Reaction patterns of monoclonal antibodies to DNA

Starting with spleen cells from MRL/lpr, NZB/W, and graft-vs-host-diseased mice, we prepared a total of 57 hybridomas that produce antibodies to DNA. Using various approaches, we studied the avidity of these monoclonals in relation to their behavior in four anti-DNA assays. From the results obtained, we postulate that on the basis of anti-DNA avidity the anti-DNA ELISA, the polyethylene glycol assay, the indirect immunofluorescence test on Crithidia luciliae, and the Farr assay (in this order) detect a decreasing amount of anti-dsDNA, the Farr assay being strictly selective for high avidity anti-dsDNA. mAb selected by the anti-DNA ELISA generally were of a low avidity toward DNA. Using cardiolipin and dextran sulfate, a polyanion that bears a resemblance in charge to DNA, we studied the cross-reactivity of the monoclonals. A total of 6 of the 57 monoclonals were found to cross-react with cardiolipin, and 26 with dextran sulfate. We observed an inverse relationship between anti-DNA avidity and cross-reactivity: the lower the avidity of the antibody, the more cross-reactive it is. Based on these findings, we postulate that it is at least questionable whether low avidity, cross-reactive (monoclonal) anti-DNA is representative for the anti-DNA found in patients with SLE.     

Antibody-nucleic acid complexes. Conformational and base specificities associated with spontaneously occurring poly- and monoclonal anti-DNA antibodies from autoimmune mice

An enzyme-linked immunosorbent assay (ELISA) was developed to characterize spontaneously occurring, mono-and polyclonal anti-DNA antibodies. The assay consists of adsorbing single- (ss) and double-stranded (ds) DNA and various nucleoside-bovine serum albumin conjugates (e.g., A-, G-BSA, etc.) to microtiter wells and assesses the ability of various antibodies to bind to these immobilized antigens. The conformational and base specificity of two monoclonal antibodies (designated MRss-1 BWds-3) was examined in this manner. The exclusive binding of MRss-1 to ssDNA and guanosine-BSA (G-BSA) confirms our previous findings [Munns, T.W., Liszewski, M.K., Tellam, J.T., Ebling, F. M., & Hahn, B.H. (1982) Biochemistry 21,2929-2936] that this antibody recognizes single-stranded nucleic acids by virtue of their guanine content. The extensive binding of BWds-3 to dsDNA, its limited binding to ssDNA, and complete absence of binding to nucleoside-BSA antigens implied a double-stranded conformational specificity. Further, competitive studies with naturally occurring and synthetic alternating copolymers indicated that BWds-3 preferentially recognized the native dsDNA antigens. ELISA analysis of the spontaneously occurring, polyclonal anti-DNA antibodies from MRL/lpr and NZB/NZW-F1 mice revealed that the majority of anti-ssDNA antibodies bound to nucleoside-BSA conjugates. Anti-G antibodies were most prominent in both strains of mice, yet lesser and more variable quantities of anti-A, -C, -U, and -T antibodies were also detected. Preadsorption of serum with G-BSA/Sepharose resulted in the complete removal of anti-G antibodies and a 60% reduction in anti-ssDNA antibodies.(ABSTRACT TRUNCATED AT 250 WORDS)     

Polyspecific binding of Escherichia coli beta-galactosidase by murine antibodies to DNA

To characterize further polyspecific interactions of antibodies to DNA, the binding of sera from autoimmune MRL-lpr/lpr mice to Escherichia coli beta-galactosidase (beta-gal) was analyzed. This protein was selected for study because of preliminary observations that sera from autoimmune mice bound unexpectedly to cloned fusion protein constructions containing beta-gal. Using ELISA assays, sera from MRL-lpr/lpr mice demonstrated high levels of antibodies to both DNA and beta-gal, in titers significantly greater than those of BALB/c controls. Affinity chromatography using beta-gal-Sepharose demonstrated that antibodies enriched for anti-beta-gal activity bound both DNA as well as beta-gal, indicating the presence of a population of cross-reactive anti-DNA antibodies. Furthermore, anti-DNA mAb of MRL-lpr/lpr strain origin also bound beta-gal by ELISA, although these levels were lower than those to DNA. Together, these results extend the range of polyspecific binding of murine anti-DNA antibodies to bacterial proteins. They further suggest caution in the interpretation of immunoassays using fusion protein constructions containing beta-gal, especially with sera from autoimmune mice.     

