Human Anti-Rh(D) Monoclonal Antibodies
The present invention relates to human monoclonal antibodies to the Rh(D) antigen of human red blood cells. In particular, it relates to such antibodies of the IgG1 sub-class which may be used to detect not only the normal Rh(D) antigen, but also important variants of this antigen.
Of the antigens of the so-called Rh blood group system, the Rh(D) antigen is responsible for some of the most severe reactions following transfusion to a patient with corresponding antibody. Since an Rh(D-) individual with anti-Rh(D) who receives Rh(D+) blood is liable to suffer substantial red blood cell (RBC) destruction due to the Rh(D) phenotype incompatibility, blood of donors and blood transfusion recipients is routinely classified as Rh(D+) or Rh(D-) by agglutination tests with anti-Rh(D) antibody. The Rh phenotype of RBCs is commonly further defined with reference to the Fisher-Race system, which is based on the assumption that the inheritance of the Rh antigens is determined by three pairs of allelic genes, C-c, D-d and E-e, acting at very closely linked loci. According to this theory, a person may inherit a set of three Rh genes from each of his parents (i) C or c , (ii) D or d , (iii) E or e (no d antigen has as yet been identified, but the symbol 'd' is used to indicate the presence of a gene, allelic to the D gene, which does not produce D antigen). For example, an Rh(D+) person may inherit CDe from one parent and cde from the other. The frequencies of the commonest Rh gene combinations as determined with reference to the Fisher-Race system for an English population, together with the 'short symbols' which
are used, particularly in speech, are given in Table 1 below.
Despite expansion over the years, the FisherRace system has not been adequate to account for all the reactions that have been observed with the Rh system (Mollison, P.L. (1983) Blood Transfusion In Clinical Medicine, 7th edn., Blackwell Scientific, Oxford). Nevertheless, the World Health Organisation has recommended that in the interest of simplicity and uniformity this nomenclature should be universally adopted and all Rh genotypes given hereinafter are defined on the basis of the conventional FisherRace system.
In addition to the need for anti-Rh(D) antibody for Rh-typing of RBCs, such antibody is also importantly required for passive immunisation of Rh(D-) mothers to prevent haemolytic disease of the newborn (HDN). This condition arises in newborn Rh(D+) infants of Rh(D-) mothers previously sensitized to Rh(D) antigen as a result of IgG anti-Rh(D) antibodies crossing the placenta during pregancy and causing
foetal RBC destruction. Sensitization of the Rh(D-) mother to Rh(D) antigen may have occurred at the birth of an earlier Rh(D+) child due to some foetal RBCs entering the maternal circulation and being recognised by the maternal immune system. To reduce the incidence of HDN, it is routine practice in the United Kingdom and many other countries to give anti-Rh(D) antibodies to Rh(D-) mothers immediately after the birth of an Rh(D+) infant so that any Rh(D+) RBCs which have entered the maternal circulation are rapidly removed (Mollison, P.L. (1983) loc. cit.; Laros Jr., R.K. (1986), "Erythroblastosis Fetalis" in "Blood Group Disorders In Pregnancy", Ch. 7, p. 103). At the present time, anti-Rh(D) antibody for use in both Rh-typing of RBCs and passive immunisation of Rh(D-) mothers is largely obtained directly from female donors immunised during pregnancy or from immunised male volunteers. The success of the programme of post-partum prophylactic administration of human anti-Rh(D) immunoglobulin to Rh(D-) women has, however, resulted in a dramatic reduction in the number of naturally alloimmunised women (Urbaniak, S.J., "RhD haemolytic disease of the newborn: the changing scene", Br. Med. J. (1985)
291, 4-6). Also, deliberate immunisation of individuals with Rh(D+) RBCs carries the risks common to receiving any transfusion of RBCs, e.g. risk of transmission of hepatitis viruses and HIV. Hence, there is much interest in obtaining human monoclonal antiRh(D) antibodies for both diagnostic and therapeutic purposes.
