CA1212913A - Human monoclonal antibodies - Google Patents

Abstract

ABSTRACT OF THE DISCLOSURE
Human monoclonal antibody reagents for specific therapeutic and diagnostic use against such diseases as hepatitis, rabies and tetanus can be prepared from human peripheral blood lymphocytes especially selected for their capability to bind to a selected antigen. The selected lymphocytes are infected with Epstein Barr Virus and after culture are cloned by limiting dilution on irradiated feeder layers of autologous human mononuclear cells in the presence of T cell growth factor. Clones producing the antibody to the selected antigen are recovered and grown in bulk culture and the desired reagent prepared therefrom.
Clones thus produced may be fused with thioguanine-resistant ouabain-resistant human myeloma cell lines so as to produce a hybridoma which is more stable, which has a higher cloning efficiency, and which produces more antibody.

Description

FIELD OF THE INVErlTIO~
This invention relates to human monoclonal antibodies and processes for producin~ the same. More particularly this invention relates to the production of a human monoclonal anti-body reagent against tetanus toxoid using an Epstein Barr Virustechnique.
The clones described herein may be obtained, upon written request from the Department of Microbiology and Immunology, Queen's University at Kingston, Canada where they are maintained in the cell bank of the University. The KR4 and B6 clones have also been deposited in the American Type Culture Collection (ATCC) in the ~nited States under accession Nos. HB8528 and HB852~ respectively.
BACKGROUND OF THE INVENTION
A hybridoma technique for producing monoclonal anti-bodies in mice and rats is known to the art (Kohler & Milstein Nature 256: 495, 1975; Eur. J. Immunol. 6: 511, 1976) and has revolutionized immunology in recent years. Several monoclonal reagents are available and find applicability for both thera-peutic use and diagnostic use. Heretofore, however, it has not been possible to produce high titered human monoclonal antibodies. Such antibodies are desirable and have advantages over the conventional murine myeloma fusion products for several reasons ~i) human monoclonal antibodies are prefer-able for gamma globulin therapy (i.e. rabies or hepatitis virus,tetanus, and prophylactic an~i-D Irhesus)) due to the risk of ~2~2~
, .

sensitization with xeno-antisera, (ii) autoantibodies or naturally occurring human antibodies could be used as antigens to select and develop human monoclonal anti-idiotypic anti-bodies which would potentially be useful for suppressing the response to autoantigens or transplant antigens, (iii~ con-sidered simply from the standpoint of cell markers, radio-immunoassays, HLA typing and diagnostic reagents it is clear that human antibodies offer advantages over antibody from murine x murine or murine x human hybridomas, and (iv) from the biological standpoint it i5 more relevant to make human monoclonal antibodies in order to determine the mechanism and spectrum of the human B cell specificity repertoire. Most importantly purely human monoclonal antibodies are vastly superior ~or passive therapy or anti-idiotypic regulation of the human immune response.
OBJECT OF THE INVENTION
It is, there~ore, an object oE the present invention to provide a method for the production of human monoclonal antibody reagents in relatively high yield.
Another object of the invention is to provide human monoclonal anti`body reagents specific (i) against (a) dangerous viruses such as rabies or hepatitus, (b) tetanus toxoid, (c) cancer antigens, or (d) idiotypes on autoantibodies such as rheumatoid arthritis systemic lupus erythematosus, Hashimoto's thyroiditis or multiple sclerosis, or (ii) for prophylactic anti-D use, such as Rh disease.
Another object of the invention is to provide a hybridoma comprising the monoclonal antibody of the present invention used with KR-4 or other thioguanine resistant, 12~2~13 ouabain resistant myeloma cell lines.
By one aspect o~ this invention there is provided a process for producing human monoclonal antibodies in relatively high yield from human blood comprising:
(a) selecting human B-lymphocytes, capable of binding to a selected antigen, from a sample of human blood;
(b) exposing a preparation of said selected B-lymphocytes to purified said selected antigen so as to produce a proportion of antigen specific B-cells in said preparation;
(c) killing antigen non-specific B-cells in said preparation thereby enriching the antigen specific B-cells in said preparation by negative selection;
(d) infecting said antigen specific B-cells in said enriched preparation with Epstein Barr Virus (e) cloning said EBV infected B-cells, on irradiated feeder layers of human blood mononuclear cells, by limiting dilution; and (f) recovering human monoclones secreting antibodies specific against said selected antigen.
~y another aspect of this invention there is provided a human monoclonal antibody reagent specific against a selected antigen selected from the group comprising (a) tetanus toxoid, (b) dangerous viruses, (c) cancer antigens, (d) idiotypes on autoantibodies and (e) blood group D (rhesus) antigens.
By yet another aspect of the invention there is provided a process for producing a hybridoma in which human monoclones secreting antibody specific against a selected antigen as hereinbefore described is fused with a thioguanine-resistant, ouabain-resistant human myeloma cell line.

