Mar 16, 1981 - Thirty hybridomas that secrete immunoglobujlins against the simian virus 40 tumor antigens were isolated and cloned. Of these, 28 produced ...
Vol. 39, No. 3
JOURNAL OF VIROLOGY, Sept. 1981, p. 861-869
0022-538X/81/090861-09$02.00/0
Monoclonal Antibodies Specific for Simian Virus 40 Tumor Antigens ED HARLOW,* LIONEL V. CRAWFORD, DAVID C. PIM, AND NICOLA M. WILLIAMSON Department of Molecular Virology, Imperial Cancer Research Fund Laboratories, London WC2A 3PX, England
Received 16 March 1981/Accepted 13 May 1981
Thirty hybridomas that secrete immunoglobujlins against the simian virus 40 tumor antigens were isolated and cloned. Of these, 28 produced antibodies which bound to simian virus 40 large-T, and 2 produced antibodies which bound to the host 53,000-dalton protein. As in previous work, large-T antigen was found to have at least one determinant that it shared with small-t antigen and to have a miniimum of six unique determinants. Several of the monoclonal antibodies from the L series hybridomas recognized determinants that were present on a subset of the large-T antigen from simian virus 40-transformed mouse cells. These monoclonal antibodies should be useful in studies of the structure and function of the simian virus 40 tumor antigens.
Sera from animals bearing simian virus 40 (SV40)-induced tumors contain antibodies which specifically bind to a group of proteins known collectively as the SV40 tumor antigens. The SV40 tumor antigens may be viral proteins or host proteins which have become immunogenic after SV40 transformation. The SV40coded large-T and small-t antigens are members of the first class, and these polypeptides have relative molecular weights of 94,000 and 20,500, respectively. At least one other member of this class of viral antigens has yet to be identified. The viral polypeptide which functions as the tumor rejection antigen has not been determined, but it probably shares some sequence homology with large-T (27). In addition to these viral proteins, at least two host antigens are specifically recognized by tumor sera. These are the S surface antigen (10) and the host 53,000dalton (53K) protein found in a complex with large-T (16, 19, 21). Much of our knowledge about the SV40 tumor antigens is based on the antibody activities present in tumor sera. Many advances in our understanding of the SV40 tumor antigens have followed the development of better antisera or immunochemical methods. Recently, several laboratories have reported the production of monoclonal antibodies raised against the SV40 tumor antigens (7, 17, 20). These immunoglobulins represent the most recent and sophisticated development in immunochemical analysis of the SV40 tumor antigens. Studies with monoclonal antibodies have shown that large-T can be divided into subclasses by the presence or
absence of a single antigenic determinant (7). This suggests that large-T, in addition to having multiple activities, may have several topological forms. Monoclonal antibodies to large-T have also shown that large-T shares at least one antigenic determinant with a host 68K protein (17). It seems likely that detailed mapping of the SV40 tumor antigens will provide important information about the tertiary structure and, ultimately, about the functions of these antigens. We have expanded the existing collection of monoclonal antibodies by raising large numbers of hybridomas that produce antibodies specific for the SV40 tumor antigens. To date, we have isolated and cloned 28 hybridomas that secrete anti-large-T antibodies and 2 hybridomas that produce anti-53K immunoglobulins. We present here the initial characterization of these antibodies. Because of their potential uses, we have selected only antibodies which both function efficiently in immunoprecipitation and bind to staphylococcal protein A. MATERIALS AND METHODS Cells and viruses. CeUs were grown as previously described (8). The SV40-transformed mouse cells (SVA31E7 and VLM) were kindly provided by Y. Ito and S. Tevethia, respectively; the SV40-transformed rat cells (14B), by D. Lane; the SV40-transformed hamster cells (H65 90B), by S. Tevethia; the SV40transformed rabbit cells (TRK54), by P. Black; the SV40-transformed human cells (SV80), by E. Gurney; and the methylcholanthrene-transformed mouse cells (L929), by D. Cox. All viruses were propagated on CV1 African green monkey kidney cells as described previously (8). The SV-S strain was used as wild-type 861
862
HARLOW ET AL.
