GB2113713A - Forming monoclonal antibodies to carbohydrate hapten antigens - Google Patents

Abstract

Hybridoma cells, capable of forming monoclonal antibodies to antigens having selected carbohydrate haptens, have been prepared. The carbohydrate haptens are preferably selected from synthetic oligosaccharides coupled via a bridging arm, particularly where the oligosaccharide acts as a human blood group determinant. The carbohydrate hapten alternatively may be a determinant of tumor-associated antigens. The hybrid cells have been screened, cloned, and cultured to produce the specific antibodies to the selected carbohydrate hapten.

Description

SPECIFICATION Forming monoclonal antibodies to carbohydrate hapten antigens This invention relates to the systematic generation of monoclonal antibodies that specifically binds selected carbohydrate haptens. Preferably the haptens are synthetic oligosaccharides. Hybridoma cells have been prepared by fusing antibody-producing animal cells primed by the antigen bearing the carbohydrate hapten, with myeloma cells. The resulting hybrid cells are screened for specificity to the selected carbohydrate hapten using the same hapten coupled to a carrier protein in a solid phase binding assay, and the desired hybridoma cells recovered.

Background Monoclonal antibodies in general, and the hybrid cells which make them have been described by C. Milstein in Scientific American, Oct. 1980, p. 66. H. Koprowski et al in United States Patent 4,172,1 24 and 4,196,265 have described the production of monoclonal antibodies to malignant tumor antigens and to viral antigens. G.L. Mayers et al in Transplantation Proceedings, Vol. XII, No. 3, Sept.

1980, have described the immunochemistry of monoclonal antibodies to certain synthetic benzoate or phthalate haptens.

The Milstein group (see reference above) has produced a monoclonal antibody reagent that recognizes blood Group A antigen. A monoclonal anfiCd3 has been prepared by Ortho Diagnostics Co.

These monoclonal antibodies apparently have been prepared using natural group A or group Cd3 antigens which may bestow less specificity than desired for typing purposes.

Synthetic oligosaccharide haptens corresponding to the antigenic determinants of human ABO and Lewis blood-groups have been synthesized1,2 in a manner which permits the preparation of soluble haptenprntein antigens23, hapten-red cell indicator cells3 and immunoadsorbents4. The soluble antigens give rise to specific antibodies in rabbits and goats. Refinement of these heterogeneous antibody populations is facilitated by the availability of immunoadsorbents bearing the appropriate ABO, Lewis or incomplete determinants5,6. In this way, reliable Lewis typing sera5 and superior immunohistochemical reagents have been produced6.

The exploitation of these synthetic determinants has so far been limited to conventional procedures for inducing high-titre immune sera2,6. The value of these unique chemical structures, however, far exceeds the existing methodologies of haematology and blood-banking. For example, extracorporeal immunoadsorption of ABO antibodies has been used for patients receiving allogeneic bone marrow transplants from ABO-incompatible donors7. The immunoadsorbents (SynsorbsR) used in this instance were composed of synthetic blood group-A and B determinants covalently bound to crystalline silica.

We have found that these and other related synthetic carbohydrate structures, artificial antigens and haptenated red-cells are highly effective and specific for the generation of blood-group specific monoclonal antibodies. For example, several human blood-group B specific IgM producing hybridoma clones have been obtained through the use of the synthetic blood-group B trisaccharide determinant (see graphic formula-page 2).

Summary of the Invention The invention includes a method of preparing hybridoma cells which produce monoclonal antibody having high specificity for a single carbohydrate hapten comprising: a) providing antigen consisting essentially of a single carbohydrate hapten coupled to one of a soluble carrier protein and an antigenic microparticle; b) priming by immunization, animal cell capable of producing antibodies, with said antigen; c) fusing said primed cells with myeloma cells to form hybridoma cells; d) screening said hybridoma cells for the specific antibody to said carbohydrate hapten, by using said same single hapten coupled to a carrier in a solid phase binding assay; and e) recovering the desired hybridoma cells.

The method may include the additional step of culturing the hybridoma cells and recovering the antibodies specific to the carbohydrate hapten from the liquid produced. Preferably the method includes the further additional step of purifying the specific antibodies by immunoadsorption employing the same carbohydrate determihant coupled to an insoluble matrix. The selected carbohydrate hapten is preferably a synthetic oligosaccharide, coupled via a bridging arm having from 4-1 8 carbon atoms. As well as human blood group determinants, the carbohydrate hapten may be of the type found in tumorassociated antigens.

