HYBRIDOMA Volume 11, Number 5, 1992 Mary Ann Liebert, Inc., Publishers
Development of Human Monoclonal Antibodies Against Human Cytomegalovirus YOSHIAKI KANOH,1 MIKIHIRO YUNOKI,1 TOMOKUNI TANIGUCHI,1 YUKIO SUZUKI,1 SHOJI IDENO,1 ETSUKO MORITA,1 MINORU HIRAMA,1 KAZUMASA YOKOYAMA,1 and KOICHI YAMANISHI2 'Central Research Laboratories, The Green Cross Corporation, Hirakata 573, Japan 2Research Institute for Microbial Diseases, Osaka University, Suita 565, Japan
ABSTRACT Human monoclonal antibodies
(HCMV) have been developed by fusion of human spleen cells and human lymphoblastoid cell lines (NP101 and NP197). The cell line NP101 had great advantages in its high fusion frequency and the stability of
The specificity of HMAbs was confirmed by immunosorbent assay (ELISA) and immunofluorescence enzyme-linked six of the HMAbs Two staining. obtained, which were IgG3 subclass, neutralized viral infectivity in the absence of complement. The neutralizing activity of one of these two HMAbs was enhanced in the presence of human complement, whereas the other was not. Another IgGl subclass HMAb neutralized viral infection only in the presence of complement. The remaining three HMAbs showed no neutralizing activity. Those HMAbs may provide an important approach to studying human immune responses to HCMV. HMAbs having neutralizing activity may prove to be useful for passive immunotherapy of HCMV diseases. the resultant
INTRODUCTION Human cytomegalovirus (HCMV) is an important pathogen in humans. Immunocompromised hosts including organ allografts recipients, patients with acquired immunodeficiency syndromes, and human fetuses
and sometimes fatal HCMV infection (1). While immunity is said to be critical in recovery from HCMV there have been a number of reports suggesting that pasinfections, sive immunizations with human immunoglobulin containing antibodies to HCMV reduce the incidence of symptomatic infections and modify its
often develop cell-mediated
patients (2,3,4). With the limited availability of globulin containing a high titer of antibody to HCMV, the development of HMAbs to HCMV has been investigated in many laboratories (5,6,7,8,9). However, few of these HMAbs were able to neutralize HCMV (6,7,8). We report here the development of HMAbs to HCMV produced by human-human hybridomas. These human-human hybridomas were obtained by the fusion of splenic lymphocytes and two newly established human fusion partner cell lines, NP101 and NP197. Our approach uses neither Epstein-Barr virus transformation nor mouse myeloma or severity in
fusion partners. The HMAbs obtained have are thought to be good candidates strong neutralizing activity for immunotherapy of HCMV disease. Furthermore, we describe the participation of human complement in a virus neutralization assay. human-mouse
MATERIALS AND METHODS Cell fusion were obtained from the spleens of stomach cancer These cells were cultured for 6 days in Iscove's MEM (IMDM) supplemented with 20 % fetal calf serum (FCS), 0.025 % pokeweed mitogen (Gibco Lab., Inc.) and 0.001 % Zysorbin (Zymed Lab., Inc.) at a density of 107 cells/ml. Stimulated cells were fused with human partner cell line NP101 or NP197 at a ratio of 2:1 in the presence of 45 % polyethylene glycol 4000 for 1 min. The cell lines NP101 and NP197 were hypoxanthine-aminopterin-thymidine (HAT)-sensitive and Ouabainresistant mutants of IM-9. These cell lines did not secrete IgG. The fused cell lines were dispensed into 96-well microplates at a concentration of 5 x 104 lymphocytes/well and selected in IMDM containing 20 % FCS and HAT. Three to four weeks after cell fusion, the culture supernatants from wells showing cell growth were harvested to test for anti-HCMV activity.
assay for anti-HCMV
Anti-HCMV activity was screened by an enzyme-linked immunosorbent assay (ELISA) using HCMV antigen and infected cells as target antigens. HCMV strain AD169 was used throughout this study. MRC-5 cells were infected with HCMV at a multiplicity of infection (moi) of 0.5 to 1.0. After 7 to 10 days cultivation, cell-free culture supernatants were collected and centrifugea at 100,000 g for 1 hr. The HCMV antigen obtained was diluted to 10 ¿jg/ml with 0.05 M bicarbonate buffer (pH 9.6) and coated on EIA plates. HCMV-infected cell plates were prepared from MRC-5 cells grown in 96-well microplates and infected with HCMV (moi 0.05). Eight days later, the cells were washed and fixed with methanol. Control plates were prepared without virus infection. The ELISA procedure was according to the method previously described (10), except that horseradish peroxidase (HRP)-conjugated goat anti-human IgG (Tago Inc.) was used as the second antibody. Hybridomas secreting antibodies reacting with HCMV antigen or =
with infected cells and not with control cells limiting dilution method.