SV40. The SVGT12 and SVGT,BG viruses were grown with tsB201 helper virus and were the gift of P. Southern and P. Berg. The nondefective adenovirus-SV40 hybrid viruses Ad2+ND1 and Ad2+ND2 were provided by A. Lewis, and the defective adenovirus-SV40 hybrid virus Ad2+D2 was the gift of D. Lane. The SV40 deletion mutants d11263 and dl1265 were from P. Berg, and the tsA1499 mutant was from T. Shenk. Radioactive labeling, immunoprecipitation, and electrophoresis of proteins. Virus-infected cultures were pulse-labeled for 3 h with [uS]methionine or [32P]phosphate when cytopathic effects were seen. The labeling conditions and the methods of immunoprecipitation and sodium dodecyl sulfate-polyacrylamide gel electrophoresis have been described previously (8). Preparation of monoclonal antibodies. Syngeneic tumors were induced in BALB/c female mice by subcutaneous injections of 106 or 107 VLM cells. Tumors were excised, and tumor fragments were injected into a second BALB/c female, either by a subcutaneous or an intraperitoneal route. When small tumors appeared, the mice were boosted via the tail vein with 0.1 ml of either a whole-cell lysate from 107 VLM cells or a partially purified large-T-53K complex from 107 SVA31E7 cells (the kind gift of F. McCormick). Both methods for the final boost have been used successfully. The spleen was removed 3 days after the final boost, and splenocytes were fused with 107 P3-NS1-1Ag4-1 BALB/c plasmacytoma cells as described by Galfre et al. (6). Cells were plated in microtiter wells in RPMI 1640 medium supplemented with 20% fetal calf serum (prescreened to support hybridoma growth), 10% NCTC 135 medium (GIBCO Laboratories), 0.15 mg of oxalacetate per ml, 0.05 mg of pyruvate per ml, and 0.2 U of insulin per ml (RPMI medium). The medium was changed 1 day later to RPMI medium plus 0.1 mM hypoxanthine, 0.01 mM methotrexate, and 0.01 mM thymidine. Culture supernatants were ready to be screened 7 to 10 days after fusion. Approximately 50,l of tissue culture supernatant was removed from each microtiter well and added to lysate from 106 SV40-transformed cells labeled with [32P]phosphate in 0.5 ml of 150 mM NaCl-5 mM EDTA-50 mM Tris-0.25% gelatin (pH 7.4) (NET/ GEL). [nP]phosphate-labeled lysates were prepared from either SV80 human or SVA31E7 mouse cells. The mixture was incubated overnight at 4°C, and then 20 p4 of a 10% suspension of fixed Staphylococcus aureus Cowan I (SAC) (12) was added. After 15 min, SAC was pelleted and suspended in 30 pd of 2% sodium dodecyl sulfate-100 mM dithiothreitol-10% glycerol20 mM Tris (pH 6.8, sample buffer). The SAC and sample buffer mixture was heated to 850C for 10 min, and 20 pl was loaded onto a 10% polyacrylamide gel. Positive culture supernatants were determined by the presence of the phosphate-labeled large-T or 53K bands on autoradiographs of the dried gels. Positive cultures were transferred to 25-cm2 flasks in RPMI medium with 0.1 mM hypoxanthine and 0.01 mM thymidine and then were single-cell cloned three times either by a limiting dilution or by pipetting single cells. Individual hybridoma cells could not form colonies in RPMI medium without the addition of preconditioned RPMI medium or a feeder cell layer to provide con-
J. VIROL.
ditioned medium. Spleen cells from normal BALB/c females served as effective feeder cells (ca. 106 spleen cells per ml). Preconditioned medium was prepared by incubating spleen cells in RPMI medium for 3 days. The spleen cells were removed by centrifugation, and the supernatant was used as preconditioned medium. Purification of monoclonal antibodies. Antibodies were purified from large volumes of tissue culture supernatants on protein A-Sepharose CL4B beads (5). Tissue culture supernatants were adjusted to pH 8.0 with 1 M Tris, and protein A-Sepharose CL4B was added. The mixture was stirred overnight at 40C, and the beads were collected on a sinteredglass filter. The beads were washed with 10 volumes of 100 mM Tris (pH 8.0), followed by 10 volumes of 20 mM Tris (pH 8.0). The immunoglobulins were eluted with 10 volumes of 100 mM glycine (pH 3.0), and the eluate was adjusted to approximately pH 8.0 with 1 M Tris. Preparations were at least 95% pure immunoglobulin as judged by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Coomassie blue
staining. Radioactive labeling of purified antibodies. Iodination of purified monoclonal antibodies was done by a modification of the method of Syvanen et al. (26). Metabolically labeled antibodies were prepared by growing cloned hybridoma cells in the presence of [35S]methionine. Cells were washed and suspended in Dulbecco modified Eagle medium without methionine. The medium was then supplemented with [3S]methionine (0.1 mCi/ml), and the culture was incubated overnight. Cells were removed by centrifugation, and antibodies were purified from the supernatant as described above. Immunoassays. Immunoglobulin classes and subclasses were determined either by standard doublediffusion agarose gels (1% agarose in 150 mM NaCl-50 mM barbital buffer, pH 9.6) or by competition radioimmunoassays. For double-diffusion gels, rabbit anti-mouse immunoglobulin class and subclass sera (Miles Laboratories, Inc.) were loaded into wells adjacent to tissue culture supernatants or purified antibodies, and precipitin lines were read after 24 h at room temperature. For the competition radioimmunoassays, antibodies from rabbit anti-mouse immunoglobulin class and subclass sera were purified on protein A-Sepharose. These antibodies were bound to polyvinyl chloride microtiter wells by adding 50 pl of a 10-,ug/ml solution in 10 mM phosphate (pH 7.0) to each well and incubating it overnight. Unbound antibody was removed by three washes with NET/GEL buffer, which also filled any remaining binding sites on the plate. An unknown tissue culture supernatant (20 p1) was then mixed with approximately 10 ng of an iodinated monoclonal antibody of a known immunoglobulin class and subclass. This mixture was added to the well and incubated for 3 h at 20°C. The plate was washed in NET/GEL, and the counts bound to the well were determined. This test was both specific and sensitive. Positive tissue culture fluid diluted 100-fold still blocked the binding of the iodinated probe to a significant extent. The ability of antibodies to bind to SAC was measured by adding equal concentrations of metabolically labeled purified antibodies to equal amounts of SAC.
