EP0831714A1 - Monospecific antibodies and methods using them - Google Patents

Abstract

Monospecific antibodies are produced which are specific for the carboxyl-terminal region of human inducible nitric oxide synthase (hiNOS). These monospecific antibodies are used in an assay systems to detect hiNOS. Elevations of hiNOS occur in hepatitis and inflammatory bowel disease, as well as numerous inflammatory diseases. The monospecific antibodies and the assay systems are used to quantitate hiNOS as a readout of hiNOS activity and to evaluate potential hiNOS modulators.

Description

TITLE OF THE DISCLOSURE

MONOSPECIFIC ANTIBODIES AND METHODS USING THEM

BACKGROUND OF THE DISCLOSURE The emergence of nitric oxide (NO), a reactive, inorganic free radical gas as a molecule contributing to important physiological and pathological processes is one of the major biological revelations of recent times. This molecule is produced under a variety of physiological and pathological conditions by cells mediating vital biological functions. An example is endothelial cells lining the blood vessels. Nitric oxide derived from these cells relaxes smooth muscle and regulates blood pressure and has significant effects on the function of circulating blood cells such as platelets and neutrophils as well as on smooth muscle, both of the blood vessels and also of other organs such as the airways. In the brain and elsewhere nitric oxide serves as a neurotransmitter in non-adrenergic non-cholinergic neurons. In these instances nitric oxide appears to be produced in small amounts on an intermittent basis in response to various endogenous molecular signals. In the immune system nitric oxide can be synthesized in much larger amounts on a protracted basis. Its production is induced by exogenous or endogenous inflammatory stimuli, notably endotoxin and cytokines elaborated by cells of the host defense system in response to infectious and inflammatory stimuli. This induced production results in prolonged nitric oxide release which contributes both to host defense processes such as the killing of bacteria, parasites and viruses as well as pathology associated with acute and chronic inflammation in a wide variety of diseases. The discovery that nitric oxide production is mediated by a unique series of three closely related enzymes, named nitric oxide synthases, which utilize the amino acid arginine, molecular oxygen and NADPH as co-substrates has provided an understanding of the biochemistry of this molecule and provides distinct pharmacological targets for the modulation of the synthesis of this mediator, which should provide significant beneficial effects in a wide variety of diseases. Nitric oxide and L-citrulline are formed from L-arginine via the dioxygenase activity of specific nitric oxide synthases (NOSs) in mammalian cells. In this reaction, L-arginine, θ2 and NADPH are cosubstrates while FMN, FAD and tetrahydrobiopterin are cofactors. NOSs fall into two distinct classes, constitutive NOS (cNOS) and inducible NOS (iNOS). Three isoforms of NOS have been identified comprising two cNOS isoforms and one iNOS isoform. The two cNOS isoforms are characterized as follows:

(i) a constitutive, Ca⁺⁺/calmodulin dependent enzyme, located in the endothelium (ecNOS or NOS 3), that releases NO in response to receptor or physical stimulation, (ii) a constitutive, Ca⁺⁺/calmodulin dependent enzyme, located in the brain (ncNOS or NOS 1) and elsewhere, that releases NO in response to receptor.

The inducible NOS isoform (NOS 2) is characterized as follows: (iii) a Ca⁺⁺ independent enzyme which is induced after activation of vascular smooth muscle, macrophages, hepatocytes, endothelial cells, chondrocytes, and a large number of other cells by endotoxin and cytokines. Once expressed, this inducible NO synthase produces NO in relatively large amounts for long periods of time.

Spectral studies of both the mouse macrophage iNOS and rat brain ncNOS have shown that these enzymes (which have been classified as P-450-like from their CO-difference spectra) and amino acid sequence homology contain a heme moiety. The structural similarity between NOS and the P-450/flavoprotein complex suggests that the NOS reaction mechanism may be similar to P-450 hydroxylation and/or peroxidation. This indicates that NOS belongs to a class of flavohemeproteins which contain both heme and flavin binding regions within a single protein in contrast to the multiprotein NADPH oxidase or Cytochrome P-450/NADPH Cyt c reductase complexes.

Two distinct cDNAs accounting for the activity of NOS 1 and NOS 3 in man have been cloned, one for NOS 1 (Nakane et. al, FEBS Letters, 316, 175-182, 1993) which is present in the brain and a number of peripheral tissues, the other for an enzyme present in endothelium (NOS 3) (Marsden et. al., FEBS Letters, 307, 287-293, 1992). This latter enzyme is critical for production of NO to maintain vasorelaxation. The second class of enzyme, iNOS or NOS 2, has been cloned from human liver (Geller et. al., PNAS, 90, 3491-5, 1993), and identified in more than a dozen other cells and tissues, including smooth muscle cells, chondrocytes, the kidney and airways. As with its counterpart from the murine macrophage, this enzyme is induced upon exposure to cytokines such as gamma interferon (IFN-γ), interleukin-lβ (IL-lβ), tumor necrosis factor (TNF-α) and LPS (lipopolysaccharide). Once induced, iNOS expression continues over a prolonged period of time. The enzyme does not require exogenous calmodulin for activity, but does contain calmodulin as a tightly bound cofactor.

Endothelium derived relaxation factor (EDRF) has been shown to be produced by NOS 3 (Moncada et. al., Pharmacol. Reviews, 43, 109-142, 1991). Studies with substrate analog inhibitors of NOS have shown a role for NO in regulating blood pressure in animals and blood flow in man, a function attributed to NOS 3. NO has also been shown to be an effector of the cytotoxic effects of activated macrophages (Nathan, FASEB J., 6, 3051-64, 1992) for fighting tumour cells and invading microorganisms (Wright et al., Card. Res., 26 ,48-57, 1992 and Moncada et al, Pharmacological Review, 43, 109-142, 1991). It also appears that the adverse effects of excess NO production, in particular pathological vasodilation and tissue damage, may result largely from the effects of NO synthesized from the NOS 2.

NO generated by NOS 2 has been implicated in the pathogenesis of inflammatory diseases. In experimental animals hypotension induced by LPS or TNF-α can be reversed by NOS inhibitors and reinitiated by L-arginine (Kilbourn et. al, PNAS, 87, 3629-32, 1990). Conditions which lead to cytokine-induced hypotension include septic shock, hemodialysis (Beasley and Brenner, Kidney Int., 42, Suppl., 38, S96--S100, 1992) and IL-2 therapy in cancer patients (Hibbs et. al., J. Clin. Invest., 89, 867-77, 1992). NOS 2 is implicated in these responses, and thus the possibility exists that a NOS inhibitor would be effective in ameliorating cytokine-induced hypotension. Recent studies in animal models have suggested a role for NO in the pathogenesis of inflammation and pain and NOS inhibitors have been shown to have beneficial effects on some aspects of the inflammation and tissue changes seen in models of inflammatory bowel disease, (Miller et. al, J. Pharmacol. Exp. Ther., 264, 11-16, 1990) and cerebral ischemia and arthritis (Ialenti et. al, Br. J. Pharmacol ., 110, 701-6, 1993; Stevanovic-Racic et al., Arth. & Rheum., 37, 1062-9, 1994). Moreover transgenic mice deficient in NOS 1 show diminished cerebral ischemia (Huang et. al, Science, 265, 1883-5, 1994).