Murine monoclonal anti-DNA autoantibodies bind to endogenous bacteria

Several bacterial species (including Streptococcus faecalis, Bacillus cereus, Staphylococcus aureus, and Escherichia coli) were tested for their ability to react with monoclonal anti-DNA antibodies that were derived from MRL-lpr/lpr mice. S. faecalis reacted with 8/15 of such antibodies. The binding was unaffected by DNase, but it was competitively inhibited by DNA. F(ab')2 fragments of the monoclonal antibodies reacted with the bacteria, but Fc fragments did not. Phospholipids extracted from the bacterial cells were able to bind to three representative anti-DNA antibodies that also bound to whole bacteria. The results suggest that bacterial phospholipids might provide an immunogenic stimulus for the production of antibodies that cross-react with DNA. We propose that some anti-DNA auto-antibodies and anti-bacterial antibodies evolve from a restricted group of antibodies with high avidity for the phosphodiester groups that occur in DNA and bacterial cells walls.     

High levels of anti-mitochondrial antibodies in MRL mice. Influence of anti-DNA antibodies on anti-mitochondrial antibodies measured by an enzyme-linked immunosorbent assay

We developed an easy enzyme-linked immunosorbent assay (ELISA) for anti-mitochondrial antibodies, and detected high levels of anti-mitochondrial antibodies in MRL/Mp-lpr/lpr (MRL1/l) mice. The influence of the presence of anti-DNA antibodies in the tested sera on this assay was also evaluated. The monoclonal anti-DNA antibodies, which were made from non-immunized MRL1/l mice, did not react with the mitochondrial antigens adhering to polystyrene plates. Absorption with DNA had little effect on the levels of mitochondrial antibodies in MRL1/l sera. Our assay for mitochondria-binding antibody was very easy and sensitive, although it could not differentiate among heterogeneous anti-mitochondrial antibodies.     

VH gene analysis of spontaneously activated B cells in adult MRL-lpr/lpr mice. J558 bias is not limited to classic lupus autoantibodies

To determine the genetic origins of lupus auto-antibodies, we analyzed the relationship between VH gene usage and auto-Ag-binding properties of 352 B cell hybridomas derived from MRL-lpr/lpr mice. The hybridomas were derived from neonatal, 1-month-old, 3-month-old, and 6-month-old mice. The experimental strategy provided that the hybridomas were monoclonal at initial evaluation, so the Ag binding and V gene frequencies of the entire population could be determined. Initially, 1032 Ig-producing hybridomas were evaluated for binding to six Ag; VH gene family use was determined in 119 anti-DNA and anti-rabbit thymus extract (RTE) antibodies (autoantibodies) and in 233 age-matched Ig that did not bind to any of the six Ag (nonbinders). Neonatal B cells, including cross-reactive IgM autoantibodies and nonbinder IgM, used relatively 3' VH genes. The majority of B cells in adult mice used VH genes of the J558 family. Although J558 use was significantly higher among the autoantibodies (anti-DNA and anti-RTE) than among the nonbinder Ig, this difference was due to a higher frequency of J558 use by 1-month-old mice. At 3 months, J558 use by the nonbinder Ig increased to the same frequency of J558 use as in the autoantibody population. J558 use in both groups of antibodies exceeded a previously reported estimation of J558 expression in the functional B cell repertoire of young adult MRL-lpr/lpr mice. Several subgroups of antibodies that share properties with pathogenic Ig, including IgG, cross-reactive Ig, and anti-dsDNA autoantibodies, demonstrated a marked preferential expression of the J558 family. These results suggest that there is an age-related bias in the activation of B cells using J558 VH genes in MRL-lpr/lpr mice that is under the influence of a selective force distinct from, or in addition to, an ssDNA or RTE auto-Ag-driven response.     

