Tumor-specific P450 protein

Abstract

The discovery that CYP1B1 protein is detectable in a wide range of human cancers of different histogenetic types, but is not detectable in non-cancerous tissues, gives rise to diagnostic methods for detecting tumors based on this protein as a marker, and to the possibility of tumor therapies involving the protein. A diagnostic method may include the steps of (a) obtaining from a patient a tissue sample to be tested for the presence of cancer cells; (b) producing a prepared sample in a sample preparation process; (c) contacting the prepared sample with an antibody that reacts with human CYP1B1 protein; and (d) detecting binding of the antibody to CYP1B1 protein in the prepared sample.

Description

TECHNICAL FIELD

[0002] This invention relates to tumor diagnosis and therapy, and to materials and methods for use therein.

[0003] More particularly, the invention is based on the identification of a cytochrome P450 form, specifically CYP1B1, in a wide range of tumors, with a high frequency of expression in each type, and proposes the use of this enzyme as a tumor marker, and as the basis of a selective therapeutic approach involving the design of drugs, e.g., which are activated to a cytotoxic form by the action of CYP1B1.

BACKGROUND

[0004] The major goal of cancer chemotherapy is the development of anti-cancer drugs that are effective in a wide range of cancers and produce no toxic effects in normal tissues. The target of such drugs should be expressed only in tumor cells and not normal cells. However, to date, no such tumor specific target, general to all types of cancers, has been identified.

[0005] The cytochromes P450 are a multi-gene family of constitutive and inducible enzymes, which have a central role in the oxidative metabolic activation and detoxification of both a wide range of xenobiotics (2-4) and several groups of endogenous compounds active in cell regulation and cell signaling including arachidonic acid (5), steroid hormones (6) and fatty acids (7). The major families of P450 involved in xenobiotic metabolism each consist of several individual forms with different regulatory mechanisms and substrate specificities (2). The majority of P450s are primarily expressed in liver (2) although individual P450 forms are also expressed in specific extra-hepatic tissues (8) including small intestine, kidney and lung.

[0006] The human CYP1 gene family (individual P450 forms are identified by the prefix CYP in accordance with the current P450 nomenclature (3)), which is one of the major P450 families involved in the metabolism of xenobiotics, is now known to consist of three individual forms classified into two sub-families. The CYP1A subfamily contains two highly homologous and well characterized but distinct members, CYP1A1 (9) and CYP1A2 (10). CYP1A1 is an inducible P450 expressed primarily in extrahepatic tissues (11) while CYP1A2 is a major form of P450 that is constitutively expressed in liver (12). Recently a second human CYP1 subfamily has been identified which to date contains one member, CYP1B1 (1). This P450 is dioxin-inducible, and sequence analysis of CYP1B1 shows 40% homology with both CYP1A1 and CYP1A2. Although CYP1B1 is assigned to the CYP1 family on the basis of its sequence, it appears to be structurally distinct from both CYP1A1 and CYP1A2.

[0007] Several forms of P450 are considered to have an important role in tumor development since they can metabolize many potential carcinogens and mutagens (13). Moreover, P450 activity may influence the response of established tumors to anti-cancer drugs; several cancer chemotherapeutic agents can be either activated or detoxified by this enzyme system (14). The presence of individual forms of P450 had previously been investigated in different types of cancer including breast cancer (15), lung cancer (16), colon cancer (17) and head and neck cancer (18) to determine if intra-tumor metabolism of anti-cancer agents by P450 could occur and thus influence the response of tumors to these agents. These studies have generally shown that the level of the P450 forms investigated is significantly reduced or absent in tumors when compared with the adjacent normal tissue in which the tumors have developed. However, our recent studies of several different types of cancer (19) including breast cancer, esophageal cancer and soft tissue sarcomas have shown that there may be tumor-specific expression of a CYP1 form of P450.

[0008] Although CYP1B1 mRNA had previously been identified by Northern blotting in several normal human tissues (1), the presence of CYP1B1 protein itself had not been demonstrated.

SUMMARY

[0009] The present invention is based on the discovery that CYP1B1 is a tumor-specific form of P450, present in a wide range of malignant tumors and not detected in normal tissues.

[0010] Accordingly, a first aspect of the present invention provides a method for the identification of tumor cells, which method comprises the use of a recognition agent, for example an antibody, recognizing CYP1B1 protein to contact a sample of tissues, cells, blood or body product, or samples derived therefrom, and screening for a positive response. The positive response may, for example, be indicated by an agglutination reaction or by a visualizable change such as a color change or fluorescence, e.g. immunostaining, or by a quantitative method such as in the use of radio-immunological methods or enzyme-linked antibody methods.

[0011] The method therefore typically includes the steps of (a) obtaining from a patient a tissue sample to be tested for the presence of cancer cells; (b) producing a prepared sample in a sample preparation process; (c) contacting the prepared sample with a recognition agent, such as an antibody, that reacts with human CYP1B1 protein; and (d) detecting binding of the recognition agent to CYP1B1 protein, if present, in the prepared sample. The human tissue sample can be from for example the bladder, brain, breast, colon, connective tissue, kidney, lung, lymph node, esophagus, ovary, skin, stomach, testis, and uterus.

