camel immune response so that diagnosis and control of infectious diseases, ... (Sigma) medium with 20% gamma-globulin free horse serum, in addition to pen-.
Veterinary immunology
Veterinary Immunology and Immunopathology 45(1995)175-184
ELSEVIER
and immunopathology
Monoclonal antibodies against camel (Camelus dromedarius) IgG, IgM and light chains S.M. Azwai, S.D. Carter*, Z. Woldehiwet Department
cf l’eterinar?, Pathologyand Deparrrtmt qfl’eterir~ar:~ Clir~~ul Sucr~~~md. Husbandry. University qfliverpool. P.O. Bo.\- 147. Liverpool L69 3B.K I.K
Ittttttd
Accepted I? April I994
Abstract
Monoclonal antibodies specific for camel IgG and IgM heavy chains and immunoglobulin light chains were produced by a simple, time-saving and efficient method. Poplitcal lymph nodes isolated 9 days after a primary foot-pad immunisation were used as the source of antibody producing hybridoma cells. Ascites was induced in mice and ascitic fluid collected. The specificity of anti-IgG and anti-IgM monoclonal antibodies was determined b)
enzyme-linked immunosorbent assay (ELISA) and Western blotting. Monoclonal antibodies specifically reacting to IgG, subclasses were demonstrated; no monoclonal antibodies specific to IgG2 or IgG, were generated. An unexpected finding was that some classspecific monoclonal
antibodies
were light chain, and not heavy chain, reactive.
1. Introduction Dromedaries form an integral part of the desert environment, where their ability to survive and produce milk, meat, fibre and provide transport for nomadic peoples is unique. Most veterinarians in these developing areas are being presented with an increasing number of camels suffering from a wide range of infectious diseases, but they lack the proper serodiagnostic aids to help in diagnosing these diseases. The ultimate aim of this study was to devise ways to monitor the camel immune response so that diagnosis and control of infectious diseases, which could affect camel welfare and productivity, could be properly performed. Thus, well defined antiglobulin reagents are essential. In addition to the absence of monoclonal antibodies specific for camel immunoglobulins, it has been shown * Corresponding author Ol65-2427/95/$09.50 0 1995 Elsevier Science B.V. 411rights reserved XSDlO165-2427(94)05334-O
176
SM. Azwai et al. / Vet. lmmunol. Immunopathol. 45 (1995) 175-184
that monoclonal reagents prepared for other animal species do not react with camel immunoglobulins ( Azwai et al., 1993 ). Since hybridoma technology for the purpose of eliciting monoclonal antibodies of predetined specificity was introduced by Kiihler and Milstein ( 1975 ), it has become a mainstay in most laboratories that utilise immunological techniques to study problems in basic, applied, or clinical research. Most techniques for producing monoclonal antibodies are complex and require multiple immunisations. Here, we describe a technique, which is very efficient for good immunogens (such as immunoglobulins) and produces hybridomas with only one low volume immunisation and uses very few mice, a distinct benefit when the use of experimental animals is being questioned. In this paper, we report the first successful production of monoclonal antibodies specific for camel immunoglobulin light chains and IgG and IgM heavy chains. Analysis of the binding of the monoclonal antibodies showed IgG and IgM class reactivity not only in heavy chains, but in light chains as well.
2. Materials and methods 2.1. Mice immunisation BALB/c inbred male mice, aged 8 weeks were used. In initial experiments, an ammonium sulphate precipitate (40% vol/vol saturation) of camel serum was prepared. This has previously been shown to precipitate camel IgG and IgM (Azwai et al., 1993). The immunogen, dialysed against phosphate buffered saline (PBS) and at 1 mg ml-‘, was emulsified with an equal volume of complete Freund’s adjuvant (CFA, Sigma, Poole, UK). Mice were immunised for monoclonal antibody production by the method of Holmdahl et al. ( 1985) where the hind limb foot pads were the site of immunisation and the popliteal lymph nodes were taken as the source of plasma cells. Mice received 25-50 ~1 of the CFA emulsion, by subcutaneous injection into each hind foot pad. After 9 days, the mice were killed by cervical dislocation and the draining popliteal lymph nodes dissected out under aseptic conditions, and used as a source of antibody producing cells. 2.2. Myeloma cell line A BALB/c plasmacytoma cell line (NS-0) was used for fusion, after being tested for resistance to 8-azaguanine (Sigma) and sensitivity to hypoxanthine, aminopterine and thymidine (HAT, Sigma). It was maintained in RPMI-1640 (Sigma) medium with 20% gamma-globulin free horse serum, in addition to penicillin/streptomycin solution (Sigma) and Hepes and sodium pyruvate buffer (Sigma).