VOL. 39, 1981
MONOCLONAL ANTIBODIES FOR SV40 TUMOR ANTIGENS
After 10 min at 200C, SAC was pelleted by centrifugation, and the supernatants and pellets were counted. A modified version of the direct-binding radioimmunoassay of Lane and Robbins (18) was used to measure the ability of monoclonal antibodies to bind to gelpurified proteins. After the monoclonal antibodies were incubated overnight with the radiolabeled gelpurified polypeptide, 5 pl of rabbit anti-mouse immunoglobulin serum was added. After 1 h, 50 pl of SAC was added, and the percentage of counts bound to SAC was determined. Indirect immunofluorescence was done as previously described (21). Cascade immunoprecipitations. SVA31E7 cells were pulse-labeled with [nP]phosphate for 3 h at 370C. Tissue culture supernatant from one of the L series hybridomas was added to a detergent lysate of the labeled culture. After 1 h at 40C, 0.5 id of rabbit antimouse imnmunoglobulin was added, and the incubation was allowed to continue for an additional hour. The antibody-antigen complexes were collected on SAC, and the supematant was added to a fresh sample of tissue culture supernatant from the same hybridoma. The procedure was repeated twice more, and the resulting supernatant was immunoprecipitated with mouse anti-tumor serum. The individual SAC pellets were washed, and the polypeptides were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis.
RESULTS Production of monoclonal antibodies. The application of cell fusion techniques to lymphoid cells by Kohler and Milstein (14) has allowed the production of hybridoma cell lines which secrete antibodies, each of which recognizes a single antigenic determinant. This approach has been widely used over the past few years to produce monoclonal antibodies with a chosen specificity. We sought to devise a useful procedure for making large numbers of monoclonal antibodies that are specific for SV40 tumor antigens. A number of different regimes were tested to yield both successful fusions and high numbers of hybridomas that produced antibodies against the tumor antigens. The most successful procedure is outlined above. To avoid the problem of inoculating mice with immunogenic proteins present in tissue culture media, we injected mice with fragments of syngeneic tumors. Peritoneal tumors gave better yields of positive hybridomas than did subcutaneous tumors. In a typical fusion, 1 in 5 to 1 in 10 of the hybridoma clones produced antibodies against large-T. Because we wanted antibodies which would be useful in the widest variety of assays, we selected antibodies which were efficient in immunoprecipitation and could be collected on SAC. A typical screening assay is shown in Fig. 1. Cells from wells which were positive in these assays were transferred and single-cell cloned
863
three times. Hybridomas were then assigned an L series number, and large volumes of tissue culture supematant were collected. Most hybrid cells adapted easily to Spinner culture, and this facilitated large-volume culture. Antibody was purified from supematants by affinity chromatography on protein A-Sepharose. Localization of binding sites. During the study of the biology of SV40 infections and transformations, many viruses have been isolated which contain subgenomic fragments of the SV40 genome. Many of these viruses are either naturally occurring isolates of cocultivation with adenovirus (Ad2) or are viruses with DNA sequences which have been altered in vitro. Several of these viruses have well-defined lesions in the early region of SV40 DNA and produce polypeptides that have limited amino acid homology with large-T. Figure 2 shows the regions of shared homology between polypeptides from a collection of these viruses and authentic large-T. Semiconfluent monolayers of CV-1 cells were infected with these viruses, and the cultures were radiolabeled when cytopathic effects were seen. Samples of lysates from these infected cultures were immunoprecipitated with either purified monoclonal antibody or tissue culture supernatant. All precipitations were adjusted to receive equal antibody input. The results are summarized in Table 1. Using only immunoprecipitation data that were positive for a given polypeptide, we mapped the location of the antigenic determinants that were recognized by the L series monoclonal antibodies to sectors of large-T (Fig. 3). Because the large-T-related proteins could contain the amino acid residues that comprise an antigenic determinant common to large-T but fold in such a way as to mask this determinant, we did not use any negative immunoprecipitation data to further localize the binding sites. Although similar polypeptides are not available for the host 53K protein, Table 1 shows that at least some monoclonal antibodies raised against the mouse 53K protein also immunoprecipitated similar proteins from a wide range of mammalian species. In separate experiments, these 53K proteins have been shown to be analogous to the mouse 53K protein not only in shared antigenic determinants, but also because they form a complex with large-T (8; unpublished data). Characterization of L series monoclonal antibodies. Tissue culture supernatants from cloned hybridomas were assayed by double-diffusion techniques or competition radioimmunoassays to determine the immunoglobulin class and subclass (Table 2). All of the antibodies had gamma-heavy chains. This was not surprising,