~

~l2~L3 DESCRIPTION OF DRAWINGS
The invention will be describ~d in more detail herein-after with reference to the drawings in which:-Figure 1 is a graph illustrating secretion of antibodyversus days in culture; and Figure 2 is a qraph illustrating the rate of production of anti TT antibody by the hybridomas of the present i~vention.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS
Epstein Barr Virus (EBV~ is a lymphotrophic herpes virus which is known to selectively infect human B cells and transform them into cell lines producing a variety of poly-clonal antibodies (Rosen et al, Nature 269: 420, 1977), against TNP hapten (Kozbor et al, Scand. J. Immunol. 10: 187, 1979), or strep A carbohydratP (Steinitz et al, Immunobiology 156: 41, 1979). Thus, if human B cells that bind to an antigen of choice are selected and then infected by EBV they are "immortalized" and when grown in tissue culture, in vitro, they secrete antibody of the same specificity as their surface receptor (Nature ibid, and Scand. J. Immunol. ibid).
Clones of lymphoblastoid cell lines which are stable producers of tetanus toxoid antibody have been detected and reported by Zurawski et al ("Clones of Human Lymphoblastoid Cell Lines Producing Antibody to Tetanus Toxoid", Current Topics in Microbiology and Immunology, Vol. 81, 1978) but the monoclonality of these clones has not been established, and the yield is so low that only 10 ng/ml of anti-tetanus toxoid was detected in a sensitive radioimmunoassay. It has now been found that, by the techniques of the present invention that highly specific, highly homogeneous human ~2~ 13 monoclonal antibodies of high affinity can be produced in relatively high yield and which are extremely useful and superior for the purposes and reasons set forth above.
The invention will be described hereinafter with particular reference to human monoclonal anti-tetanus toxoid antibody and to human monoclonal anti-trinitrophenol antibody although it will be appreciated that the specific antibody is one of choice as noted hereinabove. Viral antigens (rabies, hepatitis), cancer antigens, idiotypes on auto-antibodies such as those found in rheumatoid arthritis,systemic lupus erythematosus, Hashimoto's thyroiditis and multiple sclerosis or blood group antigen D (rhesus), are all protein antigens which are, in principle, no different than tetanus toxoid described herein and for which successful derivation of monoclonal antibodies of human origin has been achieved. In every case the blood donor will already have an ongoing immune response to the said antigens. Furthermore, success in deriving human monoclonal antibodies to tetanus toxoid (Example 1) as w~ll as to the chemicall~ defined hapten, TNP, (Fxample 2) illustrates that the technique is applicable to the entire range of antigens from large to small, that might be encountered.
Human volunteers with a history of recent tetanus toxoid booster injections are boosted again with tetanus toxoid vaccine about two weeks prior to donating blood for use in the present process. A donation of peripheral blood is then taken by simple venipuncture and the lymphocyte fraction is separated by density centrifugation on Ficoll-hypaque gradients. Macrophages are removed from thelymphocyte fraction by adherence to plastic surfaces and carbonyl iron and T cells are removed by rosetting with sheep erythrocytes. The remaining B-lymphocytes are then exposed to purified tetanus toxoid followed by washing and incubation in order to allow capping and shedding of the surface immunoglobulin receptors on the tetanus toxoid specific binding B cells~ ~fter this treatment, the B-cells in the preparation which are not specific for tetanus toxoid are killed by treament with rabbit antihuman immunoglobulin and complement. This negative selection technique is believed to be the key to relatively high yields because it enriches for tetanus toxoid specific B cells, which can then be infected with EBV, such as that released by the marmoset cell line B95-8, and bulk cultured. After culturing for about 1 week the cells are cloned by limiting dilution on irradiated feeder layers of autologous mononuclear cells in the presence oE T cell growth factor (Interleukin-2). After the cloning treatment, the supernatants of wells containing growing cells are tested for the presence of anti-tetanus toxoid antibody by an enzyme linked immunosorbent assay tELISA).
Those clones producing the specific anti-tetanus toxoid anti-body are then recovered and grown in bulk cultures. The supernatants are collected and the desired antibody reagent may be precipitated with ammonium sulfate. The monoclonality of the antibody may be checked by isoelectric focussing.
Example 1 A 100 ml blood sample from a donor recently boosted with tetanus toxoid vaccine was separated by density centri-fugation on Ficoll-hypaque gradients to produce a lymphocyte ~ - 6 -~Z~
fraction. Macrophages were removed by adherence to plastic surfaces and carbonyl iron and T cells were removed by roset-ting with sheep erythrocytes. The resulting B lymphocytes were exposed to 150 ~ g/ml of tetanus toxoid antigen which had been highly purifled by molecular sieving, ion exchange chromatography, and electrophoresis, for 30 minutes at OC.
The exposed cells were then washed and incubated at 37C for 6 hours. The cells were then treated with rabbit antihuman immunoglobulin and complement to kill those cells not specific for tetanus toxoid. The tetanus binding lymphocytes were then infected with EBV and after 1 week of bulk culture the cells were cloned by limiting dilution on feeder layers of autolog-ous human peripheral blood mononuclear cells in the presence of T cell growth factor. After 3 weeks the supernatants of wells containing growing cells were tested for anti-tetanus toxoid antibody by ELISA. 105 cells/ml were then seeded into each of 4 microwells and grown ~or 7 days in culture.
Supernatants were assayed daily by ELISA using affinity column purified human anti-tetanus antibody as standard, for anti-tetanus antibody production. The results are shown graphically in Figure 1. As can be seen, 3 clones produced up to about 1000 ng/ml anti-tetanus antibody within a week whereas the fourth clone failed to produce antibody and there~ore serves as a negative control. The class of antibody produced was lgM, and monoclonality was established by isoelectric focussing.
_ ample 2 B lymphncy~es were separated from a 100 ml blood sample from a dono- sensitized to txinitrophenol (TNP).