Classification and quantitation of immunoglobulins.
IgG subclass kits (Green Cross Corp.) were used to determine the subclass of each HMAb. Mouse MAb to each subclass was coated as the target antigen and HRP-conjugated goat anti-human IgG (Cappel Inc.) was used as the second antibody. Each HMAb was purified from the serum-free culture supernatants by protein G-Sepharose (Pharmacia A.B.). The light chain type was determined using anti-human kappa (Tago Inc.) or anti-human lambda (Tago Inc.) as a target antigen and HRP-conjugated anti-human kappa (Tago Inc.) or anti-human lambda (Tago Inc.) as the second antibody. IgG concentration was determined by ELISA using standard human serum (Behring Werke) as a control. The target antigen was goat anti-human IgG (Tago Inc.) and the second antibody was HRP-conjugated anti-human IgG (Tago Inc.). Indirect immunofluorescence assay MRC-5 cells were grown on glass coverslips and infected with HCMV (10 plaque forming units, pfu) when the cells neared confluency. The cells were cultured for 2 to 3 days until cytopathic effects were ob-
served. The infected cells were then fixed with -20 °C acetone for 10 min and air dried. They were then incubated with various HMAbs (undiluted supernatant) for 1 hr at 37 °C. After washing to remove excess antibody, fluorescein isothiocyanate-conjugated anti-human IgG (Tago Inc.) was added and incubated for an additional hour at 37 °C. After washing, the cells were mounted in 90 % glycerol in phosphate-buffered saline.
Virus neutralization Serial dilutions of HMAb, diluted in MEM containing 1 % FCS, were incubated with an equal volume of 2000 pfu/ml HCMV and a 1:10 dilution of guinea pig complement (Cedarlane Lab., Ltd.) After 1 hr incubation, 100 jul of the mixture was added to confluent MRC-5 cells grown in a 6-well plate. The plate was incubated for 1 hr at 37 °C, washed to remove excess virus and then overlaid with MEM containing 10 % FCS and 0.5 % agarose. After 12 days cultivation, the cells were fixed with formalin and stained with 0.03 % méthylène blue. The neutralizing titer was expressed as the concentration of the antibody necessary for a 50 % reduction of plaque formation (ED50). Occasionally, human serum negative for both HCMV neutralizing and anti-HCMV ELISA activity was used instead of guinea pig complement. In those cases, identical heat-inactivated (56 °C, 30 min) serum was used as the control. The concentration of the serum was 50 %.
Immunoprecipitation MRC-5 cells
infected with HCMV and cultured until 70 % of the
cytopathic effects. The infection were metabolically labeled
MRC-5 cells at 3 to 4
with 35S-methionine (Amersham) were harvested and lysed with 1 % NP-40, 1 % sofor 24 hr. The cells dium deoxycholate, 0.1 % sodium dodecyl sulfate (SDS), 1 mM phenylmethylsulfonyl fluoride (PMSF), 140 mM sodium chloride and 20 mM Tris (pH 7.8) (lysis buffer) and the lysate centrifugea at 100,000 g for 1 hr. The crude HCMV antigen supernatant was incubated with HMAb and protein G-Sepharose (Pharmacia A.B.) for 15 hr at 40 °C with gentle mixing. The protein G-Sepharose was washed with lysis buffer 20 % glycerol, 0.04 % bromphenolblue, 10 % and incubated in 3 % SDS, 2-mercaptoethanol and 63 mM Tris (pH 7.6) at 100 °C for 5 min and centrifuged. The supernatant was collected and separated by electrophoresis on 4-20 % gradient polyacrylamide gels (Tefco Co., Ltd.) in the presence of SDS, fixed in 10 % acetic acid and 30 % methanol, dried and exposed to Fuji RX (Fuji Photo Film Co., Ltd.). A 14C methylated protein mixture (Amersham) was used as a molecular weight marker.
Competitive binding inhibition Purified HMAbs for 3 hr at
incubated with NHS-biotin (Vector Lab., Inc.) to prepare biotin-labeled antibodies.