Further conditions where there is an advantage in inhibiting NO production from L-arginine include therapy with cytokines such as TNF, IL-1 and IL-2 or therapy with cytokine-inducing agents, and as an adjuvant to short term immunosuppression in transplant therapy. In addition, compounds which inhibit NO synthesis may be of use in reducing the NO concentration in patients suffering from inflammatory conditions in which an excess of NO contributes to the pathophysiology of the condition, for example adult respiratory distress syndrome (ARDS) and myocarditis. There is also evidence that an NO synthase enzyme may be involved in the degeneration of cartilage which takes place in autoimmune and/or inflammatory conditions such as arthritis, rheumatoid arthritis, chronic bowel disease and systemic lupus erythematosis (SLE). It is also thought that an NO synthase enzyme may be involved in insulin-dependent diabetes mellitus. Therefore, the antibodies of the present invention are useful in the development, evaluation and production of medicaments for the treatment of patients suffering from conditions which are mediated by hiNOS. The antibodies of the present invention are useful for diagnosing and monitoring the treatment of a number of diseases in which increased iNOS has been implicated. The implication of these diseases is well documented in the literature. For example, with regard to psoriasis, see Ruzicka et. al., J. Invest. Derm., 103: 397 (1994) or Kolb-Bachofen et. al, Lancet, 344: 139 (1994) or Bull, et al, J. Invest. Derm., 103:435(1994); with regard to uveitis, see Mandia et. al, Invest Opthalmol, 35: 3673-89 (1994); with regard to type 1 diabetes, see Eisieik & Leijersfam, Diabetes & Metabolism, 20: 1 16-22 (1994) or Kroncke et. al, BBRC, 175: 752-8 (1991) or Welsh et. al, Endocrinol, 129: 3167-73 (1991); with regard to septic shock, see Petros et. al, Lancet, 338: 1557-8 (1991),Thiemermann & Vane, Eur. J. Pharmacol, 211 : 172-82 (1992), or Evans et. al, Infec. Imm., 60: 4133-9 (1992), or Schilling et. al, Intensive Care Med., 19: 227-231 (1993); with regard to pain, see Moore et. al, Brit. J. Pharmacol, 102: 198-202 (1991), or Moore et. al, Brit. J. Pharmacol, 108: 296-97 (1992) or Meller et. al, Europ. J. Pharmacol, 214: 93-6 (1992) or Lee et. al, NeuroReport, 3: 841-4 (1992); with regard to migraine, see Olesen et. al, TIPS, 15: 149-153 (1994); with regard to rheumatoid arthritis, see Kaurs & Halliwell, FEBS Letters, 350: 9-12 (1994); with regard to osteoarthritis, see Stadler et. al, J. Immunol, 147: 3915-20 (1991); with regard to inflammatory bowel disease, see Miller et. al, Lancet, 34: 465-66 (1993) or Miller et. al, J. Pharmacol. Exp. Ther., 264: 11-16 (1993); with regard to asthma, see Hamid et. al, Lancet, 342: 1510-13 (1993) or Kharitonov, et. al, Lancet, 343: 133-5 (1994); with regard to immune complex diseases, see Mulligan et. al, Br. J. Pharmacol, 107: 1159-62 (1992); with regard to multiple sclerosis, see Koprowski et. al, PNAS, 90: 3024-7 (1993); with regard to ischemic brain edema, see Nagafuji et. al, Neuwsci., 147: 159-62 (1992) or Buisson et. al, Br. J. Pharmacol, 106: 766-67 (1992) or Trifiletti et. al, Europ. J.

Pharmacol, 218: 197-8 (1992); with regard to toxic shock syndrome, see Zembowicz & Vane, PNAS, 89: 2051-55 (1992); with regard to heart failure, see Winlaw et. al, Lancet, 344: 373-4 (1994); with regard to ulcerative colitis, see Boughton-Smith et. al, Lancet 342: 338-40 (1993); and with regard to atherosclerosis, see White et. al, PNAS, 91: 1044-8 (1994); with regard to glomerulonephritis, see Mϋhl et. al, Br. J. Pharmcol, 112: 1-8 (1994); with regard to Paget's disease and osteoporosis, see Lowick et. al., J. Clin. Invest., 93: 1465-72 (1994); with regard to inflammatory sequelae of viral infections, see Koprowski et. al, PNAS, 90: 3024-7 (1993); with regard to retinitis, see Goureau et. al, BBRC, 186: 854-9 (1992); with regard to oxidant induced lung injury, see Berisha et. al, PNAS, 91 : 744-9 (1994); with regard to eczema, see Ruzica, et al., J. Invest. Derm., 103:395(1994); with regard to acute allograft rejection, see Devlin, J. et al, Transplantation, 58:592-595 (1994); and with regard to infection caused by invasive microorganisms which produce NO, see Chen, Y and Rosazza, J.P.N., Biochem. Biophys. Res. Comm., 203:1251-1258(1994).

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 - The titer of the peptide-specific antibody in immunized rabbits is shown.

Figure 2 - The specificity of the antibody is shown with the requirement for the carboxyl-terminal Leu.

Figure 3 - The competitive displacement of radiolabelled peptide from the antibody by unlabelled peptide is shown.

Figure 4 - The antibody concentration and saturation time for peptide binding is shown.

Figure 5 - The displacement of radiolabelled peptide from the antibody by hiNOS is shown.

Figure 6 - Monitoring of column chromatographic steps in the purification of hiNOS is shown. Figure 7 - Western blot analysis is shown for the antibody, and that the detection of hiNOS is blocked by the competing peptide.

Figure 8 - Western blot analysis is shown demonstrating that the antibody recognizes recombinantly produced hiNOS.

Figure 9 - Immunohistochemical staining of hiNOS in liver tissue sample from a hepatitis patient is shown.

Figure 10 - Inhibition of hiNOS staining by preincubation with competing peptide in a tissue sample adjacent to that in Figure 9 is shown.

Figure 1 1 - Lack of inhibition of hiNOS staining by preincubation with a noncompeting iNOS-related peptide is shown in a tissue sample adjacent to that in Figure 9.

Figure 12 - Immunohistochemical staining of hiNOS in an intestine tissue sample from a patient with U.C.

Figure 13 - Immunohistochemical staining of hiNOS in a colon tissue sample from a patient with Crohn's Disease.

Figure 14 - Inhibition of hiNOS staining by preincubation with a competing peptide in a tissue sample adjacent to that in Figure 13.

SUMMARY OF THE DISCLOSURE The present invention is drawn to monospecific antibodies which bind to specific portions of human inducible nitric oxide synthase (hiNOS) and fragments thereof. Using the antibodies of the present invention, the quantity of hiNOS as well as the presence or absence of hiNOS is measured, for example, in a classical RIA or by classical immunolocalization techniques. The antiserum detects hiNOS in, for example, clinical tissue samples. As a measure of activity the antiserum is used to detect or quantify hiNOS in: (a) models in which endogenous synthesis is stimulated by various cytokines or other stimuli; (b) in the diagnosis of various human diseases including those described herein; and (c) in monitoring the course of treatment of hiNOS-mediated diseases, including those described herein. Use of this antibody also allows the evaluation of hiNOS modulators in various pharmacokinetic/pharmacological animal models as well as in various human diseases, including those described herein.

DETAILED DESCRIPTION OF THE INVENTION

The present invention is directed to the detection, quantitation and localization of hiNOS and fragments thereof as a measure of hiNOS activity. The present invention is also directed to an assay system to evaluate the potency of hiNOS modulators in vivo. The term "modulation" as used herein is defined as the increase or decrease in the amount of hiNOS protein. The term "modulator" as used herein is defined as a substance or substances which directly or indirectly causes the modulation of hiNOS as defined above. The quantitation of hiNOS and fragments thereof may be useful in the diagnosis of various diseases including but not limited to those described herein. The identification of hiNOS inhibitors may lead to the development of drugs for the treatment of diseases including, but not limited to, those described herein. More specifically, the present invention is directed to monospecific antibodies which detect hiNOS and polypeptide fragments thereof, but do not detect other NOS isoforms or non-human iNOS. One way to monitor hiNOS levels in situ, is to assay for hiNOS protein or degradation products thereof. Since hiNOS is involved in a large number of human diseases, including those described herein, the present invention is focused, in part, on developing reagents to quantify hiNOS for this purpose. This is the first time that reagents have been developed to qualify and quantify the presence and/or absence of hiNOS in clinical specimens. To date, general immunological and other assays are used to quantify hiNOS. These assays do not detect denatured hiNOS molecules. The antisera described herein not only specifically recognize hiNOS, but also recognize denatured hiNOS. Using these antisera, hiNOS and carboxyl-terminal fragments thereof can be specifically detected, quantified and localized in a wide variety of physical conditions, such as Western blot and in tissue samples.