As stated above, in routine blood testing, blood types are divided into Rh(D+) and Rh(D-) on the basis of the apparent presence or absence of Rh(D) antigen on the RBCs as indicated by agglutination tests with anti-Rh(D). However, a small number
of persons with apparently Rh(D-) blood have RBCs that are not directly agglutinated by anti-Rh(D) during such routine testing, but that do react when the D-typing test is performed using selected anti-Rh(D) reagents by the indirect antiglobulin test. Cells thus identified are designated Du. The frequency of the Du phenotype is about 0.2% overall, 0.6% among Caucasians, and about 1.5% of all Rh(D-) gravid women. At least three different mechanisms may be responsible for the expression of the Du phenotype: (1) hereditary absence of a portion of the complete Rh(D) antigen, (2) gene interaction with suppression of D by C in the trans position, and (3) a D gene producing a weak antigen. In the early 1950s, reports first appeared of the presence of anti-Rh(D) in individuals of the Du phenotype following blood transfusion with Rh(D+) blood or pregnancy resulting in the birth of a Rh(D+) infant. It later became apparent that in some individuals whose blood is classified Rh(D+) parts of the Rh(D) antigen are missing from the RBCs. When exposed by transfusion or pregnancy to Rh(D+) RBCs carrying the complete Rh(D) antigen, persons carrying an incomplete Rh(D) antigen on their RBCs are capable of making alloanti-D against the Rh(D) antigen portion they lack. The blood of such individuals is called D variant when the RBCs react directly with routine anti-Rh(D) reagents or Du variant when the cells react only by the indirect antiglobulin technique.
The observation that allo anti-Rh(D) can be produced in patients who have Rh(D+) RBCs has led to common usage of the term "D mosaic" to describe the Rh(D) antigen in its complete native form. Routine anti-Rh(D) reagents, generally cannot differentiate those RBCs that lack part of the D mosaic from those that have all the D components.
The D variant phenotypes have been categorised by Tippett and Sanger (Vox. Sang. (1962)7, 9-13). This system is based on the interaction of RBCs and serum from D- and Du variant individuals. The six categories (see Table II below) allow for expansion; subdivisions are already recognised in categories III, IV and V. Categories I and II have been found to have so many similarities that they are now generally considered as a single sub-group.
Table II
Tippett and Sanger categories for D- or
Du positive blood with anti-Rh(D)
Category Racial origin Usual haplotyp*
I White DCe
II
IIIa Black
IIIb Usually Black Dce
IIIc White
IVa Mostly Black, some White
IVb White Dce Va Black and White
Vb White DuCe
Vc Black and White
VI Nearly all White DuCe
An alternative, but lesser used, classification by Wiener uses letters A,B,C,D instead of Roman Numerals. Although there is no direct correlation between the two systems, it is often considered that DB and DV1 are interchangeable. Although the frequency of D and Du variant individuals within the human population is relatively low, the total number of individuals of these blood types who potentially have some risk of effective
anti-Rh(D) formation as a result of exposure by blood transfusion or pregnancy to non-variant Rh(D+) cells is far from insignificant. Moreover, in addition to Rh(D-) women who give birth to Rh(D+) or Du infants, Du variant women who give birth to an Rh(D+) infant may also benefit from postpartum anti-Rh(D) treatment to reduce the risk of HDN (White, C.A. et al. (1983) Am. J. Obstet. Gynecol. 145, 1069-1073). Anti-sera capable of distinguishing D and Du variant RBCs are not widely available. Hence, provision of anti-Rh(D) monoclonal antibodies with a range of binding specif icities for D and Du variant RBCs is seen as useful in enabling the more ready identification and categorisation of individuals possessing such cells (especially D or Du variant pregnant females who are suitable candidates for prophylactic anti-Rh(D) treatment) as well as for providing further structural information on the Rh(D) antigen complex. Human monoclonal anti-Rh(D) antibody production has previously been achieved by:-
(a) directly cloning Epstein Barr virus transformed B lymphocyte cell lines (hereinafter referred to as EBV-transformed LCL) derived from B lymphocytes of anti-Rh(D) positive donors (see GB-A 2127434; Crawford et al. (1983) Lancet 1, 386-388 and Paire et al (1986) Immunol. Lett. 13, 137-141),
(b) cloning hybridoma cell lines formed by fusing anti-Rh(D) producing, EBV-transformed LCL with mouse, mouse-human or human myeloma cell lines (see co-pending British application no. 8709748, Thompson et al. (1986) Immunol. 58, 157-160 and EP-A-0162918) , or
(c) by fusion of a human LCL with immune B cells (Lowe et al (1986) Vox. Sang. 51, 212-216).
