Induction of plasmacytomas secreting antigen-specific monoclonal ...

Proc. Nati. Acad. Sci. USA Vol. 88, pp. 8735-8739, October 1991

Immunology

Induction of plasmacytomas secreting antigen-specific monoclonal antibodies with a retrovirus expressing v-abl and c-myc EVA M. WEISSINGER*, HARALD MISCHAK*, DAVID A. LARGAESPADAt, DONAL A. KAEHLERt, THOMAS MITCHELLt, SANDRA J. SMITH-GI4+*, REX RISSERt*, AND J. FREDERIC MUSHINSKI* *Molecular Genetics Section, Laboratory of Genetics, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892; and tMcArdle Laboratory for Cancer Research, University of Wisconsin, Madison, WI 53706

Communicated by Michael Potter, June 24, 1991 (received for review May 15, 1,991)

ABSTRACT ABL-MYC, a recombinant murine retrovirus that expresses v-abl and c-myc, rapidly induces transplantable mono- or oligoclonal plasmacytomas in BALB/c mice. To determine if the targets for trnsformation of this retrovirus are antigen-committed B lymphocytes and to explore this system as an alternative technique for producing antigenspecific monoclonal antibodies, plasmacytomas were induced in mice that had beep immunized with two different types of immunogens, hen egg white Iysozyme and sheep red blood cells. The maority of these plasmacytomas secreted immunogenspecific antibodies. Plasmacytomas induced in unimmunized mice did not react with hen egg white lysozyme or sheep red blood cells. The specific antibodies were comparable in concentration, specificity, and affinity to monoclonal antibodies obtained with conventional hybridoma technology, but, in addition to IgGs and IgMs, they included specific IgA antibodies, which are rare among splenic-derived hybridomas. Our results demonstrate that a principal target for ABL-MYC is an antigen-committed B Iymphocyte. In addition this procedurem provides an alternative method for the production of monoclonal antibodies, without a requirement for heterocaryon formation by cell fusion techniques.

Plasmacytomas typically arise with a mean lhtency of 240 days in up to 60%o of mice from susceptible strains (e.g., BALB/cAnPt) given an i.p. injection of mineral-oil (pristane) (1). Pristane (tetramethylpentadecane)'leads to the development of oil granulomas in the mesenteries of the mice, and plasmacytomas arise in these tissues. The appearance of the tumors can be accelerated dramatically if Abelson murine leukemia virus (A-MuLV) 'is administered in addition to pristane (2). Infection of pristane-primed mice with A-MuLV leads to the development of plasmacytomas in 10-30% of the mice between 60' and 120 days. Molecular genetic studies of plasmacytomas have revealed that 190%o of all plasmacytomas induced by i.p. injection of mineral oil, with or without A-MuLV, show a deregulation of the c-myr protooncogene due to chromosomal alterations (3, 4). This universal deregulation of c-myc suggests that c-myc plays a major role in plasmacytomagenesis, although it appears to belonly one step in a multistep scenario. This multistep hypothesis was reinforced by the finding that plasma cell tumors were also generated rapidly by retroviral constructs created to study the cooperation of c-myc with different oncogenes. These oncogene combinations are Elw-c-myc (E. is the immunoglobulin heavy-chain enhancer) plus v-Ha-ras (5) and v-myc plus v-raf (6, 7). Both combinations induced plasmacytomas, but only if mice were pristane-primed, and neither increased the tumor incidence above 60% in adult BALB/c mice. Transgenic mice carrying Eg-v-abl develop plasmacytomas in concert with an activated c-myc gene, due to secondThe publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. 8735

ary chromosomal alterations. In addition, plasmacytomas occur rapidly in E-,v-abl x Ep-myc transgenic hybrid mice, without' pristane and in strains resistant to plasmacytoma

induction with pristane alone (8). Therefore we predicted that a retrovirus expressing these two oncogenes would rapidly and efficiently induce plasmacytomas. Thus ABL-MYC, a replicltion-defective retrovirus coexpressing v-abl and c-myc was constructed by inserting the protein-encoding portion of c-myc under the control of a herpes simplex thymidine kinase promoter into the A-MuLV backbone (9). This virus rapidly induced plasmacytomas in 100o of adult BALB/c mice with or without helper virus and even in the absence of pristane (9, 10). The average latency of tumor development was shorter than that observed with any other retroviral constructs. Since dividing cells are particularly good targets for retroviral integration and transformation (11), we reasoned that antigen-stimulated B cells might be particularly good targets for ABL-MYC. Therefore, BALB/c mice were immunized twice with either hen egg white lysozyme (HEL) or sheep red blood cells (SRBC) before ABL-MYC infection with the intention of generating antigen-specific plasmacytomas.