As used herein, all amino acid three letter and single letter designations conform to those designations which are standard in the art, and are listed as follows:

Alanine Ala A Leucine Leu L

Arginine Arg R Lysine Lys K

Asparagine Asn N Methionine Met M

Aspartic acid Asp D Phenylalanine Phe F

Cysteine Cys C Proline Pro P

Glutamic acid Glu E Serine Ser S

Glutamine Gin Q Threonine Thr T

Glycine Gly G Tyrptophan Trp w

Histidine His H Tyrosine Tyr Y

Isoleucine He I Valine Val V

The antibodies and assays of the present invention for hiNOS are used as a diagnostic tool to demonstrate increased or decreased levels of hiNOS in various diseases as well as to monitor the efficacy of specific and selective hiNOS modulators in various animal models and man. In animal models hiNOS modulator compound efficacy is evaluated by quantifying and/or determining the presence or absence of hiNOS. Therefore, assays to monitor both the quantity and/or presence or absence of hiNOS are useful to characterize hiNOS modulators. In man, it is difficult to assay hiNOS as a readout for biochemical efficacy since human tissue samples cannot be readily obtained from a patient. The approaches taken in the present invention to develop these assays are described herein. The hiNOS amino acid sequence has been identified by Geller et al (1993), P.N.A.S., 90, pp 3491-3495. The peptides around the hiNOS carboxy-terminal seven amino acids are synthetically prepared and polyclonal antisera against those peptides are generated to use as immunoreagents in the assays of the present invention.

Monospecific antipeptide antibodies are generated which recognize the carboxy-terminal seven amino acids of hiNOS. These antibodies are used to develop radioimmunoassays (RIA) to quantify the hiNOS molecule at this site. These antibodies recognize hiNOS in a wide variety of samples and under a variety of physical conditions.

Using the antibodies of the present invention, several assay systems have been developed. The assays are used to detect and quantify hiNOS in human tissue, synovial fluid, blood, urine or other biological fluids, and Western blots. The assay has a limit of detection of about 50 fmoles in Western blots.

Monospecific antibodies to hiNOS are purified from mammalian antisera containing antibodies reactive against hiNOS or are prepared as monoclonal antibodies reactive with hiNOS using the technique of Kohler and Milstein, Nature 256: 495-497 (1975). Monospecific antibody as used herein is defined as a single antibody species or multiple antibody species with homogeneous binding characteristics for hiNOS without binding to the other NOS isoforms or non-human iNOS. Homogenous binding as used herein refers to the ability of the antibody species to bind to a specific antigen or epitope, such as those associated with the hiNOS, as described above. The hiNOS specific antibodies are raised by immunizing animals such as mice, rats, guinea pigs, rabbits, goats, horses and the like, with rabbits being preferred, with an appropriate concentration of hiNOS or a synthetic peptide conjugate based on sequences in this fragment either with or without an immune adjuvant.

Preimmune serum is collected prior to the first immunization. Each animal receives between about 0.1 μg and about 1000 μg of hiNOS or peptide conjugate associated with an acceptable adjuvant. Such acceptable adjuvants include, but are not limited to, Freund's complete, Freund's incomplete, alum-precipitate, water in oil emulsion containing Corynebacterium parvum and tRNA. The initial immunization consists of hiNOS or synthetic peptides based on the C- terminus thereof conjugated to bovine thyroglobulin in, preferably, Freund's complete adjuvant injected at multiple sites either subcutaneously (SC), intramuscular (IM), intraperitoneally (IP) or a combination of the above. The hiNOS or synthetic peptides may also be conjugated to other carrier molecules which include those known in the art, such as keyhole limpet hemocyanin and BSA. Each animal is bled at prescheduled regular intervals, to determine antibody titer. The animals may or may not receive booster injections following the initial immunization. After the initial immunization, animals with no response or low titers are given booster injections. These animals receiving booster injections are generally given an equal amount of the hiNOS or peptide conjugates in Freund's incomplete adjuvant by the same route. Booster injections are given at about three week intervals until maximal titers are obtained. At about 10 to 14 days after each booster immunization or about bi-weekly after a single immunization, the animals are bled, the serum collected, and aliquots are stored at about -20°C.

Monoclonal antibodies (mAb) reactive with hiNOS or peptide conjugates are prepared by immunizing inbred mice, preferably Balb/c, with hiNOS or peptide conjugates. The mice are immunized by the IP or SC route with about 0.1 μg to about 10 μg, preferably about 1 μg, of hiNOS or peptide conjugates in about 0.5 ml buffer or saline incorporated in an equal volume of an acceptable adjuvant, as discussed above. Freund's complete adjuvant is preferred. The mice receive an initial immunization on day 0 and are rested for about 3 to 30 weeks. Immunized mice are given one or more booster immunizations of about 0.1 to about 10 μg of hiNOS or peptide conjugates in a buffer solution such as phosphate buffered saline by the intravenous (IV) route. Lymphocytes, from antibody positive mice, preferably splenic lymphocytes, are obtained by removing spleens from immunized mice by standard procedures known in the art. Hybridoma cells are produced by mixing the splenic lymphocytes with an appropriate fusion partner, preferably myeloma cells, under conditions which will allow the formation of stable hybridomas. Fusion partners may include, but are not limited to: mouse myelomas P3/NSl/Ag 4-1 ; MPC-11 ; S-194 and Sp 2/0, with Sp 2/0 being preferred. The antibody producing cells and myeloma cells are fused in polyethylene glycol, about 1000 mol. wt., at concentrations from about 30% to about 50%. Fused hybridoma cells are selected by growth in hypoxanthine, thymidine and aminopterin supplemented Dulbecco's Modified Eagles medium (DMEM) by procedures know in the art. Supernatant fluids are collected from growth positive wells on about days 14, 18, and 21 and are screened for antibody production by an immunoassay such as solid phase immunoradioassay (SPLRA) using hiNOS or peptide conjugates as the antigen. The culture fluids are also tested in the Ouchterlony precipitation assay to determine the isotype of the mAb. Hybridoma cells from antibody positive wells are cloned by a technique such as the soft agar technique of MacPherson, Soft Agar Techniques, in Tissue Culture Methods and Applications, Kruse and Paterson, Eds., Academic Press, 1973. Monoclonal antibodies are produced in vivo by injection of pristane primed Balb/c mice, approximately 0.5 ml per mouse, with about 2 x lθ6 to about 6 x 106 hybridoma cells about 4 days after priming. Ascites fluid is collected at approximately 8-12 days after cell transfer and the monoclonal antibodies are purified by techniques known in the art.

In vitro production of anti-MNOS mAb is carried out by growing the hydridoma in DMEM containing about 2% fetal calf serum to obtain sufficient quantities of the specific mAb. The mAb are purified by techniques known in the art. Antibody titers of ascites or hybridoma culture fluids are determined by various serological or immunological assays which include, but are not limited to, precipitation, passive agglutination, enzyme-linked immunosorbent antibody (ELISA) technique and radioimmunoassay (RIA) techniques. Similar assays are used to detect the presence of hiNOS in body fluids or tissue and cell extracts.

It is readily apparent to those skilled in the art that the above described methods for producing monospecific antibodies may be utilized to produce antibodies specific for hiNOS. The antibodies are useful in the production of a diagnostic assay system for the detection and measurement of hiNOS which is involved in the diseases described herein, and other diseases, from analysis of the patients tissues or biological fluids. In addition, the antibodies are useful for the immunolocalization of hiNOS in biopsy/tissue samples. Such analysis allows the determination of sites of hiNOS activity in situ.

