Bioconjugate Chem. 1993, 4, 134-138
134
Peroxidase-Linked Anti-Basic Fibroblast Growth Factor Monoclonal Antibody Fab’ Conjugates: Application for Two-Site Enzyme Immunoassay and Immunohistochemical Detection Masayuki Kurobe,+JYoshifumi Takei,t Takashi Fukatsu,§ Akemi Kato,+and Kyozo Hayashi’,? Department of Molecular Biology, Gifu Pharmaceutical University, Gifu 502, Japan, Department of Clinical Biochemistry, Gifu College of Medical Technology, Seki, Gifu 501-32, Japan, and Gifu Prefecture Tajimi Hospital, Tajimi, Gifu 507, Japan. Received June 30, 1992
After conjugating thiol groups in the hinge region of monoclonal antibody (mAb) Fab’ fragments specific for basic fibroblast growth factor (bFGF) with maleimido-horseradish peroxidase HRP) complexes synthesized by incubation of HRP with the heterobifunctional reagent N-succinimidyl-4-(maleimidomethy1)cyclohexane-1-carboxylate,we developed a fluorometric enzyme immunoassay method based on the sandwiching of the factor between anti-bFGF IgG-coated polystyrene beads and the conjugates, and also an immunohistochemical method for detection of the location of the factor. The discriminatory detection limit by the developed enzyme immunoassay (EIA) was as low as 30 pg/mL. The reproducibility of within- and between-assay series was 6.07-9.18% and 6.28-6.82%, respectively, and the recovery of exogenous bFGF from serum was approximately 98%. The curves generated by the concentrated fraction that eluted a t the same position as standard bFGF by size-exclusionchromatography on a TSK 2OOOSW column were parallel to the curve for standard bFGF. From these results, we consider the developed EIA method to be acceptable in regard to sensitivity, precision, and specificity. Also, without the introduction of any additional signal amplification system, positive immunohistochemical reactions were successfully detected by the HRP-linked anti-bFGFmAb Fab’ in fibroblastic and endothelial cells, which have already been shown to synthesize and secrete bFGF, indicating that these conjugates provide a useful means for direct immunohistochemical detection of the factor.
INTRODUCTION Basic fibroblast growth factor (bFGF),which is a single chain protein composed of 146 amino acid residues ( I ) , is identified in a variety of normal and malignant tissues (2) and characterized as an exclusive mesodermal and neuroectodermal cell mitogen ( 3 , 4 ) . The biological functions of bFGF are mediated by a specific cell-surface receptor, a 130-kDa polypeptide with an extracellular domain that binds bFGF and an intracellular domain that contains intrinsic tyrosine-kinase activity (5-7), pointing to an extracellular mechanism of action of bFGF. However, all forms of its primary translational products lack of hydrophobic signal sequence (8,9), which would direct their secretion through the classical secretory pathway. Recently, the oncogenic character of FGF has been pointed out in view of its highly conserved homology with the FGF family, which includes int-2 (10, l l ) , hst-1 (12), and a oncogene obtained from Kaposi’s sarcoma (13). Two cDNAs encoding the human FGF receptor were isolated and found to be closely related to the oncogene bek (1416). In mammalian cells, the expression of bFGF cDNA results in the transformation of the host cells ( 1 7, 18). Despite many investigations, the precise molecular mechanism by which bFGF is exported out of the cell and the pathophysiological mechanism including the tumorigenesis elicited by it remain unexplained. Knowledge of the exact distribution of bFGF and its alterations in the