MATERIALS AND METHODS ABL-MYC Virus Stocks. Helper virus-containing ABLMYC [ABL-MYC(M-MuLV)] stocks were obtained from NIH 3T3 cells that had been cotransfected with the ABLMYC plasmid (pABL-MYC) and the Moloney murine leukemia virus (M-MuLV) Plasmid (pMOV-3) (9). Helper virusfree ABL-MYC [ABL-MYC(f2)] stocks were obtained by transfecting the 4k2 packaging line (12) with pABL-MYC. In both cases cells were maintained in Dulbecco's modified Eagle's medium (Applied Biosystems) supplemented with 10% fetal calf serum (FCS) and 2 mM glutamine at 370C. Culture supernatant was harvested after 24 hr, passed through a 0.22-I&m filter, and stored at -70'C until use. Titers of transforming ABL-MYC with and without helper virus, determined by NIH 3T3 transformation asrsays (13), were 4 X 105 and 4 x 106 focus-forming units per ml, respectively. The titer of replication-competent M-MuLV was determined by the UV-XC plaque assay (14) and was negative in the case of

ABL-MYC(4/2) stocks.

Animals. Adult (>60 days old) BALB/cAnPt mice maintained at Hazelton Laboratories (under National Cancer Institute contract no. NO1/CB/71085; Rockville, MD) were used in these studies. Abbreviations: HEL, hen egg white lysozyme; SRBC, sheep red blood cells; RBC, red blood cells; A-MuLV and M-MuLV, Abelson and Moloney murine leukemia virus, respectively; FCS, fetal calf serum; PBS, phosphate-buffered saline; BSA, bovine serum albumin; IL-6, interleukin 6; PCFIA, particle-coupled fluorescence immunoassays. tDeceased, September 27, 1990.

8736

Immunology: Weissinger et al.

Immunization Protocols. Two- to 3-month-old BALB/c mice were immunized i.p. either with 50 gg (per mouse) of HEL in complete Freund's adjuvant followed by a boost in incomplete Freund's adjuvant 14 days later (Table 1, groups la and lb) or with 5 x 108 SRBC in phosphate-buffered saline (PBS) on days 0 and 14 (Table 1, groups 2a, 2b, and 2c). Animals immunized with SRBC were injected with pristane 7 days after the initial immunization to provide the optimal environment for plasmacytoma growth, whereas those immunized with HEL and mineral oil-containing Freund's adjuvant did not receive additional pristane. Control groups were treated identically, but without immunogens in the immunizing mixture (Table 1, groups 3a, 3b, and 3c). All mice were infected with 105 focus-forming units of ABL-MYC(qk2) or ABL-MYC(M-MuLV) on day 16 or 18 after the first immunization. On day 28, mice in groups la, lb, 2b, and 2c received an additional boost with either 50 ,ug of HEL or 5 x 101 SRBC in saline, as appropriate. Mice in group 3c received 5 x 108 SRBC as a primary immunization 10 days after the virus. Immnunoassays. Automated particle-coupled fluorescence immunoassays (PCFIA), described by Jolley et al. (15), were performed using a "screen machine" (Baxter/Pandex Division, Mundelein, IL). Polystyrene beads (0.9 um; Baxter/ Pandex) were coated with antigens either by direct absorption or by using 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide coupling to carboxylated beads as described in detail by Rousseau et al. (16). Coated beads were resuspended in PBS with 0.1% sodium azide. Antibody isotypes were determined on 1:104 dilutions of ascites fluid or serum samples from individual mice in an antigen-independent PCFIA as described (16). Fluoresceinated and peroxidase-labeled anti-a, anti-,u, anti-y, anti-K, and anti-A antibodies were obtained from Southern Biotechnology Associates (Birmingham, AL). Peroxidase reagents were purchased from Kirkegaard and Perry Laboratories (Gaithersburg, MD). For determination of antigen specificity, all ascites samples were diluted to 1 ,ug of immunoglobulin per ml and incubated with polystyrene beads coated with a panel of control antigens: HEL, human lysozyme, mouse serum albumin, bovine serum albumin (BSA), lens crystallin, phosphorylcholine, Staphylococcus nuclease, mouse IgG1, inulin, and galactan. Protein antigens were purchased from Sigma or obtained as described elsewhere (16). Ascites samples containing antibodies of different heavy or light chain classes were incubated individually with antibodies specific to their immunoglobulin subclasses. This procedure allowed us to distinguish binding patterns of these antibodies. In addition to these protein antigens, the ascites and serum samples were also screened against red blood cells (RBC) of several species in addition to SRBC. For SRBC specificity, serum and ascites samples were screened in an ELISA using plates coated with 1% sonicated RBC in PBS. In addition, direct and indirect hemolytic plaque assays were performed with cells from the ascites fluid and, as a control, with cells from a peritoneal lavage from immunized mice. Affinity Analysis and Antigen Inhibitio Assays. Anti-HEL IgG antibodies were partially purified by two ammonium sulfate precipitations at 4°C (final concentration of ammonium sulfate = 40%o). Fifty nanograms of purified plasmacytoma antibodies were incubated overnight with various concentrations of HEL (inhibitor). The percentage of free antibody was determined by incubating the antigen-antibody mixture for 2 min with HEL bound to polystyrene beads and subsequent screening for bound antibody by PCFIA as described (17). Association constants (K.) were calculated by using Scatchard plot analysis modified for ELISA as described (18, 19) and as modified for PCFIA by Lavoie et al. (17). In Vitro Culture of Plasmacytomas. Plasma cells of tumors were cultured in vitro by mincing the tumors in a collagenase/