To evaluate the in vivo or in vitro efficacy of hiNOS modulators, patients or cells in culture are treated with hiNOS modulators and the level of hiNOS is quantified using the immunoassays of the present invention, including the RIA and Western blots, as a monitor of hiNOS activity. Patients or cells in culture are treated with hiNOS modulators and the amount of hiNOS in the cells, cell extracts, conditioned medium, tissue, blood, or other biological fluid is detected or quantified using the immunoassays of the present invention, including the RIA or Western blot. The amount of hiNOS modulation is calculated as a percentage of the hiNOS with modulator treatment as a proportion of hiNOS without modulator treatment. The same approach is used to quantify the modulation of hiNOS activity upon treatment with cytokines. The same assay is used to evaluate treatment progress in human hiNOS-mediated diseases by monitoring the modulation of hiNOS in tissues, blood or other biological fluids. This modulation is determined by quantifying the level of hiNOS prior to drug treatment followed by quantitation of the hiNOS level during or after drug treatment. This assay is also used to quantify the level of hiNOS in tissues, blood or other biological fluids from animal models of hiNOS- mediated diseases as well as humans.

The antiserum generated against hiNOS is also used to localize, detect or quantify hiNOS by standard immunolocalization techniques. This approach is used to localize sites of hiNOS activity in situ. In the presence of a modulator, hiNOS levels may be either reduced or elevated compared to levels of hiNOS in samples not treated with modulator or prior to treatment. By comparing the distribution of hiNOS in tissues from patients treated with drug (hiNOS modulator) to tissue from patients not treated with drug or pretreatment, the distribution of hiNOS modulation is localized and monitored. This approach is used in tissues from humans or animals injected with hiNOS modulators, humans or animals injected with cytokines to generate hiNOS endogenously, or any disease mediated by hiNOS in humans or animal models.

The components of the assay described herein can easily be assembled into a package or kit. For example, the components can be used to coat the surface of a substratum or solid carrier. Suitable carrier materials include cellulose, cross linked dextrose, silicone rubber, microcrystalline glass, and a wide variety of plastics. The most common types of carriers or substratum are plastics including but not limited to polyethylene, polyvinyl chloride, polyamides, polystyrene, polypropylene and other polymers. The carrier or substratum can be formed into various shapes including but not limited to tubes, dishes, plates, multiwell plates, beads or any vessel. The assay components may be covalently bonded to the carrier or substratum, cross-linked, or physically coupled thereto. A variety of detection techniques may be utilized in the assay kit of the present invention, including but not limited to, a colorimetric detection, fluorescence detection, ultraviolet radiation detection, and isotope detection. Kits suitable for immunodiagnosis, or for any other purpose described herein containing the appropriate reagents are constructed by packaging the appropriate materials, including the antibodies of the present invention, in a suitable container along with the remaining reagents and materials required for the assay. A set of instruction should be included.

Recombinant human inducible Nitric Oxide Synthase (rhiNOS), originally cloned from human hepatocytes (David A. Geller, Charles J. Lowenstein, Richard A. Shapiro, Andreas K. Nussler, Mauricio Di Silvio, Stewart C. Wang, Don K. Nakayama, Richard L. Simmons, Solomon H. Snyder and Timothy R. Billiar. 1993, Proc. Natl Acad. Sci., 90, pp 3491-3495), was overexpressed in a baculovirus system and purified to homogeneity in a two-step process using immunoaffinity and DEAE chromatography. The column was constructed using a rabbit polyclonal antipeptide antibody raised against the peptide immunogen, a sequence cognate to the C-terminus of human hepatocyte iNOS. The column binds the active dimer and inactive monomer forms and both were eluted with a 25 uM solution of the cognate peptide. Immunoblots and silver-stained SDS-PAGE analysis of the protein exhibited a band with an apparent molecular weight of 130 kDa. The specific activity varied from 60-200 nmoles citrulline/min/mg at 37°C as measured in an HPLC-based ^H-arginine to -^H-citrulline conversion assay (ACCA). Cell cultures supplemented with serum and delta-amino levulinic acid yielded enzyme with the highest specific activity.

The following examples are provided as an illustration of the present invention without, however, limiting the same thereto.

EXAMPLE 1

Peptide Immunogens

Knowing the specific hiNOS amino acid sequence and the amino acid sequence of the other NOS isoforms allowed the identification of antigenic peptides and peptide probes representing the carboxy terminus of hiNOS. Antibodies raised against these peptides are specific for hiNOS and do not recognize the other NOS isoforms or non-human iNOS. The specific amino acid sequence associated with the carboxy terminus of hiNOS that was used as a basis for the peptide immunogens was Ser-Leu-Glu-Met-Ser- Ala-Leu [SEQ.ID.NO.:l].

Peptide antigens and peptide probes were synthesized using (Fmoc) f uorenylmethoxy-carbonyl chemistry on an ABI 430A peptide synthesizer (Applied Biosystems, Inc.). Peptide acids were synthesized on standard Wang resins, while peptide amides were synthesized on Rink amide resins. Syntheses were carried out according to the FastMoc™ protocols for benzotriazoltetramethyl-uronium hexafluorophosphate (HBTU) mediated couplings described in detail in the ABI 430A Synthesis Notes (Applied Biosystems, 1992). Peptidyl resins were cleaved and deprotected with TFA according to the procedures described in the ABI 430A Operators Manual. Peptides were purified by reversed phase HPLC on a Waters DeltaPak C18 column with an acetonitrile gradient of 2-50% in aqueous 0.1% trifluoroacetic acid (TFA). Purity of individual peptides was assessed by reversed phase HPLC on an Applied Biosystems Spheri-5 C18 column. The structure of the peptides was confirmed by mass spectrometry utilizing either fast atom bombardment or electrospray ionization. The primary antigen associated with the carboxyl terminus of hiNOS is synthesized with four additional amino acid residues. Cysteine-arginine-norleucine-omithine is attached to the synthetic peptide Ser-Leu-Glu-Met-Ser- Ala-Leu (SEQ.ID.No.:l) to give the following antigen: Cys-Arg-Nle-Orn-Ser-Leu-Glu-Met-Ser-Ala-Leu (SEQ. ID. NO.:2). The cysteine is a linking amino acid because it allows the antigen to be linked or attached to an immunogenic carrier. The norleucine was added as an internal marker to determine the actual number of antigen molecules attached to a single immunogenic carrier. The arginine and ornithine were added to increase the aqueous solubility of the peptide.

Preparation of N-acetylated peptide: Peptide (SEQ. ID. NO.: 2) was dissolved in a 1.0-ml solution of 70:30 methano acetic anhydride and incubated for 10 min at 25°C. This was then diluted with an equal volume of H2θ followed by evaporation to dryness. Reconstitution in H2θ and evaporation was repeated twice to remove excess acetic acid. The resulting powder was redissolved in 0.075M PBS-azide with a pH of 7.0. EXAMPLE 2

Attachment of Antigen to Carrier

Attachment of the antigenic peptides to the carrier protein was carried out according to a modification of the method of Lerner et al, Proc. Nat. Acad. Sci. USA 78: 3403-3407 (1981) using the heterobifunctional coupling reagent, Sulfo-MBS (Pierce Chemical Co.). The peptide antigen was attached to the carrier bovine-thyroglobulin (TG), by combining 10 mg of TG dissolved in 2.5 ml degassed phosphate buffer, 20 mM, pH 8.0 with 4.2 mg Sulfo-MBS and incubating for 30 minutes at room temperature with stirring. The carrier-coupling reagent mixture was then applied to a disposable PD-10 Sephadex G-25 column (Pharmacia) which had been equilibrated with de-gassed 50 mM phosphate buffer, pH 7.0. A small vial containing 6 micromoles of the purified, lyophilized peptide antigen was placed under the column outlet and the activated TG fraction was eluted into the vial with an additional 3.5 ml of the pH 7.0 buffer. The peptide antigen-activated carrier complex was allowed to react overnight at 4°C with gentle stirring. The degree of coupling for the peptide-carrier immunogen conjugate was determined by removing an aliquot of the final reaction mixture and passing it through a PD-10 Sephadex G-25 column equilibrated with PBS to remove any remaining free peptide and/or reaction by-products.