* Correspondence and reprint requests should be addressed to Kyozo Hayashi, Ph.D., Gifu Pharmaceutical University, Department of Molecular Biology, Mitahora-higashi, Gifu 502, Japan. Phone: 0582-(37)-3931,Ext. 318. Telefax: 0582437)5844. +
1
Gifu Pharmaceutical University. Gifu College of Medicinal Technology. Gifu Prefecture Tajimi Hospital. 1043-1802/93/2904-0 134$04.00/0
bFGF-producing cells may aid in further understanding of the yet poorly defined role of this factor. In order to study these unique biochemical and pathophysiological features of bFGF, it is necessary to develop a sensitive and specific assay and immunohistochemical methods for detection of the factor. In this paper we describe the preparation of HRP-linked anti-bFGF monoclonal antibody (mAb) Fab’ conjugate and its efficiency for development of a monoclonal antibody-based two-site enzyme immunoassay method and also for immunohistochemical detection of the factor. The developed enzyme immunoassay (EIA) method for bFGF was deemed acceptable in regard to sensitivity, precision, and specificity. Without any amplification step, the positive reactions were successfully detected by the conjugates in fibroblastic and endothelial cells in unfixed and frozen sections of thyroid gland (histologically normal). EXPERIMENTAL PROCEDURES
Materials. Bovine serum albumin (Fr. V) was purchased from Armour Pharmaeutical Co. (Kankakee, IL); pepsin (EC 3.4.23.11, from Calibiochem-Behring Co. (San Diego, CA); peroxidase (HRP, EC 1.11.1.7), from Sigma Chemical Co. (St.Louis, MO). Sephadex G-25, and G-100 were obtained from Pharmacia Fine Laboratories (Uppsala, Sweden). An Affigel Protein A MAPS kit was purchased from Bio-Rad Laboratories (Richmond, CAI. N-Succinimidyl4-(maleimidomethy1)cyclohexane-1-carboxylate (SMCC) was bought from Pierce Chemical Co. (Rockford, IL); 3-(p-hydroxyphenyl)propionic acid (HPPA), from Sigma Chemical Co. (St. Louis, MO). All other chemicals were from Wako Pure Chemical Ind. Ltd. (Osaka, Japan). Polystyrene beads (3.2 mm in diameter) were obtained from Wako Pure Chemical Ind., Ltd. (Osaka, Japan). 0 1993 American Chemical Society
PeroxidasaLinked Anti-bFGF mAb Fab’ Conjugates
bFGF a n d mAbs against bFGF. Two monoclonal antibodies (mAbs 12 and 98) (19)and recombinant human bFGF (20) used for the standard were generously provided by the Takeda Chemical Ind., Ltd. Immunological properties of the mAbs such as specificity were characterized in detail by Seno et al. (19). Preparation of HRP-Linked Anti-bFGF mAb Fab’ Conjugates. The procedures for synthesis of HRP-linked anti-bFGF mAb Fab’ conjugates were similar to those described earlier (21). Briefly, the IgG fraction of mAb 98 (subclassof IgG1) against bFGF was purified from mouse ascites by affinity chromatography with an Affigel Protein A MAPS kit according to the manual published by BioRad Laboratories, Co. From the purified IgG thus obtained, we prepared the Fab’ fragments by digestion of the IgG fraction with pepsin following reduction with 2-mercaptoethylamine. On the other hand, HRP was treated with SMCC to introduce maleimido groups, and the maleimido-HRP thus obtained was allowed to react with thiol groups in the hinge region of the Fab’ fragments. The HRP-linked anti-bFGF mAb Fab’ conjugates were finally purified by gel filtration on an Ultrogel AcA 44 column equilibrated and developed with 0.1 mol/L phosphate buffer, pH 6.5. The presence of a thiol group was determined by using 4,4'-dithiodi pyridine.