By cloning EBV-transformed LCL from anti- Rh (D) positive donors , we have been able to obtain,
however, monoclonal anti-Rh(D) antibodies of the IgG class which have a particularly useful binding specificity spectrum not shown for any previously disclosed anti-Rh(D) monoclonal antibody reagent. According to the present invention, we provide human monoclonal antibodies having the following essential characteristics:
(a) exhibiting activity against Rh(D) antigen, but not C, c, E or e antigens of the Rh blood group system;
(b) being IgG1 proteins;
(c) having kappa light chains;
(d) being of the allotype Glm (1, 2, 17);
(e) exhibiting activity against D V, DTar ,DVI and DB variant antigens; and
(f) being substantially non-reactive with non- papam treated DIV cells in an IAG test, and antigen-binding fragments thereof.
In addition monoclonal antibodies of the invention have been shown to exhibit activity against
DU, and papain-treated DIV RBCs in an IAG test and to be inactive against the R =N antigen.
It has been established that amongst individuals classified as D-positive or Du by a conventional agglutination test, but who are capable of making anti-D, a high percentage have the D VI or DB variant antigen (Mollison, P.L. (1983) in "Blood Transfusion In Clinical Medicine", Ch. 8, p 339). The antibodies of the present invention will not only importantly enable classification of such individuals as RhD+, but will also bind a number of other important D variant antigens. Such monoclonal antibodies may be employed as routine anti-D reagents to classify
RBCs as D-positive, Du or D-negative and may also find use in identifying individuals of the DIV type.
Preferably, for use in Rh-typing an anti-Rh(D) monoclonal antibody of the present invention will be combined with one or more additional anti-Rh(D) monoclonal antibodies having one or more additional binding specificities, including anti- DIV activity, e.g. an IgG1 anti-Rh(D) monoclonal antibody of our co-pending International Application of even date herewith claiming priority from GBA 8722018 or an IgG3 anti-Rh(D) monoclonal antibody of our co-pending International Application of even date herewith claiming priority from GB-A 8722019, the contents of the aforementioned applications being incorporated by reference into the present specification. Thus, for example, an anti-Rh(D) monoclonal antibody of the present invention, especially the monoclonal antibody of the cell line hereinafter designated B7, may be blended with an IgG1 or IgG3 anti-Rh(D) monoclonal antibody having the following binding characteristics:
(a) exhibiting activity against Rh(D) antigen, but not C, c, E or e antigens of the Rh blood group system;
(b) exhibiting activity against Du cells by an indirect antiglobulin test, (c) eexxhhiibbiittiinngg aaccttiivviittyy aagainst D IV, DV and DTar variant antigens; and (d) being inactive against D VI and DB variant antigens, e.g. a monoclonal antibody selected from the monoclonal antibodies of the deposited EBV-transformed lymphocyte cell lines ECACC 86091604, ECACC 86091605 and ECACC 86091606.