An aliquot (50 μl) of the fraction containing the thyroglobulin immunogen complex (determined by A280 was evaporated to dryness for amino acid analysis. For amino acid analysis, the samples were hydrolysed using 200 μL of 6.0 N HCl containing 0.1% phenol maintained at 1 10°C for 24 hours. The sample was analyzed using a Beckman Model 6300 amino acid analyzer. The analysis showed that there was greater than about 20 moles of antigen peptide per mole of TG. Synthetic Probes And Specificity Peptides

Antigen probes to determine antibody specificity and to evaluate the presence and amount of hiNOS were synthesized by the process described herein. Antigenic probes used to determine the presence and amount of hiNOS were designed to include a tyrosine residue at the terminus distal to the epitope so that the probe could be coupled to 125ι. The initial probe used to determine antibody titer was a synthetic peptide based on the amino acid sequence of hiNOS, Ser-Leu-Glu-Met-Ser-Ala-Leu [SEQ.ID.NO.:l], plus an amino terminal tyrosine residue.

Radioiodination of the assay probe was accomplished by reaction with chloramine T. The peptide probe was dissolved in water at a concentration of 220 μg/ml. A 50 μl volume of this solution (containing 11 μg) was added to 10 μl of 0.5 M phosphate (K+) buffer, pH 7.5 and then combined with 2mCi of 125l a and 10 μl freshly prepared chloramine T (0.1 mg/ml) in water. The mixture was allowed to react for 30 seconds and the reaction was stopped with 10 μl of 1 mg/ml Nal plus 1 mg/ml sodium thiosulfate. The radioiodinated probe was purified by HPLC using a Supelco C-8 column (0.4 x 25 cm). The iodinated probe was eluted by a 35 minute 1% per-minute gradient of 99% eluant A-1% eluant B to 36% eluant A-64% eluant B at a flow rate of 1 ml per minute. Eluant A consisted of 0.1 % trifluoroacetic acid in water and eluant B consisted of 0.1 % trifluoroacetic acid in acetonitrile.

Production of Monospecific Antibody

New Zealand White Rabbits were immunized with the peptide immunogen. The initial immunizations employed 333 μg of the immunogen conjugate per 1 ml Freund's complete adjuvant (FCA) given intramuscularly per rabbit. On day 7, animals were again given 333 μg of immunogen in FCA and on day 35 a total of 333 μg of immunogen was given subcutaneously at 6-10 sites. On day 45 the animals were bled, then boosted with 333 μg of immunogen. On day 55, animals were again boosted and 10 days later bled. This boosting and bleeding schedule was continued 3-5 times to obtain an adequate supply of antiserum. All antisera were stored at -20°C.

A specific antibody which recognizes only hiNOS was produced. This antibody was designated N053 and has been utilized in a wide variety of immunoassays and immunopurification processes. High titer antisera was produced in rabbits, which reached titers of approximately 1:500,000 (Figure 1). Titers were greater than 1 :50,000 generally.

EXAMPLE 3

Radioimmunoassay For hiNOS Products

YRASLEMSAL [SEQ.ID.NO.:3] (also referred to herein as N054) was radioiodinated utilizing the iodogen method. Briefly, YRASLEMSAL [SEQ.ID.NO.:3] (0.2 nmoles) was incubated for 12-15 minutes with 400-500 mCi Nal25j j_n an iodogen (Pierce Chemical Co.) coated microfuge tube. The radiolabeled peptide was separated from the unreacted Na^5ι by chromatography on a 5 ml bed volume G-10 Sephadex column eluted with 25% acetic acid containing 0.1 % Triton X-100. The non-retained material, which contained the radiolabeled peptide, was lyophilized to dryness. The 125ι_p_eptide was further purified on RP-HPLC. The peptide was reconstituted into a minimal volume of H2O and injected onto a Zorbax C8 reverse phase column and eluted with an acetonitrile/H2θ gradient at 1 ml/minutes. Buffer A was 0.1 % TFA in H2O and buffer B was 0.1 % TFA in acetonitrile. The column was eluted for 10 minutes with 90% buffer A/10% buffer B; for 10 minutes with a linear gradient to 70% buffer A/30% buffer B; and for 30 minutes with 70% buffer A/30% buffer B. Fractions of 500 μl, were collected and the radioactivity determined. The fractions containing the monoiodinated peptide were combined and lyophilized to dryness in serum vials. The vials were stored at -20°C until use. The final probe routinely had a specific activity of approximately 2,000 Ci/mmole. Competitive Radioimmunoassav

An RIA buffer solution was prepared containing NaCl 100 mM, 47 mM NaHPθ4, 0.5% bovine serum albumin, 0.1 % Tween 20 and 0.1% NaN3. The pH was adjusted to 7.2±0.1. It is readily apparent that a variety of other buffer solutions are suitable for use in the assay described herein and the success of the assay is not limited to the above buffer.

Protocol 1. The initial experiments were performed in 12x75 mm borosilicate disposable culture tubes (Fisher Scientific) The assay was conducted in a total volume of 300 μl of the RIA buffer. One hundred microliters of RIA buffer or standards or samples diluted in this buffer were mixed with 100 μl of N053 antiserum (final dilution 1:60,000) and 100 ul of the l²⁵I-probe (-30,000 CPM) and incubated overnight at 4°C.

The synthetic peptide YRASLEMSAL [SEQ.LD.NO.:3] (N054) was used to generate the standard curve. A 10 nM stock solution of standard peptide was prepared in the RIA buffer and stored frozen at -20°C. In each experiment 10 concentration standards (2-1000 fmoles/100 μl) were prepared by serially diluting 100 μl aliquots of the stock solution 1 :2 in the RIA buffer 9 times. Therefore, the standard curve concentrations of peptide were 2, 4, 8 16, 31.25, 62.5, 125, 250, 500, and 1000 f moles/100 μl. Tubes without standard (to measure the total amount of antiserum binding) and tubes without antiserum (to measure the total amount of radioactivity added and background) were also included. The standard curve is performed in duplicate and samples analyzed with multible dilutions.

The antibody-bound radiolabeled probe was separated from unbound radiolabeled material by an anti-rabbit-IgG antibody attached to magnetic beads. A suspension of BioMag beads (goat anti-rabbit IgG (H&L) (PerSeptive Diagnostics, Cambridge MA) was diluted 1:2 just prior to use with PBS. One ml aliquots of the suspension were added to all tubes, except tubes not containing antiserum which were used for the determination of total radioactivity, and incubated at room temperature for 5 minutes. During the 5 minute incubation period, the tubes containing beads were transferred to a rack with compression springs (Amersham, Arlington Heights, IL) which securely hold the test tubes. The rack was applied to a magnetic plate (Amersham, Arlington Heights, IL) and incubated for an additional 10 minutes. The rack was then inverted and the supernatant fluid containing the unbound 125j. probe was collected in vermiculite and discarded. The ends of the tubes were blotted onto paper towels. The antiserum bound 125l-probe peptide was "held" to the bottom of the test tubes by the magnetic field. It is readily apparent that other methods of separating bound probe from unbound probe can be used, such as dextran coated charcoal and double antibody.

The radioactivity associated with the 125ι_p_rob_e was determined in a 10- well Cobra II Auto Gamma Counter (Packard Instruments). The percent binding of probe to antisera (B/Bo) in the presence of the competitive unlabeled peptide was determined based on the amount bound in the absence of the competitive unlabeled peptide. The B/Bθ as a function of standard peptide concentration was plotted by a 4-parameter algorithm (Immunofit Packard Instruments). The B/Bo of the samples were similarly determined and the concentration of rH- iNOS expressed as fmoles of unlabeled peptide, was determined from the standard curve. Only values falling between 20-80 B/Bθ% binding were considered to be valid for calculations.

The standard curve parameters (EC80, EC50, EC20) and the percent binding were recorded for each assay. In addition, quality control standard solutions of the standard peptide were run in each assay. The quality control standards were required to be within 2 standard deviations of mean of the initial 5 assays to be considered valid.