~Z~ 3 Those lymphocytes specific for TNP were preselected by bind-ing to autologous erythrocytes coupled to TNP and were isolated by density centrifugation on Ficoll-Hypaque gradients, and then infected with Epstein-Barr virus (EBV).
A~ter 2 weeks of bulk culture the cells were cloned by limiting dilu-tion on feeder layers of autologous human blood mononuclear cells in the presence of T cell growth factor.
After 3 weeks the supernatants of wells containing growing cells were tested for anti-TNP antibody by an enzyme-linked immunosorbent-assay (~LISA) with TNP coupled to methylated bovine serum alburnin (TNP mBSA) in the solid phase. Two clones out of ~00 produced 1500 ng/ml anti-TNP antibody of the IgM class. Monoclonality was established by isoelectric focussing.
Human monoclonal antibodies offer several advantages over those of murine origin since ~hey are more likely to exhibit specificities against private rather than polymorphic human antigens, and also because the risks of xenosensitiza-tion against murine antibodies being eliminated, they are preferred for therapeutical use. However, several investi-gators have observed that the EBV ~echnique used for generat-ing human antibodies yields unstable cell lines in which specific antibody titers tend to decline progressively over extended periods of cell proliferation. On the other hand, the human x human hybridoma technique has only been performed with lymphocytes isolated from hyperimmune patients with viral infections or Hodgkin's disease. Acoordingly, there is a considerable restriction in the spectrlm ~f human antibodies that can b~come available and limitations exist ~2~ L3 in obtaining sufficient yields to afford large-scale antihody preparation. In order to overcome the difficulties inherent in each technique, the possibility of combining them together has been examined. In Example 1 above a technique where lymphocytes from noxmal donors immunized with tetanus toxoid (TT) were preselected for antigen-binding cells, subsequently transformed with EBV and cloned has been described. In Example 3 below, an EBV-transformed clone B6 was fused with a human myeloma (KR-4) to rescue high amounts cf anti-TT antibody production. The resulting hybridomas were found to be more stable, to have a higher cloning efficiency and to secrete approximately 8-fold more specific antibody than the parental B6 clone.
Example 3 -Cell _ es B6 is an anti-tetanus toxoid producing, EBV-trans-formed, cloned cell line which was previously established in the laboratory at Queen's University at Kingston, Canada, from which it is available on written request. GM1500 6TG A-ll is a 6-thioguanine-resistant, (HAT-sensitive) human myeloma which was generously provided by Dr. ~. Koprowski at The Wistar Institute in Philadelphia.
Selection of a Oua Myeloma -106 cells/ml of the human myeloma GM1500 6TG A-ll were mutagenized with either 60-150 ~g/ml ethyl methane sulEonate (EMS) (Sigma) for 24 hours, or 100-300 R of Y
irradiation . Following mutagenic treatment, cells were allowed a 10-day period of expression in normal medium. 2 x 106 cells/ml were then seeded in 96 well microtiter plates (Linbro) in 0.1 ml volumes in the presence of 10 7 M ouabain (Sigma). Cultures were fed every 4 days with ouabain contain-ing medium and wells with viable colonies were scored after 2 weeks. The surviving cells were then subcultured in 24 well tissue culture plates (Costar) and subsequently expanded in flasks (Falcon) in gradually increasing concentrations of ouabain. Dead cells were removed by the Ficoll-Isopaque method. The resulting clone KR-4 was found resistant to 5 x 10 4 M ouabain 5 months after mutagenesis.
Cell Fusion and Hybrids Selection Cryopreserved, anti-TT antibody secreting, EBV-transformed cells of clone B6 were culutred in RPMI-1640 supplemented with 20% FCS, L-glutamine (4 mM), ~-mercaptoethanol (50 ~M) and gentamycin (50 ~g/ml). Prior to fusion, the Thg , Oua human myeloma, (KR-4), was grown in the same medium as B6, supplemented with 30 ~g/ml 6-thioguanine and 10 4 M
ouabain to counterselect possible revertants.
For fusion, 107 B6 cells and 107 KR-4 cells were washed in serum-free medium and then mixed in 50 ml tubes (Falcon). After centrifugation at 150 g at room temperature, cells were fused as described by G. Kohler and Milstein (Nature (London)) 256, 495-497, 1975) with the following modifications: 0.5 ml of pre-warmed (37C), 45% (w/v) polyethylene glycol (M.W. - 4000) (Sigma) in RPMI, p~ = 7.4, was added slowly to the cell pellet over a one-minute period with gentle shaking. The cells were then incubated further for 60-90 sec and 10 ml of warm (37C) RPMI was slowly added over a period of 10 min. Cells were then incubated for 15-20 min at 37C followed by centrifugation for 4 min at 150 g.