Because of batch-to-batch variation, a preliminary test was necessary to assess the optimal magnitude of dilution of the stock conjugate solution in the assay. Preparation of Anti-bFGF mAb IgG-Coated Polystyrene Beads. The method for coating polystyrene beads was similar to that described earlier (21). One hundred beads were immersed in a solution of the affinitypurified mAb 1 2 (subclass IgG1; 200 pg/mL) in 20 mL of 50 mmol/L Tris-HC1 buffer, pH 8.5, overnight a t 4 OC. The coated beads were stored in buffer A (0.1 mol/L phosphate buffer, pH 7.0, containing 0.1 5% BSA, 0.3 mol/L NaC1, 0.1% NaN3, and 1 mmol/L MgC12) at 4 OC until used. Two-Site Enzyme Immunoassay Method. For all dilutions buffer A was used. The typical assay procedure is as follows: an aliquot of the standard in buffer A or the samples (100 pL) and buffer A (150 pL) were added to each tube containing one mAb IgG-coated bead. After incubation a t 4 “C overnight the aqueous solution was aspirated off and the beads were extensively washed with buffer B (0.1 mol/L phosphate buffer, pH 7.0, containing 0.1% BSA). The beads were next incubatd a t 4 OC overnight with HRP-linked anti-bFGF mAb Fab’ conjugate in 250 PL of buffer B (300-fold dilution of the stock conjugate solution). After extensivelywashing of the beads with buffer B, they were transferred to clean tubes. HRP activity fixed on the beads was fluorophotometrically measured by use of HPPA as coupling reagent as described earlier (21). In any assay of samples, a series of standard solution was always assayed together with the samples. Experiments were carried out in duplicate except where noted otherwise. Blood Samples. Serumandplasma(anticoagulatedwith heparin) samples collected from normal subjects were generously donated by Ms. K. Tanno of the Gifu RedCross Blood Bank. Immunohistochemical Staining. Unfixed frozen sections of thyroid gland (histologically normal) were stained as a model tissue. Histopathological diagnoses were made by Dr. Tetsuro Nagasaka, a histopathologist
Bloconjugate Chem., Vol. 4, No. 2, 1993 135
a t the Department of Laboratory Medicine, Nagoya University School of Medicine (Nagoya, Japan). Sections of thyroid gland freshly obtained surgically were directly fixed with phosphate-buffered saline (PBS) containing 4% paraformaldehyde for 10 min a t room temperature. The specimens were incubated with PBS containing 0.3 % hydrogen peroxide for 3-5 h in order to block endogenous peroxidase, and then with PBS containing the HRP-linked anti-bFGF mAb Fab’ conjugates overnight a t 4 “C. Between each step, the specimens were washed four times with PBS. Peroxidase activity, indicating antigen location, was visualized by incubation of the slides for 10 min with a freshly prepared solution of 0.02 % 3,3‘-diaminobenzidine (DAB) in PBS containing 0.005 % hydrogen peroxide. The specimens were counterstained with Mayer’s hematoxylin. RESULTS AND DISCUSSION
Thiol groups in the hinge region of the mAb Fab’ fragments prepared by digestion of the purified IgG with pepsin followed by reduction with 2-mercaptoethylamine were conjugated with the maleimido-HRP complexes by using the Fab’ and the modified HRP a t a molar ratio of 1:2. The maleimido-HRP complexes were synthesized by incubation of HRP (2 mg, 50 nmol) in 0.1 mol/L phosphate buffer, pH 6.0, containing a molar excess of the heterobifunctional reagent SMCC (1.3 mg, 5 mmol). Under the same conditions as described earlier (211,pepsin digestion of the anti-bFGF mAb 98 IgG fraction (subclass IgG1) was found to have been further hydrolyzed into smaller peptide fragments. To obtain a higher yield of the F(ab’)Z fragments, we digested the antibody by 1/50 (w/w) pepsin for 4 h. Under the optimized reaction conditions, recovery after this step was estimated to be 85-90 % . The number of thiol group in the molecule of Fab’ prepared was estimated to be 1.09 (range = 1.07-1.12, n = 3). Within 24 h of incubation in 0.1 mol/L phosphate buffer, pH 6.0, containing 5 mmol/L EDTA, the number of thiol groups of Fab‘ decreased less than 3 7%. When HRP was allowed to react with SMCC, the average number of maleimido groups introduced per enzyme molecule was 1.67 (range = 1.40-1.93, n = 3). The enzyme activityafter the reaction was 98.6 % (range = 96.5-101.5, n = 3) of that before the reaction. A t the last step, the HRP-linked Fab’ conjugates were purified by gel filtration on an Ultrogel AcA 44 column. Elution profiles of the protein and the enzyme activity (data not shown) indicated that free forms of HRP and Fab’ were almost completely separated from the conjugates, since HRP and Fab’ were not polymerized during the conjugation reaction. Fractions eluted at higher molecular weight region containing both immune and enzyme activities were stored a t 4 “C in the presence of 0.1% BSA and 0.0015% merthiolate. The molar ratio of Fab’ to the enzyme in the conjugates obtained was calculated to be 0.92 based on the molar extinction coefficients at 403 and 280 nm of Fab’ and the enzyme, respectively. The ratio of absorbance at 403 nm to that at 280 nm for the enzyme used was 3.1 (22). And the enzyme activity decreased about 20 % upon conjugation with Fab’. We did not systematically examine a change in the binding ability of antibody, e.g. the affinity constant, during the conjugation reaction. Overall yield of the conjugates was approximately 40 % . In order to develop a sensitive and valid EIA for bFGF requiring fewer mAbs, we chose the method of sandwiching
136
Bioconjugate Chem., Vol. 4, No. 2, 1993
Kurobe et al.