S.M. kulai
et al. / CT’t.Iwnur~ol. Immunopathol.
45 (1995) 17%154
177
2.3. Cell fusion Popliteal lymph nodes were mashed gently through sterile gauze into warm RPMI- 1640 medium, washed and mixed with plasmacytoma cells at a ratio of 8: 1 (lymphocytes: plasmacytoma cells). Cell suspensions were centrifuged for 5 min at 300 xg and polyethylene glycol (PEG, mol. wt. 1300- 1600; Sigma) was added, for cell fusion, to a concentration of 45%. Fused cells were resuspended in HAT medium to approximately 1 x 1OSmyeloma cells ml- ‘. The suspended fused cells were distributed in 96-well flat-bottomed microtitre plates (Cel-Cult, Hounslow, UK) ( 100 ~1 per well). The plates were incubated at 37 ‘C in 5% CO, and fed with 100 ~1 of fresh HAT medium. Supernatants were initially assayed for specific antibodies at 14 days post fusion and subsequently, following expansion. 2.4. Enzyme-linked immunosorbent assay (ELISA) For initial screening of clones, non-activated microtitre ELISA (Dynatech, UK) plates were coated with 10 pg ml-’ of ammonium sulphate precipitate of camel serum (50 ~1 per well in PBS pH 7.2) at 37°C for 1 h and overnight at 4°C. Unbound proteins were washed off with PBS three times, and 50 ~1 from each well of the tissue culture supernatants was added. Following incubation at 37 ‘C for 1 h and three washes with PBS/Tween 20 (0.05%), bound antibodies were detected with alkaline phosphatase-conjugated goat anti-mouse IgG (Sigma ). Following incubation at 37°C for 1 h, bound conjugate was detected with substrate (p-nitrophenylphosphate, Sigma). An ELISA reader (Titertek, UK) was used to measure the absorbance at 405 nm. 1.2
9 0
0.6
0.0 0
10
20
40 30 Fraction
50
60
70
Fig. 1. ACA-34 gel filtration of 40% ammonium sulphate precipitate of camel serum. Two peaks were identified: IgM, fractions 24 + 25; IgG, fractions 39 + 40.
178
S.M.
kwai
et al. / I’et. Inmunol.
Itwnunopathol.
45 (1995) 175-184
As a result of ELISA data, all positive wells were transferred to 24-well flat bottomed plates (Cel-Cult ) where hybridomas were fed hypoxanthine and thymidine (HT, Sigma) medium. All positive hybridomas growing in the 24-well plates were limit diluted at least twice to ensure positive clones were growing up from a single cell. Resulting single clones were first tested for production of monoclonal antibodies against purified camel IgG containing (IgG,, IgGz and IgG,; Fig. 1, fractions 39 and 40) and IgM (Fig. 1, fractions 24 and 25) (Azwai et al., 1993) in an ELISA system similar to the previously mentioned one, where 10 pg ml-’ (50 ,ul per well in PBS pH 7.2) of the purified IgG or IgM was used to coat the ELISA plates; then positive clones were tested for reactivity against heavy and light chains of IgG and IgM on sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) /Western blotting. ELISA results were used to classify single clones as reactive with whole IgG, whole IgM or both. Supernatant and ascites titres were obtained by double dilutions in ELISA. 2.5. Maintenance of antibody producing clones Classified clones were transferred to 25-cm’ tissue culture flasks (Corning, New York, USA) in RPMI-1640 (Sigma) medium with 20% gamma globulin-free horse serum and incubated at 37°C in 5% COz. Supernatants of the growing clones were collected, stored at - 70” C and replaced by fresh medium. Clones were split and frozen in liquid nitrogen.
2.6. Biotin conjugation of monoclonal antibodies Tissue culture supernatants of selected clones were twice precipitated with 40% saturated ammonium sulphate, the precipitates were obtained by centrifugation at 10000 rev min- ’ for 15 min, dissolved in distilled H,O and dialysed against 0.1 M sodium borate buffer (pH 8.8 ) overnight. The monoclonal antibodies were biotinylated as described by Harlow and Lane ( 1988), using N-hydroxysuccinimidobiotin (Sigma) as a linking agent.