Serial dilutions of HMAbs were added to each well of HCMV-coated 96well plates and incubated for 15 hr at room temperature. After washing to remove excess antibody, biotinylated antibody was added to the plate, incubated for 2 hr at 37 °C and washed. HRP-conjugated avidin (Vector Lab., Inc.) was then added. After 1 hr incubation at 37 °C, the plates were washed and the substrate solution added. The reaction was carried out for 10 min at room temperature. The inhibitory rate was expressed as the percentage of the absorbance obtained with each HMAb to that obtained without HMAb.
RESULTS Generation of
Table 1 presents the summary of the cell fusion experiments. Human spleen cells were obtained from two stomach cancer patients and fused with either NP101 or NP197. Cell line NP101 was superior to NP197 in its fusion frequency. In the first screening, about 10 % of the wells
Summary of Fusion
TK YF YF
NP197 NP101 NP197 NP101
1152 1200 1152 2112
67 576 75 637
showing cell growth were positive for antibody to HCMV antigen or HCMV infected cells. Cells from those wells were expanded and subcloned by limiting dilution in 96-well microplates. Eleven hybridomas were established and HMAbs from six of them were further characterized.
Neutralizing activity of HMAbs
Hybridomas were grown in serum-containing medium and then incubated in serum-free medium for 7 to 10 days. HMAbs were purified from these supernatants and analyzed in a virus neutralization assay. Table 2 shows the neutralizing activity of each HMAb with and without guinea pig complement. Two IgG3 subclass HMAbs, E46 and K357 showed neutralizing activity without complement. The reactivity of E46 was enhanced about 10 times in the presence of complement, whereas K357 was not. One IgGl subclass HMAb, K633, showed neutralizing activity only in the presence of complement. The other three HMAbs did not show neutralizing activity. The neutralizing activity of E46 and K357 was further investigated in the presence of human serum. As shown in Table 3, the activity of both HMAbs and polyclonal antibody was enTABLE 2.
of Monoclonal Antibodies
E46 K115 1351
TK x NP197 TK x NP101 YF x NP101
IgG3 IgGl IgGl
K357 1393 K633
TK YF TK
NP101 NP101 NP101
IgG3 IgGl IgGl
X X x
ED50 (^/g/rnl) without C' with C 0.03 0.26 ND ND
0.17 ND ND
ND: Not detected
Neutralizing Activity Antibodies with
with with heat inacti- untreated
Antibody E46 K357 IVIG IVIG
0.28 0.09 1136.00
0.02 0.08 188.00 167.00
Intravenous Immunoglobulin (IVIG) used was human IgG prepared from pooled plasma by polyethylene glycol fractionation. The ED50 of IVIG was determined with human serum (1) or a
guinea pig complement (2). 573
ND 0.19 ND 0.20
K115 FIGURE 1.
1351 of Indirect ImmunofluoresPhotomicrographs cence of HMAbs with CMV-lnfected Cells. 574
E46 K115 -i i-1 M.W. Inf. Control I. C.
K357 1393 K633 M.W. inf. Control[ C. [ C. C. M.W.(kDa) m *—200
+—92.5 *—69 -46
H—14.3 FIGURE 2. Molecular Weight of HCMV
hanced in the presence of human heat-inactivated human serum.
and not in the presence of
antigens recognized by HMAbs
The specificity of the HMAbs was confirmed by an indirect immunofluoresence assay. As shown in Figure 1, E46, K357 and K633 stained the cytoplasm of infected cells, with a weaker, diffuse nuclear staining. HMAb 1351 intensely stained nuclear regions. Uninfected cells gave no staining with these HMAbs (data not shown). HCMV antigens recognized by these HMAbs were determined by immunoprecipitation (FIGURE 2). The molecular weight of the target proteins recognized by each HMAb were calculated as followed: E46; 78kD, K115; 50kD, 1351; 50kD, K357; 80kD, 1393; 50-85-120kD, K633; 50-85-120kD. HMAbs were classified into 5 groups based upon their antigenic epitopes. A competitive binding inhibition assay showed that, with the exception of 1393 and K633, there seemed to be no mutual binding inhibition between them (TABLE 4). TABLE 4. Mutual Binding Inhibition of Monoclonal Antibodies
_E46 E46 K115 1351 K357 1393 K633 +: -:
Showed more than 50 % inhibition Showed less than 50 % inhibition