A monoclonal antibody of the present invention, may also be of particular value for use in an anti-Rh(D) typing reagent additionally including an IgM anti-Rh(D) with weak or no anti-Du activity, i.e. insufficient activity against Du cells to be able to reliably
distinguish such cells from D-negative cells in a conventional agglutination test. Such a combination anti-Rh(D) reagent will contain the following antibody components:
(a) an IgM monoclonal antibody as hereinbefore defined, preferably an IgM monoclonal anti- Rh(D) selected from the monoclonal IgMs of the deposited hybridoma cell lines MAD-2 (ECACC 86041803) and FOM-1 (ECACC 87021301) , which form inter alia the subject matter of published European Patent Application 0251440;
(b) an antibody of the present invention, especially the monoclonal antibody of the cell line hereinafter designated B7; and optionally (c) one or more further IgG monoclonal anti-Rh(D) antibodies which individually exhibit activity with Du red cells by the indirect antiglobulin test, such that the blended reagent reacts by the same test with Du, DIV, DV and DVI cells. When Rh-typing is carried out with such a reagent having the binding capability defined in
(c) above, D-positive cells will firstly be directly agglutinated by the IgM anti-Rh(D). The remaining non-agglutinated cells (apparently D-negative) may then be subsequently divided into truly D-negative and Du cells by addition of conventional Coomb's reagent for an antiglobulin test, whereupon Du cells binding IgG antibody will be agglutinated and thus distinguished. The monoclonal anti-Rh(D) antibodies of the invention can be made by conventional methods known for the production of monoclonal antibodies and in particular by the culture of EBV-transformed human B-lymphocytes selected on the basis of secretion
of anti-Rh(D) immunoglobulin having the characteristics set out above for the required antibodies. The culture supernatants so produced constitute a further feature of the present invention. We have now investigated in detail 3 cloned EBV-transformed lymphocyte cell lines which produce IgG1 anti-Rh(D) monoclonal antibodies as defined above, hereinafter referred to as B7, B8 and B9. All these cloned cell lines were obtained by starting with peripheral B lymphocytes from a single anti-Rh(D) donor and carrying out the following steps (i) infecting the peripheral B lymphocytes with EBV;
(ii) culturing the EBV- transformed cells in RPMI- 1640 medium supplemented with 10% (v/v) phytohaemagglutinin (PHA); (iii) enriching for anti-DVI producing cells by rosetting four times at 3-4 weekly intervals with DVI RBCs; and (iv) cloning cells producing anti-Rh(D) of the desired D variant specificity.
In continuous culture using RPMI-1640 medium supplemented with 10% (v/v) mycoplasma free-foetal calf serum, 0.2 mg/ml arginine and antibiotics to prevent mycoplasma growth, the above-mentioned specific cell lines have been found to be highly stable and to provide a culture supernatant suitable for use in Rh-typing without the need for concentration. Indeed, these culture supernatants may be diluted for use.
Table III below gives the results of indirect anti-globulin tests (IAG) tests (3% RBCs in low ionic strength saline) with untreated supernatants from continuous cultures of the above-mentioned specific clones. The degree of agglutination was graded in conventional manner on a scale of 0 to 6.
Table IIIa
Culture B7 B8 B9
Supernatant IAG reactivity of supernatant with 3% RBCs in low ionic strength saline.
(Grade: 0 to 6)
RBC phenotype R1R1 6 5 6
R1R2 6 5 5
R2R2 6 6 6 R1 uwr' 6 5 5
Ror 6 5 6 r'r 0 0 0 r"r 0 0 0 rr 0 0 0
Table Illb gives further data in respect of monoclonal antibody B7 :
Table IIIb
IAG titre v. R1R1 = 256
IAG titre v. R1 Br = 128
Microtitre v. R1R2 = 1 X 106
Anti-Rh(D) = 3 .3Iu/ml
IgG = 0 . 8 ug/ml
It was further found that those monoclonal antibodies of the invention that were tested reacted with RBCs of phenotype R2rG-, hrs-, R1Rz and R2Rz but are negative with r"Gr, rm, rG, r'sr, hrB-, r'wr and Rh33+.
According to a still further aspect of the present invention, we thus provide a method of Rh-typing of RBCs wherein an aqueous solution of a monoclonal anti-Rh(D) immunoglobulin of the present invention is employed. The monoclonal immunoglobulin is preferably contained in a culture supernatant which may be used directly or, more usually, after dilution. As hereinbefore indicated, it may be desirable to blend an IgGl antibody of the present invention with one or more further anti-Rh(D) monoclonal antibodies of different specificity, e.g. a further IgG antibody having anti-DIV activity. Suitable diluents include physiological saline or phosphate buffered saline advantageously containing bovine serum albumin and a surfactant or suspending agent such as Tween 80 or methyl cellulose.