B -- trisaccharide R' = O(CH2)BCO2CH3 R2 = OH A -- trisaccharide R1 = O(CH2)8CO2CH3 R2 = NHAc

B -- disaccharide R1 = O(CH2)8CO2CH3

H -- disaccharide R1 = O(CH2)8CO2CH3 The animal cell immunization may be carried out in a selected live animal, e.g. rodent, rabbit or goat, or in cell culture. In the latter case, the cells preferably are selected iymphocytes, and the hapten preferably is coupled to antigenic carrier proteins of the type bovine serum albumin (BSA) and horse haemoglobin (Hgb). In the case of live animals being immunized, primed spleen or lymph node cells are produced and isolated in step (b).For immunization in vivo, the hapten may be coupled to an antigenic microparticle of the type of red blood cells, latex beads or agarose beads.

Preferably the method includes an additional step between steps (b) and (c) of pre-fusion enrichment of cells which include lymphocytes bearing lg receptors for the carbohydrate determinant by using fluorescent probes coupled to the carbohydrate hapten and selecting the desired cells using a fluorescent activated cell sorter.

The myeloma cells used in step (c), preferably are mouse myeloma cells of the type SP2/0, P3x63Ag8 and NS 1. The actual fusing is carried out as known in the prior art.

The screening in step (d) is carried out by using the same single hapten (preferably synthetic) coupled to a carrier in a solid phase binding assay (as described in detail below). The binding assay preferably includes radioimmunoassay or ELISA (enzyme-linked immunosorbent assay) techniques.

The hybrid cells (yielding antibodies specific to the single hapten) isolated and recovered, preferably are cloned in semi-solid agar containing suitable nutrients. For production of antibodies, the clone may be mass cultivated in tissue culture or injected into the peritoneal cavity of selected animals.

Antibodies can be recovered from the resulting culture fluid or ascitic fluid by known techniques.

The invention includes hybridoma cells primed to produce mono-clonal antibodies to antigens, said antigens consisting essentially of single synthetic carbohydrate haptens coupled to antigenic carriers. The invention also includes resulting monoclonal antibodies to antigens whose determinants consist essentially of single synthetic carbohydrate haptens.

Description of Drawings Figure 1 is a graph showing the gel filtration elution profile of ascites fluid from mouse injected intraperitoneally with two immunized clones.

Figure 2 is a graph showing the % inhibition of precipitation of B erythrocytes by monoclonal antibody with synthetic B-trisaccharide hapten as inhibitor, in a radioimmunoassay.

Example and Preferred Embodiments Antigens Haptens were available from chemical synthesis12 and were coupled to bovine serum albumin (BSA) and horse hemoglobin (Hgb) [Miles Laboratories, Kankakee, illinois], to provide synthetic antigens, SyntagensR [Chembiomed, Edmonton, Alberta]. The haptens were also coupled to human red blood cells according to published procedures3.

Immunization Female BALB/c mice [Jackson Laboratory, Bar Harbor, Maine] 12-1 6 weeks of age, were injected intraperitoneally with either (a) 100,ug of the group B-BSA antigen (SyntagenR-B, Chembiomed) in 0.1 ml of phosphate buffered saline (PBS) and 0.1 ml of complete Freund's adjuvant, or (b) 0.25 ml of a 20% suspension in PBS of human blood group Oneb(+) red cells haptenated3 with the synthetic blood group B-trisaccharide hapten2. Red cells haptenated in this manner were agglutinated by commercial blood group B sera [Ortho, Don Mills, Ontario] diluted 1:128. Fourteen days after the priming injection B-haptenated red cells (0.25 ml) were injected intravenously via the tail vein.

Media andPlasmacytoma Cells The non-lg-producing Sp2/08 plasmacytoma cell line [Institute for Medical Research, Camden, N.J.] was maintained in Dulbecco's modified Eagle's (DME) medium (DMEM) containing 6-thioguanine and 10% Freund's complete adjuvant (FCS) at 370, 8% CO2 and 95% relative humidity. For one week prior to fusion, the cell-line was grown in DME medium without 6-thioguanine. All media employed, routinely contained 30 yg/ml of gentamicin [Sigma, St. Louis, Mo.]. Hypoxanthine thymidine (HT) medium was composed of DMEM supplemented to contain 20% FCS and 10% Nat. type Culture Coll.