Monoclonal antibodies to DNA and RNA from NZB/NZW F1 mice: antigenic specificities and NH2 terminal amino acid sequences

Two anti-DNA hybridoma autoantibodies ( A52 , D42 ) were prepared by fusing spleen cells from unimmunized NZB/NZW F1 female mice with BALB/c myeloma cells. The monoclonal antibodies were purified to homogeneity and were analyzed for their antigen-binding specificities. The two anti-DNA antibodies bound single-stranded, double-stranded, and supercoiled DNA, with a marked preference for the single-stranded conformation. Competition experiments performed with synthetic polynucleotides, as well as chain reconstitution experiments, indicated that both the sugar-phosphate backbone and the heterocyclic bases of the nucleic acid are essential for antibody recognition. Amino terminal sequence analysis of A52 and two RNA-binding hybridoma proteins revealed that the heavy chains from all three were members of the VHII subgroup and that the A52 light chain was homologous to the VK8 subgroup. The D42 heavy chain was found to be similar to a phosphocholine-binding hybridoma of the VHIII subgroup.     

Murine monoclonal anti-DNA antibodies bind directly to glomerular antigens and form immune deposits

The capacity of monoclonal anti-DNA antibodies, derived spontaneously from MRL-lpr/lpr mice, to bind directly to intrinsic glomerular antigens and form immune deposits was evaluated. Two antibodies, H130 (IgM-kappa) and H241 (IgG2a-kappa), bound to normal glomeruli in vitro. This binding was not inhibited by DNAase, but it was, in the case of H130, inhibited by the anti-idiotype anti-H130. Both antibodies also bound to glomerular digests on nitrocellulose. After i.v. injection, however, H241 bound to glomeruli and formed glomerular immune deposits, whereas H130 did not. Similarly, after i.p. injection of H241 hybridomas to normal mice, all mice developed glomerular immune deposits. In contrast, administration of H130 hybridomas, other anti-DNA-producing hybridomas, and other unrelated hybridomas did not lead to glomerular immune deposit formation. We conclude that certain lupus auto-antibodies can form glomerular immune deposits by binding directly to non-DNA antigenic structures that are normally present in extracellular locations within normal glomeruli.     

Specificity of anti-polynucleotide monoclonal antibodies from human-human hybridomas

Summary Nine human-human hybridoma clones, secreting monoclonal antibodies reactive with nucleic acids, were generated by fusing with lymphocytes of lung cancer or systemic lupus erythematosus patients. These hybridoma antibodies were classified into 5 types, in terms of reactivities with DNA, RNA, various synthetic nucleic acids and cardiolipin. Hybridoma clone SU-1 secreted antibody reacting with dsDNA, ssDNA and RNA (type I). Clone HL-321 did not react with these, but with poly (dT), poly (I) and poly (G) (type II). Clone HL-349 was reactive with almost all nucleic acids tested and also with cardiolipin (type III). Clones HF-4, HF-7, HB-7 and HL-259 reacted with ssDNA, poly(A), poly(G) and cardiolipin, but not with RNA (type IV). HB-5 and SH-9 antibodies were reactive only with poly (dT) (type V). Editor’s Statement This paper describes isolation and characterization of human-human hybridoma clones producing antibodies to nucleic acids. Isolation of such hybridomas from lymphocytes of cancer patients and the similarity of some isolates to those obtained from mice exhibiting autoimmune disease represent interesting observations that may lead to future insights.     