Proc. Natl. Acad Sci. USA 88 (1991)

dispase mixture (3:1, vol/vol) and incubating this mixture at 370C for 1 or 2 hr. After washing the cells twice in FCScontaining medium, the recovered mixed cell population was maintained in RPMI 1640 supplemented with 15% FCS, 4 mM glutamine, 50 ,uM 2-mercaptoethanol, and 2000 units of interleukin 6 (IL-6) per ml. IL-6 was obtained from a-ChyJ2, a macrophage cell line established in our laboratory (20). 3_S-Labelig of Antibodies and Immunoprecipitation. Plasmacytoma cells (106) from ascites fluid or tissue culture were cultivated in 5 ml of methionine-free RPMI 1640 supplemented with 1%o FCS, 4 mM glutamine, 50 ;LM 2-mercaptoethanol, and 2000 units of IL-6 per ml. Cells were labeled for 4 hr with 0.1 mCi (1 Ci = 37 GBq) of [35S]methionine, and the supernatants were collected. One hundred microliters of supernatant was incubated overnight with 20 ul of HELcoated polystyrene beads or with 20 1ul of BSA-coated polystyrene beads as a negative control. After extensive washing, the proteins were released from the beads by boiling the beads in sample buffer (21). Twenty microliters of the boiled samples and 1 ,ul of untreated supernatant were electrophoresed on a SDS/109/ PAGE gel. The gel was dried and exposed to x-ray film. Isolation and Hybridization of RNA and DNA. Poly(A)+ RNA and genomic DNA were isolated as described (22), and Southern and Northern blots were hybridized as described (23). Both Northern and Southern blots were probed with a 2.0-kilobase (kb) Sac 1-HindIII v-abl fragment (24) and a 1.6-kb HindIll fragment of c-myc (25). Southern blots were additionally probed with a 4.5-kb HindIII-BamHI fragment of pK, a genomic subclone of the mouse K constant region gene (CK; ref. 26). Autoradiography was performed with Kodak x-ray film for 1-2 days at -70°C using intensifying screens.