2.7. SDS-PAGE IgG and IgM fractions corresponding to I50 kDa and more than 600 kDa, respectively, from gel filtration (ACA-34) of an ammonium sulphate precipitate of camel serum (Fig. 1) were run in vertical reducing SDS-PAGE mini-gels of 12% acrylamide, as described by Laemmli ( 1970)) where 120 ,a1of (25 ,ugper 10 ,A) reduced (2-mercaptoethanol) total protein of either IgG or IgM fractions were loaded in a single trough comb. Low range molecular weight markers (MW range 14.4-97.4 kDa; Bio-Rad, CA, USA) were used to determine sizes of proteins.
S.M. .kwai
et al. / Ikt. I~wnurlol.
Imrm~nopathol.
4.5 (199.5) I75- 184
179
2.8. Western blotting Proteins were transferred electrophoretically from SDS-PAGE gels to nitrocellulose membranes in a semi-dry blotter with a discontinuous buffer. After blocking for 1 h with 1% gelatin at room temperature and washing three times ( 10 min) in PBS/Tween 20 (0.05%), the nitrocellulose membrane strips were probed with biotinylated monoclonal antibodies prepared from tissue culture supernatants, for 1 h at room temperature, followed by three IO-min washes of PBS/ Tween 20 (0.05%). Bound biotinylated antibodies were detected with extravidin-alkaline phosphatase (Sigma). Development of the solid phase reactants was visualised with naphthol phosphate/fast red substrate (Sigma). 2.9. Ascites production Mice were primed with an intraperitoneal injection of 0.5 ml 2,6,10,14_tetramethylpentadecane (Pristane, Sigma). Fourteen days later, ( 1- 10 )x 1O6hybridoma cells of the different growing single clones in 0.5 ml culture medium, were injected intraperitoneally into pristane-primed mice. Ascitic fluid obtained 1O14 days later was clarified by centrifugation and stored frozen at - 70 “C prior to testing and analysis.
3. Results 3.1. Monoclonal antibody production Monoclonal antibody-secreting hybridomas were successfully established from popliteal lymph node cells following immunisation with an ammonium sulphate precipitate of camel serum. Only two mice were needed to generate large numbers of blastoid cells ( 105- 106) from the popliteal lymph nodes 9 days after immunisation. Initial screening of supernatants generated by fusion of lymphocytes from the popliteal lymph nodes of mice, injected with the ammonium sulphate precipitate of camel serum, revealed 74 positive wells from a total of 576 wells seeded. As a result of ELISA screening, using the ammonium sulphate precipitate of camel serum as screening antigen, 13 mother clones were selected, according to their ELISA data. Two rounds of limiting dilution cloning of each mother clone generated a library of 28 subclones, each generated from a single cell. 3.2. Reactivity of monoclonal antibodies
Reactivity with IgG and IgM in ELBA Camel IgG and IgM, prepared by gel filtration in ACA-34 (Fig. 1) as previously described (Azwai et al., 1993) were used as antigens in ELISA. The IgG peak (tubes 39 and 40) contained all three subclasses of IgG described by Hamers-Casterman et al. ( 1993), as shown by identification of heavy chains of 50
SM. kwai
180
et ul. / Lkt. Immunol.
lmmw~opathol.
45 (1995) 175-184
Fig. 2. SDS-PAGE/Western blotting of camel IgG from ACA-34 separation. Lane 1, amido black (protein) staining; Lane 2, binding of monoclonal antibody ( 1.B6) to the heavy chain of camel IgG,; Lane 3. binding of monoclonal antibody (4.C3) to camel IgG light chain. Arrows mark the position of molecular weight markers (kDa). Table 1 Reactivity of monoclonal antibodies to camel IgG and IgM (whole molecules, heavy and light chains) Clone
I.B6 5.A4 1.B3 2.B6 4.B1 l.C2 5.A4D8 1.A4 1.C2D8 4.c3 4.B2
Titre
ELISA (OD 405)
Western blotting
SN
Ascites
IgG
IgM
IgG,,
IgG,2
IgG,3
IgMp
IgG L.C. IgM L.C.
l/6400 l/800 l/400 l/400 1/200 l/1600 l/l600 l/400 l/400 l/3200 l/1600
ND ND ND ND ND l/51200 l/51200 ND ND l/12800 ND
0.822 0.616 0.506 0.564 0.480 0.030 0.598 0.568 0.060 0.464 0.379
0.042 0.010 0.014 0.019 0.028 0.612 0.084 0.029 0.597 0.284 0.252
+ + + + + -
+ + + -
+ + + -
+ -
+ + + +
_ _ _ _ + I+
SN, tissue culture supernatant. IgG,,. IgG 50 kDa heavy chain; IgG,*, IgG 46 kDa heavy chain: IgG,,, IgG 42 kDa heavy chain; IgM,u, IgM heavy chain: IgG L.C., IgG light chain; IgM L.C., IgM light chain. ND, not done.
