Pilot studies with azide- treated endothelial cells have suggested that the Factor X activator expressed in the presence of the inhibitor is similar to that induced by hypoxia: antibodies to coagulation factors (as in table 1) did not block Factor X activation, mercury chloride but not PMSF blocked Factor X activation, and SDS-PAGE demonstrated that the activator migrated with Mr «100 kDa. In this context, azide also appeared to induce several of the other ORPs observed on two-dimensional gels. This suggests that the signals resulting in de novo biosynthesis under hypoxic conditions are a consequence of perturbation of the electron transport pathway in mitochondria.

These studies indicate that a part of the endothelial response to hypoxia includes induction of the synthesis of a range of new proteins which can contribute to novel functions of hypoxic cultures. The Factor X activator is an example of a protein induced in the presence of only low concentrations of oxygen which may activate the coagulation system and could provide a marker for hypoxemic vasculature.

Other proteins induced by hypoxia may also contribute to the spectrum of altered properties, including decreased barrier function and changes in secreted products observed in hypoxic endothelial cells (16-20) .

REFERENCES

1. Stelzner, T., R. O'Brien, K. Sato, and J. Weil. 1988. Hypoxia-induced increases in pulmonary transvascular protein escape in rats. J. Clin. Invest. 82:1840-1847.

2. Hultgren, J. 1978. High altitude pulmonary edema. In Lung Water and Solute Exchange. N. Staub, editor. Marcel Dekker, Inc., New York. 437-469.

3. Kinasewitz, G., L. Groome, R. Marshall, W. Leslie, and H. Diana. 1986. Effect of hypoxia on permeability of pulmonary endothelium of canine visceral pleura. J. Appl. Physiol. 61:554-560.

4. Olesen, S-P. 1986. Rapid increase in blood-brain barrier permeability during severe hypoxia and metabolic inhibition. Brain Res. 368:24-29.

5. Fanburg, B. and S.-L. Lee. 1987. Effects of anaerobiosis on the endothelium. In Pulmonary Endothelium in Health and Disease. U. Ryan, editor. Marcel Dekker, Inc., New York. 447-456.

6. Sevitt, S. 1967. The acutely swollen leg and deep vein thrombosis. Brit. J. Surg. 54:886-890.

7. Hamer, J. , P. Malone, and I. Silver. 1981. The p0₂ in venous valve pockets: its possible bearing on thrombogenesis. Brit. J. Surg. 68: 166-170.

8. Roca, J. , M. Hogan, D. Story, D. Bebout, P. Haab, R. Gonzalez, O. Ueno, and P. Wagner. 1090. Evidence for tissue diffusion limitation of V0₂max in normal humans. J. Appl. Physiol. 67:291-299.

9. Subjeck, J., and S. Thung-Tai. 1986. Stress protein systems of mammalian cells. Am J. Physiol.

₅ 250:C1-C17.

10. Anderson, G., D. Stoler, and L. Scarcello. 1989. Normal fibroblasts responding to anoxia exhibit features of the malignant phenotype. J. Biol. Chem.

10 264:14885-14892.

11. Anderson, G., D. Stoler and L. Scarcello. 1989. Retrotransposon-like VL30 elements are efficiently induced in anoxic rat fibroblasts. J. Mol. Biol. 205:765-769.

15

12. Sciandra, J. , J. Subjeck, and H. Hughes. 1984. Induction of glucose-regulated proteins during anaerobic exposure and of heat-shock proteins after reoxygenation. PNAS (USA) 81:4843-4847. 0

13. Wilson, R. , and R. Sutherland. 1989. Enhanced synthesis of specific proteins, RNA, and DNA caused by hypoxia and reoxygenation. J. Radiation Oncology Biol. Phys. 16:957-961.

14. Cummiskey, J. , L. Simon, J. Theodore, U. Ryan, and E. Robin. 1981. Bioenergetic alterations in cultivated pulmonary artery and aortic endothelial cells exposed to normoxia and hypoxia. Exp. Lung Res. 2:155-163.