Table 11. Recovery of Exogenous bFGF from Serum.
serum 1 serum 2 serum3 serum4 serum 5
endogenous bFGF (pg/mL)
exogenous bFGF added (pg)
152.3 195.0 210.5 160.5 180.3
100 100 100 100 100
increase in bFGF over % endogenous (pg) recovery 105.5 105.5 91.2 91.2 89.9 89.9 108.8 108.8 93.5 93.5
a Assay conditions are as follows: a 900-pL volume of serum from normal subjects and a 100-pL volume of the standard bFGF (lo00 pg/mL) in buffer A were mixed and incubated at 4 “C overnight. A 100-pL volume of serum or a 100-pLvolume of the sample added the standard bFGF was assayed by our EIA system, and each recovery value was calculated.
.01
.I
1
10
100
r b F G F (ngiinl)
Figure 1. Dose-response curve of recombinant bFGF by sandwich EIA. Fluorescence intensity of specifically bound peroxidase activitywas calculatedby subtractionof fluorescence intensityobtained in the presence of bFGF from that in ita absence (background).Eachpointindicatesthe mean of duplicateassays. Table I. Within- and Between-Assay Variations bFGF level
(pg/mL)
assay
Within-Assay Variations 110.2 10.12 127.7 f 7.75
6.07
Between-Assay Variations 110.6 f 7.54
6.28
serum 1 (n = 10) serum 2 (n = 10) serum 1( n = 5) serum 2 ( n = 5 )
coefficient of variation (%)
133.4 f 8.38
9.18
1
6.82
the antigen between HRP-labeled Fab‘ conjugatesand mAb IgG-coated polystyrene beads and the use of HPPA as a coupling reagent, to detect HRP activity. To assess its sensitivity and validity as a two-site monoclonal antibody-based enzyme immunoassay system, we initially prepared a calibration curve with recombinant bFGF in buffer A used as a standard. Typical procedures used for the fluorometrical two-site enzyme immunoassay were the same as those described earlier (21). When 100pL aliquots of various concentrations of recombinant bFGF as a standard were assayed, the dose-response curve (Figure 1)was obtained. When the fluorescence intensity of 1pg/mL quinine in 50 mmol/L HzS04 was adjusted to a scale of 100, relative fluorescence intensity obtained in the complete absence of bFGF (background) was measured to be 14.51 f 0.20 (mean f SD, n = 10). As shown in Figure 1,the linearity of the curve was observed in between 30 pg/mL and 10 ng/mL. Thus the sensitivity is great enough to measure the concentration of bFGF in human body fluids such as serum and urine. In proportion to the length of the incubation time, the fluorescence intensity increased. An incubation a t 4 “C overnight was sufficient to obtain constant formation of each immune complex. The anti-bFGF antibody IgGcoated beads stored in buffer A a t 4 OC for a t least 3 months gave almost the same calibration curve, and also the HRPlinked anti-bFGF mAb Fab’ conjugates stored a t 4 “C for the same time gave the same results, demonstrating that the beads and the conjugates can be stored safely for a t least that period of time. Reproducibility of the developed EIA was tested by use of clinically available normal sera. As shown in Table I, the coefficients of variation of within- and between-assay series were 6.07-9.1876 (n = 10) and 6.28-6.8276 ( n = 5), respectively.
,
2
2’
4
3’
3‘
i’
2’
2 . 2 ,
scruiii ( A , e 1
3O
3 ’
standard b F C F
(
0 I
Figure 2. Dilution curve of each fraction of immunoreactive bFGF obtained by HPLC size-exclusion chromatography of human serum on a TSK 2000SW column. Each fraction that eluted at the same positions as standard bFGF was pooled and concentrated. Each concentratedsolution (serum 1,bFGF level of 850 pg/mL, 0 ;or serum 2, bFGF level of 400 pg/mL, A) was serially diluted 2-fold with buffer A and a 100-pLvolume of each diluent was assayed. The standard bFGF solution in the above buffer (bFGFlevel of 2000 pg/mL,0)was diluted 3-fold with the above buffer and a 100-pL volume of each diluent was used for assay. -J
’
:.: ......s.... ... . . s
Serum
Plasm2
0 : :
:, *::&I;.:.;.