- ~Z~

The cells were washed and resuspended in RPMI (20~ FCS~ at a density of 2 x 106 cells/ml and distributed in 0.1 ml volume to 96 well microtiter plates (Linbro). 3000 R-irradiated mouse spleen cells (0.5 x 106 cells/well) or unirradiated mouse peritoneal cells (5 x 103 cells/well) were used as a feeder layer. After 24 hours, H~T (Flow Laboratories, Inc., Canada) and ouabain (at a 5 x 10 6 M final concentration) were added. Seventy-two hours after fusion, the medium was removed and replaced with fresh HAT medium containing 10 5M
ouabain. Cells were fed every 4-5 days with the same selective medium for a period of 14 days; with a medium con-taining only hypoxanthine (10 4 M) and thymidine (1.6 x 10 5 M) for the next 7 days and finally with RPMI 1640 plus 20%
FCSn Control cultures of each parental cell line (B6 and KR-4) co~tained no surviving cells after 12 days in HAT medium plus ouabain. The first putative hybridoma cells could be detected approximately 12 days after fusion, on the basis of their size which was approximately twlce that of each parental fusion partner.
Chromosome Analysis Cells in the exponential phase of growth were treated with 0.1 ~g/ml Colcemid (Trademark of Grand Island Biological Co.
for colchicine) for 1 hour, swollen for 5-10 min in 75 mM KCl hypotonic solution at 37C and fixed twice in methanol-glacial acetic acid mixture (3:1). Air-dried slides were stained with 4% Giemsa (trade name for a commercially available hiological stain sold by Fisher Scientific Co~, New Jersey) in Gurr's buffer ~pH 6.8), according to the method of Worton and Duff (Methods of Enzymology, 58, 322-344, 1979). A minimum of 40 indi-vidual metaphases were recorded for each individual cell line.

Anti~en Preservative-free tetanus toxoid (lot AS1070) was obtained from Connaught Laboratories Ltd., Willowdale, Ontario, Canada. There were 2300 Lf/mg nondialyzed nitrogen, and the stock protein concentration was 3.5 mg/ml.
Anti-tetanus Toxoid Standard The immunoglobulin (Ig) fraction of serum from tetanus toxoid (TT) boosted human donors was generously provided by Dr. Wye of Connaught Laboratories, and was ]O purified by affinity chromatography on a TT-coated Sepharose 4B (Pharmacia Fine Chemicals) column. Eighty-one percent of the recovered anti-TT antibody was specific for TT as described previously. Titers of anti-~T antibody using this standard are therefore slightly overestimated.
Enzyme-Linked Immunoadsorbent Assay (ELISA) - Supernatants from cultures (density = 106 cells/ml) were tested for anti-TT antibody by the ELISA technique with TT in the solid phase as described in detail elsewhere. OD400 readings were converted to ng protein by comparison to the standard curve made with affinity column purified human anti-TT antibody. For detection of class specific Ig, microtiter wells were coated with 10 ~g/ml of the F(ab')2 fragment of goat anti-human Ig (Cappel Laboratories Inc.) in sodium bicarbonate buffer (50 mM, pH~9.6). Culture super-natants or standard (chromatographically purified human IgG, IgM respectively, Cappel) were diluted with 0.05% Tween in phosphate buffer saline and incubated for 8 hours at 20C.
The wells were washed and incubated overnight at 4C with 1:
lOOO dilution of heavy-chain-specific goat anti-human IgG