5 15. Lee, S-L. , and B. Fanburg. 1987. Glycolytic activity and enhancement of serotonin uptake by endothelial cells exposed to hypoxia/anoxia. circ. Res. 60:653-688.

16. Hassoun, P., P. Pasricha, E. Teofel, S. Lee, and B. Fanburg. 1989. Hypoxia stimulates the release by bovine pulmonary artery endothelial cells of an inhibitor of pulmonary artery smooth muscle cell growth. Am. J. Respir. Cell Mol. Bio. 1:377-384. 0

17. Bounelis, P., W. Magargal, S. King, F. Booyse, S. Oparil, and D. Miller. 1989. Hypoxia stimulates platelet-derived growth factor gene expression by pulmonary artery endothelial cells. Clin. Res. 5 36:503A.

18. King, S., F. Booyse, P-H. Lin, M. Traylor, A. Narkates, and S. Oparil. 1989. Hypoxia stimulates endothelial cell angiotensin-converting enzyme 0 antigen synthesis. Am. J. Physiol. 256:C1231- C1238.

19. Krulewitz, A., and B. Fanburg. 1984. The effect of oxygen tension on the in vitro production and 5 release of angiotensin-converting enzyme by bovine pulmonary artery endothelial cells. Am. Rev. Respir. Dis. 130:866-869.

20. Ogawa, S., H. Gerlach, C. Esposito, A. Pasagian- 0

Macaulay, J. Brett, and D. Stern. 1990. Hypoxia modulates the barrier and coagulant function of cultured bovine endothelium: increased monolayer permeability and induction of procoagulant properties. J. Clin. Invest.-, in press. w9 21. Schwartz, S. 1978. Selection and characterization of bovine aortic endothelial cells. in Vitro 14:.966-984.

22. Brett, J., S. Steinberg, P. deGroot, P. Nawroth, and D. Stern. 1988. Norepinephrine down-regulates the activity of protein S on endothelial cells. J. Cell Biol. 106:2109-2118.

_ 23. Stern, D. , J. Brett, K. Harris, and P. Nawroth. 1986. Participation of endothelial cells in the protein C-protein S anticoagulant pathway: the synthesis and release of protein S. J. Cell Biol. 102:1971-1978. 5

24. Jaffe, E., L. Hoyer, and R. Nachman. 1973.

Synthesis of antihemophilic factor antigen by cultured human endothelial cells. J. Clin. Invest. 52:2757-2765.

25. Esmon, C. 1987. The regulation of natural anticoagulant pathways. Science 235:1348-1452.

26. Loike, J. , V. Kozler, and S. Silverstein. 1979. Increased ATP and creatine phosphate turnover in 5 phagocytosing mouse peritoneal macrophages. J. Biol. Che . 254:9558-9564.

27. O'Farrell, P. 1975. High resolution two-dimensional electrophoresis of proteins. J. Biol. Chem. 0 250:4007-4021.

28. Garrels, J.^* 1989. The QUEST system for quantitative analysis of two-dimensional gels. J. Biol. Chem. 264:5283-5298. 5 29. Fujikawa, K. , M. Legaz, and E. Davie. 1972. Bovine ffaaccttoorrss XX,, aanndd XX₂₂.. IIssoolation and characterization.

Biochem. 11:4882-4891.

30. Bajaj, P., and K. Mann. 1973. Simultaneous purification of bovine prothrombin and factor X. J. Biol. Chem. 248:7729-7741.

31. Laemmli, U. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage

T4. Nature 227:680-685.

32. Henderson, L. , and W. Konigsberg. 1979. A micromethod for complete removal of dodecyl sulfate from proteins by ion-pair extraction. Anal. Biochem. 93:153-157.

33. Mahler, H. and E. Cordes. 1971. in Biological Chemistry. Harper and Row, Publishers, New York. PP« 503, 505, 512, 686.

34. Galanga, A., and S. Gordon. 1985. Isolation and characterization of cancer procoagulant; a cysteine proteinase from malignant tissue. Biochem. 24:5558-5567.

35. Van Dieijen, G. , G. Tans, J. Rosing, and H. Hemker. 1981. The role of phospholipid and Factor Villa in the activation of bovine Factor X. J. Biol. Chem. 256:3433-3442.