.,
’
0
0
I
.
1.:
,
:e!:
?;.
I
.
s.:e
:i
.
190 f 32 ? g / m l (n = - 8 )
-50
T
(3 =
42 pg/ml 50) ,
Peroxidase-LinkedAnti-bFGF mAb Fab’ Conjugates
Bioconjugate Chem., Vol. 4, No. 2, 1993
137
Figure 4. Immunohistochemical staining of normal region of ectomized thyroid gland. Unfixed specimens of normal thyroid tissue showing scattered bFGF-positive reactions that appear diffusely in the cytoplasm of capillary endothelial cells (arrows in A, left) and of fibroblast cells (arrows in B, right) X 100. 2000 SW column. As shown in Figure 2, the slope of the curves were parallel to the slope of the curve for standard bFGF in buffer A, indicating that the antigenicity is similar. From these results, we consider the developed EIA to be acceptable in regard to linearity, sensitivity, recovery from the serum, and precision. Next, we applied this EIA method for the estimation of bFGF levels in serum and plasma of normal subjects. The results are seen Figure 3. As shown in Figure 3, the level of immunoreactive bFGF in normal serum was 190 f 32 pg/mL (mean f SD, n = 48), which agrees with the range of 30-206 pg/mL reported by Watanabe et al. (23). We further measured the immunoreactive bFGF level in normal plasma (anticoagulated with heparin) and found it to be 150 f 42 pg/mL (n= 50), not significantly different from the serum value. Structurally,there existed an amino terminal truncated form of bFGF. The sites in the bFGF molecule recognized by the mAb 12 used for coating the beads was located within the sequence of the first nine amino acids of the N-terminal region (19). Thus, our EIA would not detect any N-terminal truncated form(s) of bFGF that might retain biological activity. Conventionally, methods for immunohistochemicaldetection of antigens employing monoclonal antibodies include an additional signal amplification step such as the peroxidase-antiperoxidase (PAP) or avidin-biotin complex (ABC) system. As a model, we examined an application of the HRP conjugates for immunohistochemical detection of the antigen in the sections of normal organs and tissues. As shown in Figure 4, a diffuse positive
reaction was observed in the cytoplasm of capillary endothelial cells and of fibroblastic cells located in normal parts of unfixed frozen sections of thyroid gland surgically obtained. The reaction product varied in density even on the same slide. These results are clearly in support of the numerous reports on production and secretion of bFGF by endothelial cells (24-27). Neutralization of the conjugates with an excess molar amount of recombinant bFGF resulted in a disappearance of the positive reactions (data not shown). From these results, it is apparent that HRPlinked mAb Fab’ conjugates specific for bFGF could detect the antigen in sections of tissues and organs without introduction of any additional signal amplification system. Studies on the distribution and levels of biologically active substances found in only minute amounts in tissues require sensitive, quantitative, and specific analytic methods. The earliest method to detect bFGF was based on its biological activity, i.e., enhanced uptake of 3H-labeled and the radioreceptor assay (RRA) method thymidine (a), (29). The RRA method, however, has various inherent disadvantages such as less specificity and sensitivity, less stability of radioactive materials, and requirement of comparatively expensive equipment. And the biological method is not specific and sensitive enough to measure endogenous bFGF in various tissues and body fluids including serum. Recently, Watanabe et al. (23)established an EIA method for bFGF, the principle of which is based on sandwiching of the factor between a mixture of two mAb IgGs fixed on the surface of wells of a multiwell