[0012] A preferred sample preparation process includes tissue fixation and production of a thin section. The thin section can then be subjected to immunohistochemical analysis to detect binding of the recognition agent to CYP1B1 protein. Preferably, the immunohistochemical analysis includes a conjugated enzyme labeling technique. A preferred thin section preparation method includes formalin fixation and wax embedding. Alternative sample preparation processes include tissue homogenization, and preferably, microsome isolation. When sample preparation includes tissue homogenization, a preferred method for detecting binding of the antibody to CYP1B1 protein is Western blot analysis. Alternatively, an immunoassay can be used to detect binding of the antibody to CYP1B1 protein. Examples of immunoassays are antibody capture assays, two-antibody sandwich assays, and antigen capture assays. Preferably, the immunoassay is a solid support-based immunoassay. When Western blot analysis or an immunoassay is used, preferably it includes a conjugated enzyme labeling technique.

[0013] Although the recognition agent will conveniently be an antibody, other recognition agents are known or may become available, and can be used in the present invention. For example, antigen binding domain fragments of antibodies, such as Fab fragments, can be used. Also, so-called RNA aptomers may be used (36, 37). Therefore, unless the context specifically indicates otherwise, the term “antibody” as used herein is intended to include other recognition agents. Where antibodies are used, they may be polyclonal or monoclonal. Optionally, the antibody can produced by a method so that it recognizes a preselected epitope of said CYP1B1 protein.

[0014] A second aspect of the invention lies in the presence of CYP1B1 protein selectively in tumors, e.g. in kidney tumors and not normal renal tissue, combined with the absence of CYP1B1 protein expression in normal liver, which provides a mechanism for the selective targeting of anti-cancer drugs based on CYP1B1 metabolism in tumors. Drugs can be designed for, or screened for, specific metabolism by CYP1B1 in tumors whereby this metabolism converts a non-toxic moiety into a toxic one, which kills or inhibits the tumor or makes it more susceptible to other agents.

[0015] A third aspect of the invention provides for the targeting of cytotoxic drugs or other therapeutic agents, or the targeting of imaging agents, by virtue of their recognition of CYP1B1 epitopes on the surface of a tumor cell, whether as part of the complete CYP1B1 protein itself or in some degraded form such as in the presentation on the surface of a cell bound to a MHC protein.

[0016] Another aspect of the invention provides stimulation of the immune system of cancer patients, for example by activating cytotoxic or helper T-cells which recognize CYP1B1 epitopes so as to implement a cell-mediated or humoral immune response against the tumor. The activation of the immune system can be achieved by immunization with CYP1B1 sequences.

[0017] Because the expression of CYP1B1 is very common in tumors of many different types, it is likely that this enzyme performs an essential function for the tumor cells, for example by inactivating endogenous anti-tumor compounds such as 2-methoxyestradiol.

[0018] Consequently, another aspect of the invention is the reduction of CYP1B1 levels in tumor cells, for example, by the use of suicide inhibitors or by using antisense RNA methods to decrease the synthesis of the protein. Down-regulation of the CYP1B1 promoter could also achieve the reduction in CYP1B1 levels.

DESCRIPTION OF DRAWINGS

[0019] FIG. 1 shows SDS-PAGE and immunoblotting procedures using an anti-CYP1B1 antibody to look for CYP1B1 protein in normal and tumorous kidney tissue and in normal liver tissue.

[0020] FIG. 2 shows SDS-PAGE and immunoblotting to detect CYP1B1 protein in breast tumor and in normal liver tissue.

[0021] FIG. 3 shows an immunoblot of CYP1B1 in different types of tumors and normal tissues.

[0022] FIG. 4 shows CYP1B1 and β-actin mRNA in normal (A and B) and corresponding tumor (C and D) samples which have been detected by RT-PCR.

DETAILED DESCRIPTION

[0023] Expression of CYP1B1 was investigated in different types of cancers that had developed in a broad range of different anatomical sites (bladder, breast, colon, kidney, lung, esophagus, ovary, skin, stomach, uterus, bone and connective tissue, lymph node, brain and testis). Primary malignant tumors of these tissues constitute different histogenetic types (carcinomas, lymphomas, sarcomas, neuro-epithelial tumors and germ cell tumors) each with a different biological behavior. These tumors also represent a range of both common and less common types of cancer. The presence of CYP1B1 was also investigated in a wide variety of normal tissues.

[0024] Immunohistochemistry for CYP1B1 showed that in all the different types of tumor there was strong immunoreactivity for CYP1B1. CYP1B1 immunoreactivity was localized specifically to tumor cells. Non-tumor cells including stromal cells, inflammatory cells, and endothelial cells present in the sections of tumor showed no immunoreactivity for CYP1B1. There was no significant intra-tumor heterogeneity of CYP1B1 immunoreactivity and only in five out of 133 tumors was CYP1B1 not detected. There was no immunoreactivity for CYP1B1 in any of the normal tissues studied which included liver, kidney, small intestine and lung.

[0025] The absence or low level of individual forms of P450 in most studies of human cancer (15-18), combined with extrapolation from studies of rodent hepatic carcinogenesis (25), had led to the general belief that tumor cells do not significantly express P450. However, we have now shown that CYP1B1 is expressed in a wide variety of malignant tumors of different histogenetic types and is not present in normal tissues, indicating that this P450 is a tumor specific form of P450. Tumors are composed of a variable proportion of tumor cells and non-tumor cells. To identify that a protein is tumor specific, it is important to demonstrate that the protein is localized only to tumor cells. Immunohistochemistry allows the direct visualization of tumor cells and has the spatial resolution to separate tumor cells from non-tumor cells. Furthermore, it is important to show there has been no differential degradation of proteins in normal tissue samples compared with tumor samples, and immunoblotting for β-actin (as a positive control protein) showed it to be present in every normal and tumor sample indicating there was no protein degradation. In addition, Coomassie blue staining of the polyacrylamide gels showed no evidence of protein degradation. Moreover, immunohistochemistry of the tumor samples provides its own internal control as sections of tumor contain non-tumor cells.