S.M. Azwai et al. / Vet. Immunol. Imtnunopathol. 45 (1995) 175-184
181
-b
25-
_)
Fig. 3. SDS-PAGE/Western blotting of camel IgM from ACA-34 separation. Lane 1, amido black (protein) staining; Lane 2, binding of the monoclonal antibody (1.C2) to the heavy chain of camel IgM; Lane 3, binding of monoclonal antibody (4.B2) to camel IgM light chain. Arrows mark the position of molecular weight markers (kDa),
1
2
Fig. 4. SDS-PAGE/Western blotting of camel IgG from ACA-34 separation. Lane 1, amido black (protein) staining; Lane 2. binding of monoclonal antibody (2.B6) to camel IgG,, lgGz and IgG, heavy chains. Arrows mark the position of molecular weight markers (kDa).
182
S.M. kwai
et al. / Vet. Itnmunol. Immunopathol.
45 (1995) 175-154
kDa, 46 kDa and 42 kDa in SDS-PAGE (Lane 1; Fig. 2). Reactivity of monoclonal antibodies with IgG and IgM is shown in Table 1. It can be seen that monoclonal antibodies specific to either IgG or IgM were generated and that reactivity with both classes was also detected. Western blotting reactivity
Further analysis by Western blotting (Table 1 ), confirmed that some IgM-only reactive (by ELISA) clones were binding the 70 kDa IgM heavy chain (Clone 1.C2; Lane 2; Fig. 3). Analysis of IgG reactivity showed two clones ( 1.B6 and 5.A4) specifically reacting to IgGi (50 kDa) heavy chain only (Lane 2; Fig. 2). Other IgG reactive clones (1 .B3,2.B6 and 4.Bl) were shown to react with IgG,, IgG2 (46 kDa) and IgGJ (42 kDa) heavy chains (Lane 2; Fig. 4). Clones (4.C3 and 4.B2) shown by ELISA to react with both whole IgG and IgM molecules, were shown to bind the light chains of both IgG and IgM (Lane 3; Figs. 2 and 3 ). An unexpected finding was that Western blotting revealed IgG- specific (ELISA) monoclonal antibodies (5.A4D8 and 1.A4) which reacted with none of the IgG heavy chains but with the 25 kDa light chain of IgG. This reactivity was
2s-
‘P.“(
1
2
Fig. 5. SDS-PAGE/Western blotting of camel IgM from ACA-34 separation. Lane I, amido black (protein) staining; Lane 2, binding of monoclonal antibody ( l.C2D8), which is IgM specific, to IgM light chain only. Arrows mark the position of molecular weight markers (kDa).
S.M. .gzwai et al. / C’et. Immunol.
Imnunopathol.