Monoclonal antibodies of the present invention may also find use in the provision of a prophylactic reagent for use in post-partum prevention of HDN. Thus, according to a further aspect of the present invention, we provide a monoclonal IgG1 antibody as hereinbefore described for use in passive immunisation of an Rh(D-) or D or Du variant mother after the birth of an Rh(D+) child. Generally, for this purpose a monoclonal antibody of the present invention will be blended with one or more further anti-Rh(D) monoclonal antibodies with different characteristics, e.g. one or more anti-Rh(D) monoclonal antibodies of the IgG3 sub-class and/or capable of exhibiting activity against DIV cells. A sterile solution of an antibody according to the present invention suitable for human injection may be formulated in any physiologically acceptable aqueous medium, for example isotonic phosphate-buffered saline or serum. Alternatively, such an antibody may be supplied in a freeze-dried formulation ready for reconstitution prior to use.
B7 of the above-mentioned EBV-transformed lymphocyte cell lines was deposited on 16th September 1987 at the European Collection of Animal Cell Cultures, Porton Down, U.K under accession no. ECACC 86091603.
Further details of the preparation of this deposited cell line and identifying characteristics of its continuous culture supernatant are provided in Example 1 of the following non-limiting Examples.
EXAMPLE 1
(i) Establishment and Cloning of Anti-Rh (D) Producing EBV-transformed LCL
(a) Source of B Lymphocytes
The donor employed was a male, initially immunised by transfusion in 1966, boosted 6 times since and last boosted 13 days before donating a "buffy coat" fraction (white cells) in 1985 when his serum anti-D level was 318lU/ml.
(b) Establishment of cell lines
Peripheral blood mononuclear cells from the chosen donor were separated on Lymphoprep (Nyegaard and Co) incubated in the presence of EBV (1 ml culture supernatant from filtered mycoplasma free B95-8 cell line per 107 cells) at 37°C for 1 hour and washed in phosphate-buffered saline (PBS). Aliquots were enriched for surface anti-Rh(D) positive lymphocytes by rosetting with bromelain-treated OR1R2 (CDe/cDE) RBCs and then plated at 1.0×106 cells/ml in 2 ml wells using lymphoblastoid cell culture medium (RPMI-1640 medium containing 10% (v/v) mycoplasma free fetal calf serum (FCS), 0.2 mg/ml arginine, 100 IU/ml penicillin (Glaxo), 50 μg/ml streptomycin (Glaxo), 25 IU/ml polymixin (Glaxo), 25 μg/ml kanamycin (Gibco), 20 μl/ml fungizone (Squibb), 25 μg/ml gentamycin sulphate (Sigma) and 20 μg/ml trobicin (UpJohn)), supplemented with 1% (v/v) phytohaemagglutinin (PHA).
All the cultures were subsequently incubated at 37°C in 5% CO2, 95% humidified air. Medium changes were performed every 3 to 4 days and, after 3 weeks culture, the cells were transferred to
50 ml flasks. Cell lines were enriched by rosetting four times with R1 VIr cells at 3-4 weekly intervals.
(c) Cloning
Cells were plated out at limiting dilution at 1 and 5 cells per well in flat bottomed 96-well plates over a feeder layer of mouse peritoneal macrophages (Doyle et al. (1985) Human Immunology 13, 199-209). Cultures were fed once a week and after 3-4 weeks cloned cells positive for antiD were grown up.
(d) Quantitation of anti-D activity and
IgG in culture supernatants Anti-D activity in the supernatants was quantified against British national standards by Auto Analyser. The quantitative estimation of IgG was performed by ELISA (modification of the method of Wakefield et al. in Clin. Chim. Acta. (1982) 123, 303-310) with at least eight determinations for each supernatant. Coating antibody (affinity purified goat anti-human IgG (Sigma)) was used at 1/200 in 0.05M carbonate buffer pH 9.6. Supernatants and standard (purified human IgG (Sigma)) were diluted in RPMI 1640 + 10% FCS. Peroxidase-conjugated goat anti-human IgG (Sigma) was diluted 1/500 in PBS + 0.05% Tween 20 and the substrate was TMB (3,3', 5,5'-tetramethyl benzidine).
