HYBRIDOMA Volume 11, Number 3, 1992 Mary Ann Liebert, Inc., Publishers

Characterization and Nucleotide Sequences of the Variable Regions of a Monoclonal

Antibody against a-Fetoprotein




Immunology Division, Development

Center for Biotechnology, 81 Chang Hsing St., Taipei, Taiwan, R.O.C.


a-Fetoprotein (AFP) is a well-known tumor marker of hepatocellular (HCC). Monoclonal antibodies against AFP possessing specific binding ability to HCC are potential candidates for immunoscintigraphy and immunotherapy. A new monoclonal antibody against AFP (0325-6-9) was isolated. Its specificity and targeting tumor ability were characterized by enzyme-linked immunosorbent assay (ELISA), cell immunostain and complement killing. These results suggest that 0325-6-9 is specific to hepatoma cells. The nucleotide sequences of variable regions of 0325-6-9 were determined by M13 dideoxynucleotide sequencing method. With the information of nucleotide sequence, this antibody then could be modified by recombinant technology for its usage in in vivo diagnosis and immunotherapy. carcinoma


Hepatocellular carcinoma (HCC) is the leading cause of cancer deaths in males in Taiwan and in Southeast Asia. Among various cancer markers used for HCC, -fetoprotein has been well-recognized as an outstanding tumor marker for HCC. This is evidenced by its increased secretion of AFP in HCC and teratocarcinoma (1). specifically Although AFP is a secretary it is also retarded the cell surface before release. For diagnostic at protein, radiolabelled monoclonal AFP has been used in antibody against purposes, HCC cell and monoclonal antibodies tumors (3,4). Thus, imaging germ (2) could also used in HCC AFP be and against targeting teratocarcinomas(5,6) in therapeutic application.

As described above, monoclonal antibody demonstrating specificity and binding ability to HCC could be a useful tool for immunoscintigraphy and immunotherapy. At present, most antibodies available are derived from


rodents. Because of its obvious immunologie response, it cannot be used in human for therapeutic purposes. Therefore, humanized antibody modified could be a genetically potential candidate for in vivo diagnosis and due to the minimized human anti-mouse Ig immune immunotherapy, response (7-11). The humanization of a rodent antibody can be accomplished by molecular biology technologies. In order to modify the gene, it is necessary to know the nucleotide sequences of the variable regions of the rodent antibody. To date, there is no report about the nucleotide sequences of the variable regions of a murine anti-AFP antibody. This manuscript reports a newly isolated murine monoclonal antibody (0325-6-9) against AFP. Its specificity and binding ability to hepatoma cell lines were determined by enzyme-linked immunosorbent assay (ELISA), cell immunostain and complement killing methods. In addition, the nucleotide sequences of the variable regions of this AFP monoclonal antibody are also presented.


Preparation of q -Fetoprotein (AFP): AFP was purified from HCC patients. Patients' ascites was collected and purified by anti-AFP polyclonal antibody affinity chromatography as described by Hong et al. (12). Immunization and Hybridoma Production: One new born BALB/c mouse was intraperitoneally injected with human serum albumin (HSA) with the daily dosage of 80 ¿¿g (in 100 ¿¿1 of phosphate buffered saline, pH7.2, PBS), for 10 days consecutively. After 3 months, the mouse was immunized with 6 ¿ig of AFP by intraspleen method, and boosted with the same amount of AFP by lymph node method as described (13). The splenic lymphocytes were fused with BALB/c myeloma FO obtained from American Type Cell Culture (ATCC) and hybridomas were selected with 0.1 mM hypoxanthine, 4x10"^ M aminopterin and 1.6x10"^ M thymidine (HAT) medium (13). Screen of q-AFP Hybridomas with ELISA Method: The 96-well microtiter plates (obtained from Nunc. Inc.) were coated with rabbit anti-AFP antibodies (1 /ig/100 «1 PBS, prepared in this laboratory) overnight at 4°C. Plates were then blocked with dilution buffer (PBS containing 0.05% Tween-20, 1% gelatin) and washed with PBS. The screening process was carried out by reacting the following reagents and washing steps as described consecutively (1). 100 ¿¿1 of diluted HCC ascites containing 0.01 ¿ig/ml of AFP and incubated at 37°C for one hour and washed; (2) 100 ¿¿1 of hybridoma culture supernatant was added and incubated at 37°C for one hour and washed; (3). 100 ¿¿1 of goat anti-mouse Ig antibodies-alkaline phosphatase (Southern Biotechnology Associates, Inc., 1:1000 dilution) was added, incubated at 37°C for one hour, and washed; (4). 100 ¿¿1 of p-nitrophenyl phosphate(PNPP) was added as described (13). Plates were read at O.D. 405 by an ELISA reader (SLT 210, SLT-Labinstruments, Australia).