Protocol 2. The RIA has been adapted into a 96-well format using 96- well FlashPlate™ (New England Nuclear Corp.) scintillation proximity counting technology. The wells of these plates contain a scintillant that detects radioactivity. When a radioactive ligand becomes bound to the surface of the wells the scintillant is activated and the amount of radioactivity quantified in a 96 well scintillation counter. Unbound radioligand in solution does not activate the scintillant thus obviating the need to separate "bound versus free" radiolabelled probe in RIA protocols. These plates were precoated with sheep-anti-rabbit IgG as a capture reagent.

Antiserum N053 was diluted (usually 1 :60,000) into Dulbecco's phosphate buffered saline (PBS). Aliquots, 200 ul, were added to wells of the Flashplates and incubated for 18 hours or for as long as 1 week at 4°C. Two wells on the plate received PBS only to determine non-specific binding. All wells were rinsed with 200 ul of RIA buffer and then incubated with 200 ul of this buffer for at least 1 hour to "block" non-antiserum coated areas of the well. In an identical manner as described in Protocol 1 , the synthetic peptide YRASLEMSAL [SEQ.ID.NO.:3] (N054) was used to generate the standard curve. In each experiment 10 concentration standards (2-1000 fmoles/100 ml) were prepared by serially diluting 100 ml aliquots of the stock solution 1:2 in the RIA buffer 9 times. The assay was conducted in a total volume of 200 μl of the RIA buffer. One hundred microliters of RIA buffer or standards or samples diluted in this buffer and 100 ul of the 125τ_p_robe (~30,000 CPM) were incubated at room temperature for 2.5 hours or other time intervals. The radioactivity in each well was determined in a TopCount (Packard Instruments) scintillation spectrophotometer and transformed into B/Bo. The B/Bo values as a function of standard peptide concentration were plotted by a 4- parameter algorithm (Immunofit Packard Instruments). The B/Bo of the samples were similarly determined and the concentration of rH- iNOS, expressed as fmoles of unlabeled peptide, was determined from the standard curve. Only values falling between 20-80 B/Bθ% binding were considered to be valid for calculations. Using this new technology it was possible to obtain accurate data after only 2-3 hours of incubation as compared to exisisting RIA methods (for example Protocol 1) which typically employ overnight incubations. In addition, it is readily apparent that other means of detection of a radioisotope are suitable for use in the present invention such as the SPA system utilizing scintillent- coated beads (Amersham).

Determination of Antibody Binding The antibody titer for the most active rabbit antisera against the peptide antigen was determined using a radioimmunoassay such as protocol 1 (Figure 1). The antiserum was diluted in phosphate buffered saline with dilutions ranging from 1 :10,000 to 1 :500,000 per 100 μl. The diluted antiserum was contacted with 125l-Tyr-Ser-Leu-Glu-Met- Ser- Ala-Leu (SEQ.ID.No.:3) radiolabeled probe. The radioactive probe was diluted in assay buffer to yield approximately 30,000 cpm per 100 μl aliquot. The assay volume was made up to 200 μl by the addition of 100 μl assay buffer. All determinations were made in duplicate.

Data Generated using the Immunoassays Described above

The antibody was shown to be specific for the peptide YRASLEMSAL (SEQ.ID.NO.:3), and that the carboxyl-terminal leucine is required for antibody recognition by the ability of the radiolabelled peptide to be displaced by unlabelled peptide (Figures 2 and 3). The sensitivity of the assay to amino acid substitutions at the carboxy- terminal leucine further demonstrates the specificity of this antibody for hiNOS (Figure 2). The RIA's were performed in test tubes with an 18 hour (Protocol 1) incubation as well as in Flashplate wells with a 2.5 hour incubation (Protocol 2). The competitive displacement of the peptide did not appreciably change as a function of incubation time. It was also demonstrated that the binding of the radiolabelled peptide was dependent on antibody concentration, and that saturation of antibody by the peptide was achieved within about 2.5 to 5 hours (Figure 4). It was shown that recombinant human iNOS displaced the bound radiolabelled peptide from the antibody as well (Figure 5). The RIA was also successfully used to monitor and follow the purification of hiNOS. The fractions resulting from the chromatographic steps of purifying hiNOS were analyzed by RIA and the amount of hiNOS in each fraction was measured (Figure 6). The RIA was shown to quantify the amount of hiNOS in cytokine-stimulated human hepatocytes (Table 1 ). Human hepatocytes were incubated with a mixture of cytokines. The hepatocytes were then lysed into a buffer containing protease inhibitors and the amount of hiNOS was determined in the RIA at four dilutions. The results are shown in Table 1 and demonstrate a substantial increase in hiNOS following cytokine-stimulation of the hepatocytes.

TABLE 1

nM hiNOS

Treatment Experiment 1 Experiment 2 cytokine stimulated 238 + 5(a) 292 + 59 untreated 0 0 (a) - mean ± SD.

EXAMPLE 4

Western Blots Using hiNOS Antisera Gel electrophoresis was performed on 4-20% (w/v) poly aery lamide gels using the procedure of Laemmli {Nature 227, 680- 685, 1990). The protein components were electrophoretically transfered onto nitrocellulose membranes. The rH-iNOS components were visualized with the iNOS specific antiserum N053 diluted 1 :40,000 as the detection reagent and sheep anti-rabbit IgG conjugated to alkaline phosphatase. It was demonstrated that the antibody detects hiNOS in Western blots, and that the binding was blocked in a concentration- dependent manner (Figures 7 and 8). Immunostaining of hiNOS was detected at dilutions of antiserum as low as 1 :160,000. The immunostaining was specific for hiNOS and was demonstrated by Western blot analysis. No staining of mouse iNOS, or the other constitutive isotypes of human NOS was seen on Western blots. The antibody was also able to detect hiNOS in 3T3 cells producing hiNOS, but not in 3T3 cells not producing hiNOS (Figure 8). Protein concentration was determined using the method of Bradford (Bradford M.M. 1976, Analytical Biochemistry, 73, p 248) with bovine serum albumin as the standard.

EXAMPLE 5

Modulator Screening Assay

Cells or tissues are treated with a modulator of hiNOS activity prior to collection of a sample for hiNOS evaluation. Control cells or tissues are treated with vehicle alone prior to collection of a sample for hiNOS evaluation. After a period of time sufficient to allow the modulator to produce its effect of hiNOS modulation the samples for analysis are collected. The level of epitope in the sample is determined by the immunoassays disclosed herein, including the RIA. Upon treatment with a hiNOS modulator, there is an increase or reduction in the amount of epitope in the treated samples compared to untreated samples. The result is confirmed by a standard enzyme activity assay. NOS activity is measured as the formation of L-[2,3,4,5-3H]Citrulline from L-[2,3,4,5-3H]Arginine. The incubation buffer (100 μL) contains: 100 mM TES, pH 7.5, 5 μM FAD, 5 μM FMN, 10 μM BH4, 0.5 mM NADPH, 0.5 mM DTT, 0.5 mg/mL BSA, 2 mM CaC12, 10 μg/mL calmodulin (bovine), 1 μM L-Arg, 0.2 μCi L-[2,3,4,5-³H]Arg, and the modulator in aqueous DMSO (max. 5 %). The reaction is initiated by addition of enzyme. Incubations are performed at room temperature for 30 minutes and stopped by the addition of an equal volume of quenching buffer consisting of 200 mM sodium citrate, pH 2.2, 0.02% sodium azide. Reaction products are separated by passing through a cation exchange resin and quantitated as cpm by scintillation counting. Percent modulation is calculated relative to enzyme incubated without modulator according to: % modulation = 100 x (cpm L-[2,3,4,5-3H]Cit with modulator/cpm L-[2,3,4,5-^H]Cit without modulator). EXAMPLE 6