138
Bioconjugate Chem., Vol. 4, No. 2, 1993
plate and HRP-linked mAb conjugates, and found an increased level of bFGF i n sera of patients with renal cell carcinoma. The discriminatory detection limit of bFGF, being as low as 30 pg/mL, is t h o u g h t t o be adequate for measurement of the concentration of endogenous bFGF i n human body fluids such as serum, plasma, and urine. Sixteen o u t of 20 patients with benign m a m m a r y gland carcinoma showed generally a higher level of immunoreactive bFGF i n their serum (range = 125-1080 pg/mL) than that found in the normal subjects (Figure 3). Although i t appears that bFGF m a y act exclusively as an autocrine or paracrine factor functioning as an angiogenic and mitogenic factor ( 4 ) , details of the mechanism of these actions remain t o be elucidated. Our immunoassay system for bFGF using the HRP-linked anti-bFGF Fab’ conjugates described here provides a useful means for detection of bFGF i n various fluids, and the conjugates can also be used directly for immunohistochemical detection of the factor. ACKNOWLEDGMENT
The authors wish t o thank the T a k e d a Chemical Ind. Ltd., for generously providing ascites fluids containing each mAb. We also would like t o thank Ms. K. T a n n o of the Gifu Red-Cross Blood Bank, for providing sera and plasma of normal subjects, and Dr. T. Nagasaka of the Department of Laboratory Medicine, Nagoya University School of Medicine (Nagoya), for generously donating specimens of thyroid gland. LITERATURE CITED (1) Esch, F., Baird, A., Ling, N., Ueno, N., Hill, F., Denoroy, L., Klepper, R., Gospodarowicz, D., Bohlen, P., and Guillmin, P. (1985) Primary structure of bovine pituitary basic fibroblast growth factor (FGF) and comparison with the amino terminal sequence of bovine brain acidic FGF. Proc. Natl. Acad. Sei. U.S.A. 83, 6507-6511. (2) Gospodarowicz, D., Ferrara, N., Schweigerer, L., and Neufeld, G. (1987) Structural characterization and biological functions of fibroblast growth factor. Endocr. Rev. 8, 95-114. (3) Burgess, W. H., and Maciag, T. (1989) The heparin-binding (fibroblast) growth factor family of proteins. Ann. Rev. Biochem. 58, 575-606. (4) Gospodarowicz, D., Neufeld, G., and Schweigerer, L. (1986) Molecular and biological characterization of fibroblast growth factor, and angiogenic factor which also controls the proliferation and differentiation of mesoderm and neuroectoderm derived cells. Cell Differ. 19, 1-17. (5) Lee, P. L., Johnson, D. E., Cousens, L. S., Fried, V. A., and Williams, L. T. (1989) Purification and complementary DNA cloning of a receptor for basic fibroblast growth factor. Science 245, 57-60. (6) Mansukhani, A., Moscatelli, D., Talarico, D., Levytska, V., and Basilico, C. (1990) A murine fibroblast growth factor (FGF) receptor expressed in CHO cells is activated by basic FGF and Kaposi FGF. Proc. Natl. Acad. Sei. U.S.A. 87, 4378-4382. (7) Keegan, K., Johnson D. E., Williams, L. T., and Hyman, M. (1991) Isolation of an additional number of the fibroblast growth factor receptor family, FGFR-3. Proc. Natl. Acad. Sei. U.S.A. 88, 1095-1099. (8) Jaye, M., Howk, R., Burgess, W., Ricca, G. A,, Chiu, I. M., Ravera, M. W., O’Brien, S. J., Modi, W. E., Maciag, T., and Drohan, W. N. (1986) Human endothelial cell growth factor. Science 233, 541-545. (9) Schweigerer, L., Neufeld, G., Friedmann, J.,Abraham, J. A., Fiddes, J. C., and Gospodarowicz, D. (1987) Capillary endothelial cells express basic fibroblast growth factor, a mitogen that promotes their own growth. Nature 325, 257-259. (10) Dickson, C.,and Peters, G. (1987)Potentialoncogene product related to growth factors. Nature 326, 833.
Kurobe et al.