[0026] The presence of CYP1B1 in many types of tumor suggests that this P450 may have a crucial endogenous function in tumor cells and CYP1B1 may contribute to drug resistance that is observed in many types of tumor. CYP1B1 is also likely to be important in tumor development and progression. Its identification in a diverse range of cancers of different histogenetic types and its absence from normal tissues appears to make CYP1B1 one of the common changes of a gene product in malignancy (26).

[0027] A previous investigation (1) found mRNA in normal tissues. In some tumors, increased (2-4×) mRNA was found compared with normal. This might be due to increased transcription mediated by hypoxia inducible factor. This is a novel heterodimeric transcription factor which is induced by hypoxia, and the stimulus in this case may be the hypoxic micro-environment that can exist in tumors, and this factor can have as one of its components the Ah receptor nuclear translocator (27). However, regulation of other forms of P450 is complex (2, 28) and the regulation of CYP1B1 in tumors is likely to be complex also, with multiple mechanisms including transcriptional and post-transcriptional factors involved.

[0028] The tumor-specific expression of CYP1B1 has important consequences for both the diagnosis and treatment of cancer. New diagnostic procedures based on the presence of CYP1B1 in cancer cells can be developed, while the expression of CYP1B1 in tumor cells provides a molecular target for the development of new anti-cancer drugs that are selectively activated by CYP1B1 in tumor cells. Since CYP1B1 is found in a wide range of tumors it would be expected that such drugs would be effective in treating many different types of cancer. An important feature is that it would be anticipated these drugs would not be associated with the systemic toxicity that limits the use of current anti-cancer drugs as CYP1B1 is not present in normal tissues especially liver, small intestine and kidney that are the main tissues involved in drug metabolism. Thus, a major problem to targeting anticancer drugs at tumors based on their selective activation by P450 has been marked hepatic P450 metabolism of drugs resulting in decreased bioavailability and/or undue toxicity. The absence of CYP1B1 protein in liver overcomes this problem.

[0029] As regards tumor diagnosis, numerous methods for using antibodies to detect a specific protein, including CYP1B1, in a biological sample are known and can be used in the present invention. Any of the various antibody methods can be used alone in practicing the present invention. If desired, two or more methods can be used to complement one another.

[0030] A preferred method for use in the present invention is immunohistochemical analysis. Immunohistochemical analysis advantageously avoids a dilution effect when relatively few cancer cells are in the midst of normal cells. An early step in immunohistochemical analysis is tissue fixation, which preserves proteins in place within cells. This prevents substantial mixing of proteins from different cells. As a result, surrounding normal cells do not diminish the detectability of CYP1B1-containing cancer cells. This is in contrast to assay methods that involve tissue homogenization. Upon tissue homogenization, CYP1B1 protein from cancer cells is mixed with proteins from any surrounding normal cells present in the tissue sample. The concentration of CYP1B1 protein is thus reduced in the prepared sample, and it can fall below detectable limits. Immunohistochemical analysis has at least three other advantages. First, it requires less tissue than is required by alternative methods such as Western blot analysis or immunoassay. Second, it provides information on the intracellular localization and distribution of immunoreactive material. Third, information on cell morphology can be obtained from the same thin section used to test for the presence of CYP1B1 protein. Preferably, when immunohistochemical analysis is employed in the practice of this invention, several thin sections from each tissue sample are prepared and analyzed. This increases the chances of finding small tumors.

[0031] Another preferred antibody method for use in the present invention is Western blot analysis, i.e., sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), followed by immunoblotting. Sample preparation for Western blot analysis includes tissue homogenization, and optionally isolation of microsomes. Western blot analysis has the advantage of detecting immunoreactivity on proteins that have been separated with high resolution, according to (apparent) molecular weight.

[0032] Immunoassays such as antibody capture assays, two-antibody sandwich assays, and antigen capture assays can also be used in the present invention. Sample preparation for immunoassays includes tissue homogenization, and optionally isolation of microsomes. Immunoassays have the advantage of enabling large numbers of samples to be tested relatively quickly, and they offer quantitative precision.

[0033] Principles and practice of immunohistochemistry, Western blot analysis, and immunoassays are well known. One of ordinary skill in the art can select suitable protocols and carry out immunohistochemical analysis, Western blot analysis, or an immunoassay, in the practice of the present invention (30).

[0034] Experimental Details

[0035] 1. Preparation of Antibodies

[0036] As already noted, the diagnostic aspects of the present invention can conveniently use an antibody that recognizes human CYP1B1. Antibody specificity for CYP1B1 protein is preferable, but not required. Preferably, any non-CYP1B1 protein recognized by the antibody is readily distinguished from CYP1B1, e.g., according to apparent molecular weight on a Western blot. With selection of an appropriate assay-protocol, which is within ordinary skill in the art, the invention can be practiced with a polyclonal antibody or a monoclonal antibody.

[0037] A polyclonal antibody or monoclonal antibody suitable for use in the present invention can be obtained according to conventional procedures (30). Preparation of antibodies that react with CYP1B1 protein is known (31). Procedures for obtaining antibodies that react with human CYP1B1 protein can be carried out using a preparation of non-human CYP1B1 protein, e.g., murine CYP1B1 protein. A CYP1B1 protein preparation suitable for eliciting antibodies useful in the present invention can be obtained according to various procedures, including those described (31).

[0038] Antibodies useful in the present invention can be obtained by immunizing an animal with a preparation containing intact CYP1B1 protein. Alternatively, useful antibodies can be obtained by immunizing an animal with a polypeptide or oligopeptide corresponding to one or more epitopes on the CYP1B1 protein.