(NCTC) 109 medium [Microbiological Associates, Bethesda, Maryland] and with the additional constituents in the concentrations indicated: 0.16 mg/ml sodium pyruvate, 0.2 units/ml bovine insulin [Sigma], 0.1 5 mgs/mi sodium oxaloacetate, thymidine 16,aM and hypoxanthine 0.1 mM. Selective HT medium was identical to the above HT medium but contained in addition, aminopterin (0.8 yM).

Fusions The fusion protocol was the modification described by Kennett9, which incorporates the essential elements of the procedures developed by Milstein and co-workersl0 and by Scharff and his co workers1'. Briefly,- spleen cells from two immunized mice were collected by perfusion with DMEM containing 20% FCS in 60 mm petri dishes. Red cells were lysed by a 10 min treatment with coid 0.17 M ammonium chloride. Sp2/0 Plasmacytoma cells (1 x107), in logarithmic growth were pelleted (200xg) with 1 x 103 viable, spleen cells, and the cell mixture washed with serum free DMEM.A 30% solution (0.2 ml) of 30% polyethylene glycol MW 1 000 (PEG 1000) [Baker, Canadian Laboratories, Ottawa, Ontario] in serum free DMEM at 370 was added to the loosened pellet which was gently resuspended by swirling and tapping the centrifuge tube. The mixture was centrifuged for 6 min. at 200xg and after 8 min. total exposure to the PEG, the solution was slowly diluted by addition of 5 ml of serum free DMEM at 370. A further 5 ml of DMEM containing 20% FCS was added, and the cells then pelleted at 130xg and resuspended in 30 ml of HT medium at 370. After about 30 min. at room temperature, the suspension was distributed, 1 drop/well, into 96 well Linbro [trademark] microplates [Flow Laboratories, Mississauga, Ontario] and 24 hr later, 1 drop of HT 2X aminopterin was added to each well.Microscopic examination at day 7 was used to assess wells with putative hybrids and the number of clones to each well. Following this, two drops of HT medium were added to each well and depending upon cell density, putative hybrids were screened by ELISA assay between day 1014.

Those wells becoming 5070% confluent were transferred with HT medium to 24 well Linbro plates [Flow Laboratories]. Hybnds with supernatant that gave an optical density (O.D.) greater than 0.1 against negligible background in ELISA screening were cloned for further sceening of antibody specificity.

Cloning Cloning of hybrid cells was performed in semi-solid agar using mouse spleen cells as feeders'2 '3.

Two volumes of a 4.5% Difco agar solution, at 450, was mixed with 25 volumes of HT medium, also at 450. Hybrid cells (1 x 103 and 1 x 104) in 0.5 ml of HT medium together with BALB/c spleen cells (5 x 1 06) in 0.5 ml of HT medium were pipetted into a universal container. Agar-HT solution (9 ml) was then pipetted directly onto this suspension to ensure mixing. The agar was allowed to cool and incubated under the same conditions as the parental Sp2/0 cell line for 7-14 days. Clones were picked with a pasteur pipette and transferred to microplates for screening.

ELISA screening Linbro ELISA plates [Flow Laboratories] were coated with artificial antigen B-Hgb by incubation of 1 00 ,ul of antigen solution/well (1 0 g/ml) for 3 hours at room temperature (RT) and stored sealed at 40 until use. Between each operation, plates were washed three times with phosphate buffered saline containing detergent Tween 20 + na azide (PTA). Culture supernatant (100,,t1), from wells containing putative hybrids were incubated for 3 hours at RT in the ELISA plate. Goat antimouse IgM and Igri [Miles] conjugated with alkaline phosphates'4 [Sigma] was diluted to working strength (1:100) with PBS and 100,ul was added to each well. After 1 hr at RT, 100 l of p-nitrophenolphosphate [Sigma] substrate solution (1 mg/ml of p-nitrophenolphosphate disodium salt in 0.05 M sodium carbonate buffer pH 9.8, containing magnesium chloride 1 0-3M) was added and absorbance at 405 nm was read 1 hr later using a multichannel spectrophotometerTitertek Multiscan [trademake-Flow Laboratories].

Clones hybridomas remaining positive in ELISA were injected for ascites. Sera and ascitic fluids were assayed in the same manner as culture supernatants.

Ascitic fluid Pristane (2,6,10,14-tetramethylpentadecane)-primed 8ALB/c mice were injected with 1 x 107 hybridoma cells in DMEM (0.5 ml). After 7-10 days ascitic fluid was tapped, centrifuged and stored at 70C.