36. Bach, R. 1988. Initiation of coagulation by tissue factor. CRC Critical Reviews in Biochem. 23:339- 368.

Claims

What is claimed is:

1. A purified Factor X activator isolated from endothelial cell membranes.

2. The purified Factor X activator of claim l expression of which by a cell is reversibly induced by hypoxia or by sodium azide.

3. The purified Factor X activator of claim 2 comprising a polypeptide characterized by the ability to migrate as a single band and by an apparent molecular weight of between about 50,000 and about 150,000 daltons on a non-reduced SDS- polyacrylamide gel, loss of activity on a reduced SDS-polyacrylamide gel, loss of activity from exposure to mercury chloride, and an isoelectric focussing peak at a pH of about 5.

4. The purified Factor X activator of claim 3 characterized by an apparent molecular weight of about 100,000 daltons on a non-reduced SDS- polyacrylamide gel.

5. The purified Factor X activator of claim 4 having an optimum Factor X activating effect at a pH range of approximately 6.0 - 6.8.

6. A purified nucleic acid molecule encoding the Factor X activator of claim 1.

7. A cDNA molecule of claim 6.

8. An isolated genomic DNA molecule of claim 6.

9. An isolated RNA molecule of claim 6.

10. An antibody directed to the Factor X activator of claim 1.

5

11. A monoclonal antibody of claim 10.

12. An antibody of claim 10 coupled to a toxin.

_Q 13. The antibody of claim 12 wherein the toxin is diptheria toxin.

14. The antibody of claim 10 directed to an epitope which is an active site of Factor X activator.

15

15. A monoclonal antibody of claim 14.

16. An inhibitor directed to Factor X activator.

-_Q 17. A pharmaceutical composition comprising an amount of the antibody of claim 14 effective to inhibit the activity of Factor X activator and a pharmaceutically acceptable stabilizer.

__ 18. The pharmaceutical composition of claim 17 wherein the stabilizer is albumin.

19. A pharmaceutical composition comprising an amount of the antibody of claim 12 effective to necrotize a tumor, and a pharmaceutically acceptable

30 stabilizer.

20. The pharmaceutical composition of claim 19 wherein the stabilizer is albumin.

35

21. A pharmaceutical composition comprising an amount of the inhibitor of claim 16 effective to inhibit the activity of Factor X activator, and a pharmaceutically acceptable carrier.

22. An oxygen-regulated cell membrane protein which is expressed in endothelial cells after exposure to hypoxic conditions for a period of at least 23 hours.

23. An isolated nucleic acid molecule which encodes the oxygen-regulated cell membrane protein of claim 22.

24. A cDNA molecule of claim 23.

25. An isolated genomic DNA molecule of claim 23.

26. An isolated RNA molecule of claim 23.

27. An antibody directed to the oxygen-regulated cell membrane protein of claim 22.

28. A monoclonal antibody of claim 27.

29. The oxygen-regulated cell membrane protein of claim 22 identified as #311 in Figure 2(A).

30. The oxygen-regulated cell membrane protein of claim 22 identified as #2718 in Figure 2(A).

31. The oxygen-regulated cell membrane protein of claim 22 identified as #1530 in Figure 2(A).

32. The oxygen-regulated cell membrane protein of claim 22 identified as #1333 in Figure 2(A).

33. The oxygen-regulated cell membrane protein of claim 22 identified as #2326 in Figure 2 (A) .

34. The oxygen-regulated cell membrane protein of claim 22 identified as #3009 in Figure 2(A).

35. The oxygen-regulated cell membrane protein of claim 22 identified as #5417 in Figure 2(A).

36. The oxygen-regulated cell membrane protein of claim 22 identified as #5418 in Figure 2(A).

37. The oxygen-regulated cell membrane protein of claim 22 identified as #3328 in Figure 2(A).

38. The oxygen-regulated cell membrane protein of claim

22 identified as #8327 in Figure 2(A).

39. The oxygen-regulated cell membrane protein of claim

22 identified as #403 in Figure 2(B).