~2~ L3 or anti-IgM conjugated to alkaline phosphatase (Sigma Chemicals Co.) followed by 1 mg/ml p-nitrophenyl-phosphate (Sigma) in carbonate buffer containing 10.5 mM MgC12. The reaction was stopped by adding 3 M NaOH and the subsequent development of color was measured at 400 nm in a Perkin-Elmer spectrophotometer.
Surface Immunoglobulin and HLA Typing Cells were stained by immunofluorescence using class specific reagents as described in detail elsewhere. The HLA-A
and -B antigens were kindly assayed by the Immunology Laboratory of the Kingston General Hospital, Kingston, Canada using a micxocytotoxicity method, as described by Terasaki and McClelland (Nature (London), 204, 998-1000, 1964).
Cloning Cloning was performed by limiting dilution (0.3 cells per well) in 96 well microtiter plates (Linbro), on feeder layers consisting of irradiated (3000 R) mouse spleen cells (5 x 105 cell~/well). Clones were fed by replacing the medium once a week. Positive wells were screened by ELISA
for anti-TT antibody present in the culture supernatants.
Biosynthetic_Labelling of Immuno~lobulins Cells, at a density of 2 x 106 per ml, were cultured for 12 hours (viability 70-75%) and 24 hours (viability 20-30~) in methionine-free medium containing 15~ dialyzed FCS, 50 ~M ~-mercaptoethanol and 50~uCi of ~ 35S J-methionine (1004.9 Ci/mmol) per ml (lCi = 3.7 x 101 becquerels; New England Nuclear). Cells were pelleted and immunoglobulins from the tissue culture superna-tants were precipitated with rabbit anti-human IgG, IgA, IgM (Dako) by the Staphylococcus :~23L~3 aureus technique (S. W. Kessler, J. Immunol. 115, 1617-1624, 1975) modified as described previously. The immune complexes were resuspended in Laemmli buffer, analyzed on 12.5% SDS-polyacrylamide gels (U.K. Laemmli, Nature (London) 227, 680-6~3, 197Q) and subjected to flurography for 4 days at -70C. Chromatographically purified human IgM and IgG
(Cappel) were used as markers.
Selection of OuaR Variants of GM1500 6TG A-ll Myeloma Cells The human myeloma G~1500 6TG A-ll is deficient for hypo-xanthine phosphoribosyl transferase activity (E.C. 2.4.

2.8) as the result of a selection for 6-thioguanine-resistance (~hgR) previous]y performed. Cells were mutagenized using low levels of gamma irradiation (100-300 R) or ethylmethane sulfonate (EMS) (60-150 ~ g/ml) and selected for ouabain resistance (OuaR) as detailed in Materials and Methods. As shown in Table 1 below, the highest frequency of OuaR mutants (1 x 10 6j was obtained following 200 R irradiation treatment and was 5 times higher -than the OuaR frequency in untreated ; controls (2 x 10 7) (Experiment 2 and 5~ Exposure of the cells to EMS up to 150 ~ g~ml for 24 hours, the highest concen-tration tested, did not result in enrichment for OuaR variants (Experiment 4) and might be due to the naturll resistance of the GM 1500 myeloma to this alkylating agent, a phenomenon occasionally encountered with permanent tumor cell lines.
Mutant cells were grown in gradually increasing concentrations of the drug and after 5 months, one of them (KR-4) was found resistant up to 5 x 10 4 M ouabain. Since we have not tested the level of resistance at the end of each successive step, it is no-t clear whether the mutants selec-ted initially were already highly resistant to the drug or not.

~2~13 ~, ,~
o~ U~ ~ ~
~ o ~,, * al 3 0 ~ h ~J
* ~ I O
~ ta ~ u~ o o _~ u a~
r~ ~
I ~D ~ ~ ~
O ~ o ~J ~ co,~ 1 3 ~ ~ ~ O
O O O ~ ~ a~ ~n U O X . .
~ ~~ o U~ 1 0-rl a~ u, ,J o ~-~J o ~ ~ o Ul ' ~ ~ rl OIn ~ ra h ~ ~ IJ o O ~ ~`1 -- ~ U~ U ~ .~
~ .,1 Z, tn ~ ~ ~ a) a) s~
~ (~
" a) R ~ '1 ~ '1 O
* ~ O a~
.C ~ 0 le ~1 r~ rl ,~
t ~~1 IJ a) u~ l l l l l ~
U ~ ~ ; ~ ~ ~ ~ u o 1~1 ~d ~ ' ~ ; K ~; ~; ~ U ~1 ~1 0 5-1 U 1~1 u~ h a) U~
u~ a)~ ~1 ~ 1 0 a~ tn ~ u ~
~: ::~ 1 1~
~1 O . ~ ~ 0 3 ~ O
,_, , .,, a~ 3