864
J. VIROL.
HARLOW ET AL.
Large-T-
4
53K-.
FIG. 1. Immunoprecipitation screen of hybridoma supernatants. A confluent monolayer of SV80 cells was pulsed with [12P]phosphate, and samples of the lysate representing iO' cells were mixed with approximately 50 1dl of tissue culture supernatant from an 8-day culture of hybridoma cells. Immunoprecipitates were collected fr-om 20 lAof SAC, and polypeptides were separatedby sodium dodecyl sulfate-polyacr-ylamide gel electrophoresis. The gels were dried and autoradiographed on Kodak SB-S film.
SV40
Larg.-T
SV40smi- t
SVGT14FG 35K SVGTI2 65K
-
dl 1263 Large -T
-
-
U
-
_
di 1265 Large-T-U
-
tsA 1490 Lags-T
-
Ad2* ND1
_
28K
Ad2 N02 42K
Ad2*ND2
56K
Ad2+D2 107K
-
_
-
FIG. 2. Homology between SV40 large-T antigen and large-T-related polypeptides. The shaded areas include the maximum lengths of amino acid homology between SV40 large-T antigen and viral proteins containing a defined segment of the SV40 coding capacity. Nucleotide numbering is according to the SVsystem (29). The location ofsmall-t sequences was computed from Tooze (29), of SVGT12 and SVGT14fBG from P. Southern (personal communication), of d11263 and d11265 from Van Heuverswyn et al. (30), of tsA 1499 from Pintel et al. (22), ofAd2+ND1 and Ad2+ND2 from Khoury et al. (13), and ofAd2+D2 from Hassell et al. (9). Conversions between numbering systems were done by the formulas given by Tooze (29), and included the 17-nucleotide EcoRIIfragment
P.
because tumors should serve as a continuous source of stimulation for the humoral response as local tumor necrosis occurs. The ability of different monoclonal antibodies to bind to SAC was measured by radioimmunoassay. Metabolically labeled antibodies were purified on protein A-Sepharose, and the percentage of antibody that could be bound to SAC in the absence of antigen was determined. As expected, the binding was pH dependent, and at pH 8.0 immunoglobulins from the immunoglobulin G2a (IgG2a), IgG2b, and IgG3 subclasses could be essentially removed (>90%) from the supernatant by the addition of saturating amounts of SAC. Immunoglobulins from the IgGl subclass bound less well. Therefore, all immunoprecipitations and immunoassays using IgGl antibodies had an added incubation with rabbit anti-mouse antibodies to ensure the complete removal of the IgGl immunoglobulins with SAC. Characterization of antibody-antigen interaction. The ability of the monoclonal antibodies to bind to antigenic determinants which survive sodium dodecyl sulfate-polyacrylamide gel electrophoresis was measured by a modification of the direct-binding radioimmunoassay of Lane and Robbins (18). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis destroyed the antigenic determinant recognized by 16 of the 28 anti-large-T monoclonal antibodies tested. All 12 monoclonal antibodies that bound
VOL. 39, 1981
MONOCLONAL ANTIBODIES FOR SV40 TUMOR ANTIGENS
865
TABLE 1. Immunoprecipitation of large-T and 53K-related proteins by the L series monoclonal antibodies Immunoprecipitation
;4
Monoclonal antibody
~
-
E-
E-
LO
e
pd Rat 14B Z Et ~~+ +
+
+
+
+
>5
>
+0
+
-
+
+
-
+
+
+
+
+
+
+
-
+
-
+
+
+
+
+
+
+
-
L6
L5
A
+Lo+
+
+%
0
+
-
+ + . +ete . . .++ ++ . ++ .++ ++ . NDt L'L14 +D, . + + +
L10
+
L18
L14 L16
L21
f **:
.. +
+
+
-+
+-
+ + +-+ND+
+
+
+
-
+
+
-
-
+
+
-
-
-
+
-
+
+
+
+
-
-
+
-
-
-
-
-
-
-
-
+
+
-
-
ND
+
-
-
-
-
-
-
-
-
+
L19 + + + + + +-ih 3Smtitie coceL++ LOs-aee a~~~~~~~~~~ Infcte L270 + + + + + +
CV- celuselbldwih[2]popae b~~~~~~C Infcte L21 + + + + +--ND-+ L23 + + + ..+ + + + + + L27 + + ..+ + + + + + puse>bee + + + gL130 MonkeyCV-1 + clls wih[3>mehoni L31 L32 L33 L35 L36 TAO7
+
-
+
-
+
-
+
+
+
-
-
-
+
-
-
+
-
+
-
-
L38
+ +
IL4
+ +
+
-
+
L42
L44
+-
+
+
+
+
+
+
+--
+
+
+
+
+
+
+ +
+
+
+
+
-
-
+
+ -
--
+
-
-
-
-
-
-
+
+
+
+
+
+
+
+
+
+
+
.. + .. ++
+ +
+ +
+ +
+ +
-
-
-
-
-
-
-
_
-
-
__
-
-
-
.. -
+
-
+
-
-