(11) Moore,R., Casey, G., Brookes S., Dixon, M., Peters, G., and Dickson, C. (1986) Sequence, topography and protein coding potential of mouse int-2: A putative oncogene activated by mouse mammary tumor virus. EMBO J. 5, 919-924. (12) Taira, M., Yoshida, T., Miyagawa, K., and Sakamoto, H. (1987) cDNA sequence of human transforming gene hst and identification of the coding sequence required for transforming activity. Proc. Natl. Acad. Sei. U.S.A. 84, 2980-2984. 3) Delli Bovi, P., Curatola, A. M., Kern, F. G., Greco, A,, Ittman, M., and Basillico, C. (1987) An oncogene isolated by transfection of Kaposi’s sarcoma DNA encodes a growth factor that is a member of the FGF family. Cell 50, 729-737. 4) Dionne, C. A., Crumley, G., Bellot, F., Kaplow, J. M., Searfoss, G., Ruta, M., Burgess, W. H., Jaye, M., and Schlessinger, J. (1990) Cloning and expression of two distinct high-affinity receptors cross-reacting with acidic and basic fibroblast growth factors. EMBO J. 9, 2685-2693. (15) Kornbluth, S., Paulson, K. E., and Hanafusa, H. (1988)Novel tyrosine kinase identified by phosphotyrosine antibody screening of cDNA libraries. Mol. Cell Biol. 8 , 5541-5544. (16) Seno, M., Sasada, R., Watanabe, T., Ishimura, K., and Igarashi, K. (1991) Two cDNAs encoding novel human FGF receptor. Biochem. Biophys. Acta 1089, 244-246. (17) Rogelj, S., Weiberg, R. A., Fanning, P., and Klagsbrun, M. (1988) Basic fibroblast growth factor fused to a signal peptide transform cells. Nature 33, 173-175. (18) Sasada, R., Kurokawa, K., Iwane, M.,andIgarashi, K. (1988) Transformation of mouse BALB/c3T3 cells with human basic fibroblast growth factor cDNA. Mol. Cell Biol. 8, 588-594. (19) Seno, M., Iwane, M., Sasada, R., Moriya, N., Kurokawa, T., and Igarashi, K. (1989) Monoclonal antibodies against basic fibroblast growth factor. Hybridoma 8, 209-221. (20) Iwane, M., Kurokawa, T., Sasada, R., Seno, M., Nakagawa, S.,and Igarashi, K. (1987)Expression of cDNA encoding human fibroblast growth factor in E. coli. Biochem. Biophys. Res. Commun. 146,470-477. (21) Kurobe, M., Kono, M., Yoshida, M., andHayashi, K. (1988) A sensitive twesite enzyme immunoassay for human pancreatic secretory trypsin inhibitor (PSTI)using monoclonal antibodies. Clin. Chim. Acta 178, 205-214. (22) Yoshitake, S., Imagawa, M., and Ishikawa, E. (1982)Efficient preparation of rabbit Fab’-horseradish peroxidase conjugates using maleimide compounds and its use for enzyme immunoassay. Anal. Lett. 15, 147-160. (23) Watanabe, H., Hori, A., Seno, M., Kozai, Y., Igarashi, K., Ichimori, Y., and Kondo, K. (1991) A sensitive enzyme immunoassay for human basic fibroblast growth factor. Biochem. Biophys. Res. Commun. 175, 229-235. (24) Shing,Y., Folkman, J.,Sullivan, R., Butterfield, C., Murray, J., and Klangsbrun, M. (1984) Heparin affinity: purification of a tumor-derived capillary endothelial cell growth factor. Science 223, 1296-1299. (25) Gospodarowicz, D., Morgan, M., Braun, D, and Birdwell, C. R. (1976) Clonal growth of bovine vascular endothelial cells: fibroblast growth factor as a survival agent. Proc. Natl. Acad. Sei. U.S.A. 73, 4120-4124. (26) Klagsbrun, M., and Shing, Y. (1985) Heparin affinity of anionic and cationic capillary endothelial cell growth factors: analysis of hypothalamus-derived growth factors and fibroblast growth factors. Proc. Natl. Acad. Sci. U.S.A. 82, 805-809. (27) Klagsbrun, M., Sasse, J., Sullivan, R., and Smith J. A (1986) Human tumor cells synthesize an endothelial cell growth factor that is structurally related to basic fibroblast growth factor. Proc. Natl. Acad. Sei. U.S.A. 83, 2448-2452. (28) Gospodarowicz, D. (1975)Purification of a fibroblast growth factor from bovine pituitary. J. Biol. Chem. 250,2515-2520. (29) Baird, A., Florkiewicz, R. Z., Maher, P. A,, Kaner, R. J., and Hajjar, D. P. (1990) Mediation of virion penetration into vascular cells by association of basic fibroblast growth factor with herpes simplex virus type 1. Nature 348, 344-346.