Agglutination Hemagglutinations were performed with a 1% suspension of freshly drawn human B and 0 red cells in BSP. Culture supernatants or ascitic fluids (50 ul) were serially diluted with microdiluters [Cooke, Alexandria, Virginian in V-well Takatsy [trademark] microtiter trays. Red cells (50 jul of 1 % suspension) were added to each well and agglutination was scored after 1 hr at RT.

Gel Flltration Chromatograph Ascitic fluid (0.5 ml) was characterised by gel filtration with cross-linked agarose (S--300 Sephacryl-trademark) in pH 8.0, 0.2 M Tris buffer containing 0.5 M NaCI. Fractions (3 ml) were collected from a column 45 x 2.5 cm. Tubes were assayed for antibody activity by either ELISA or agglutination tests.

Radioimmunoassays Radiolabeling: B-BSA was radiolabeled with 1251 by using iodogen procedure15.

Antibody and inhibition assay: Ascitic fluids were diluted with saline (20 jul) in small test tubes in duplicate. 1251 labelled antigen (1251 B-BSA, Hl in saline, 25,000 cpm) was added to each tube and the mixture was diluted with normal rabbit serum (0.2 ml) and 0.2 M Tris-HCI buffer, pH 7.4 (0.2 ml). The soluiton was left at 40C overnight and the immune complex was precipitated by addition of 7.5% polyethylene glycol 6000 (0.5 ml) in 0.2 M Tris-HCI buffer, pH 7.4, as described by Desbuquois and Aurbach16. The mixture was centnfuged at 800x g for 1 hour at 4 C and supernatant and pellet were separated. The pellet was washed once with 3.5% PEG in 0.2 M Tris-HCI buffer, pH 7.4, (1 ml) and the washings were added to the supernatant.The pellet and supernatants were counted in a gamma counter. The percentage of radioactivity in the pellet is a measure of the antibody level in the ascitic fluid.

Inhibition assay: Synthetic haptens were diluted in serial twofold dilution with PBS in a volume of 20 jul per tube set-up in duplicate. An equal volume of diluted ascitic fluid (1:1000) was added to each tube together with normal rabbit serum (0.2 ml) and 0.2 M Tris-HCI buffer, pH 7.4, (0.2 ml). The mixture was left at 40C for 4 hrs. Following this, 1251 labeled antigen (20 jul) was added to each tube and left at 40C overnight. The pellet and supernatants were separated and counted as described previously.

Test Results Two fusion experiments were performed using the non-lg-producing plasmacytoma cell line Sp208 and spleens from mice immunized according to identical protocols. The primary injection employed a soluble artificial antigen, B-BSA (Syntagen-B), to which approximately twenty haptenic groups were bound. These synthetic haptens1 possessed the terminal trisaccharide structure of the human blood group B (see formulae) and were coupled to BSA via an amide bond to the lysine residues of the carrier protein23. Human blood group 0 erythrocytes, to which the B-hapten was covalently linked, were injected intravenously four days prior to fusion.Serum collected at the time spleens were excised contained blood group B specific antibody titers between 1 0103, as judged by ELISA employing the coating antigen, B-Hgb. The somatic cell fusion experiments were conducted according to standard procedures6,11.

The results of the two fusion experiments are summarized in Table 1. Of the 290 putative hybrids from experiment 1, only 19 secreted antibody specific for the blood group B determinant. Experiment 2 yielded 26 such clones but, after re-cloning, only 4 clones from experiment 1, and 12 from experiment 2, retained binding activity toward B-Hgb coated ELISA plates.

Ascites fluid was introduced by injection of the cloned hybridoma lines, 3 from experiment 1 and 6 from experiment 2. These were selected on the basis of the ELISA titers of culture supernatants. The actual titers of the ascites fluids determined in Takatsy microtiter trays are recorded in Table 2. The two clones 3E-4 and 3E-7 showed no cross reactivity with human A or 0 cells, while clone 1 A-1 2 cross-reacted with type-A cells. All three ascites had titers in the range 1:6,400-1:12,800. None of the ascitic fluids derived from hybrid clones generated by experiment 2 showed agglutinating activity.

Class specific ELISA anti-globulin reagents indicated that antibody produced by clones from the first fusion were exclusively of the IgM class, while those from experiment 2 were IgG and IgM. In order to confirm the antibody class of the B-specific hybridoma, ascites fluid from clones 3E-4 and 3E-7 were subjected to gel-filtration. The elution profile is shown in Figure 1, together with the location of agglutinating activity. This activity was restricted to peak I, which elutes in the excluded volume of the S--300 column indicating a molecular weight in excess of 300,000 daltons. Peaks II and 111 correspond to proteins with the molecular weights of IgG and serum albumin. Neither peak possessed detectable agglutinating activity toward blood group B erythrocytes.