RESULTS All immunized mice developed plasmacytomas with a mean latency of 28 ± 7 days after receiving virus. Control mice, treated with the same regimen of adjuvants and virus but without immunogens, also developed plasmacytomas, but with a longer mean latency (38 ± 7 days) (Fig. 1). To ensure that the antibodies being studied were products of the plasmacytoma cells, 10i-106 washed ascites cells were transplanted into pristane-primed, unimmunized, uninfected, syngenic mice. Ascites tumors developed in each recipient after 10-20 days. All tumors secreted high amounts of immunoglobulin, producing between 5 and 30 mg per ml of ascites fluid. Only antibodies from transplant recipients were 1080 I.co E 600

4020

o

0 0 20

40 30 Days after virus

50

FIG. 1. Comparison of appearance of plasmacytomas. Latent periods of plasmacytomas developing in mice after immunization with HEL in adjuvant (o) and in mice that had received only adjuvant

and the

same

dose of ABL-MYC

(*) are indicated.

Immunology: Weissinger et al. used in our studies for determination of antigen specificity and affinity. Of 13 mice immunized with HEL and infected with ABLMYC(M-MuLV), 11 produced antibodies strongly reactive with the antigen. These plasmacytomas continued to produce reactive antibodies following transplantation into unimmunized, pristane-primed mice. Plasmacytomas were considered positive if the ascites fluid of the transplant recipients showed reactivity (fluorescence intensity at least 3 times the background) with the antigen at dilutions of 1:104 or greater. The HEL-reactive antibodies secreted by the tumors included 4 IgAs, 5 IgMs, and 2 IgGs. To evaluate the specificity of these antibodies, the ascites fluids were screened against a panel of antigens developed for antibody repertoire studies (16). Ascites samples reacting with two or more of these antigens were considered to be multireactive, with the exception of antibodies reacting with HEL and human lysozyme, which were considered to be specific for HEL and crossreacting with human lysozyme. As shown in Table 1, six antibodies (one IgA, two IgG, and three IgM) showed reactivity with HEL alone, and five (two IgM and three IgA) reacted with at least two antigens and were classified as multireactive (16). None of 12 tumors from unimmunized mice (Table 1, groups 3a and 3b) or of those specific for SRBC (Table 1, group 2) secreted antibodies specific for HEL. The IgG antibodies of two of the ABL-MYC-induced plasmacytomas, AMHEL-8 and AMHEL-11, were purified as described above for more detailed characterization. Both AMHEL-8 and AMHEL-11 were monospecific with respect to the antigen-screening panel. Their binding to HEL was strong and HEL-inhibitable (Fig. 2). The binding curves of AMHEL-8 and AMHEL-11 were compared to those of two hybridoma-derived IgG1 K monoclonal antibodies (HyHEL 12 and HyHEL 501) in Fig. 2. Ka values for AMHEL-8 and AMHEL-11 were 2 x 109 and 1.3 x 109 liters/mol, respectively. These affinities are within the range (5 x 107-2 x 1010 liters/mol) of those determined by the same method for HEL-specific monoclonal antibodies generated by hybridoma technology (17, 27). To clone oligoclonal tumors to monoclonality, some of the plasmacytomas were adapted to growth in vitro in medium supplemented with 2000 units of IL-6 per mi. After a period of 1-2 months on an autologous feeder layer obtained from tumor nodules, the cells grew without a feeder layer, but remained strictly IL-6-dependent. After adaptation to in vitro culture, the ABL-MYC-induced plasmacytomas could be cloned by limiting dilution. The amount of antibody produced by the in vitro cloned plasma cell lines was similar to that obtained from conventional hybridomas (1-10 ,ug/ml). Table 1. Reactivity of transplanted antigen-specific plasmacytomas NonHelper Anti- Anti- MultiGroup n Antigen virus HEL* SRBC* reactivet reactivet + 5 HEL 2 0 2 1 la + lb 8 HEL 4 0 1 3 0 2a 16 SRBC 8 4 4 0 5 SRBC 2b 3 1 1 + 5 SRBC 2c 0 3 1 1 + 0 3a 8 None 0 1 7 0 3b 4 None 0 2 2 + 0 0 3c 6 SRBC* 5 1 This table summarizes the reactivity and specificity of all antibodies produced by transplanted ABL-MYC-induced plasmacytomas. Mice in groups 1, 2, and 3 were immunized and infected following different protocols as described in Materials and Methods. *Reacts with HEL or SRBC only. tReacts with at least two antigens in the panel. *Immunized only after infection with ABL-MYC.