3 r -1 o a~
i_,t~ rl rl O ~
l I
U o ~J ~1 o ~1 a) ~ 1 u ~1 D CO ~ I` In ~ _Q) ~I
~3 * x . . . . . ~ ~
O ~ U~ ~ et~ t~) O ~1 ~1 o o ~ c~ 3 H r~ O ~ ~ O
1:~ + I + I + I ~ i + I O o o ~ a~
a) ~ ~ ~ ~ u ~m ,~ ~ U ~ . ~~ r o ~ ~ U ~I-rl C~ ~ . , , . ~ rl a) o u ~ ~
oo 0 ~ 3 ~ U~rl u~ ~ ~ (~1 E~ ~ ~ + l ~ ~ a) ~
t~K . ~ 3 h U 3 O ~ 0 u, a) o P. ~ ~ o S~ N ~I h t4 0 ra ~1 ~ a.) 4~ a) ~;-,~ ~ ~1 o a~ ~ 3 0 a) 4 ~_ ~ O~ ~I rl ~ ~ ~
o ~ o , a)~
,~ ~ ~ ~ OX ~ ~ U
Ul u~ U~ ~ ~
~1 ~ ~ ~ ,_ O a)41 0 u~ h a~ 11~ 1 u~ P; P~ a) O ~ ~ u 3 ~ 3 O O O tJ
* O o o ;~ U~
U ~ ~1 ~ ~ O ~1 0 0~-rl rl ~ _~ ~ ~ O ~ ~1 ~) U r~$ t~ O ~1 ~
~ a) Lr~ o o 4~ o a~ ~ . . . ,~ ~ o ~ ~ ~a ~a ~ ~- o ,~ ,1 ,1 ~
,~ ,~ u ~ a~ 0 a) 5~ ~ ~
S I h 5~ 5-1 X U~ U~ U 11~ r~ O
E~ rl r~ r~ W
I ~ a~ 3 a) ~ ~ O
O r~ ,~
O ~
~D ~ U a~ u u a~ O o u~
~ 0 4~O ~ U
~1 0 ~-1 ~ ~ ~r ul ~ Q
~I Z, ~ O U (1~ U~
a) ~ ~ ~ ~ O ~ 3 ~:1 ~t* *
**

.. ~

.Z~3~3 Selection of Hybrid Cells One Oua myeloma clone isolated from each of the five experiments listed in Table 1 was fused with the EB~-trans-formed B~ clone in the presence of polye-thylene glycol and hybrids were selected in a HAT medium, supplemented with 10 5 M ouabain. This intermediary concentration of ouabain was chosen since OuaR is known to beha~e as a co-dominant trait in somatic cell hybrids. The frequency of hybrid formation between different myeloma variants (KR-l to KR-5) and B6 was estimated twelve days after selection was initiated, by scoring the fraction of negative wells and applying the Poisson's equation. The highest hybridization frequency (1.10 5) was obtained with the KR-4 cell line possibly due to its high level of ouabain resistance (Table 1). No clone arose from control plates containing each of the parental cell lines grown in the same selective medium. Fifty-six hybrids were expanded and four of them were subcloned by limiting dilution.
Characterization of Hybrids As shown in Table 2 below, the parental B6 cell line has a modal chromosome number of 46 (range 40-51) similar to that o~ the KR-4 myeloma (range 40-50)~ Approximately 3-4%
of the metaphases in both cell lines had a near-tetraploid karyotype in the range of 84-91 chromosomes. Nine independent hybrids and 12 independent subclones obtained by limiting dilution were examined 2 months after isolation for chromosomal content. Although some variation was noticed, due to the chromosome number sca-ttering of each parental cell line, all hybrids were near tetraploid (range 74-96), attesting -that only moderate chromosomal segregation took place.
However, in only a few hybrids was a mode clearly identifi-able. The best examples are HYB-20.42 and HYB-10.3 which had a mode of 91 and 92 chromosomes, respectively, very close -to the expected tetraploid number (Table 2~. No clear chromosome number difference was found between secretor and non-secretor hybrid subclones.
Further confirmation of the hybrid nature of the clones came from an analysis of ~LA antigens expressed on parental cells and hybridomas. As expected, several HLA
haplotypes were co-dominantly expressed in the hybrids (Table 2). In addition, KR-4 was found to secrete IgG and B6 secreted IgM exclusively whereas KR-4 x B6 hybrids produced both IgM and Ig~, in agreement with the class of Ig detected on the cell surfaces by immunofluorescence (Table 2).
Proteins produced by each cell line were biosyntheti-cally labelled with 355-methionine and the immunoglobulins were precipitated with class specific rabbit anti-human Ig and Sta~h~lococcus aureus protein. Precipitates were eIectro--phoresed in lZ.5% SDS-polyacrylamide gels and a representative KR-4 x B6 hybrid clone secreted high amounts of both~ and ~
chains from each parent, as well as L chain after 24 hours.
The KR-4 parent secreted low levels ofy and L chains whereas the B6 paren-t secreted a large amount of~lchain and little L
chain. The co-expression of Ig heavy and light chains characteristic of each parental cell line further substantiates the hybrid nature of the selected fusion products.