TABLE 1 Specific and Putative Hybrids Generated by Artificial B-antigens Fusion Putative hybrids/ B-Specific Recloned B-lg class Exp. wells plated hybrids* of specific hybrids+ IgG IgM 1 290/576 19 0 4 2 200/576 26 8 4 *Determined by ELISA assay with B-Hgb antigen coated plates +lg class by class specific ELISA reagents and SDS-PAGE.

(SDS-PAGE = Na dodecylsulphate-polyacrylamide gel electrophoresis) TABLE 2 Agglutination titres of re-cloned hybridoma lines Titre vs. Human Red Blood Cell Type Cell Line A B C 1-3E-4 n.a. 1:12,800 n.a.

l-3E-7 n.a. 1:6,400 n.a.

1-lA-12 1:512 1:12,800 n.a.

2-4C-1 n.d. n.a. n.a.

2-3A-1 n.d. n.a. n.a.

2-2E-4 n.d. n.a. n.a.

2-2G-10 n.d. n.a. n.a.

2-4A-1 n.d. n.a. n.a.

2-3D-10 n.d. n.a n.a.

n.d. = not determined n.a. = no agglutination Inhibition of the agglutination of B erythrocytes by hybridoma antibody 3E-4 with synthetic ligands (legend Table 3) shows dramatically the specificity of the monoclonal antibody for the B determinant. At the concentrations tested (up to 95 molar) the B-trisaccharide was the only active inhibitor. The terminal trisaccharide determinant of blood group A was inactive at one hundred times the concentration, at which the B-trisaccharide inhibited agglutination. Since the artificial antigen, B-BSA is polyvalent, both 3E-4 and 3E-7 antibodies precipitate this antigen. In a radioimmunoassay system based upon '251 labeled B-BSA antigen4, quantitative inhibition of precipitation was studied, Figure 2.The synthetic structures (see graphic formuiae) ranging from monoto tri-saccharides were used as inhibitors together with an AB blood group substance obtained from pigs and horses. As with agglutination experiments,the B-trisaccharide was the only synthetic ligand to show inhibition as in Figure 2. The AB blood group substance was also active as an inhibitor.

Conclusions One of the most significant advantages of the hybridoma technique has been the ability to generate a single antibody specificity from a complex antigenic mixture. The utility of such antibodies is limited only by the capability to screen out and identify redundant specificities. This task is greatly simplified by the availability of pure antigen for use in a binding assay4,14,17 or by employing a more classical approach, such as a serological matrix of hybridomas clones with a panel of cells or virus types.

The task of generating monoclonal antibodies specific for cell surface receptors is resolved to simplicity if chemically defined antigens can be used to immunize mice and then to screen putative hybrids. The approach in this example has been to immunize mice with synthetic glycoconjugate bearing the antigenic determinant of the human B blood group and following somatic cell fusion, to select IgM and IgM secreting hybrid clones for use in blood banking, serotyping and tissue staining6.

TABLE 3 Inhibition Assay of Crude Ascitic Fluid and purified antibody with synthetic sugars

INHIBITORS Minimum amounts (mg /ml) needed for complete (Sugars) inhibition of hemagglutination activity Purified monoclonal Ascitic fluid antibody B-Trisaccharide#OR 0.0625 (95 ma) 0.0625 α;#D-Gal (1#3)-ss-D-Gal#OR 1 (2mMa) 1 p-D-GaleOR' 2 (6mb1 ) 2 .= A-Trisaccharide#OR 5 (7.3mMa) 5 a-L-Fuc(12)ss-D-Gal(1e3)-13-D 1 (1.4mNa) 1 ClcNAceOR α-D-Gal#-OR 1 (3mMa) 1 α;-L-Fuc(1#2)B 1 (2mMa) 1 galactose grease OR' P1-Disaccharide#OR 1 (2mM 1 When R = (CH2)8CO2Me R' = (CH2)aCONHNH2 a) molar concentration at end point of inhibitor activity.

Three distinct advantages of the combination of synthetic antigen methodology with hybridoma may be noted with respect to generation of monoclonal antibodies directed against defined carbohydrate determinants. The first of these relates to the synthetic methods of artificial preparation.