Monoclonal autoantibodies with interleukin 3-like activity derived from a MRL/lpr mouse

Hybridomas that secrete monoclonal antibodies with interleukin 3 (IL-3)-like activity were established from spleen cells of a nonimmunized autoimmune MRL/lpr mouse. Five of the monoclonal antibodies thus obtained bound selectively to IL-3-dependent cells and supported their growth. These monoclonal antibodies inhibited the binding of IL-3 to FDC-P2 cells and vice versa. Thus, these antibodies were probably directed to IL-3 receptor sites, or at least to some cell surface proteins related to the growth of the IL-3-dependent cells. These MRL/lpr-derived monoclonal antibodies reacted strongly with cells from bone marrow, spleen, and lymph node of MRL/lpr mice, but minimally with such cells of MRL/+ or BALB/c mice. The findings were consistent with our earlier suggestion that the IL-3-like activity in MRL/lpr sera is not caused by IL-3 itself but is associated with IgG that is probably an autoantibody directed to the IL-3 receptor.     

Anti-idiotypic antiserum to monoclonal anti-Sm inhibits the autoantigen-induced proliferative response

Anti-idiotypic sera were produced in BALB/c mice against three established monoclonal anti-Sm antibodies. Inhibition assays showed that the anti-idiotypic antibodies recognized determinants that were present on all three monoclonal antibodies but not on normal mouse IgG from unimmunized BALB/c mice or myeloma proteins. Normal (+/+) and autoimmune (lpr/lpr) MRL/MpJ or C3H/HeJ mice were immunized with Sm in complete Freund's adjuvant. Immune T cells from the draining lymph nodes proliferated in response to the addition of Sm in vitro. Anti-idiotypic serum added to these cultures inhibited the proliferative response by 50 to 70%, whereas normal BALB/c serum had no effect. This inhibition of proliferation was antigen specific, because the anti-idiotypic serum did not inhibit the T cell proliferative response to an irrelevant antigen, TNP-KLH, or ovalbumin. Kinetic studies showed that the anti-idiotypic serum inhibited an early event in antigen-induced proliferative response, because the addition of serum late in culture did not cause any significant reduction in proliferation. The reduced proliferative response was due to direct action of the anti-idiotypic serum on the Lyt-1+, 2- T cell population.     

Base specificity and idiotypy of anti-DNA autoantibodies reactive with synthetic nucleic acids

Synthetic nucleic acid reactivities and the distribution of idiotypes associated with poly(dA) and poly(dT) specificities were evaluated among both monoclonal and polyclonal anti-DNA antibodies from autoimmune New Zealand mice. Ten monoclonal anti-DNA antibodies (IgG2a or IgG2b), derived from NZB/NZW mice and reactive with natural DNA (duplex and/or heat-denatured), were found to collectively exhibit a diverse binding pattern with six deoxyribohomopolymers. Several monoclonal antibodies displayed reactivity with poly(dT) comparable to that with natural DNA. Serologic studies indicated that polyclonal anti-DNA autoantibodies from NZW/NZW mice and both parental strains also cross-reacted with various homopolymers and bound preferentially with those containing pyrimidines, particularly poly(dT), relative to purines. Detailed binding analyses with two poly(dT)-reactive monoclonal antibodies demonstrated that stable DNA/anti-DNA complexes were formed with synthetic oligomers containing six to 10 nucleotides; binding to such antigens was relatively insensitive to ionic strength and inversely dependent on temperature. Both antibodies exhibited preferential binding (greater than or equal to 10-fold) with poly(dT) relative to poly(dU), suggesting the importance of the C5-methyl group and/or helical conformation in pyrimidine base recognition. Idiotypes on poly(dA)-specific and poly(dT)-specific monoclonal antibodies were found to be reciprocally distinct, localized at or near active site residues, and expressed at low levels (less than 10 to 130 ng/ml) in anti-DNA sera from all three New Zealand strains. These findings suggest that: nucleotide base determinants are significantly involved in DNA/anti-DNA interactions; poly(dT) represents a major cross-reactive synthetic antigen; and idiotype expression among lupus autoantibodies which recognize such determinants may be diverse.     