nected CV-1 cells pulse-labeled with ['S]methionine.'Infected CV- 1 cells pulse-labeled with Mouse- L929 cells pulse-labeled with [TMS]methionine. Rat 14B cells pulse-labeled with [TMS]methionine. H65 90B cells pulse-labeled with ['S]methionine. f Rabbit TRK cells pulse-labeled with ['S]methionine. ' Monkey CV-1 cells pulse-labeled with [nS]methionine. Human SV80 cells pulse-labeled with ['S]methionine. 'ND, Not detected.
-
[~P]phosphate.
eHamter
the denatured large-T could be mapped close to the termini of large-T. This suggests that largeT may have structures free from the extensive secondary or tertiary structure at its amino and carboxy termini or that any secondary or tertiary structure in these areas can be easily reformed after denaturation. Both anti-53K monoclonal antibodies bound to the gel-purified 53K probe. Gumey et al. (7) have reported the production of an anti-large-T hybridoma which divides large-T into two classes: that which displays a particular antigenic determinant and that which does not. We screened our collection of mono-
clonal antibodies for this property by using cascade immunoprecipitations. All large-T-related polypeptides which displayed an antigenic detem inant recognized by a monoclonal antibody were removed from a sample of a [32P]phosphate-labeled lysate of the SV40-transformed mouse cells (SVA31E7) by repeated immunoprecipitation with tissue culture supematant from a single hybridoma. After all of the immunoreactive material had bVn removed from the lysate, mouse anti-tumor serum was added as in a standard immunoprecipitation. Several of the hybridomas in this set divided large-T into sub-
866
J. VIROL.
HARLOW ET AL.
did not assay the L series monoclonal antibodies for binding to the 68K protein described by Lane and Hoeffler. I-I IEIIU L2 DISCUSSION L4 Monoclonal antibodies as specific sera. L5 The construction of a complete family of hybridL6 omas that secrete antibodies that recognize all L7 of the immunogenic determinants of the SV40 L9 tumor antigens would obviate the need to raise L13 antisera in animals. Surrogate sera with the desired antigenic specificity and avidity could be LN constructed from a set of purified and characterLi6 ized monoclonal antibodies. We sought to design 119 a procedure that would provide this set for the _ L20 study of the SV40 tumor antigens. Syngeneic L23 tumors were used as a primary immunization, L27 and the final boost was an intravenous injection of crude or partially purified antigen. Our initial concern was to raise anti-large-T immunoglobL31 ulins, and we constructed and cloned 28 hybridL32. omas that produced anti-large-T antibodies. The L33 -aI - I - I initial characterization of these antibodies L35 showed that large-T antigen had at least seven I - I - I L36 antigenic areas. We hesitate to describe these as I- I - I -~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ L37 antigenic determinants because of the problem L38 of confusing the binding site of a monoclonal L39 I I antibody with a potentially larger topographical L40 *I-.area that is immunogenic. If the number of the L41 antigenic areas of myoglobin and lysozyme sugL42 gested by Atassi (1, 2) is used as an estimate, L43 and the number of antigenic determinants is proportional to molecular weight, large-T should have approximately 25 antigenic areas. The FIG. 3. Localization of the binding sites of the L seven antigenic areas suggested by the mapping series monoclonal antibodies. The shaded areas are the maximum segments of SV40 DNA that could code of antibody binding sites are the largest number for the antigenic deterninants recognized by the L allowed by the resolution of this approach. The series monoclonal antibodies. The fragment boundprecise number and identification of the amino aries are determined by the ability of a monoclonal acids that are involved in antibody must antibody to immunoprecipitate the proteins shown in await more detailed understandingbinding of the imFig. 