[0039] Preparation

[0040] To prepare an antibody according to the latter approach, two 15-mer peptides corresponding to epitopes on the human CYP1B1 protein were synthesized. Each corresponded to a different putative surface loop region of the CYP1B1 enzyme. The first peptide (designated 217A) consisted of 14 amino acids, i.e., ESLRPGAAPR DMMD (SEQ ID NO: 1). Peptide 217A represented amino acid positions 312-325 of the deduced amino acid sequence. A carboxy terminal cysteine was included for use in a conjugation reaction. The second peptide (designated 218A) consisted of 14 amino acids, i.e., EKKAAGDSHG GGAR (SEQ ID NO: 2). Peptide 218A represented positions 332-345 of the deduced amino acid sequence. A carboxy terminal cysteine was added for use in a conjugation reaction. Each of these peptides was conjugated directly to KLH.

[0041] Male New Zealand rabbits were immunized at several anatomical sites using 100 μg of the 217A peptide conjugate or the 218A peptide conjugate. The conjugates were dissolved in 300 μl of PBS mixed with 300 μl of Freund 's Complete Adjuvant. Three weeks after the initial immunization, the rabbits were boosted with 50 μg of one of each of the conjugates (contained in 300 μl PBS mixed with 300 μl of Freund 's Incomplete Adjuvant, injected in several sites). One week later (four weeks after the initial injection), the rabbits were boosted again, using the same protocol. One week after this second boost, the first serum sample was collected. Rabbits were subsequently boosted, and serum samples were collected, weekly. Serum samples were screened for anti-CYP1B1 titer and specificity by Western blotting against a human CYP1B1-maltose binding fusion protein expressed in E. coli, and human CYP1B1 protein expressed in COS-1 cells.

[0042] Anti-CYP1B1 IgG was purified by immunoafftinity chromatography. The chromatography was carried out using the appropriate CYP1B1 peptide linked directly to a commercial N-hydroxysuccinamide ester of a derivatized, cross-linked agarose gel bead support (AffiGel 10; Biorad, Richmond, Calif.). Conjugation and chromatography were performed according to the vendor 's recommended protocols.

[0043] 2. Detection of CYP1B1 Protein and its mRNA

[0044] In general in the experiments that follow, samples of normal tissue were obtained from tissue specimens that were removed from patients undergoing surgery for malignant disease. Normal liver, stomach and small intestine were also obtained from organ transplant donors. All the tissue samples were processed immediately after excision to prevent any degradation of protein or mRNA and ensure no deterioration in tissue morphology. We have previously shown that human liver obtained in this way shows no loss or degradation of individual forms of hepatic P450 (20). Tissue blocks for immunohistochemistry were fixed in 10% neutral buffered formalin for 24 hours and then embedded in wax, while tissue samples for immunoblotting and mRNA analyses were rapidly frozen in liquid nitrogen and stored at −80° C. prior to use.

[0045] The presence of CYP1B1 protein in tissue samples was investigated using immunohistochemistry (21) and sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) combined with immunoblotting (22). Immunohistochemistry ensures the identification of specific types of cells containing CYP1B1 and is an ideal technique for investigating the presence of CYP1B1 in tumor cells as tumors are composed of a variable proportion of tumor cells and non-tumor cells.

[0046] Immunohistochemistry was used to determine the cellular localization and distribution of CYP1B1 and was performed on formalin fixed wax embedded sections using two antibodies which recognize CYP1B1. Sites of immunoreactivity were detected using an alkaline phosphatase anti-alkaline phosphatase (APAAP) technique (21). Samples of tumor and normal tissue were fixed in 10% neutral buffered formalin for 24 hour and then embedded in wax. Sections were cut on to glass slides and for immunohistochemistry sections of tumors and normal tissues were dewaxed in xylene, rehydrated in alcohol and then washed sequentially in cold water and 0.05 M Tris-HCl (pH 7.6) containing 0.15 M sodium chloride (TBS). The sections were then immunostained with the CYP1B1 antibodies. Subsequently, monoclonal mouse anti-rabbit immunoglobulin (1/100, Dako Ltd, High Wycombe, Bucks; UK), mouse anti-rabbit immunoglobulin (1/100, Dako) and mouse monoclonal APAAP (1/100, Dako) were sequentially applied to the tissue sections for 30 minutes each. Between antibody applications the sections were washed with TBS to remove unbound antibody. Sites of bound alkaline phosphatase were identified using bromo-chloro-indolyl phosphate and nitro blue tetrazolium as the enzyme substrate. After incubating the sections for 30 minutes at room temperature, the reaction was stopped by washing the sections in cold tap water. The slides were then air-dried and mounted in glycerine jelly. The sections were examined using bright field light microscopy in order to establish the presence or absence of immunostaining, and its distribution. SDS-PAGE and immunoblotting was followed by enhanced chemihuuinescence (ECL) technique as described below. Immunoblotting was also performed with a monoclonal antibody to β-actin (clone no. AC-15, Sigma, Poole, Dorset, UK) to show the presence of a positive control protein in tumor and normal samples and indicate that there was no evidence of protein degradation in any of the tissue samples.

[0047] The presence of CYP1B1 in each tumor that showed CYP1B1 by immunohistochemistry, and the absence of detectable CYP1B1 in normal tissues, was confirmed by Western blot analysis. The proteins subjected to Western blot analysis were from isolated microsome preparations.