DISCUSSION In this report, we described the production and initial characterization of six stable HMAbs to HCMV. Though several techniques have been reported (11), the production of HMAb is rather difficult when compared with mouse MAb. There seem to be two major issues to be resolved. One is the selection of the lymphoid tissue and the other is the choice of the fusion partner cell line. At present, the in vitro immunization procedure is not well-established, so that the selection of lymphocytes sensitized to the desired antigen is important. We used spleen cells from cancer patients as the source of lymphocytes based on cell recovery and on our previous unpublished observations that lymphocytes from cancer patients secrete higher titer antibodies to some pathogenic antigens, including HCMV, than those from healthy donors. Bieber et al.(12) also reported the advantages of utilizing spleen cells compared to peripheral blood cells. The fusion frequency of NP101 is about 10~5 when fused with lymphocytes (TABLE 1) and 10"4 when fused with transformed B cells (unpub-
observation). The hybridomas show stable production of immunoglobulin and grow well. These characteristics of NP101 are comparable to those of previously reported fusion partners (13,14,15). lished
From the six HMAbs
obtained, two IgG3 subclass HMAbs (E46 and K357) showed neutralizing activity in the absence of complement. As for the subclass distribution of antiviral antibodies, Beck et al. (16) reported that specific anti-viral activity is about 10 times higher in the IgG3-fractions compared to the other IgG subclass fractions. Wahren et al. (17) showed that the loss of IgG3 to HCMV would indicate progression of HCMV disease. In addition to the fact that the preferential induction of IgG3 was often detected in primary disease (18), these show the importance of antibodies of the IgG3 subclass in preventing HCMV disease. The neutralizing activity of E46 is enhanced about 10 fold in the presence of guinea pig complement or human serum. When the human serum was heat inactivated, the enhancing effect disappeared. This indicates that the effect of human serum depends on human complement. Furthermore, this shows that the antigenic epitope is not masked by plasma proteins such as ß-2 microglobulin (19). These characteristics of E46 strongly suggest that this HMAb would serve as an alternative source of antibodies for the prevention of life-threatening HCMV infections. The neutralizing activity of K357 was not influenced by the presence of complement. Rasmussen et al. (20) reported a mouse MAb recognizing an antigen with a molecular weight similar to that recognized by K357. The antiviral activity of their MAb was also complement-independent. HMAb K357 will be useful for identification of the association site on the virus envelope for host cells. HMAbs 1393 and K633 seem to recognize the same protein
antigen designated gel (gB) (21,22). Their antigenic epitopes
thought to be very close but not identical, because K633 does, but 1393 does not show neutralizing activity in the presence of complement. HMAb 1351, which detects a nuclear antigen, strongly reacted with infected cells. Although whether the HMAbs E46 and K357 were directed to the immediate early(IE), early (E) or late (L) class of viral proteins was not defined in the present study, immunoprecipitation analysis indicated that the HMAbs were likely to recognize the same neutralizing determinant within gcIII (gH) envelope protein (23). This suggests the possible use of the antibody for the identification of HCMV infection. REFERENCES
Griffiths, P.D. and Grundy, J.E. (1988) The status of CMV as a pathogen. Epidem. Infect. 100, 1-15. (2) Winston, D.J., Pollard, R.B., Ho, W.G., Gallagher, J.G., Rasmussen, L.E., Huang, S.N.Y., Lin, C.H., Gossett, T.G., Merigan, T.C. and Gale, R.P. (1982) Cytomegalovirus immune plasma in bonemarrow transplant recipients. Ann. Intern. Med. 97, 11-18. (3) Snydman, D.R., Werner, B.G., Heinze-Lacey, B. and Berardi, V.P. (1987) Use of cytomegalovirus immune globulin to prevent cytomegalovirus disease in renal-transplant recipients. New Eng. J.
Bron, D. and Klastersky, J. (1989) 577
infections in Clin. Oncol. 25, 1365-1368.
Emanuel, D., Gold, J., Colacino, J., Lopez,
(1984) A human monoclonal Immunol. 4, 2202-2206.
(6) Matsumoto, Y., Sugano, T., Miyamoto, Generation of
cytomegalovirus (CMV). C. and
hybridomas producing human monoclonal antibodies cytomegalovirus. Biochem. Biophys. Res. Commun.
137, 273-280. Redmond, M.J., Leyritz-Wills, M., Winger, (7)
L. and Scraba, D.G. The selection and characterization of human monoclonal antibodies to human cytomegalovirus. J. Virol. Methods 14, 9-24.
