HYBRIDOMA Volume 10, Number 1, 1991 Mary Ann Liebert, Inc., Publishers
Monoclonal Antibodies to the Soluble Human IL-6 Receptor: Affinity Purification, ELISA, and Inhibition of Ligand Binding DANIELA NOVICK,1 HARTMUT ENGELMANN,1 M. REVEL,1 O. LEITNER2 and MENACHEM RUBINSTEIN1 'The Weizmann Institute
of Science, Dpi. of Molecular Genetics and Virology,
Rehovot 76100, Israel
2MonoYeda, Weizmann Institute, Rehovot, Israel
ABSTRACT
Soluble IL-6 receptor (IL-6-R) purified to homogeneity from normal human urine used for irrrnunization of mice and rabbits. Spleen cells derived from a mouse shewing a high binding titer to IL-6-R in an inverted solid phase rad-Loijrrnunoassay (IsRIA) and in a Western blotting analysis were fused to mouse myeloma cells. The hytaidemas were screened by the IsRIA, and 30 positive clones were isolated and characterized. They were suitable for affinity purification of the IL-6-R and for its detection by Western blot analysis, by ELISA and by sandwich type sRIA. Most of them inhibited the binding of labeled IL-6-R to IL-6 in a solid phase RIA. was
DJTTÎODUC-TON
Although IL-6 (1-7) was identified more than seven years ago, it is only within the past few years that the multiplicity of biologic activities of this molecule have begun to be recognized. IL-6 has been reported to be involved in (8, introduction): a. induction of ijiirajnoglobulin production in activated B cells; b. induction of proliferation of hybrid_rre/pl-__rracyTaxra/nTyelcma cells; c. induction of IL-2 production, cell growth and cytotoxic T cell differentiation; d. stimulation of multipotent colony formation in hematopoietic stem cells; e. regulation of acute phase response; f. growth inhibition and induction of differentiation into macrephages of myeloid leukemic cell lines and g. induction of neural differentiatien. A general question was raised how dees IL-6 with its multitude of biologic activities fit into the network of interactive cytokine that is so irnportant in the regulation of the hematopoietic and irnmune system. In order to address these issues, the relationship between this cytokine and its receptor is extensively studied. IL-6-R was cloned (9) and it was found that its ligand, the IL-6, triggers the association of the extracellular portion of the receptor with a r__n-ligand-binding membrane glyecprotein (gpl30, 8). Moreover, evidence was provided by us that the soluble part of the IL-6-R increases the HGF activity of IL-6 in mouse cells and also enhances IL-6's growth inhibitory effect on human breast carcinoma cells (10). A repertoire of antibodies directed against the extracellular portion of this receptor will enable to define the epitope responsible for this interaction and will help to shed light en whether the same mechanism is involved in the multifunction nature of IL-6. Two monoclonal antibodies against human IL-6-R prepared by iirrnunizaticn of mice with murine transfectant
137
cell line were described (11). They were used for identification and detection of the intact IL-6-R and the expression of IL-6-R on normal B cell T cell subsets. The present study describes the pxreparation of a battery of 30 anti IL-6-R monoclonal antibodies suitable for affinity purification, Western analysis and ELISA
of the IL-6-R and for inhibition of
binding
of IL-6 to its
receptor.
MATERIALS AND METHODS
Reccrnbinant IL-6 (rIL-6)
Affinity purified (5x10 units/mg, 12) rIL-6 from E. coli (5) or rIL-6 from was used t_hroughout this study.