Proc. Natl. Acad. Sci. USA 88 (1991)

8737

100

a

0 m

75

z

w w IL II-

0

50 25 0

10

100

1000

INHIBITOR CONCENTRATION (ng/ml) FIG. 2. Inhibition of antibody binding by antigen. Purified antiHEL antibody at a concentration of 50 ng/ml from two ABL-MYCinduced plasmacytomas, AH8 (-v-) and AH11 (- -0- -), was incubated overnight with different concentrations of HEL (inhibitor) as described above. The curve of AH8 was also compared (Inset) to two high-affinity IgG1 K anti-HEL monoclonal antibodies from the hybridomas HyHEL 12 (HH12; ref. 17) and HyHEL 501 (HH501; M. A. Newman, C. R. Mainhart, C. P. Mallett, and S.J.S.-G., unpublished data).

AMHEL-2, one of the oligoclonal tumors cloned in vitro, secreted a monospecific IgM antibody and a multireactive IgA. After initial cloning in vitro, about two-thirds of the clones secreted only HEL-specific IgM antibodies. Some of these were cloned two more times by limiting dilution and then grown up for further analysis. Southern blot analysis was performed to verify the clonality of the tumor and the lines in tissue culture. Fig. 3 shows a Southern blot of DNA isolated from the original tumor (AMH2 gO) and the first transplant generation (AMH2 gi) as well as from individual clones selected for production of the monospecific IgM antibody. Whereas the original plasmacytoma and the first transplant contain at least three K lightchain rearrangements, each individual clone shows only one K light-chain rearrangement. To demonstrate directly that the HEL-specific antibodies were secreted by the ABL-MYC plasmacytomas, cells from ascites fluid and tissue culture were metabolically labeled Q z CV

I

T

germ-

C:

a

c)

z

z

Cs

l-E

ZJ czS -27-

I~z-

IT

V

CA

CSZi "r

>

line

FIG. 3. Southern blot analysis of in vitro-cloned tumors. Genomic DNA was isolated from tumor tissue (AMH2 gO and AMH2 gi) and from tissue culture clones obtained by cloning AMH2 gi in vitro (AMH2 H9, AMH2 H10, AMH2 G11, and AMH2 Dli). Ten micrograms of DNA was digested with BamHI, electrophoresed through a 0.8% agarose gel, blotted onto nitrocellulose, and probed with a 4.5-kb HindIII-BamHI fragment of pK, a subclone of mouse CK (25). The germ-line band is indicated and is at about 23 kb. AMH2 H9, AMH2 H10, AMH2 Gil, and AMH2 D11 represent individual clones selected for production of the monospecific IgM antibody.

Immunology: Weissinger et al.

8738

Proc. Natl. Acad. Sci. USA 88 (1991)

with [355]methionine, and the total proteins in the supernatants were resolved by SDS/PAGE (Fig. 4A). The major labeled proteins secreted were of appropriate size for the expected classes of immunoglobulin heavy (a = 55 kDa, 1L = 80 kDa, and y = 48 kDa) and light chains (23 kDa). The antibodies could be precipitated with HEL-coated polystyrene beads (Fig. 4B), but did not show any cross-reactivity with BSA-coated polystyrene beads (Fig. 4C). Binding to HEL beads was specific, since antibodies specific for antigens other than HEL did not bind to the HEL-coated beads (Fig. 4D). Similar experiments were performed with the particulate antigen SRBC. Immunized mice were infected with ABLMYC in the presence or' absence of replication-competent helper virus (M-MuLV) to assess the effect of viral spread on antigen-specific tumor formation. ABL-MYC(qk2) can be used to generate plasmacytomas that will not produce additional ABL-MYC. Preventing the replication of ABL-MYC in vivo would limit the number of target cells that could become'infected in immunized animals. Thus the use of ABL-MYC(qk2) should allow us to more firmly distinguish target cell populations and to determine whether a cell is transformed before or after antigen commitment. From 26 mice immunized with SRBC, 21 -were infected with ABL-MYC(qi2) (Table 1, groups 2a and 2b), and 5 (Table 1, group 2c) received ABL-MYC(M-MuLV). All 26 mice developed plasmacytomas. Antibodies from the ascites fluid of transplant recipients were tested for SRBC reactivity and immunoglobulin isotype by ELISA using plates coated with 1% sonicated SRBC in PBS. Among 20 tumors found to produce SRBC-reactive antibodies, 14 (5 IgM, 6 IgG, and 3 IgA) reacted only with SRBC. Six (a11 IgM) reacted with SRBC and RBC antigens from other species and/or with protein antigens in the PCFIA screening panel and were classified as multireactive (16). In addition, ascites cells were also 'screened for SRBC reactivity by direct and indirect plaque-forming assays. Cells from 12 of 26 mice formed at least 1000-fold more direct or indirect hemolytic plaques than those found in the peritonea of immunized but uninfected animals. Six tumors produced antibodies that did not react with SRBC (Table 1, group 2). Unimmunized but infected animals (Table 1, groups 3a and 3b) and animals that received immunization after virus infection (Table 1, group 3c) or antibodies secreted from tumors specific for HEL (Table 1, group 1) did not show any specific reactivity with SRBC. Northern blot analysis of the plasmacytomas showed expression of both v-pbl and exogenous c-myc in the expected sizes. Probing of the blot with myc exon 1 revealed that the endogenous c-myc gene is barely expressed in these tumors (data not shown).