The method developed for hapten synthesis125 and subsequent coupling to carriers2,3,5 is well suited to the use of both soluble protein and cells as the antigenic carriers. Since cells are often more immunogenic carriers than soluble protein'8, the ability to amplify the recruitment of specific Blymphocytes through variation of the carrier and its degree of hapten substitution can lead to an enhancement of the immune response of mice to the hapten. This enrichment of antibody-forming cells at the pre-fusion stage has a major influence on the production of antibody-forming hybrids since there is a correlation of antibody-forming cell (AFC) number with hybridoma frequency19.Although not illustrated here, augmentation of the humoural response via 'education' of T cells by carrier20 could be exploited to modulate further the specific response either in vivo or in vitro. For the in vitro priming system, the ability to generate soluble antigens employing a variety of carrier proteins bearing the relevant carbohydrate determinant is of great value. In conjunction with a fluorescent activated cell sorter (FACS) employing fluorescent probes coupled to the carbohydrate hapten, specific enrichment of B-lymphocytes bearing Ig receptors for the carbohydrate determinant permits a more sophisticated prefusion enrichment of the fusing lymphocytes.

Heterologous protein carriers bearing the carbohydrate determinant facilitate the selection and rapid screening of those antibody producing clones which bind carbohydrate. In conjunction with class specific antiglobulin reagent conjugated to alkaline phosphatasel4, the single assay distinguishes carbohydrate specific antibody of the IgM and IgG classes. In this way, we have selected three clones suitable for use in human blood-grouping. One of these clones, 1 A-1 2, showed anti-blood group A activity of much lower titer 1:51 :512vs.1 2,800 for B red cells.The other two clones, 3E-4 and 3E-7, titered at 1:6,400-1:12,800 in microtiter trays, while against a panel of human erythrocytes of differing phenotype the ascitic fluid titered somewhat lower 1:1600-3200. This is due to the difference between microtiter titration and the normal blood bank routine of tube agglutination. In any event, Table 4 shows the very clear specificity of this monoclonal product for human blood grouping.

The final advantage of synthetic haptens resides in the ability to precisely define the binding characteristics of the monoclonal antibody. This is illustrated by inhibition of agglutination and, more precisely, by inhibition of precipitation. In both cases, the absolute requirement by the antibody for the full branched trisaccharide sequence of the B-determinant is illustrated by the failure of either component disaccharide, a-D-Gal(1e3)-p-D-Gal or cg-L-Fuc(1o2)-,B-D-Gal, to bind even partially in the binding site.

This observation is in sharp contrast to published data for inhibition of agglutination21 or precipitation22 performed with heterogeneous immune sera. In these instances the B-disaccharide, a-D- Ga 1(1 -t3)-D-Gal, was an effective inhibitor in both systems, as were more elaborate linear sequences possessing this terminal disaccharide but lacking the branching fucose residue. This different behaviour in inhibition assay indicates the precise recognition achieved by the hybridoma antibody for the B trisaccharide and is to be expected for the other truly monoclonal antibodies.For some purposes, such precise specificity for antigenic determinants devoid of the conventional spectrum of cross-reactivities may be a disadvantage but in the choice of reagents to recognize cell receptors, such as blood-grouping, exquisite selection of this type is the ultimate goal. Furthermore, the linear relationship of the Btrisaccharide as inhibitor (Figure 2) clearly demonstrates the monoclonal and homcgeneous nature of the anti-8 binding site. Since all other synthetic structures bearing the 8-methoxy-carbonyloctyl 'bridging arm' also failed to exhibit even minimal inhibitory activity, it is clear that the specificity is directed exclusively toward the carbohydrate determinant.

In conclusion, we have demonstrated a flexible method for generating carbohydrate binding hybridoma antibodies which is both rapid and rigorous in terms of a well defined binding specificity. The method is well suited to the identified problem areas of hybridoma technology, i.e. seak immunogens resulting in low AFC in the pre-fusion immunization of mice and at the screening stage where specific but rapid selection of active hybrids is essential. Although we have concentrated here upon the human B blood-group determinant, the procedure described is general for the production of monoclonal antibodies with anti-carbohydrate specificity and has been successfully applied to other antigenic determinants of human and bacterial origin.