2. Nucleotide numbering is by the SVsystem (29) munochemistry of large-T. We can confirm that and includes EcoRII fragment P. The sequences from SV40 large-T and small-t antigens share at least 2770 to 2693 of the SVGT14,BG DNA have been omitone antigenic determinant because three of the ted because translation terminates within the betaL series monoclonal antibodies precipitated both globin sequences upstream from the fragment. large-T and small-t. Surprisingly, we found that large-T had at least one antigenic determinant classes based on the presence or absence of an that was not found on either small-t or the antigenic determinant (Table 2). Ad2+D2 107K protein. If the current mapping Monoclonal antibodies against large-T are po- data are correct, small-t and the Ad2+D2 107K tential probes to search for similar determinants protein together contain all of the amino acid among host proteins. Lane and Hoeffler (17) residues of large-T. We would then suggest that have reported the presence of an antigenic de- these antibodies recognize sequences masked by terminant present on both large-T and a host the unique portions of small-t or 107K, that they 68K protein. Although we did not systematically recognize amino acids that flank the large-T assay all of the monoclonal antibodies in the L splice, or that they recognize a post-translational series for cross-reactivity to host proteins, we modification unique to large-T. Expansion of the noted that seven impunoprecipitated host pro- number of hybridomas and detailed characteriteins (Table 2) (manuscript in preparation). We zation of their monoclonal antibodies will make
51
0
R
qr
0
0
I
114
I_
-
t
MONOCLONAL ANTIBODIES FOR SV40 TUMOR ANTIGENS
VOL. 39, 1981
867
TABLE 2. Characterization of the L series monoclonal antibodies Characterization Monoclonal antibody
L2 L4 L5 L6 L7 L9 L10 L13 L14 L16 L18 L19 L20 L21 L23 L27 L30
L31 L32 L33 L35 L36 L37 L38
Ig class and subclas
IgGl IgGl IgGl IgGl IgGl IgGl IgG2a IgG2b IgG2b IgG2a IgG1 IgG2a IgGl IgG2a IgGl IgGl IgG3 IgG2a IgGl IgG2a IgG2a IgGl IgGl IgG2a
Ra.i.
noassay of aT
(denatured)
+ + + + + + + + + + +
Radioimmunoassay of ceS3K (dena-
Large-T subclass
tured)
Precipitates host Nuclear immuprotein
nofluorescence
NDa ND
+ +
+C
+
ND +d ND
+ + +
-
+ + + + + +
+
NTb
+e
+
+
+ + + + NT + + +
ND ND ND ND
+ + + + + + + + + + + + + +
-
-
+ +
+ L39 IgGl + L40 IgG2a + + L41 IgGl + L42 IgGl + + L43 IgGl + + L44 IgGl a ND, Not detected. b NT, Not tested. c Immunoprecipitates a 150K protein from PTK marsupial cells. d Immunoprecipitates a 150K protein from PTK marsupial cells. e Immunoprecipitates a 90K protein from mouse fibroblasts. f Immunoprecipitates a 35K protein from mammalian cells. ' Immunoprecipitates a 90K protein from mouse fibroblasts. h Immunoprecipitates an 80K protein from 3T3 and 3T6 mouse cells. Immunoprecipitates a 35K protein from mammalian cells.
it possible to determine the total antigenic structure of large-T. Monoclonal antibodies also should be important tools to identify unknown tumor antigens. Of initial interest is the SV40-coded polypeptide that serves as the tumor rejection antigen. The identification of the tumor rejection antigen has been difficult because of the large excess of nuclear large-T compared to any potential membrane tumor rejection antigen. If the tumor rejection antigen has any structural differences from the nuclear large-T, it may be possible to raise monoclonal antibodies that can be used to recognize this distinction. Conversely, monoclonal antibodies will be usable for identifying
ND +' ND +h ND ND ND ND ND ND ND +i ND ND ND ND ND ND
+ + + + + + + + +
determinants common to the tumor rejection antigen and nuclear large-T. Monoclonal antibodies will also be helpful in the identification and characterization of truncated and super-
large-T antigens often found in transformed cells (3, 4, 15, 25). These antigens are found in numerous SV40-transforned cell lines, but little is known about their structure or function.
Monoclonal antibodies as probes of antigen structure and function. When a lysate of SV40-transformed mouse cells is immunoprecipitated with antibodies from several members of the L series, only a fraction of the large-T present can be precipitated. This does not appear to be due simply to their having low affinity (E.
868
HARLOW ET AL.