Immunoaffinitv column construction and use: The ImmunoPure IgG Orientation Kit protocol was followed in toto except for the following modifications: 6.0 ml of antiserum was diluted with an equal volume of 0.075 M phosphate buffered saline -0.02% NaN3. 1 ml of 3.0 mg/ml bovine thyroglobulin in PBS-azide was added and allowed to stand for 18 hours at 4°C. The solution was centrifuged at 3000 rpm for 10 minutes and the precipitate discarded. The supernate was added to a 25- ml bed of Protein- A agarose packed in a Bio-Rad glass column, capped, slurried and rotated end-over-end for 30 minutes at 25°C. Following 10 volumes of PBS-azide wash, the column was mixed with 15 ml containing 100 umol of N-acetylated peptide for 30 minutes and thereafter drained and washed with 2 volumes of coupling buffer (0.2 M triethanolamine, pH 8.2 with 0.02% NaNβ. 100 mg of DMP

(dimethylpimelimidate) dissolved in 14 ml of coupling buffer was added, the column resuspended and rotated end-over-end for 90 minutes at 25°C. The column was drained and 14 ml of blocking buffer (0.2 M ethanolamine, pH 8.2 with 0.02% NaN3) was added and mixed for an additional 10 minutes. The column was drained and then stripped for non-crosslinked proteins using 0.2 M sodium citrate, pH 3.0. The final wash and storage consisted of 5 volumes of PBS-azide. The precipitation of the polyclonal antiserum using bovine thyroglobulin eliminated a significant amount of non-specific IgGs (~4.0 mg/ml of antiserum). Initial attempts to couple this clarified antiserum to a solid support using the ImmunoPure IgG Orientation Kit protocol resulted in excellent coupling efficiency but there was a concomitant loss of binding for the N054 peptide and/or the enzyme. This was presumably due to the modification of the antigen binding sites of the antibody with the reagent dimethylpimelimidate (DMP) during the crosslinking step. To obviate this, an acetylated form of the cognate peptide N054 was preincubated for 30 minutes with the antibody-Protein A complex inorder to provide a steric protection for the antibody binding sites. The acetylation of the cognate peptide was intended to prevent possible crosslinking of its free N-terminus to the antibody. From earlier specificity studies, no loss of recognition by the antibody after modifications distal to the C-terminus of the peptide was seen. Following the preincubation, excess peptide was removed by washing the column with 2 volumes of the crosslinking buffer, then DMP was added. The resulting column showed high selectivity and binding capacity for the cognate N054 peptide and its acetylated form.

Immunopurification of the enzvme: The entire procedure was done at 4°C. Two separate batches of 150-ml 100,000xg crude Sf9 lysates were processed identically. The lysates were diluted with equal volumes of running buffer (20 mM TES, pH 7.4, 1 mM CHAPS, 100 mM BH4, 0.1 mM DTT and 250 mM NaCl) then clarified using a Microgon Filtration System 0.22 micron hollow fiber filters (Laguna Hills, Ca). The clarified lysates were infused onto identical 25-ml columns using

Hamilton syringe pumps at a flow rate of 3.0 ml/min. The wash steps consisted sequentially of 8 volumes of the running buffer and equilibrated back with 2 volumes of the running buffer containing 0.1 M NaCl. For elution, 8 ml of a 25 uM solution of the cognate peptide dissolved in TES running buffer (0.1 M NaCl) were added, the columns resuspended, capped and rotated end-over-end for 1.5 hours. The columns were then drained and washed with an additional 8 ml of the peptide solution and both fractions pooled. This step was repeated 3 times, the final desorption was with an overnight incubation. Overall recovery of enzyme activity was 40-60% of the total amount loaded. For subsequent purifications, the columns were regenerated by stripping off the bound cognate N054 peptide using 0.1 M sodium citrate of pH 3.0.

Analysis of the flow-through and wash fractions using the ^H-arginine-to-^H citrulline conversion assay (ACCA) showed no detectable signals suggesting total depletion of enzymatic activity from the lysates. Figure 6 shows the immunoaffinity purification steps for recombinant human hepatocyte iNOS presented as enzyme activity measured in a 30-minute ^H-arginine to ^H citrulline conversion assay. (Fractions 1-12, flow-throughs; 13-28, wash steps; 29-32, elution steps. (Inset) Silver-stained SDS-PAGE of the purification steps. Lane A, molecular weight markers, lane B, crude lysate; lane C, flow-through; lanes D-E, wash steps; lanes F-I, fractions 29-32. Molecular weight markers are myosin (200,000), β-galactosidase (116,000), phosphorylase b (97,400), bovine serum albumin (66,000) and ovalbumin (45,000). A predominant 130 kDa band was evident in all the eluted fractions, and this profile was identical for both batches. Minor bands were noted at molecular weights above and below the 130 kDa band. By densitometric measurement, the 130 kDa band was deemed >95% pure. The corresponding Western blots using the anti- N053 antiserum as the primary antibody confirmed the presence of the 130 kDa band in both the eluted active fractions and the crude lysate. The minor bands were also recognized suggesting that these species were truncations and/or degraded forms of the 130 kDa band that still contained the intact C-terminus. This was confirmed in blocking experiments where the anti-N053 antibody was first incubated to saturation with the cognate peptide N054 and then used in the immunoblots resulting in all the bands disappearing. The appropriate control using an unrelated peptide for the preincubation failed to block staining of the bands.

DEAE chromatography:

The immunoaffinity-purified enzyme from the two batches were each diluted 1 :3 with TES buffer containing 0.01 M NaCl up to a total volume of 190 ml. This brought the final salt concentrations to 0.03 M, allowing for binding to a DEAE-5PW ion exchange column. A linear gradient of 0.01 M-0.5 M NaCl in TES buffer eluted the enzyme at -0.175 M into a 2.0-ml aliquot while the cognate N054 peptide bound more tightly to the columns and was eluted at the 0.5 M NaCl point. This step provided highly concentrated enzyme solutions, with recoveries of -80-90%. EXAMPLE 7

Immunoperoxidase Localization of iNOS in Human Tissues

Inducible nitric oxide synthase (iNOS) has been implicated in the pathophysiology of a large number of diverse chronic inflammatory diseases, as the source of nitric oxide and its stable toxic endproduct nitrotyrosine. However, it is difficult to decide which disease should be treated with a selective iNOS inhibitor without direct evidence for the presence of iNOS in effector cells and at sites of tissue pathology. Therefore, it would be advantageous if one could localize iNOS directly at sites of tissue damage and within leukocytes that cause chronic inflammatory disease. Nevertheless, detection of iNOS activity in normal or challenged human leukocytes in vitro has been impossible to sporadic. Available iNOS antibodies do not specifically recognize iNOS epitopes thought to be present in fixed and embedded human tissues obtainable for retrospective analysis. We hypothesized that generation of small anti-peptide IgG that specifically recognizes iNOS would solve this problem. Accordingly, rabbit anti-peptide antibody (NO-53) was raised against the C-terminus of iNOS as described above. It specifically detects a single 130 kD species in immunoblots of iNOS- expressing cells, but does not react with the other constitutive isoforms of NOS (ncNOS and ecNOS). We have used NO-53 to immunolocalize iNOS in histological sections of human tissues as follows:

Protocol for immunoperoxidase localization of iNOS in human tissue sections using NQ53

Appropriate human tissues were fixed in 10% neutral buffered formalin, paraffin embedded, and 5 μM sections cut and attached to precoated glass slides (polylysine or Vectabond). Designated slides were treated with proteinase K (Dako Corp; diluted 1:1 with 0.05M Tris-HCL, pH7.5) for 3 minutes at 23°C. Endogenous peroxidases were blocked with 3% hydrogen peroxide in methanol (10ml 30% peroxide/90 ml methanol) for 20 minutes. The slides were washed three times for one minute in 3 changes of 0.1 M phosphate buffer. The sections were treated with 0.1% Triton X100 in PBS at 4°C for 5 minutes, washed two times for one minute in several changes of 0.1M phosphate buffer. Endogneous biotin-binding sites were blocked by incubation with the avidin solution (Vector) for 15 minutes at 23°C. The slides were then washed three times for one minute with several changes in 0.1 M phosphate buffer (OXOID tablets pH 7.8) containing 0.2% bovine serum albumin (BSA) and 0.01 % Tween-20 for lhour at 23°C in moist chambers. This buffer is used for all washing steps below unless otherwise indicated. Endogenous avidin-binding sites were blocked by incubation with biotin solution (Vector) for 15 minutes at 25°C. The slides were then washed three times for one minute with several changes in 0.1 M phosphate buffer. Nonspecific binding was blocked for 20 minutes with normal serum obtained from the species used to generate the secondary antibody, and dissolved in 0.1 M phosphate buffer. The slides were then washed three times for one minute in several changes of 0.1 M phosphate buffer.