TABLE 4 Hemagglutination Activity a (Titer) of Ascitic Fluid-(3 E-4)with Human erythrocytes of various phenotypes bearing different ABO & Lewis Antigens

Blood Group on Dilution of Ascitic Fluid Antigens on Human Erythrocytes 1:2 1:10 1:50 1:100 1:200 1:400 1:800 1:1600 1:3200 1::6400 A1Lea 0 0 0 0 0 0 0 0 0 0 A1Leb 0 0 0 0 0 0 0 0 0 0 BLeb 4+ 4+ 4+ 4+ 4+ 3+ 3+ 2+ 2+ 1+ BLea-b- 4+ 4+ 4+ 4+ 4+ 3+ 3+ 2+ 2+ 1+ OLea 0 0 0 0 0 0 0 0 0 0 Oleb 0 0 0 0 0 0 0 0 0 0 OLea-b- 0 0 0 0 0 0 0 0 0 0 a0, 1+, 2+, 3+ abd 4± degree of agglutination 0 - no agglutination 4+ - cell button remains in the clump 3+ - cell button dislodges into several clumps 2+ - cell button dislodges into amny small clumps of equal size 1+ - cell button dislodges into finely granular, but definite small slumps References 1. R. U. Lemieux, "Human Blood Groups and Carbohydrate Chemistry", Chem. Soc. Rev. 7: 423, 1978.

2. R. U. Lemieux, D. R. Bundle and D. A. Baker, "The properties of a 'Synthetic' Antigen related to the human blood-group Lewis", J. Amer. Chem. Soc. 97: 4076, 1975.

3. R. U. Lemieux, D. A. Baker and D. R. Bundle, "A methodology for the production of carbohy drate-specific antibody", Can. J. Biochem. 55: 507, 1977.

4. P. H. Boullanger, A. Nagpurkar, A. A. Naijaim and R. U. Lemieux, "Application of 1251 radioimmunoassay to measure inhibition of precipitation reactions using carbohydrate-specific antibodies", Can. J. Biochem. 56:1102, 1 978.

5. R. U. Lemieux, D. R. Bundle and D. A. Baker, "Artificial Oligosaccharide Antigenic Determinants", U.S. Patent 4,238,473,1980.

6. R. U. Lemieux, D. A. Baker, W. M. Weinsein and C. M. Switzer, "Artificial Antigens. Antibody preparations for the localisation of Lewis determinants", Biochemistry 20:199, 1981.

7. W. I. Bensinger, D. A. Baker, C. D. Buckner, R. A. Clift and E. D. Thomas, "lmmunoadsorption for removal of A and B blood-group antibodies", New England J. Med. 304:160, 1981.

8. M. Shulman, C. D. Wilde and G. Kohler, Nature. 276:269, 1978.

9. R. H. Kennett, K. A. Denis, A. S. Tury, and N. R. Klinman, "Hybrid Plasmactyoma Production: Fusions with adult spleen cells, monoclonal spleen fragment, neonatal spleen cells and human spleen cells", Curr. Top. Microbiol. and Immunol. 81: 77, 1978.

1 0. G. Gale' S. C. Howe, C. Milstein, G. W. Butcher and J. C. Howard, Nature. 266: 550, 1977.

11. M. Gefter, D. H. Magulies and M. D. Scharff, Somatic Cell Genet. 3: 231, 1977.

1 2. J. Paul, "Cell and Tissue Culture", E and J Livingstone, London, 1970, p. 239.

13. P. Coffino, R. Baunal, R. Laskar and M. D. Scharff, "Cloning of Mouse Myeloma cells and detection of rare variants", J. Cell. Physiol. 79: 429, 1 972.

14. E. Engvall, K. Jonnson and P. Perlmann, "Quantitative assay of protein antigen, immunoglobulin G, by means of enzyme-labelled antigen and antibody-coated tubes", Biochim.

Biophys. Acta. 251:427, 1971.

15. P. J. Fraker and J. A. Speck, "Protein and cell membrane ordinations with a sparingly soluble chloramide 1 ,3,4,6-tetrachlorn-3a, 6a-diphenylglycourial", Biochem. Biophys. Res. Commun. 80: 849, 1978.

1 6. B. Desbuguois and G. D. Aurbach, "Use of polyethylene glycol to separate free and antibody bound peptide hormones in radioimmunoassays", J. Clin. Endocrinol. Melab. 33: 73L, 1971.

1 7. H. E. Carlsson, A. A. Lindberg and S. Hammarstrom, "Titration of antibodies to Salmonella O antigens by Enzyme-iinked immunosorbent assay", Infection and Immunity. 6: 703, 1972.

1 8. C. Stakli, T. Staekelin, V. Miggiano, J. Schmidt and P. Haring, "High frequencies of antigen specific hybridomas: Dependence on immunisation parameters and prediction by spleen cell analysis", J. Immunol. Methods 32:297, 1980.