Harlow, unpublished data), and it seems likely the determinant recognized by these monoclonal antibodies is either hidden or destroyed during extraction. Microheterogeneity of the large-T population is not unreasonable if one considers the multiple physical associations large-T is known to assume. SV40 large-T binds to SV40 DNA (11, 23), the host 53K protein (16, 19, 21), and itself. Any of these associations could mask an antigenic site. Likewise, allosteric changes associated with the binding of DNA, 53K protein, or large-T, as well as with enzymatic activities, could account for the microheterogeneity of the large-T population. An extension of this argument suggests that monoclonal antibodies could become useful probes to mark the binding sites for DNA, 53K, and large-T as well as for the enzymatic sites for the intrinsic ATPase activity (28). It should be noted that several of the L series monoclonal antibodies also immunoprecipitated host proteins. In several cases, these antibodies specifically precipitated different host proteins. At present we are unable to determine whether this binding to host proteins is fortuitous or whether these proteins share functional domains with large-T. However, we should note that all of the monoclonal antibodies that immunoprecipitated host proteins bound to the antigenic determinants that survived sodium dodecyl sul-
fate-polyacrylamide gel electrophoresis. Monoclonal antibodies in protein purification. The specificity of monoclonal antibodies for individual sites should be an added asset when immunoassays are used as screens in protein purification. Because purified immunoglobulins can easily be bound to solid substrates, radioimmunoassays and enzyme-linked immunoabsorbent assays will be useful techniques in SV40 tumor antigen purification. Also, several IgG subclasses will fix complement, and this will allow workers to avoid the confusing multiple activities present in tumor sera when assaying column fractions by complement fixation. Several workers have successfully used monoclonal antibodies for affinity columns. Of particular note is the isolation of active interferon, using an affinity column (24). Monoclonal antibodies specific for SV40 tumor antigens. Immortalizing the humoral response of mice to SV40-induced tumors by producing hybridomas alleviates many of the cumbersome aspects of immunochemical studies of the SV40 tumor antigens. Both as replacements for specific sera and as well-characterized probes for antigenic determinants, monoclonal antibodies have already proven their importance. Whether anti-large-T monoclonal anti-
J. VIROL.
bodies will be useful in studying the functions of large-T will depend on the number and location of antigenic determinants. Comprehensive studies of large-T structure and function, using monoclonal antibodies, will require the construction of hybridomas that secrete antibodies that bind to strategic antigenic determinants. One approach to making these hybridomas is to select the desired antibodies from a large family of hybridomas that secrete anti-large-T immunoglobulins. We have presented here a regimen to produce such a set of hybridomas and have characterized the initial 30 isolates. ACKNOWLEDGMENTS We thank P. Beverly for teaching us hybridoma technology, D. Lane for helpful discussions, K. Osborn and S. Neves for technical assistance, P. Southern and P. Berg for making virus stocks available before publication, and B. Kuypers and K. Leppard for allowing us to cite their unpublished results. E.H. is a fellow of the Lady Tata Memorial Trust. LITERATURE CITED 1. Atassi, M. Z. 1975. Antigenic structure of myoglobin: the complete immunochemical anatomy of a protein and conclusions relating to antigenic structures of proteins. Immunochemistry 21:423-438. 2. Atassi, M. Z. 1978. Precise determination of the entire antigenic structure of lysozyme: molecular features of protein antigenic structures and potential of 'surfacesimulation' synthesis-a powerful new concept for protein binding sites. Immunochemistry 15:909-936. 3. Chang, C., D. T. Simmons, M. A. Martin, and P. T. Mora. 1979. Identification and partial characterization of new antigens from simian virus 40-transformed mouse cells. J. Virol. 31:463-471. 4. Edwards, C. A., G. Khoury, and R. G. Martin. 1979. Phosphorylation of T-antigen and control of T-antigen expression in cells transformed by wild-type and tsA mutants of simian virus 40. J. Virol. 29:753-762. 5. Ey, P. L., S. J. Prowse, and C. R. Jenkin. 1978. Isolation of pure IgGl, IgG2a and IgG2b immunoglobulins from mouse serum using protein A-Sepharose. Immunochemistry 15:429-436. 6. Galfre, G., S. C. Howe, C. Milstein, G. W. Butcher, and J. C. Howard. 1977. Antibodies to major histocompatibility antigens produced by hybrid cell lines. Nature (London) 266:550-552. 7. Gurney, E. G., R. 0. Harrison, and J. Fenno. 1980. Monoclonal antibodies against simian virus 40 T antigens: evidence for distinct subclasses of large T antigen and for similarities among nonviral T antigens. J. Virol. 34:752-763. 8. Harlow, E., D. C. Pim, and L. V. Crawford. 1981. Complex of simian virus 40 large-T antigen and host 53,000-molecular-weight protein in monkey cells. J. Virol. 37:564-573. 9. Hassell, J. A., E. Lukanidin, G. Fey, and J. Sambrook. 1978. The structure and expression of two defective adenovirus 2/simian virus 40 hybrids. J. Mol. Biol. 120:209-247. 10. Hayry, P., and V. Defendi. 1970. Surface antigen(s) of SV40-transformed tumor cells. Virology 41:22-29. 11. Jessel, D., J. Hudson, T. Landau, D. Tenen, and D. M. Livingston. 1975. Interaction of partially purified simian virus 40 T antigen with circular viral DNA molecules. Proc. Natl. Acad. Sci. U.S.A. 72:1960-1964. 12. Kessler, S. W. 1975. Rapid isolation of antigens from cells
VOL. 39, 1981
13.