IgG antinuclear antibodies with cross-reactive rheumatoid factor activity

To investigate whether IgG antinuclear antibodies have cross-reactive rheumatoid factor activity, monoclonal IgG antibodies to DNA and Sm from autoimmune MRL-lpr/lpr mice were assayed by ELISA for binding to IgG antigens. Of the nine anti-DNA and anti-Sm monoclonals tested, six showed significant binding to affinity-purified rabbit IgG (RIgG) and human IgG (HIgG). To confirm that cross-reactivities were due to a single antibody, immunoabsorption of a representative polyspecific monoclonal termed C11 (anti-DNA, anti-Sm) on either Sepharose-DNA or Sepharose-RIgG resulted in marked loss of activity to the three antigens DNA, Sm and RIgG compared with immunoabsorption on Sepharose-bovine serum albumin. The monomolecular nature of the cross-reacting antibody was also suggested by inhibition analysis of C11; DNA inhibited C11 binding to RIgG 64%, whereas Sm inhibited binding to RIgG 33%. Aggregated RIgG and HIgG, however, did not inhibit binding of C11 to DNA, Sm, or solid-phase RIgG, probably reflecting the low affinity of this antibody for fluid phase Ig. Together, these findings suggest that antinuclear autoantibodies of the IgG, as well as the IgM, class have polyspecific IgG binding activity and suggest that IgG antinuclear antibodies may emerge from rheumatoid factor responses.     

Pathogenic antibodies in systemic lupus erythematosus: anti-LAMP antibodies

We recently demonstrated that a monoclonal anti-DNA antibody, spontaneously produced in lupus B/W mice, recognizes the same protein(s) at the surface of several human cell types involved in lupus pathogenesis including normal human erythrocytes, normal platelets and rat neuronal tissue. This cell-surface protein(s) cross-react(s) with double-stranded DNA. We suggest to call this protein(s) LAMP [lupus associated membrane protein(s)]. Here we show that: immunoglobulins eluted from kidneys of autoimmune MRL/lpr/lpr mice strongly react with LAMP. Anti-LAMP antibodies are present in large amount in MRL/lpr and B/W mice sera. Anti-LAMP are present in 25 out of 25 human SLE sera ranged as SLE on the basis of revised American Rheumatism Associated classification. Interestingly, two of these sera did not display anti DNA anti-body activity. Taken together, these results strongly suggest a role of LAMP in the pathogeny of SLE.     

The regulatory role of NZB T lymphocytes in the production of anti-DNA antibodies in vitro

Murine lupus is characterized by the production of numerous autoantibodies and immune complex glomerulonephritis. Anti-DNA antibodies are the hallmark of this disorder and may be associated pathogenetically with the glomerulonephritis. The cellular mechanisms underlying the regulation of the production of anti-DNA antibodies may prove to be the fundamental abnormalities responsible for the lupus syndrome seen in these mice. By using a system of spontaneous anti-DNA antibody production in vitro, we have determined that such production is characteristic of autoimmune NZB and MRL-lpr/lpr mice but not of the nonautoimmune control strains. Additional examination of the cellular mechanisms involved in the regulation of this response in NZB mice revealed: 1) this response is markedly T cell dependent, 2) NZB T cells are essential for maximal production of this autoantibody, and 3) NZB T cells actively interfere with normal immune regulatory mechanisms that lead to the production of anti-DNA antibodies spontaneously in vitro by nonautoimmune syngeneic B lymphocytes. Although these studies of anti-DNA antibody production in vitro disagree with previous work by others they successfully reproduce the results obtained earlier in experiments performed in vivo.     