Western Blot

Analysis.: Purified AFP, HSA and culture supernatant of HepG2 analyzed on native and reduced SDS 8% polyacrylamide gel electrophoresis(PAGE)(14,15). Western blot was performed following the procedures described by the supplier (Hoefer Mini Transfer System). Cross-Reactivity Test. Purified AFP (0.003 ¿£g in 100 ¿¿l/well), human were


immunoglobulin (HIg, 0.5 ¿ig in 100 ¿il/well) and HSA (0.5 ¿ig in 100 ¿il/well) was attached onto separate set of microtiter plates, respectively, and used as antigens for testing the cross-reactivity of 0325-6-9. Ascites serial in diluted 1% bovine albumin serum [1:10^, 1:10^, 1:10^ antibody were (BSA)-PBS]. Coated wells were incubated at 37° C for one hour and washed with PBST. Aliquots of ascites antibody from each dilution were added, incubated at 37°C for one hour and washed. Goat anti-mouse Ig-HRP (Southern Biotechnology Inc. 100 ¿¿1 in 1:1000 dilution) was added and incubated at 37°C for one hour, and then plates were washed. O-Phenylenediamine 2HC1 (OPD, Sigma) was dissloved in 0.1 M citrate-phosphate buffer containing 0.02% H2O2, and 100 ¿¿ 1 of the above OPD solution were added and incubated at 37°C for 30 min. The reaction was stopped to each well with 100 ¿¿ 1 of 1 N H2SO4 and measured by absorbance at 492nm with an ·

ELISA reader. Cell Immunostain. Hepatoma cell lines-HepG2 (15), Hep 3B (15), HCC 36, HA leukemia cell line, CEM, and 22T/VGH (16) and SK-HEP-1 (17) and human cell obtained were line, H9, lymphoma separately from Veterans General O. C. and ATCC. R. Hospital, Taipei, Taiwan, A sample containing lx 10^ cells were transferred to chamber slides (4802, Nunc). After overnight incubation at 37°C with 5% C02 the leukemia and supernatants of the hepatoma cells were removed and lymphoma cells which attached onto slide were dried gently with cool hair¬ drier. Slides were washed with PBS at room temp, for 5 min; fixed with acetone at 4°C for 15 min. and washed with PBS. Endogenous peroxidase acitivity was quenched by dipping the slides in methanol containing 3% H2O2 Several drops of dilution buffer (1% BSA in PBS at room temp, for 15 min.. with 0.05% thimersol) were added and slides were incubated at room temp, for one hour in a humidified chamber. Primary antibody (10 ¿tg/ml in dilution buffer) was added, incubated at 37*C for 60 min. and washed with PBS. Detection was done by adding goat anti-mouse Ig-HRP (0.5 ¿ig/ml, Southern Biotechnology Association Inc.), incubated at 37°C for another hour and washed. Solution A [180 mg of 3-amino-9-ethyl-carbazole (AEC, Sigma) in 15 ml of N, N-dimethyl formamide (Sigma)] was added to solution [270 ml of distilled H2O and 7.5 ml of 2M sodium acetate, pH 5.0]. Furthermore, 150 of 30% also added to form developing solution. Slides were ¿il H2O2 was submerged in developing solution at room temp, for 4 min., and then rinsed with tap water. Nuclei were stained with hematoxylin counterstain at room temp, for 3 min. Slides were rinsed with running tap water till red nuclei turned blue. When slides were dry, they were mounted with immu-mount (Seagull Scientific Inc.). Two murine anti-thyroid stimulating hormone monoclonal antibody (2D7-5) and anti-Chlamvdia trachomatis monoclonal antibody (002-7), prepared in this laboratory, were used as control antibodies in the cell immunostain and complement killing experiments. leukemia and Complement Killing. Five hepatoma cell lines, two human cell and SW1116 human colonie lines, adenocarcinoma, obtained lymphoma (a from ATCC, maintained in 10% FCS-L15) were employed as target cells for complement killing. Hepatoma cells (about 3 lO^/flask) were grown at 37°C 5% CO2 for 2-3 days in 25T flasks. Cells were harvested by washing with 5% FCS-DMEM before use. For non-hepatoma cells, 1x10^ cells were washed ,