[0048] Microsomes were prepared essentially as described (32). Tissue samples were thawed in 25 ml 0.01 M Tris-HCl buffer, pH 7.4, containing 1.15% KCl before being homogenized in 0.01 M Tris-HCl buffer, pH 7.4, containing 0.25 M sucrose, 15% glycerol, using an Ultra-Turrax homogenizer (type TP 18/2; Janlke and Kunkel A G, Staufen Breisgau, Germany). After centrifugation at 15,000× g for 20 min, the supernatant was removed and recentrifuged at 116,000× g for 50 min. The pellet was resuspended a first time in 0.1 M Tris-HCl buffer, pH 7.4, containing 15% glycerol, 1 mM EDTA, and recentrifuged at 116,000× g for 50 min. The pellet was resuspended a second time in Tris-HCl-glycerol-EDTA buffer. Microsomal protein concentration was determined (34).

[0049] A discontinuous polyacrylamide gel system, as described (33) with modifications (32), was employed for separation of proteins in the microsomes. 20 μl of a 1 mg/ml preparation of normal samples, and 40 μl of a 0.5 mg/ml preparation of tumor samples, in 0.125 M Tris-HCl, pH 6.8, containing 2.35% (w/v) sodium dodecyl sulfate, 5% (v/v) 2-mercaptoethanol, and 0.005% bromophenol blue tracking dye were loaded onto the gel. 10 μl of a 1 mg/ml preparation of human liver microsomes were used as positive controls. Samples were run on a 10% non-gradient gel at 30 mA.

[0050] Following SDS-PAGE, resolved proteins were blotted onto a nitrocellulose membrane (Schleicher & Schuell; Dassel, Germany) overnight as described (35). Nonspecific binding site were blocked with PBS containing 2% (w/v) nonfat milk, 0.05% (v/v) TWEEN 20TM for 30 minutes at room temperature, with continuous shaking. This buffer was also used for washing stages. The nitrocellulose membrane was then incubated with CYP1B1-specific antibody (1:1000) for 90 minutes and goat anti-rabbit immunoglobulin horseradish peroxidase conjugate (1:2000 Bio-Rad Laboratories, Hemel Hempstead, Herts, UK) for 60 minutes. The membrane was washed for three successive 15-minute periods and one 60-minute period after each incubation to remove unbound antibody. Bound horseradish peroxidase was then visualized with an Enhanced Chemiluminescence (ECL) kit (Amersham International, Aylesbury, Bucks, UK). Detection was carried out as described in the ECL protocol, with the X-ray film (Hyperfilm-ECL; Amersham) being exposed for 30 seconds.

EXAMPLE 1

[0051] Expression of CYP1B1 in Normal Kidney and Kidney Tumors was Investigated.

[0052] Nephrectomy specimens (n=10) excised from primary renal cell carcinoma were used. Samples or normal kidney were taken at least several centimeters distant from the edge of each tumor, and only macroscopically viable tumor was sampled. Normal human livers (n=5) were obtained from renal transplant donors and stored at −80° C. prior to use.

[0053] Microsomes of normal kidney, kidney tumors and normal liver were prepared and subjected to the SDS-PAGE and immunoblotting procedures in an enhanced chemiluminescence technique (21, 22) using an anti-CYP1 polyclonal antibody. Recognition of human CYP1B1 was demonstrated using a maltose-binding recombinant CYP1B1 fusion protein expressed in E coli. Expressed CYP1A1 and CYP142 were supplied by Dr C L Crespi, Gentest Corp, Ma, USA. The results are shown in FIG. 1: lane 1 human liver, lane 2 expressed recombinant CYP1B1 protein, lanes 3, 5, 7, 9, 11 normal kidney samples, lanes 4, 6, 8, 10, 12 corresponding kidney tumors. The same amount of microsomal protein (30 μg) was loaded into each lane, thus allowing direct comparison between the kidney and liver samples.

[0054] As shown in FIG. 1, the kidney tumors and expressed CYP1B1 show a single immunoreactive band at 60 kDa corresponding to the molecular weight of expressed CYP1B1. In normal kidney none of the samples showed an immunoreactive band at 60 kDa. In addition, none of the kidney tumors or normal kidney samples showed the presence CYP1A1.

[0055] Immunoblotting of liver samples showed an immunoreactive band at 54 kDa corresponding to the molecular weight of CYP1A2. The intensity of the band at 54 kDa showed liver-to-liver variation, whereas there was no CYP1B1 immunoreactive band at 60 kDa in any of the liver samples.

EXAMPLE 2

[0056] The expression of CYP1B1 was also investigated in breast cancer using immunoblotting.

[0057] Samples of breast tissue were obtained from patients undergoing surgery either for primary breast cancer or non-neoplastic breast disease. Immunoblotting was performed on breast cancers obtained from six patients (age range 45-67; three non-smokers, information not available for three patients), and histologically all these tumors were carcinomas of no special type. The tissue samples were frozen in liquid nitrogen and stored at −80° C. prior to analysis.

[0058] SDS-PAGE and immunoblotting were carried out as described previously. CYP1B1 was detected using the anti-CYP1 polyclonal antibody referred to above. The results are shown in FIG. 2: lane 1 human liver, lane 2 expressed CYP1B1, lanes 3-8 breast tumors. As can be seen, a single protein band of molecular weight 60 kDa corresponding to the molecular weight of the expressed CYP1B1 protein was identified. As previously, CYP1B1 was not detectable in the liver sample, but CYP1A2 was detected.

EXAMPLE 3

[0059] Immunohistochemistry was used to demonstrate the presence of CYP1B1 specifically in a variety of normal and tumor tissues. The results are shown in Table 1 and in FIG. 3.