~2~ 3 C~ _.
,~ o . ~ r~
o~ , ,, o~
~n + + + I + I+ ~ + + + + ~ u Q) I
o r-l ~ 1~1 U ao U~
ra ., i-l . N
R o a~ r-l + + + I -~ I + + + + + + +
a) 00 u~ In I O _ r ~1 0 X
m ~ In r-l O a~
~_I ~I I C~
r~ +++I+I ++ +++++
o a~ 0~
~ ._ r-l 1~1 ~1 r-l ~ U

Pi r-l u~ O
rcs ~ I O O ~
S~ l O +I+I+I I+ ++I++ Z
0 ~ ~r o o ~ ~_ rO
r~ ~ ~
R ~ In ,~ ~ ,~
R O In c~
i 1` r~ o ~D ~ + + I I + I + I + I + I I ~
o m ~r O O
0 ~
,1 ~r ~
h ~ ~ *
U ~ *
*
~ lt * # *
0 ~ * # ~ * *
a ~ ~ ~c # ~c C~ S~ ~ ~ W * ~ *
0 ~ ~ *
r-l 0 r~ I H r-l * o~O
O
~, r~ C ~ t~ ~1 0 a~ ~ I o O ~ H H H ~ r-t W S r~ r-t ~.
~ O ~ H ~ ~ ~ ~:
.,, ~ , I, I I I _ ~ ~ a /~ o o r~ t ta ~ # ~1 S~ U~ ,~ +
E~ O ~ ~ ~)~ 41 ~ # - r t .
~ H O~) O *
O ~ h-rl S~
~~ a) o # ~
S~ rt ~ r-l ~ r-l 3 O u~ 0 ~ m m r~ h O
rl r-l h r-i I ::S I I I I I r t ~ 0r~t ~ r~ ~
r~ ~) 0 ~
O O ~ I r-l ~ ~\o\t) ~ $

-" lZ~ 3 Table 2 (continued) *A minimum of 40 metaphase spreads were counted for each cell line.
**The isotype of surface Ig was determined by immuno-fluorescence using class specific anti-Ig reagents.
In general all hybrids and KR-4 exhibited weak in~unofluorescence (Leitz UV microscope)whereas B6 was strongly fluorescent.
***Total secreted Ig was determined in an ELISA assay with F(ab')2 goat anti-human Ig in the solid phase.
****Specific antibody was determined using an ELISA
with tetanus toxoid in the solid phase. Cells were at the steady state or declining phase of growth (day 9).
*****Determined fro~ experiments described in Figure 2 during logarithmic phase of growth.
~*****HLA haplotypes were determined by antibody and complement cytolysis in Terasaki plates.
*******Cloning was performed by limiting dilution on feeder layers, inoculating 0.3 cell per well.

Antibody Production Fifty-four out of 56 hybrids between KR-4 and B6 were positive when screened by ELISA for production of anti-TT antibody. The four most positive hybrids were subsequently cloned by limiting dilution. Out of 395 subclones, 23 did not produce detectable amounts of anti-TT antibodies, the majority produced varying levels and seven secreted high levels (3-6~ug/ml) of IgM-kappa anti-TT antibody. The mean cloning efficiency of three hybrids (64%~ was at least 2-fold higher than the cloning efficiency of the B6 line (Table 2).
Three of these hybridoma clones selected for further studies produced 3, 4, 6 ~g per ml of anti-TT anti-body at the end of a growth period of 9 days, starting from an inoculum of 2 x 104 cells per ml. The B6 parent produced