(e) Immunoglobulin class and subclass determination
An immunodot assay (McDougal et al. (1983) 63, 281-290) was used to determine the reaction of the monoclonal anti-Rh(D) antibodies absorbed to nitrocellulose with anti-IgG, anti-IgM, anti-
kappa and anti-lambda antiserum (Serotec); positive reactions were detected with peroxidase-conjugated anti-sheep IgG (Serotec) followed by colour development with 4-chloro-1-naphthol. The IgG subclass was evaluated by agglutination of anti-Rh(D) coated
RBCs by monoclonal anti-subclass antibodies (Unipath).
(f) SDS-Page and Western Blotting. Iscove's supernatants (serum free) were electrophoresed under reducing conditions on 15% polyacrylamide gels (Laemmli, Nature (1970) 227, 680-685. The separated proteins were then electrophoretically transferred to nitrocellulose membranes (Brunette, Annals Biochem. (1981) 112, 195-203), which were probed with anti-IgG antiserum (Serotec) and detected as above.
(g) Protein A absorption
2 ml volume of supenatants were run twice down a 25 mm (1ml) column of Protein A Sepharose C1-4B (Sigma) and absorption of anti-Rh(D) assessed by titration. The IgGl anti-Rh(D)s of cell lines B7, B8 and B9 were all found to be absorbed by Protein A.
(h) Gm allotyping
RBCs were coated with the monoclonal anti- Rh(D) antibodies and agglutination assessed using panels of Gm allotyping reagents (Birmingham or Amsterdam).
(ii) Serology
Culture supernatants from continuous cultures of B7, B8 and B9 were tested by the IAG test using
rabbit anti-human IgG and 3% R1R1, R1R2, R2R2, R1 uwr', Ro ur, R1 Br, r'r, r"r or rr cells in low ionic strength saline (see Tables IIIa and IIIb). The same supernatants were also tested against a panel of D variant red blood cells under the same conditions (see Table IV below). The degree of agglutination was graded in conventional manner on a scale 0 to 5 or 0 to 6.
Table IV
Reaction of monoclonal anti-Rh(D) antibodies with "partial" D positive red cells using IAG
Culture supernatants B7 B8 B9
RBCs
DIV(non- - - - papain treated)
DV + + +
DTar + + +
DVI + + +
DB + + +
(iii) Tissue Typing and Karotype analysis The results of tissue type and karotype analysis of the cell lines B7, B8 and B9 as set out in Table V below confirmed that the cell lines were derived from B lymphocytes of the donor B described in Section (i) (a) above.
Table V
Cell Line B7 B8 B9
Donor B Sex male
Tissue Type HLA: A 1,3
B 8,35
DR 1,3
Karyotype sex chromosomes xy Ploidy most cells diploid
EXAMPLE 2
Solution for Rh(D)-Phenotyping of RBCS.
In general, it is preferred for the above purpose to use a blend of an anti-Rh(D) monoclonal antibody according to the invention (e.g. B7) with a further IgG anti-Rh(D) monoclonal antibody, for example B11 of our copending International patent application No. of even date herewith claiming priority from GB-A 8722018.
Solution for Manual Use
The final blend is 1:1:1 B11:B7: diluent.
Diluent 100 ml 30% Bovine Serum Alubumin 2.42g KH2PO4 2.77g Na2HPO4.2H2O 4.50g NaCl 0.2 ml Tween 20
1.00g NaN3 To 1.0 litre with distilled water: pH 6.8
This blend can be used in all manual tests for D and Du typing, e.g. microtitre, microplate, IAG.
A blend of 1:1:2 B11:B7: diluent may also be used.
Blend for Machine Use
This blend may be used, for example, in a Technicon autogrouper 16C
Pre-blend reagents B11:B7 1:1
For D-Phenotyping (D-positive v. D-negative) the solution comprises 1:1000 blend: diluent, and for Du determinations (Du v. D-negative) the solution comprises 1:5 blend: diluent.
Diluent 2% Bovine Serum Albumin in 1.3% physiological saline containing 13.5% methylcellulose.