with either 5% FCS-RPMI or 5% FCS-L-15, resuspended in 5 ml of maintaining medium containing primary antibody (10¿¿g/ml), and incubated for one hour Cells were washed with serum free at 37°C with shaking for every 20 min. medium. Five milliliters of serum free medium containing 200 ¿/l of human plasma, 50 u 1 of rat serum and 50 ¿il of hamster serum (complement) were added, incubated for 5 hours at 37°C with shaking with 20 min. interval. For one control group, no complement was added. No antibodies were added in the other control groups. The measurement of complement killing was made by counting the ratio of dead and live cells after staining with trypan blue. %

lysis= _No. of dead cells_ No. of dead cells + No. of living cells.

Cloning and Sequencing of the Variable Regions of 0325-6-9. RNA was isolated from hybridomas with RNA extraction kit (RPN. 1264, Amersham). DNA fragments of variable regions of both heavy and light chains were amplified by PCR as described (18). Amplified DNA fragment were eluted from gel and cloned into M13 mpl8, and used to transform JM103 with M13 cloning kit (BM). Single-stranded DNA was isolated and sequenced with USB DNA sequencing kit (Sequenase Version 2.0) RESULTS Isolation of Monoclonal Antibody Specific for AFP. After the splenic lymphocytes-FO cell fusion, hybrid cells were plated onto five 96-well plates. Fourteen days later, hybridoma cells appeared in 200 of the 480 wells. Culture supernatants from these 200 wells were screened by ELISA as described in Materials and Methods to detect the secretion of anti-AFP antibodies. Among these two hundred wells tested, twenty wells were These positive. twenty wells were further identified by Western blot analysis. Supernatants from four of the twenty wells contained antibodies reacted with a 70 KD protein from purified AFP and supernatant from HepG2, but did not react with HSA. The molecular weights of AFP and HSA were about 70KD (1). The best of these four wells, whose antibody demonstrating the highest affinity in ELISA, was identified. From this chosen well, cells were purified by subcloning with the limiting dilution method. The hybridoma cell which produced the highest affinity monoclonal antibody was named 0325-69 and its isotype was determined to be IgGl by the mouse subclass kit

(Southern Biotechnology Associates, Inc.). Cross-Reactivity Test. Human Ig and HSA have been known

to be human predominant Ig and HSA was examined by ELISA as described in Materials and Methods. Results are presented in Table I. Although the amount of coated HIg and HSA was 167 times higher than that of the coated AFP, the reactivity indicated by absorbance at 492nm was obviously lower. These results suggest that 03256-9 is specific to AFP, and human Ig and HSA do not interfere with the detection of AFP. 0325-6-9 Illustrating Specificity to Five Hepatoma Cell Lines. Five leukemia and lymphoma cell lines were hepatoma cell lines and two human