[0060] Immunohistological localization of CYP1B1 was investigated in tumors and normal tissues from invasive ductal carcinoma of the breast, endometrial adenocarcinoma, transitional cell carcinoma of the bladder, diffuse high grade malignant lymphoma, high grade astrocytoma of the brain, soft tissue sarcoma (malignant fibrous histiocytoma), normal liver, normal kidney, normal small intestine. The antibody used was the 218A anti-CYP1B1 polyclonal antibody described above.

[0061] FIG. 3 shows an immunoblot of CYP1B1 in different types of tumors and normal tissues. Lane 1 normal colon, lane 2 colon adenocarcinoma, lane 3 normal kidney, lane 4 carcinoma of kidney, lane 5 normal breast, lane 6 breast cancer, lane 7 normal jejunum, lane 8 normal stomach, lane 9 normal liver, lane 10 malignant mixed Müllerian tumor, lane 11 endometrial adenocarcinoma, lane 12 ovarian carcinoma, lane 13 diffuse B cell lymphoma, lane 14 transitional cell carcinoma, lane 15 lung carcinoma, lane 16 positive control (dioxin-induced ACHN kidney tumor cells (panel A only). In panel B, the same series of tissue samples have been immunoblotted for β-actin, which is present in all normal and tumor samples. A Coomassie blue stained polyacrylamide gel of the same series of tissue samples displayed no evidence of protein degradation. The results demonstrated that this P450 is specifically localized to tumor cells, and that there is no CYP1B1 immunoreactivity in normal tissues.

TABLE 1
CYP1B1 Expression in Tumor Tissues and Normal Tissues.
	Normal	Tumor
	no pos./	no pos./	Histopathological
Tissue	no tested	no tested	diagnosis
Bladder	0/8	8/8	transitional cell
			carcinoma
Brain	0/12	11/12	astrocytoma
Breast	0/10	12/12	invasive ductal carcinoma
Colon	0/10	11/12	adenocarcinoma
Connective tissue	0/9	8/9	sarcoma
Kidney	0/11	11/11	clear cell carcinoma
			n = 10;
			transitional cell
			carcinoma n = 1
Liver	0/8	not tested	not tested
Lung	0/8	7/8	squamous carcinoma
Lymph node	0/5	9/9	non-Hodgkin's
			lymphoma
Esophagus	0/8	8/8	squamous carcinoma
Ovary	not tested	7/7	Adenocarcinoma
Skin	0/6	6/6	squamous carcinoma
Small Intestine	0/5	not tested	not tested
Stomach	0/10	9/10	Adenocarcinoma
Testis	0/8	14/14	malignant germ cell
			tumors
Uterus	0/5	7/7	Adenocarcinoma n = 5;
			Malignant mixed
			Müllerian tumor n = 2
Total	0/123	128/133

EXAMPLE 4

[0062] Experiments were conducted to detect CYP1B1 RNA in various tumor and normal tissues.

[0063] Reverse transcription polymerase chain reaction (RT-PCR) experiments to detect CYP1B1 mRNA were carried out as described in McKay et al (23). RNA was extracted from tissue samples and cDNA was synthesized from the isolated RNA using oligo (dT). The CYP1B1 primers had the following sequences: Forward 5′-AAC TCT CCA TCA GGT GAG GT-3′ (SEQ ID NO:3) (nt 2104-2123); Reverse 5′-TAA GGA AGT ATA CCA GAA GGC-3′ (SEQ ID NO:4) (nt 2573-3593) giving a PCR product of 489 bp. β-actin was used as a positive control to confirm the presence and integrity of mRNA in each sample and the β-actin primers which were bought from Stratagene (Cambridge, UK) had the following sequences: Forward 5′-TGA CGG GGT CAC CCA CAC TGT GCC CAT CTA-3′ (SEQ ID NO:5) (nt 1067-1105); Reverse 5′-CTA GAA GCA TTT GCG GTG GAC GAT GGA GGG-3′ (SEQ ID NO:6) (nt 1876-1905). PCR with 35 cycles of amplification for both CYP1B1 and β-actin was performed as described (23). The positive control for CYP1B1 was a 2.78 kb CYP1B1 cDNA and the negative control was sterile water in place of cDNA. After PCR 10 μl of the PCR product was electrophoresed on a 1.5% agarose gel which incorporated 0.007% w/v ethidium bromide and visualized by UV illumination. The CYP1B1 PCR product was sequenced, after purification, by the direct dideoxy sequencing technique with a T7 sequencing kit (Pharmacia, Milton Keynes, UK) used according to the manufacturer 's protocol. To further investigate the relative amount of CYP1B1 mRNA in normal and tumor tissues, semi-quantitative RT-PCR of normal and tumor kidney samples was performed using serial dilution cDNA of (24). β-actin mRNA was used as an internal control (29).

[0064] FIG. 4 shows CYP1B1 and β-actin mRNA in normal (A and B) and corresponding tumor (C and D) samples which have been detected by RT-PCR. Lane 1 kidney, lane 2 colon, lane 3 skin, lane 4 esophagus, lane 5 stomach, lane 6 lymph node, lane 7 breast.

[0065] Analysis of the tumors by RT-PCR showed that all tumor samples in which CYP1B1 had been identified contained CYP1B1 mRNA. The PCR product was of the expected molecular size when analyzed by agarose gel electrophoresis. Sequencing of the PCR product confirmed identity with CYP1B1.