,.

96

-

84

-

58

-

48

-

V.

e

DISCUSSION Our results suggest that the major target for ABL-MYC transformation is. an antigen-committed B cell. Although B-cell clones that react specifically with a single antigen constitute a minor subset of the total B lymphocyte population, even after immunization, B cells reacting with the immunizing agent will be preferentially mitotically activated when a boost'immunization is administered (28). Since only dividing cells are capable of stably integrating retroviruses as proviruses (29), antigen-committed B cells in repeatedly immunized animals would be much more likely to be transformed by ABL-MYC. It could be argued that antigenspecific plasmacytomas develop by the selective outgrowth of pre-B cells. If this was the case, a very large number of immature B cells would have to become infected to account for the high frequency of antigen-specific plasmacytomas obtained in our experiments. Alternatively, if 1 cells become targets for transformation with ABL-MYC after commhitment to antigen specificity, then immunization will cadse them to be clonally expanded and to become subject 'to infection. In our experiments both ABL-MYC(M-MuLV) and ABL-MYC(Q2) were capable of inducing antigenspecific plasmacytonas with the same frequency. Thus the main target for ABL-MYC is a mature, antigen-committed B cell. The selective infection of activated, antigen-committed B cells generated after a boost immunization is administered may explain the highly efficient induction of antigen-specific plasmacytomas by ABL-MYC.

C

B

A

Southern blot analysis of DNA from 10 of the SRBCspecific plasmacytomas induced by ABL-MYC(qk2) revealed that 5 were monoclonal, 2 were biclonal, and 3 contained four or five different clones as judged by the numbers of immunoglobulin gene rearrangements and viral integration sites. Stability of monoclonal tumors during transplantation for at least eight generations was proven by the identity of immunoglobulin gene rearrangements and viral integration sites in the transplants. Four examples' of monoclonal anti-SRBC plasmacytomas are shown in Fig. 5. Note that monoclonal tumors (e.g., AMSRBC 4H) with only one immunoglobulin K rearrangement can contain more than one AIBL-MYC provirus. Analysis of the anti-HEL tumors induced with ABLMYC(M-MuLV) revealed a higher degree of oligoclonality. From the seven HEL-specific tumors, two were monoclonal, three were biclonal, and two showed more than five rearrangements (data not shown). Data shown in Fig. 3 indicate that polyclonal tumors grow in vitro and can be cloned to monoclonality by limiting dilution.

~ ~ '4

,I,,

VI'

VI'

D tp. >' z . #X X t X 4

a

36-

26 Ep

Culture supernatant

4

.-. ..u.

HEL beads

BSA beads

HEL beads

FIG. 4. Precipitation of antibodies with antigen. AMtI1 (IgA), AMH2 (IgM), and AMH11 (IgG) are [35Slmethionine-labeled antibodies from three anti-HEL plasmacytomas, while ,HH59 (IgM) and HH502 (IG)are monoclonal antibodies against HEL from conventional hybridomas. AMF1, AMF5, and AMY1 are the products of ABLMYC-induced plasmacytomas from unimmunized mice. (A) Crude supernatant (1 Al per lane). (B) Labeled antibody eluted from HEL-coated beads. (C) Eluates from BSA-coated beads. (D) Labeled antibody from control tumors that were eluted from HELcoated beads. Positions of prestained protein standards (in kDa) are indicated on the left.