Therefore, the cross-reactivity of 0325-6-9




Cross-Reactivity 0325-6-9 ascites dilution 1:10000 1:100000 1:1000000

Test of 0325-6-9


Human Ig and HSA O.D. 492nm

HIg (5ug/ml)

pure AFP

(0.03ug/ml) >2.0 1.13+0.078 0.20+0.010

AFP: Alpha-fetoprotein HSA:Human serum albumin



0.23+0.112 0.02+0.015 0.19+0.053 0.02+0.012 0.01+0.005 0.17+0.045 HIg: Human immunoglobulin

immunostained with 0325-6-9, anti-thyroid stimulating hormone (TSH) murine monoclonal antibody (2D7-5, IgGl) and anti-Chlamydia trachomatis murine monoclonal antibody (002-7, IgG 2a), respectively. The staining processes were described in Materials and Methods. In the control group, the primary antibody was replaced with dilution buffer only. Results of the cell immunostain experiments are presented in Table H. HepG2, Hep3B, SK-HEP1, HCC36 and HA22T/VGH gave a response from strong +++ (crimson), ++ (red), ++, ++, to weak + (pink) reactivity with 0325-6-9, respectively. None of these hepatoma cell lines reacted with either 2D7-5 or 002-7. Two human leukemia and lymphoma cell lines did not react with any of three tested monoclonal antibodies. When the primary antibody was replaced with dilution buffer, all tested cell lines appeared negative. These results illustrated the specificity of 0325-6-9 to five hepatoma cell lines and the presence of AFP in the cytoplasm of hepatoma cells. 0325-6-9 Reacting with Cell-Surface AFP. One of the most important characteristics of an anti-AFP monoclonal antibody for therapeutic use is its potential cancer cell targeting ability. The binding ability of 0325-6-9 to hepatoma cell surface was demonstrated in the following complement killing leukemia and lymphoma cell experiments. Five hepatoma cell lines, two lines and one colonie adenocarcinoma cell line were reacted with 0325-6-9, 2D7-5 and 002-7 separately, and then treated with complement as described in Materials and Mothods. In one seperate control experiments, no added. In the other control experiment, no antibody was was complement added. All the experiments were repeated three times. Results are presented in Table The percent of lysis was the mean of three individual . For 0325-6-9, the percent of lysis of HepG2, Hep 3B, HCC 36, experiments. SK-HEP-1 and HA22T/VGH were 64.66% +24.98%, 76.80% ±10.90%, 53.6% + 14.6%, 56.83% + 3.89% and 67.23% ± 5.53%, respectively. For both 2D7-5 and 002-7, the percent of lysis of all five hepatoma cell lines was lower than 10% which was about the same as the control group. The percent of lysis of leukemia and lymphoma cell lines, one colonie adenocarcinoma was also two lower than 10%, regardless of the treatment with any of three tested antibodies, or no complement was added. The percent of lysis of five hepatoma cell lines was lower than 5% when only complement was added. These results suggest that 0325-6-9 is specific to five hepatoma cell lines and AFP is present on the cell surface of hepatoma cells. Nucleotide and Amino Acid Sequences of both Heavy and Light Chain 343


W Cu

o o

) O X>










t t í


X O >

c o

U o



Xt a o



o w O.








< in









-s U

Ë o


c u

(S ( O




Ö +






gì -e-




+1 g?




+1 g?







ce .

c« *

!2 ce





'5 "Sj







gì ri


un >—'

6S vO




* o en s



S #35 un vo "* gì

S £ r? CN +|







3 C ce





Characterization and nucleotide sequences of the variable regions of a monoclonal antibody against alpha-fetoprotein.

alpha-Fetoprotein (AFP) is a well-known tumor marker of hepatocellular carcinoma (HCC). Monoclonal antibodies against AFP possessing specific binding ...
1MB Sizes 0 Downloads 0 Views