CONCLUDING REMARKS

[0066] The absence or low level of individual for P450 in most studies of human cancer (15-18), combined with extrapolation from studies of rodent hepatic carcinogenesis (25), had led to the general belief that tumor cells do not significantly express P450. However, we have now shown that CYP1B1 is expressed in a wide variety of malignant tumors of different histogenetic types and is not present in normal tissues, indicating that this P450 is a tumor specific form of P450. Tumors are composed of a variable proportion of tumor cells and non-tumor cells. To identify that a protein is tumor specific, it is important to demonstrate that the protein is localized only to tumor cells. Immunohistochemistry allows the direct visualization of tumor cells and has the spatial resolution to separate tumor cells from non-tumor cells. Furthermore, it is important to show there has been no differential degradation of proteins in normal tissue samples compared with tumor samples and immunoblotting for β-actin (as a positive control protein) showed it to be present in every normal and tumor sample indicating there was no protein degradation. In addition, Coomassie blue staining of the polyacrylamide gels showed no evidence of protein degradation. Moreover, immunohistochemistry of the tumor samples provides its own internal control as sections of tumor contain non-tumor cells.

REFERENCES

[0067] 1. T. R. Sutter et al., J. Biol. Chem. 269, 10392 (1994)

[0068] 2. S. A. Wrighton and J. C. Stevens, Crit. Rev. Toxicol. 22, 1 (1992)

[0069] 3. D. R. Nelson et al, Pharmacogenetics 6, 1 (1996)

[0070] 4. T. Shimada and F. P. Guengerich, Chem. Res. Toxicol. 4, 391 (1991); V. Nedelcheva and I. Gut, Xenobiotica 24, 1151 (1994); B. K. Park, M. Pirmohamed, N. R. Kitteringham, Pharmac. Ther. 58, 385 (1995).

[0071] 5. J. H. Capdevila, J. R. Falck, R. W. Estabrook, FASEB J. 6, 731 (1992)

[0072] 6. W. L. Miller, Endocrine Rev. 9, 295 (1988)

[0073] 7. E. H. Oliw, Prog. Lipid Res. 33, 329 (1994)

[0074] 8. G. I. Murray and M. D. Burke, Biochem. Pharmacol. 50, 895 (1995); L. S. Kaminsky and M. J. Fasco, Crit. Rev. Toxicol. 21, 407 (1992); E. G. Scheutz et al., Arch. Biochem. Biophys. 294, 206 (1992)

[0075] 9. A. K. Jaiswal, F. J. Gonzalez, D. W. Nebert, Science 228, 80 (1985)

[0076] 10. A. K. Jaiswal, D. W. Nebert, F. J. Gonzalez, Nucl. Acid Res. 14, 6773 (1986)

[0077] 11. D. Sesardic, M. Pasanen, O. Pelkonen, Carcinogenesis 11, 1183 (1990)

[0078] 12. H. Schweikl et al., Pharmacogenetics 3, 239 (1993)

[0079] 13. F. J. Gonzalez and H. V. Gelboin, Drug Metab. Rev. 26, 165 (1994); F. P. Guengerich, Cancer Res. 48, 2946 (1988); K. Kawajiri and Y. Fujii-Kuriyama, Jpan. J. Cancer Res. 82, 1325 (1991); F. P. Guengerich, Pharmac. Ther. 54, 17 (1992)

[0080] 14. G. A. Le Blanc and D. J. Waxman, Drug Metab. Rev. 20, 395 (1989); K. T. Kivisto, H. K. Kroemer, M. Eichelbaum, Brit. J. Clin. Pharmacol. 40, 523 (1995)

[0081] 15. N. Albin et al., Cancer Res. 53, 3541 (1993)

[0082] 16. C. Toussaint et al., Cancer Res. 53, 4606 (1993); M. Czerwinski et al., Cancer Res. 54, 1085 (1994)

[0083] 17. W. H. M. Peters et al., Gastroenterology 103, 448 (1992); I. de Waziers et al., Carcinogenesis 12, 905 (1991); L. Massad et al., Cancer Res. 52, 6567 (1992); K. Mekhail-Ishak et al., Cancer Res. 49, 4866 (1989)

[0084] 18. F. Janot et al., Carcinogenesis 14, 1279 (1993)

[0085] 19. G. I. Murray et al., Br. J. Cancer 63, 1021 (1991); G. I. Murray et al., J. Pathol. 171, 49 (1993); G. I. Murray et al., Gut 35, 599 (1994)

[0086] 20. R. J. WeaveR et al., Biochem. Pharmacol. 47, 763 (1994)

[0087] 21. J. A. McKay et al., J. Histochem. Cytochem. 43, 615 (1995)

[0088] 22. D. C. Spink et al., Arch. Biochem. Biophys. 293, 242-348 (1992)

[0089] 23. J. A. McKay et al., FEBS Letts. 374, 270 (1995)

[0090] 24. M. J. Dallman and A. C. G. Porter, In PCR, A practical approach, eds. J. McPherson, P. Quirke and G. R. Taylor (IRL Press, Oxford) p 215-224.

[0091] 25. A. Buchmann et al., Carcinogenesis 6, 513 (1985); M. W. Roomi, R. K. Ho, D. S. R. Sarma, E. Farber, Cancer Res. 45, 564 (1985); A. Buchmann et al., Cancer Res. 47,2911(1987)

[0092] 26. T. G. Krontiris, N. Engl. J. Med. 333, 303 (1995); L. M. Ponz de Leon, Recent Res. Cancer Res. 136, 35 (1994); S. K. Bahra, B. K. A. Rasheed, S. H. Bigner, D. D. Bigner, Lab. Invest. 71, 621 (1994)

[0093] 27. G. L. Wang, B-H, Jiang, E. A. Rue, G. L. Semenza, Proc. Natl. Acad. Sci. USA 92, 5510 (1995)

[0094] 28. J. P. Whitlock et al., FASEB. J. 10, 809 (1996)