40. The oxygen-regulated cell membrane protein of claim 22 identified as #688 in Figure 2(B).

41. The oxygen-regulated cell membrane protein of claim 22 identified as #1605 in Figure 2(B).

42. The oxygen-regulated cell membrane protein of claim 22 identified as #1308 in Figure 2(B).

43. The oxygen-regulated cell membrane protein of claim

22 identified as #2227 in Figure 2(B).

44. The oxygen-regulated cell membrane protein of claim 22 identified as #2708 in Figure 2(B).

45. The oxygen-regulated cell membrane protein of claim 22 identified as #4234 in Figure 2(B).

46. The oxygen-regulated cell membrane protein of claim ₅ 22 identified as #4420 in Figure 2(B).

47. The oxygen-regulated cell membrane protein of claim 22 identified as #6220 in Figure 2(B).

0 48. The oxygen-regulated cell membrane protein of claim 22 identified as #7420 in Figure 2(B).

49. The oxygen-regulated cell membrane protein identified as #8415 in Figure 2(B). 5

50. A method of detecting expression of Factor X activator which comprises contacting a cell with a detectably labeled nucleic acid probe directed to Factor X activator under conditions which permit 0 the probe to hybridize with mRNA encoding Factor X activator, if any such is present, and detecting expression by detecting any hybridization which occurs.

- 51. A method of detecting expression of Factor X activator in a cell which comprises contacting the cell with the antibody of claim 10 labeled with a detectable marker under conditions which permit the antibody to bind to Factor X activator on the surface of the cell, if any such is present, and 0 detecting expression by detecting the marker.

52. A method of detecting expression of an oxygen- regulated cell membrane protein which comprises contacting a cell with a detectably labeled nucleic acid probe directed to the oxygen-regulated cell membrane protein under conditions which permit the probe to hybridize with mRNA encoding the cell membrane protein, if any such is present, and detecting expression by detecting any hybridization which occurs.

53. A method of detecting expression of an oxygen- regulated cell membrane protein which comprises contacting the cell with the antibody of claim 27 labeled with a detectable marker under conditions which permit the antibody to bind to Factor X activator on the surface of the cell, if any such is present, and detecting expression by detecting the marker.

54. A method for diagnosis of a hypoxemic condition in a tissue comprising detection of Factor X activator expression in a sample of the tissue by either of the methods of claims 50 or 51.

55. A method for diagnosis and localization of a hypoxemic condition comprising injecting into a subject a radioactively labeled antibody of claim 10 and determining its binding location by use of a radiation detector.

56. A method for diagnosis of a hypoxemic condition in a tissue comprising detection of expression of an oxygen-regulated cell membrane protein in a sample of the tissue by either of the methods of claims 52 or 53.

57. A method for diagnosis and localization of a hypoxemic condition comprising injecting into a subject a radioactively labeled antibody of claim 27 and determining its binding location by use of a radiation detector.

58. A method of treating a subject having a hypoxemic condition which comprises administering to the subject an amount of the pharmaceutical composition of any of claims 17, 19 or 21 effective to inhibit the activity of Factor X activator.

59. The method of claim 58 wherein the hypoxemic condition is a result of ischemia.

60. The method of claim 58 wherein the ischemia is caused by a tumor.

61. The method of claim 58 wherein the ischemia is caused by a thrombus.

62. The method of claim 58 wherein the ischemia is caused by a wound.

63. A method of treating a subject having a tumor which comprises administering to the subject an amount of the pharmaceutical composition claim 19 effective to necrotize the tumor.

64. A method for preparing the Factor X activator of claim 1 which comprises:

a. inducing cells to express Factor X activator;

b. recovering the Factor X activator from the resulting cells; and

c. purifying the Factor X activator so recovered. 91/13095

-52-

65. A method for preparing the Factor X activator of claim 1 which comprises:

a. inserting isolated nucleic acid encoding Factor X activator in a suitable vector;

b. inserting the resulting vector in a suitable host cell;

c. recovering the Factor X activator produced by the resulting cells; and

d. purifying the Factor X activator so recovered.

5

0

35