14. 15.
16. 17.
18.
19.
20.
21.
MONOCLONAL ANTIBODIES FOR SV40 TUMOR ANTIGENS
with a staphylococcal protein A-antibody absorbent: parameters of the interaction of antibody-antigen complexes with protein A. J. Immunol. 116:1617-1624. Khoury, G., J. Alwine, N. Goldman, P. Gruss, and G. Jay. 1980. New chimeric splice junction in adenovirus type 2-simian virus 40 hybrid viral mRNA. J. Virol. 36: 143-151. Kohler, G., and C. Milstein. 1975. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature (London) 256:495-497. Kress, M., E. May, R. Cassingena, and P. May. 1979. Simian virus 40-transformed cells express new species of proteins precipitable by anti-simian virus 40 tumor serum. J. Virol. 31:472-483. Lane, D. P., and L. V. Crawford. 1979. T antigen is bound to a host protein in SV40-transformed cells. Nature (London) 278:261-263. Lane, D. P., and W. K. Hoeffler. 1980. SV40 large T shares an antigenic determinant with a cellular protein of molecular weight 68,000. Nature (London) 288:167170. Lane, D. P., and A. K. Robbins. 1978. An immunochemical investigation of SV40 T antigens. 1. Production properties and specificity of a rabbit antibody to purified simian virus 40 large-T antigen. Virology 87:182193. Linzer, D. I. IL, and A. J. Levine. 1979. Characterization of a 54K dalton cellular SV40 tumor antigen present in SV40-transformed cells and uninfected embryonal carcinoma cells. Cell 17:43-52. Martinis, J., and C. M. Croce. 1978. Somatic cell hybrids producing antibodies specific for the tumor antigen of simian virus 40. Proc. Natl. Acad. Sci. U.S.A. 75:23202323. McCormick, F., and E. Harlow. 1980. Association of a murine 53,000-dalton phosphoprotein with simian virus 40 large-T antigen in transformed cells. J. Virol. 34:
869
213-224. 22. Pintel, D., N. Bouck, G. di Mayorca, B. Thimmappaya, B. Swerdlow, and T. Shenk. 1980. SV40 mutant tsA 1499 is heat-sensitive for lytic growth but generates cold-sensitive rat-cell transformants. Cold Spring Harbor Symp. Quant. Biol. 44:305-309. 23. Reed, S. I., J. Ferguson, R. W. Davis, and G. R. Stark. 1975. T antigen binds to simian virus 40 DNA at the origin of replication. Proc. Natl. Acad. Sci. U.S.A. 72: 1605-1609. 24. Secher, D. S., and D. C. Burke. 1980. A monoclonal antibody for large-scale purification ofhuman leukocyte interferon. Nature (London) 285:446-450. 25. Smith, A. E., R. Smith, and E. Paucha. 1979. Characterization of different tumor antigens present in cells transformed by simian virus 40. Cell 18:335-346. 26. Syvanen, J. M., Y. R. Yang, and M. W. Kirschner. 1973. Preparation of l"I-catalytic subunit of aspartate transcarbamylase and its use in studies of the regulatory subunit. J. Biol. Chem. 248:3762-3768. 27. Tevethia, S. S., D. C. Flyer, and R. TXian. 1980. Biology of simian virus 40 (SV40) transplantation antigen (TrAg). VI. Mechanism of induction of SV40 transplantation immunity in mice by purified SV40 T antigen (D2 protein). Virology 107:13-23. 28. Tjian, R., and A. Robbins. 1979. Enzymatic activities associated with a purified simian virus 40 T antigenrelated protein. Proc. Natl. Acad. Sci. U.S.A. 76:610614. 29. Tooze, J. (ed.). 1980. The molecular biology of tumor virus, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. 30. Van Heuverswyn, H., C. Cole, P. Berg, and W. Fiers. 1979. Nucleotide sequence analysis of two simian virus 40 mutants with deletions in the region coding for the carboxyl terminus of the T antigen. J. Virol. 30:936941.