Immunology: Weissinger et al. _e>

-

Proc. Natl. Acad. Sci. USA 88 (1991)

8739

munoglobulin comparable to those of splenic-derived hybridomas. To date, all ABL-MYC-induced plasmacytomas have been very stable in vivo and in vitro, and they can be readily

N,>

A c>-ahl t.

B

transplanted into pristane-primed mice. ABL-MYC offers another potential advantage over hybridomas for molecular genetic studies of immunoglobulin genes; namely, no fusions are required for immortalization ofcells and, therefore, no heterocaryons are produced by this technique. ABL-MYC induces plasmacytomas in all mouse strains infected so far, even in those resistant to plasmacytoma induction by pristane alone (D.A.L., H.M., E.M.W., M. Potter, J.F.M., and R.R., unpublished results). Thus ABLMYC is a very potent agent for mouse plasmacytomagenesis, and an amphotropic variant of this ecotropic virus might be useful for generating antigen-specific plasmacytomas in species other than mice, like rat, rabbit, or even man. We thank JoAnne Goodnight for expert technical assistance and Drs. Fred Finkelman, Walter Kolch, Herbert C. Morse III, Michael Potter, Bill Sugden, and Howard Temin for critically reading the manuscript.

line

FIG. 5. Southern blot of ABL-MYC plasmacytomas. Genomic DNA (10 ,ug per lane) from normal BALB/c spleen and four SRBC-specific monoclonal plasmacytomas and transplants of those (g2 and g8 = transplant generation 2 and 8) was digested with EcoRI (A) or BamHI (B), electrophoresed through 0.8% agarose gels, blotted onto nitrocellulose, and hybridized with a 2.0-kb Sac I-HindIll v-abl fragment (24) (A) or a 4.5-kb HindIII-BamHI fragment from pK, a genomic subclone of mouse CK (26) (B). The positions of DNA size markers (in kb) as well as the c-abl and the germ-line CK fragment are indicated.

A comparison of latencies in the presence (28 7 days) or the absence (38 ± 7 days) of antigen shows differences (Fig. 1). These findings support the theory of preferential expansion of antigen-committed B-cell pools and suggest that antigenic challenge accelerates the induction of plasmacytomas in this system by augmenting the pool of susceptible proliferating B cells. More detailed analysis of two anti-HEL plasmacytomas established that antibodies obtained by ABL-MYC are comparable in affinity and specificity to those obtained by conventional techniques. Both were IgG1 K, the isotype found in more than 90% of anti-HEL hybridomas (16, 30). This finding suggests that they originated from a similar anti-HEL B-cell pool as those obtained by conventional fusion techniques. Preliminary data obtained from fine-specificity analysis for HEL binding revealed that one of the IgGs showed a very rare fine specificity for HEL, suggesting that ABL-MYC can also target other B-cell pools. This notion is supported by the finding that several of the ABL-MYC tumors secreted HELspecific IgM and IgA antibodies, which are heavy-chain classes rarely found in BALB/c anti-HEL hybridomas (16, 30). Similarly three of the antigen-specific anti-SRBC antibodies were IgA antibodies, suggesting that IgA antibodies are more likely to be obtained by using ABL-MYC transformation than by conventional spleen fusion techniques. Thus ABL-MYC produces a different spectrum of monoclonal antibodies than spleen fusions. Since about 60% of the antigen-primed mice developed antigen-specific plasmacytomas, the use of this virus may prove to be a valuable alternative to the conventional hybridoma technique for production of monoclonal antibodies. When oligoclonal tumors arise, they can be cultured in vitro and cloned to monoclonality, as proven by Southern blot analysis in Fig. 3. The plasmacytomas cultured in vitro secrete amounts of im±