The sections were then incubated in NO-53 (l-5μg/ml in 1% BSA, 0.1 % NaN3 and 0.1 M phosphate buffer) for 1 hour at 23°C in moist chambers (for controls, antibodies were incubated with appropriate blocking or control peptides for 1 hour at room temperature, and clarified by centrifugation prior to incubation with the sections). The slides were then washed five times for two minutes in several changes of 0.1 M phosphate buffer.

Groups of slides were stained with biotinylated secondary antibody (Vector Labs) dissolved at l-5μg/ml in 0.1% BSA, 0.1M phosphate buffer for 30 minutes at 23°C in moist chambers, and washed five times for two minutes in changes of 0.1 M phosphate buffer.

The sections were postfixed in 3.5% formaldehyde [10ml of 16% formaldehyde (Poly sciences) plus 35ml buffer] in 0.1 M phosphate buffer (pH 7.4) for 15 minutes. The slides were then washed three times for two minutes with changes in 0.1 M phosphate buffer (pH 7.8).

Vectastain ABC solution was prepared according to the manufacturer's directions in 0.1 M phosphate buffer (pH 7.8) containing 0.1% bovine serum albumin 30 minutes before use. Slides were stained for 30 minutes at 23°C. The slides were then washed five times for two minutes in changes of 0.1 M phosphate.

GOD/DAB/Nickel substrate solution was prepared freshly before use according to Shu, S. et al "The glucose oxidase-DAB nickel method in peroxidase histochemistry of the nervous sytem" in Neuroscience Letters, 85, 169-171 , 1988, as follows: a. Stock solution A: 1.25g Nickel ammonium sulfate in 25 ml 0.2M acetate buffer (pH 6.0). b. Stock solution B: 30 mg 3,3'-diaminobenzidine tetrahydrochloride in 25 ml solution of 0.05%

Tween-20 in distilled water (8.3ml dH20, 5ml Tween-20/10mg DAB). c. Stock solution C: 10 mg glucose oxidase (Boehringer Mannhein, grade I) in 10 ml 0.1 M acetate buffer (pH 6.0) divide into 0.5ml aliquots and store of -80°C. d. 25 ml of stock A was added to 25 ml of stock B and 100 mg of b-D-glucose, 20 mg ammonium chloride, and 0.5 ml of thawed stock solution C was added. This final substrate mixture was filtered through a 0.2μM pore membrane filter.

The slides were rinsed in 0.1 M acetate buffer (pH 6.0), and sections were incubated in the GOD/DAB/Nickel substrate solution for 10 minutes at 23°C, then the reaction was stopped by washing with 0.1M acetate buffer (pH 6.0). The slides were washed in running tap water for 5 minutes, then counter stained in diluted eosin, dehydrate, cleared and coversliped.

Results: Detection of hiNOS in hepatitis and inflammatory bowel disease.

High levels of hiNOS were identified in the post-mortem liver of a patient who had hepatitis. Figure (9) shows a liver section with a venule containing conspicuous iNOS-positive mononuclear cells that comprise a large intraluminal aggregate (arrow), and are also adherent to the endothelium (arrowheads). iNOS staining was performed with 5μg/ml NO-53 IgG. The iNOS staining of mononuclear cells in an adjacent section was eliminated by pre-absorbing 5μg/ml NO- 53 IgG with 50 nM NO-54, the blocking peptide (Figure 10). The background particulate material (arrowheads) was present in unstained sections of this specimen, and is of endogenous origin. The iNOS staining of an adjacent section was unaffected by pre-absorbing 5μg/ml NO-53 IgG with 50 nM NO-70, a non-blocking control peptide lacking the natural C-terminal alanine (Figure 11 ). Therefore, the labeling of mononuclear cells in human hepatitis is specific for the C-terminal epitope of human iNOS. NO-53 antibody detected hiNOS in a section of resected colon from patient who had symptoms of ulcerative colitis for 10 years, when labeled with 2.5μg/ml NO-53 IgG (Figure 12). Intense focal iNOS labeling is localized in the epithelium (E) surrounding the villi, and within the inflamed crypts. Mononuclear cells in the lamina propria (arrows), and many polymoφhonuclear leukocytes within the lumen of inflamed crypts (arrowheads), are also strongly positive for iNOS.

The NO-53 antibody also detected hiNOS in a section of resected colon from a patient who had Crohn's disease for 27 years, labeled with 1 μg/ml NO-53 IgG pre-absorbed with 50nM NO-70 (the non-blocking control peptide) (Figure 13). Intense diffuse iNOS labeling is present throughout the epithelium surrounding the villi (arrow), and within the crypts (arrowheads). A section adjacent to that depicted in Figure 13 was labeled with 1 μg/ml NO-53 IgG pre-absorbed with 50nM NO-54 (blocking peptide) (Figure 14). Epithelial iNOS labeling has been eliminated, thus attesting to the specificy of the iNOS label in the inflamed epithelium.

The C-terminal epitope of iNOS specifically recognized by NO-53 has thus been localized in the epithelium, and within the leukocytes of the lamina propria and crypt abscesses of 8 patients with ulcerative colitis, and in 9 patients with Crohn's disease. No iNOS labeling was observed in the normal colonic epithelium adjacent to the bowel cancers of 5 patients, indicating that epithelial iNOS induction is linked to inflammation. Accordingly, we also observed diffuse iNOS localization in the colonic epithelium of 3 diverticulitis patients, suggesting that epithelial induction of iNOS is a general hallmark of the inflammed human bowel. The colonic/crypt epithelium maintains the proper ionic balance of the colon, and is the site of continuous generation of new gut lining cells. Localization of iNOS at this vital locus in patients throughout the chronic course of inflammatory bowel disease strongly suggests that a potent and selective iNOS inhibitor will be of therapeutic value for the treatment of lower bowel inflammation. However, the specific detection of iNOS in mononuclear cells within the livers of 2 hepatitis patients also indicates that iNOS plays a role in host defense, and thus any selected iNOS inhibitors must be controlled very carefully for the management of chronic inflammatory disease.

Claims

WHAT IS CLAIMED IS:

1. A method for purifying hiNOS from cultured cells comprising: a) contacting a fluid containing hiNOS with antibodies specifically binding hiNOS, and b) recovering said hiNOS.

2. The method of Claim 1 wherein said antibodies are monospecific antibodies produced by immunizing an animal with a peptide having the amino acid sequence described as SEQ.ID.NO.:l, wherein said antibody binds to an epitope located at the carboxyl terminus of hiNOS and does not bind to other isoforms of NOS or nonhuman iNOS.

3. The method of Claim 1 wherein said contacting step comprises passing a fluid containing hiNOS through a chromatographic bed having said antibodies immobilized thereon.

4. The method of Claim 3 wherein said antibodies are monospecific antibodies produced by immunizing an animal with a peptide having the amino acid sequence described as SEQ.ID.NO.:l, wherein said antibody binds to an epitope located at the carboxyl terminus of hiNOS and does not bind to other isoforms of NOS or nonhuman iNOS.

5. The method of Claim 1 wherein said fluid containing hiNOS comprises culture medium from cells producing hiNOS, or cell extracts from cells producing hiNOS.

6. The method of Claim 5 wherein said fluid containing hiNOS comprises culture medium from recombinant SF9 cells producing hiNOS, or recombinant SF9 cell extracts from recombinant SF9 cells producing hiNOS.