Treatment of autoimmune MRL/Ipr mice with monoclonal antibody to Thy-1.2: a single injection has sustained effects on lymphoproliferation and renal disease

MRL/Mp-lpr/lpr (MRL/lpr) mice spontaneously develop an autoimmune disease characterized by anti-DNA antibodies, immune-complex glomerulonephritis, and massive proliferation of a distinct population of T cells. The proliferating T cells have the phenotype Thy-1.2+, T200+, Lyt-1+,2-,3-, but Thy-1.2 and Lyt-1 are expressed in abnormally low density. These cells appear to function as helper cells, and neonatal thymectomy prevents both lymphoproliferation and autoimmunity, which suggests that autoimmunity in MRL/lpr mice is secondary to T cell proliferation. We therefore attempted to reduce lymphoproliferation by treating MRL/lpr mice with a single injection of rat monoclonal antibody (MAb) to Thy-1.2 (30-H12, IgG2b). Mice were treated at 8 wk, before the onset of overt disease. We found that MRL/lpr mice were resistant to depletion of circulating T cells (CTC) by anti-Thy-1.2; 0.6 mg of antibody totally depleted CTC from normal mice, but had little or no effect on CTC in MRL/lpr mice. However, treatment with 6 mg of MAb against Thy-1.2 reduced CTC in MRL/lpr mice by over 70%. Moreover, this single treatment markedly reduced the proliferation of CTC over the ensuing 3 mo, despite clearance of the anti-Thy-1.2 from the circulation within 3 wk. Treated mice maintained better renal function than untreated controls, as assessed by levels of blood urea nitrogen (BUN), although anti-DNA antibodies were not significantly reduced. The effect of anti-Thy-1.2 was specific; treatment with rat MAb to the common leukocyte antigen T200 produced only a transient effect on circulating lymphocytes and did not reduce renal disease. The prolonged effects of a single injection of anti-Thy-1.2 suggest that the MAb produces a sustained alteration in immune regulation. The improvement in renal disease is in accord with evidence that autoimmune disease in MRL/lpr mice is T cell dependent. Monoclonal anti-lymphocyte antibodies may be useful in the treatment of autoimmunity.     

Primary structures and chain dominance of anti-DNA antibodies

Using several anti-DNA autoantibodies, we analyzed the relative involvement of heavy and light chains in their interactions with DNA. We previously obtained eight hybridomas producing monoclonal anti-DNA autoantibodies by fusing spleen cells from an MRL-lpr/lpr mouse with myeloma cells. The chain dominance was analyzed by UV cross-linking experiments, in which the antibodies were covalently cross-linked with radioisotope-labeled oligonucleotides by short-wavelength UV-light, and the cross-linked H and L chains were analyzed by SDS-PAGE and densitometric scanning. Among these, three were found to be heavy chain dominant antibodies in which heavy chains are dominantly involved in DNA binding. The other five were co-dominant antibodies in which both heavy and light chains are involved in DNA binding. To determine the factor(s) that can explain the chain dominance in DNA binding, we determined the amino acid sequences of the variable regions of both heavy (VH) and light (VL) chains of all eight monoclonal antibodies. By analyzing the data, we were able to draw the following conclusions: (1) The arginine residues are found in the CDR3 regions of both VH and VL of the co-dominant antibodies; whereas, the same residues are found only in the CDR3s of VH, but not in VL, of the heavy chain dominant antibodies. (2) The net charges of the V regions affect the chain dominance. From the results of this study it is suggested that the presence of arginine residue in CDR3 is a critical factor in determining chain-dominance, as well as DNA binding of anti-DNA antibodies in general.     

Persistence of lpr/lpr-derived IgG2a secreting B cells in normal----lpr/lpr adult radiation chimeras or lpr/lpr----normal kidney capsule chimeras.