Foung, S.K.H., Perkins, S., Bradshan, P., Rowe, J., Rabin, L.B., Reys, G.R. and Lenette, E.T. (1989) Human monoclonal antibodies to human cytomegalovirus. J. Infect. Dis. 159, 436-443. (9) Bron, D., Delforge, A., Lagneaux, L., De Martynoff, G., Bosmans, E., Van der Auwera, P., Snoeck, R.,Burny A. and Stryckmans, P. (1990) Production of human monoclonal IgG antibodies reacting with cytomegalovirus (CMV). J. Immunol. Methods 130, 209-216. (10) Kano, Y., Nakura, K., Nakae, T., Ueda, Y., Uemura, Y., Yokoyama, K., Ishida, H., Ohyanagi, H. and Saitoh, Y. (1990) Development of
monoclonal antibodies against human gastrointestinal cancer. Tohoku J. Exp. Med. 160, 361-373. (11) James, K. and Thomas Bell, G. (1987) Human monoclonal antibody production, Current status and future prospects. J. Immunol. Methods 100, 5-40. (12) Bieber, M. and Teng, N.N.H. (1987) In vitro sensitization for the production of human monoclonal antibodies. In: A.J. Strelkauskas (Ed.), Immunology Series, Vol. 30, Human Hybridomas; Diagnostic and Therapeutic Applications (Marcel Dekker, New York)
p.39-46. (13) Teng, N.N.H., Lam, K.S., Riera, F.C.and Kaplan, H. S. (1983) Construction and testing of mouse-human heteromyelomas for human monoclonal antibody production. Proc. Nati. Acad. Sei. U.S.A. 80, 7308-7312.
Foung, S.K.H., Perkins, S., Arvin, A., Lifson, J., Mohagheghpour, N., Fishwild, D.,Grumet, F.C. and Engleman, E.G. (1985) Production of human monoclonal antibodies using a human mouse fusion partner. In: E.G. Engleman, S.K.H. Foung, J. Larrick and A. Raubitschek (Eds.), Human Hybridomas and Monoclonal Antibodies (Plenum Press, New York) p. 135. (15) Ohashi, H., Hashizume, S., Murakami, H., Aihara, K., Shinohara, K. and Omura, H. (1986) HO-323, a human lymphoblastoid cell line useful for making human-human hybridomas. Cell Biol Int Rep. 10,
77-84. O.E. (1981) Distribution of virus antibody activity among IgG subclasses. Clin. Exp. Immunol. 43, 626-632. (17) Wahren, B., Linde, A., Sundqvist, VA., Ljungman, P., Lonnqvist,
(16) Beck, human
O. (1984) IgG-subclass-specific CMV reactivity in transplant recipients. Transplant. 38, 479-483.
(18) Linde, G.A., Hammarstrom, L., Persson, M.A.A., Smoth,E.C.I., Sundqvist, VA. and Wahren, B. (1983) Virus-specific antibody ac578
tivity of different subclasses of immunoglobulins G and A in cytomegalovirus infections. Infec. Immun. 42, 237-244. (19) MacKeating, J.A., Griffith, P.D. and Grundy, J.E. (1987) Cytomegalovirus in urine specimens has host beta-2-microglobulin bound to the viral envelope. A mechanism of evading the host imresponse? J. Gen. Virol. 68, 785-792. (20) Rasmussen, L.E., Nelson, R.M., Kelsall, D.C. and Merigan, mune
antibody single protein neutralizes of human cytomegalovirus. Proc. Nati. Acad. Sei.
the infectivity U.S.A. 86, 876-880.
(21) Gretch, D.R., Kari, B., Rasmussen, L., Gehrz, R.C. and Stinski, M.F. (1988) Identification and characterization of three distinct families of glycoprotein complex in the envelopes of human cytomegalovirus. J. Virol., 62, 875-881. (22) Kari, B., Liu, Y-N.C, Goertz, R., Lussenhop, N., Stinski, M.F. and Gehrz, R.C. (1990) Structure and composition of a family of human cytomegalovirus glycoprotein complexes designated gc-I(gB). J. Gen. Virol.,71, 2673-2680. (23) Cranage, M.P., Smith, G.L., Bell, S.E., Hart, H., Brown, C, Bankier,A.T., Tomlinson, P., Barrell, B.G. and Minson, T.C. (1988) Identification and expression of a human cytomegalovirus glycoprotein with homology to the Epstein-Barr virus BXLF2 product, varicella-zoster virus gpIII, and herpes simplex virus type 1 glycoprotein H. J. Virol., 62, 1416-1422. Address reprint request to: Yoshiaki Kanoh, Ph. D. Central Research Laboratories, The Green Cross Corporation,
Hirakata, 573, Japan Received for publication: Jan. after revisions: June
31, 1992 22, 1992