1. Potter, M., Wiener, F. & Mushinski, J. F. (1984) Adv. Viral Oncol. 4, 139-162. 2. Potter, M., Sklar, M. D. & Rowe, M. D. (1973) Science 182, 592-594. 3. Mushinski, J. F. (1988) in Cellular Oncogen Activation, ed. Klein, G. (Dekker, New York), pp. 181-222. 4. Ohno, S., Migita, S., Wiener, F., Babonits, M., Mushinski, J. F. & Potter, M. (1984) J. Exp. Med. 159, 1762-1777. 5. Clynes, R., Wax, J., Stanton, L. W., Smith-Gill, S., Potter, M. & Marcu, K. B. (1988) Proc. Natl. Acad. Sci. USA 85, 6067-6071. 6. Troppmair, J., Potter, M., Wax, J. S. & Rapp, U. R. (1989) Proc. Natl. Acad. Sci. USA 86, 9941-9945. 7. Kurie, J. M., Morse, H. C., III, Principato, M. A., Wax, J. S., Troppmair, J., Rapp, U. R., Potter, M. & Mushinski, J. F. (1990) Oncogene 5, 577-582. 8. Rosenbaum, H., Harris, A., Bath, M. L., McNeall, J., Webb, E., Adams, J. M. & Cory, S. (1990) EMBO J. 9, 897-905. 9. Largaespada, D., Kaehler, D. A., Weissinger, E. M., Mischak, H., Mushinski, J. F. & Risser, R. (1990) Curr. Top. Microbiol. Immunol. 166, 91-98. 10. Weissinger, E. M., Largaespada, D., Smith-Gill, S. J., Risser, R., Mushinski, J. F. & Mischak, H. (1990) Curr. Top. Microbiol. Immunol. 166, 121-127. 11. Breindl, M., Bacheler, L., Fan, H. & Jaenisch, R. (1980) J. Virol. 34, 376-384. 12. Mann, R., Mulligan, R. C. & Baltimore, D. (1983) Cell 33, 153-165. 13. Scher, C. D. & Siegler, R. (1975) Nature (London) 253, 429-431. 14. Rowe, W. P., Pugh, W. E. & Hartley, J. W. (1970) Virology 42,1136-1139. 15. Jolley, M. E., Wang, C. H., Eckenberg, S. J., Zuelke, M. S. & Kelso, D. M. (1984) J. Immunol. Methods 67, 21-25. 16. Rousseau, P. G., Mallet, C. P. & Smith-Gill, S. J. (1990) Mol. Immunol. 26, 993-1006. 17. Lavoie, T. B., Kam-Morgan, L. M. W., Mallet, C. P., Schiling, J. W., Prager, E. M., Wilson, A. C. & Smith-Giml, S. J. (1990) in Use of X-Ray Crystallography in the Design ofAntiviral Agents, eds. Laver, W. G. & Air, G. A. (Academic, New York), pp. 213-232. 18. Figuret, B., Chaffotte, A. F., Djavadi-Ohaniance, L. & Goldberg, M. E. (1985) J. Immunol. Methods 77, 305-311. 19. Stevens, F. (1987) J. Mol. Immunol. 24, 1055-1060. 20. Mischak, H., Kolch, W., Hofer, F., Weissinger, E., Gessl, A., Davidson, W. F., Aiello, F. B., Blaas, D. & Rapp, U. R. (1990) Oncogene 5, 13771382. 21. Laemmli, U. K. (1970) Nature (London) 227, 680-683. 22. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989) in Molecular Cloning: A Laboratory Manual, ed. Nolan, C. (Cold Spring Harbor Lab., Cold Spring Harbor, NY), 2nd Ed. 23. Mushinski, J. F., Bauer, S. R., Potter, M. & Reddy, E. P. (1983) Proc. Natd. Acad. Sci. USA 80, 1073-1079. 24. Goff, S., Gilboa, E., Witte, 0. & Baltimore, D. (1980) Cell 22, 777-785. 25. Stanton, L. W., Watt, R. & Marcu, K. B. (1983) Nature (London) 303, 401-406. 26. Max, E. E., Seidman, J. G. & Leder, P. (1979) Proc. Natd. Acad. Sci. USA 76, 3450-3454. 27. Harper, M., Lema, F., Boulot, G. & Poljak, R. J. (1987) Mol. Immunol. 24, 97-108. 28. Rolink, A. G., Radaskiewicz, T. & Melchers, F. (1987) J. Exp. Med. 165, 1673-1678. 29. Miller, D. G., Adam, M. A. & Miller, A. D. (1990) Mol. Cell. Biol. 10, 4239-4248. 30. Metzger, D. W., Ching, L. K., Miller, A. & Secarz, E. E. (1984) Eur. J. Immunol. 14, 87-95.