Lethally irradiated MRL/lpr mice reconstituted with bone marrow stem cells from a normal mouse strain develop a state of split hematopoietic chimerism; erythrocytes, granulocytes, and macrophages are derived from the normal stem cell inoculum while the peripheral T lymphocytes are derived from radioresistant lpr host cells. Moreover, these mice have normal levels of serum IgM and IgG2a produced by radioresistant host B cells, even though they have relatively few sIgM+ B cells. In order to better understand the differentiation and regulation of B cells present in these chimeric mice, the current study was undertaken to localize and to assess the functional capacity of the lpr B cells producing the serum antibodies. Surface IgG2a+ cells could not be found in the spleen or lymph nodes of these mice, but large lymphocytes containing cytoplasmic IgG2 of host (lpr) allotype could be readily detected, even though they constituted less than 1% of the total spleen population. The host-derived serum IgG2 and IgG2+ cells were even present in the spleens of "leaky" mice that had relatively normal numbers of donor-derived sIgM+ B cells. These lpr B cells secreted IgG2a antibody that bound ssDNA, but they could not respond to immunization with SRBC. These results indicate that the lpr-derived radioresistant B cells have a limited capacity for proliferation and are already committed to the memory lineage. The presence of similar B cells in normal mice transplanted with neonatal lpr/lpr spleen fragments suggests that lpr/lpr B cell development is inherently abnormal.     

Variable region sequence analysis of anti-DNA antibodies: evidence for a family of closely related germ-line VH genes encoding lupus autoantibodies.

Cloning and sequencing of the V regions of the anti-DNA monoclonal antibodies (mAbs), H438 and H130, indicate that H438 is encoded by a J558 VH gene, a single D region nucleotide, and unmutated JH1, V kappa-1C and J kappa 1 genes, and the H130 L chain is encoded by a V kappa-21 subgroup gene J kappa 1 gene. Identification of VH438, which shared VH hybridization pattern with 6% of a panel of 352 MRL/lpr hybridomas, suggests that the frequency of J558 use among spontaneously activated B cells in MRL/lpr mice is greater than previously reported. The VHH438 J558 family gene is identical to VHPAR, which encodes the independently derived MRL/lpr autoantibody, MRP-2, and is highly homologous to the previously reported VHH130, which is identical to a BALB/c germ-line VH gene. Comparison of consensus sequences of homologous autoantibodies and previously reported restriction mapping suggest that a minimum of three highly related J558 germ-line genes encode lupus autoantibodies.     

Monoclonal antibodies for carcinoembryonic antigen and related antigens as a model system: a systematic approach for the determination of epitope specificities of monoclonal antibodies

A systematic approach for the determination of epitope specificities of monoclonal antibodies to a complex antigen system is described. After initial screening to identify antigen-binding monoclonal antibodies, one or more of the clones are isolated by limiting dilution cloning, grown in ascites, and the resulting antibodies secreted into the ascitic fluid are affinity purified on Sepharose-bound protein A, radiolabeled, and cross-compared with antibodies from other clones by a solid-phase competitive immunoassay. In this work, BALB/c mice were immunized with either purified carcinoembryonic antigen (CEA) or the CEA-producing cell line HC 84S. Spleen cells were fused with the mouse myeloma cell line Sp2/0-Ag14. The supernatants from 25 hybrids showed a significant binding of 125I-CEA (greater than or equal to 15%). Nine hybrids were cloned, resulting in 33 different clones. The antibodies produced by the different cloned hybrids and the remaining uncloned hybrids recognized a total of five different epitopes on CEA. All of the epitopes reside on the protein moiety of the molecule as determined by antibody binding to deglycosylated CEA. The monoclonal antibodies with five different epitope specificities were reacted with tissue sections of normal and cancerous tissues and with peripheral blood smears. Each of the five monoclonal antibodies reacted with tissue sections from colonic, gastric, lung, and mammary carcinomas, as well as from a benign colonic polyp and a resection margin from a colonic carcinoma. Four monoclonals reacted with normal liver tissue. Granulocytes in peripheral blood smears bound three antibodies strongly and one antibody weakly, and one antibody was not bound. One monoclonal antibody that reacted with normal liver tissue was not bound by granulocytes. The ability of these five monoclonal antibodies to differentially detect three different CEA-related antigens in normal and malignant tissues may have clinical utility.