4- to 8-fold less anti-TT antibody and KR-4 produced none (level of detection = 1 ng/ml). In order to estimate the 1'~ .3 rate of production of anti-TT antibody, a mode of represent~~
tion which was pioneered by Monod et al to measure the differential rate of synthesis of inducible enzymes in bacteria was adopted~ In this representation, the rate of production is given by the slope of the curve obtained by plotting the increment of antibody titer as a function of the increment of cell density. This type of representation may also be more appropriate to monitor the antibody secretion in large-scale productions. The rate of antibody production during the logarithmic phase of cell proliferation was 13-to 18-fold higher in the hybrids than the B6 parental line.
As shown in Figure 2~ the rate of antibody production was not constant but exponentially increasing for all cell lines tested. This is not unexpected since production of antibody in the supernatant is the terminal stage of a sequential process which involves synthesis, assembly, and secretion, all of them being programmed and co-ordinated during the cell cycle. The B6 parent was unusual in that higher titers were detected during the declining phase of cell growth, possibly due to the release of intracellular Ig from dying cells.
The KR-4 x B6 hybrids were stable in terms of antibody productlon for 3 months to the present ~ime whereas the EBV transEormed B6 clone titer of anti-TT antibody invariably declined during this period.
Example 3 above clearly demonstrates that high amounts of monoclonal antibody can be rescued from EBV-transformed B cells by fusion with a human myeloma, and fusion of a OuaR, ThgR human myeloma with EBV-transformed - ~Z~2~3 cell lines in a preferred method for the production of human monoclonal antibody because (i) the resulting hybrids are stable, have a higher cloning efficiency and produce antibody at a rate at least one order of magnitude higher than parental E,BV-transformed parental cells; (ii) there is little restriction as to the type of antibody produced since the antigen-specificity of the molecule is determined initially by the choice of vaccinated human donors; and (iii) rare antigen-specific B cells in the peripheral blood could conceivably be expanded by EBV transformation prior to fusion.

Claims (16)

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A process for producing human monoclonal antibodies in relatively high yield from human blood, comprising:
(a) selecting human B-lymphocytes, capable of binding to a selected antigen, from a sample of human blood;
(b) exposing a preparation of said selected B-lymphocytes to purified said selected antigen so as to produce a proportion of antigen specific B-cells in said preparation;
(c) killing antigen non-specific B-cells in said preparation thereby enriching the antigen specific B cells in said preparation by negative selection;
(d) infecting said antigen specific B-cells in said enriched preparation with Epstein Barr virus;
(e) cloning said EBV infected B-cells, on irradiated feeder layers of human blood mononuclear cells, by limiting dilution; and (f) recovering human monoclones secreting antibodies specific against said selected antigen.

2. A process as claimed in claim 1 wherein said antibody secreting human monoclones are selected by enzyme linked immunosorbent assay.

3. A process as claimed in claim 1 wherin mono-clonality of said monoclones is tested by immuno-iso-electric focussing.

4. A process as claimed in claim 1 wherein said antigen is a protein antigen selected from viral antigens, cancer antigens, idiotypes on autoantibodies, and blood group D (rhesus) antigens.

5. A process as claimed in claim 1, 2 and 3 wherein said antigen is tetanus toxoid.

6. A process as claimed in claim 4 wherein said viral antigen is selected from rabies and hepatitis.

7. A process as claimed in claim 4 wherein said idiotype is from autoantibodies in rheumatoid arthritis, systemic lupus erythematosus, Hashimoto's thyroiditis and multiple sclerosis.

8. A process as claimed in claim 4 wherein said antigen is blood group antigen D (rhesus) which is involved in Rh disease.

9. A human monoclonal antibody reagent specific against a selected antigen selected from the group comprising (a) tetanus toxoid, (b) dangerous viruses, (c) cancer antigens, and (d) idiotypes of autoantibodies, whenever produced by the process of clams 1,2 or 4 or its obvious chemical equivalent.

10. A human monoclonal antibody reagent for prophylactic anti-D use, whenever produced by the process of claim 1, 2 or 4 or its obvious chemical equivalent.

11. A human monoclonal antibody reagent for use against Rh disease, whenever produced by the process of claim 8 or its obvious chemical equivalent.

12. A human monoclonal antibody reagent specific against a virus selected from hepatitis and rabies, whenever produced by the process of claim 6 or its obvious chemical equivalent.

13. A process as claimed in claim 1, 2 or 3 wherein said monoclones are grown in bulk culture and an antibody reagent prepared therefrom by precipitation from culture supernatant with ammonium sulfate.

14. A process as claimed in claim 1, wherein said monoclones from step (f) are fused with a thioguanine-resistant, ouabain-resistant human myeloma cell line, so as to produce a hybridoma.

15. A process as claimed in claim 14 wherein said hybridomas are selected by growth in a medium containing hypoxanthine-aminopterin-thymidine and ouabain and cloned by limiting dilution.

16. A hybridoma produced by the process of claim 14 or 15.