45 (199.5) 175-184
183
not shown against IgM light chain. Likewise, an IgM-specific (ELISA) monoclonal antibody ( 1.C2D8) was shown not to bind the 70 kDa heavy chain, but instead to react with the 25 kDa light chain. Again this reaction was specific for the IgM light chain, no IgG reactivity was detected (Lane 2; Fig. 5 ) . 4. Discussion This paper extends an earlier report describing the isolation and characterisation of camel immunoglobulin classes and subclasses ( Azwai et al., 1993 ). In this paper we describe, for the first time, the production and characterisation of monoclonal antibodies specific for camel immunoglobulin light chains and IgG and IgM heavy chains. The technique used is very quick, reduces the numbers of experimental mice, the period they are used for and has been shown to produce high titre reagents. It is considered that the presence of large numbers of antigen-specific B cell blasts in the popliteal lymph node preparation is the reason for the success of this immunisation protocol. Whilst having many disease syndromes in common with other ruminants, camels also have characteristic disease problems (Higgins, 1986; Fassi-Fehri, 1987; Mustafa, 1987). Recently, it has been shown that this species also has unique immunological characteristics. Hamers-Casterman et al. ( 1993 ) have demonstrated, in camels, two IgG subclasses devoid of light chains; the first time this has been reported for any species. This paper also demonstrates the IgG subclass heavy chains of 50 kDa (IgG, ), 46 kDa (IgG,) and 42 kDa ( IgG,) reported by Hamers-Casterman et al. ( 1993). Most of the monoclonal antibodies generated against IgG heavy chains reacted with all IgG subclass heavy chains; in the initial fusion two clones reacting to IgG, heavy chain were demonstrated. We have performed other fusions with lymph node cells from mice immunised with later ACA34 fractions (Fig. 1; fractions 43 and 44) containing primarily IgGZ and IgG,. However, all secreting hybridomas which were generated were pan-IgG subclass reactive (data not shown). Our findings further indicate that some of the camel IgG and IgM specific monoclonal antibodies, as shown by ELISA against whole molecules, showed light chain (25 kDa) reactivity and had no reactivity against either IgG or IgM heavy chains. That is to say, we have monoclonal antibodies specific for immunoglobulin classes which are apparently light chain reactive, but only reactive to light chains from specific classes, e.g. some IgG-specific monoclonal antibodies bound IgG light chain, but did not bind IgG heavy chain or IgM light chain. Such data are unexpected and require an explanation. One possibility is that the 25 kDa bands are not light chains at all, but parts of the heavy chains which separate under the reducing SDS-PAGE conditions. Another possibility is that there are equivalent types of light chain to the kappa/lambda groupings seen in man and that these are not randomly associated with heavy chains and that one type of light chain associates with one type of heavy chain. It is also possible that the epitopes recognised by the monoclonal antibodies are part of the molecular sequences flanking the interchain (heavy to light) disulphide linkages which determine the light chain association with the heavy chain of IgG or IgM.
184
SM. Azwai et al. / Vet. Immunol. Itnmunopathol. 45 (1995) 175-184
The data may also suggest that the mechanism by which camel B cells construct immunoglobulins is not as described for other species. Such a radical proposal would require a great deal of extra study of gene rearrangement to determine if there are gene sequences in camel light chains similar to genes coding for the Cregion in heavy chains. Alternatively, there could be C-region flanking sequences in the light chains which have altered expression after switching from IgM to IgG production. The function of light chains in the camel will be called into question by this work; it is noteworthy that the camel heavy chains devoid of light chains (subclasses IgGz and IgG, ) are quite able to bind antigens (Hamers-Casterman et al., 1993). There has been very little development of proper serodiagnostic aids for camel diseases. As veterinary medicine expands into areas where camels are considered to be valuable domestic animals, it is necessary to develop quick and reliable serological assays for diagnosing camel diseases, and the utilisation of monoclonal antibodies is the most appropriate option. These reagents are currently being used to determine camel humoral immune responses to infectious agents. Acknowledgements
S.M. Azwai was supported by a grant from El-Fateh University, Tripoli, Libya. Note added in proof
The monoclonal antibodies generated against Camelus dromedarius IgG and IgM also bind these molecules from Camelus bactrianus (Bactrian camel). References Azwai, S.M., Carter, S.D. and Woldehiwet, Z., 1993. The isolation and characterization of camel (Camelus dromedarius) immunoglobulin classes and subclasses. J. Comp. Pathol., 109: 187-l 95. Fassi-Fehri, M.M., 1987. Diseases of camels. Rev. Sci. Tech. Off. Int. Epiz., 6: 337-354. Hamers-Casterman, C., Atarhouch, T., Muyldermans, S., Robinson, G., Hamers, C., Bajyana Songa, E., Bendahman, B. and Hamers, R., 1993. Naturally occurring antibodies devoid of light chains. Nature, 363: 446-448. Harlow, E. and Lane, D. (Editors), 1988. Antibodies: a laboratory manual. Cold Spring Harbor Laboratory, New York, 34 1 pp. Higgins, A.J. (Editor), 1986. The Camel in Health and Disease. Bailliere Tindall, London, pp. 2 I40. Holmdahl, R., Moran, T. and Anderson, M., I985 A rapid and efficient immunization protocol for production of monoclonal antibodies reactive with autoantigens. J. Immunol. Methods, 83: 379384. Kohler, G. and Milstein. C., 1975. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature, 256: 495-497. Laemmh, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227:680-685. Mustafa, I.E., 1987. Bacterial diseases of dromedaries and bactrian camels. Rev. Sci. Tech. Off. Int. Epiz., 6: 391-405.