1 9. G. Kohler and C. Milstein, "Derivation of specific antibodyproducing tissue culture and tumour lines by cell fusion", Eur. J. Immunol. 6: 511, 1976.

20. D. H. Katz and B. Benacerraf, "The regulatory influence of activated T cells or B cell responses to antigen", Adv. Immunol. 15:1,1972.

21. W. M. Watkins, "Glycoproteins: their composition, structure and function", ed. A. Gottschalk, 2nd ed. Elsevier, Amsterdam, pp. 830-91, 1972.

22. G. Schiffman, E. A. Kabat and W. Thompson, "Immunochemical studies on blood groups.

XXXI I. Immunochemical properties of and possible partial structures of blood group A, B and H antigenic determinants", Biochem. 3: 587, 1964.

Claims (26)

1. A method of preparing hybridoma cells which produce monoclonal antibody having high specificity for a single carbohydrate hapten comprising: (a) providing antigen consisting essentially of a single carbohydrate hapten coupled to one of a soluble carrier protein and an antigenic microparticle; (b) priming by immunization, animal cells capable of producing antibodies, with said antigen; (c) fusing said primed cells with myeloma cells to form hybridoma cells; (d) screening said hybridoma cells for the specific antibody to said carbohydrate hapten, by using said same single hapten coupled to a carrier in a solid phase binding assay; and (e) recovering the desired hybridoma cells.

2. The method of claim 1 including the additional step of culturing the hybridoma cells and recovering the antibodies specific to the carbohydrate hapten from the liquid produced.

3. The method of claim 2 including the further additional step of purifying the specific antibodies by immunoadsorption employing the same carbohydrate determinant coupled to an insoluble matrix.

4. The method of claim 1-3 including an additional step between steps (b) and (c) of pre-fusion enrichment of cells which include lymphocytes bearing Ig receptors for the carbohydrate determinant by using fluorescent probes coupled to the carbohydrate hapten and selecting the desired cells using a fluorescent activated cell sorter.

5. The method of claims 1-4 wherein the selected carbohydrate hapten is a synthetic oligosaccharide coupled via a bridging arm having from 4-18 carbon atoms.

6. The method of claim 1-5 wherein the carbohydrate hapten is a human blood group determinant.

7. The method of claims 1-5 wherein the carbohydrate hapten is of the type found in tummorassociated antigens.

8. The method of claim 1-7 wherein step (b) is carried out in vivo in a selected animal.

9. The method of claim 8 in which the animal is selected from rodents, rabbits and goats.

10. The method of claim 1-9 wherein the hapten is coupled to an antigenic microparticle of the type of red blood cells.

11. The method of claim 1-7 or 10 wherein step (b) is carried out in cell culture, the cells being selected lymphocytes.

12. The method of claim 11 wherein said hapten is coupled to antigenic carrier proteins of the type of BSA and horse hemoglobin.

13. The method of claims 8 or 9 wherein primed spleen or lymph node cells are produced and isolated in step (b).

1 4. The method of claim 1-1 3 wherein step (c) utilizes mouse myeloma cells of the type SP2/0, P3X63Ag8 or NS1.

1 5. The method of claims 1-14 wherein step (d) includes radioimmunoassayor ELISA techniques.

1 6.The method of claim 2 including cloning the hybrid cells in semi-solid agar.

17. The method of claims 2 or 16 including mass cultivation of the clone in tissue culture.

18. The method of claims 2, 1 6 or 17 including injection of the hybrid cells into the peritoneal cavity of a selected animal and recovering ascitic fluid therefrom containing the specific antibodies.

1 9. Hybridoma cells primed to produce monoclonai antibodies to antigens, said antigens consisting essentially of single synthetic carbohydrate haptens coupled to antigenic carriers.

20. The cells of claim 1 9, said hapten being a synthetic oligosaccharide coupled via a bridging arm having from 4-18 carbon atoms.

21. The cells of claim 20, said bridging arm having had 8-10 carbon atoms.

22. The cells of claims 19-21, said hapten being a human blood group determinant.

23. The cells of claim 22, said hapten being a human blood group B determinant.

24. The cells of claims 1921, said hapten being of the type of tumor-associated antigens.

25. Hybridoma cells of claims 19-24 in a culture medium therefor.

26. Monoclonal antibodies to antigens whose determinants consist essentially of single synthetic carbohydrate haptens as described in claims 5, 6 or 7.