[0095] 29. T. Horishika et al, Cancer Res. 52,108 (1992)

[0096] 30. Harlow et al., Antibodies—A Laboratory Manual, Cold Spring Harbor, Cold Spring Harbor, N.Y. (1988)

[0097] 31. Otto et al., Endocrinology 129:970-982 (1991)

[0098] 32. Weaver et al. Biochem. Pharmacol. 46:1183-1197 (1993)

[0099] 33. Laemmli, Nature 227:68 (1970)

[0100] 34. Lowry, J. Biol. Chem. 193:265 (1951)

[0101] 35. Towbin et al. Proc. Natl. Acad. Sci. USA 76:4350 (1979)

[0102] 36. J. R. Willimson, Nature 382:112-113 (1996)

[0103] 37. A. D. Ellington et al, Nature 346:818-822 (1990)

Claims

1. A composition comprising a compound that is selectively metabolized by CYP1B1 to generate a cytotoxic substance that kills or inhibits the growth of a tumor cell.

2. The composition of claim 1, wherein the compound is non-toxic prior to its metabolism by CYP1B1.

3. An anti-cancer drug comprising an amount of the composition of claim 1 effective to kill or inhibit the growth of a tumor cell when administered to an individual having cancer.

4. The anti-cancer drug of claim 3, wherein the tumor cell is a tumor cell of the bladder, brain, breast, colon, connective tissue, kidney, lung, lymph node, esophagus, ovary, skin, stomach, testis, or uterus.

5. A composition comprising a compound that is selectively metabolized by CYP1B1 to generate a substance that renders a tumor cell susceptible to a cytotoxic agent.

6. An anti-cancer drug comprising an amount of the composition of claim 5 effective to render a tumor cell susceptible to a cytotoxic agent when administered to an individual having cancer.

7. The anti-cancer drug of claim 6, wherein the tumor cell is a tumor cell of the bladder, brain, breast, colon, connective tissue, kidney, lung, lymph node, esophagus, ovary, skin, stomach, testis, or uterus.

8. A method of killing or inhibiting the growth of a tumor cell, the method comprising contacting a tumor cell expressing CYP1B1 with the composition of claim 1, wherein the CYP1B1 expressed by the tumor cell metabolizes the compound to generate the cytotoxic substance and thereby kill or inhibit the growth of the tumor cell.

9. The method of claim 8, wherein the tumor cell is a tumor cell of the bladder, brain, breast, colon, connective tissue, kidney, lung, lymph node, esophagus, ovary, skin, stomach, testis, or uterus.

10. A method of rendering a tumor cell susceptible to a cytotoxic agent, the method comprising contacting a tumor cell expressing CYP1B1 with the composition of claim 5, wherein the CYP1B1 expressed by the tumor cell metabolizes the compound to generate the substance and thereby render the tumor cell susceptible to the cytotoxic agent.

11. The method of claim 10, further comprising contacting the tumor cell with an amount of the cytotoxic agent effective to kill or inhibit the growth of the tumor cell.

12. The method of claim 11, wherein the tumor cell is a tumor cell of the bladder, brain, breast, colon, connective tissue, kidney, lung, lymph node, esophagus, ovary, skin, stomach, testis, or uterus.

13. A method of treating cancer, the method comprising administering to an individual diagnosed as having a tumor an amount of the composition of claim 1 effective to kill or inhibit the growth of a tumor cell in the individual.

14. The method of claim 13, further comprising detecting expression of CYP1B 1 in the tumor prior to the administration of the composition.

15. The method of claim 13, wherein the administration of the composition does not result in liver toxicity in the individual.

16. The method of claim 13, wherein the administration of the composition does not result in systemic toxicity in the individual.

17. The method of claim 13, wherein the tumor is a tumor of the bladder, brain, breast, colon, connective tissue, kidney, lung, lymph node, esophagus, ovary, skin, stomach, testis, or uterus.

18. A method of treating cancer, the method comprising: administering to an individual diagnosed as having a tumor an amount of the composition of claim 5 effective to render a tumor cell in the individual susceptible to the cytotoxic agent; and administering to the individual an amount of the cytotoxic agent effective to kill or inhibit the growth of the tumor cell.

19. The method of claim 18, further comprising detecting expression of CYP1B1 in the tumor prior to the administration of the composition.

20. The method of claim 18, wherein the administration of the composition does not result in liver toxicity in the individual.

21. The method of claim 18, wherein the administration of the composition does not result in systemic toxicity in the individual.

22. The method of claim 18, wherein the tumor is a tumor of the bladder, brain, breast, colon, connective tissue, kidney, lung, lymph node, esophagus, ovary, skin, stomach, testis, or uterus.

23. A method of preparing an anti-cancer compound, the method comprising: selecting a compound that is metabolized by CYP1B1; and preparing the compound as a non-toxic substance that is metabolized by CYP1B1 to generate a cytotoxic substance that kills or inhibits the growth of a tumor cell.

24. The method of claim 23, wherein the compound is selected by screening to identify compositions that are metabolized by CYP1B 1.

25. The method of claim 23, wherein the compound is selected by designing compositions that are metabolized by CYP1B1.

26. A method of preparing an anti-cancer compound, the method comprising: selecting a compound that is metabolized by CYP1B1; and preparing the compound as a non-toxic substance that is metabolized by CYP1B1 to generate a moiety that renders a tumor cell susceptible to a cytotoxic agent.

27. The method of claim 26, wherein the compound is selected by screening to identify compositions that are metabolized by CYP1B1.

28. The method of claim 26, wherein the compound is selected by designing compositions that are metabolized by CYP1B1.