Chinese hamster ovary cells (CHO) (13)
Purification of soluble IL-6-R by ligand affinity chrcmatography (14) Concentrated crude urine (1000 fold) or partially purified urinary proteins on carboxy methyl sepharose (CMS, 15) were passed on an IL-6 (2.5 mg) coupled to an Affigel-10 column (1 ml, BioRad Laboratories, Richmond, CA). After extensive washings with PBS, bound proteins were eluted ty citric acid (25mM, pH 2.5) and immediately neutralized. Eluted fractions from the IL-6 column were further resolved by reversed phase HPLC.
lirtnunizaticrt of Mice and Cell Fusion
Balb/c mice were first injected into the foot pads with a pure preparation of a soluble IL-6-R (2.5/ig/mouse emulsified in complete Freund's adjuvant) and three weeks later subcutaneously in incomplete Freund's adjuvant. Three additional injections were given at 10 day intervals, subcutaneously in PBS. The mouse chosen for fusion received an intraperitoneal injection of this receptor and three days later lymphocytes from toe spleen and the inguinal lymph nodes (110x10 cells) were fused with 32x10-' NSO/1 myeloma variant (NSO cells, kindly provided by C. Milstein, MRC, Cambridge, U.K.). The fused cells were distributed into microculture plates (2.5x10 cells/well). Hyhridcinas that were found to secrete anti IL-6-R antibodies were cloned and recloned by the limiting dilution technique. -jnmunization of Rabbits The
except
pattern of iirmunizaticn was used for the rabbits as for the mice, for the dose, that in the case of the rabbits was 10«g/injection.
same
Labeling of IL-6-R and anti IL-6-R antibodies HPLC purified IL-6-R as well as antibody No. 22.1 were labeled by a modification of the Chlare-tiine-T method (16). Briefly 4 ug of IL-6-R in PBS was labeled with 1 mCi of iodine in the presence of 1 mg/ml of C3nloramine-T (30 sec, 4°C) to a specific activity of 8x10 cpm/wg. 100«g of antibody No. 22.1 (ascitic fluid irnmunoglbbulins obtained by 50% amonium sulfate precipitation) was labeled to specific activity of 3x10 cpm/«g.
Screening far anti JX-6-R Specific Hybridomas by
were tested for the presence of anti IL-6-R antibodies PVC microtiter plates were coated with affinity purified
Hybridcma supematants by
an
inverted sRIA.
IsRIA
138
goat anti mouse F(ab)o antibodies (BioMakor, 10«g/ml, 80«l/well). The plates were washed with PBS attaining BSA (0.5%) and Tween-20 (0.05%), and blocked with the
solution for 2 hrs at 37°C. Hybridcrna culture supematants (50«l/well) added and the plates were incubated for 4 hrs at 37°C. The plates were then washed and -I-IL-6-R (50«1, 105 cpm) was added for further incubation of 16 hrs at 4°C. The plates were washed and individual wells were cut and counted in a gamma counter.
washing were
Preparation of -jnnunoadsorbent and Affinity Chrcmatography
Iirir__r__glbbulin fraction of ascitic fluids (10 mg) was coupled to 1 ml of agarose-polyacryl-hydrazide (17). 250 ml of partially purified urinary proteins on carbox_>methyl sepharose (CMS, 15) were passed on a 0.5 ml column at a flow rate of 0.25 ml/min. Following washings with PBS bound proteins were eluted by citric acid (25mM pH 2.5) and immediately neutralized by IM HEPES pH 8.5. ,
Reversed phase HPLC Eluted fractions from the irrtnunoaffinity column were further resolved by reversed-phase HPLC (Aquapore RP-300, 4.6x30 mm, Brownlee Labs.Santa Clara, CA) using an acetonitrile gradient in 0.3% aq. trifluoroacetic acid (TFA). 0.5 ml fractions
were
collected.
SDS-PAGE and Western
blotting
Proteins resolved by SDS-PAGE (12%) (18) were visualized either by silver staining (19) or elecLrbblotted en nitrocellulose sheets as pjreviously described (20). Electroblotted proteins were reacted with MAbs followed by 125I-goat anti mouse antibodies (0.7x10 cpm/ml). The nitrocellulose sheets were then washed, dried and autoradiographed.
Competitive
sRIA
FVC microtiter plates were coated with serum free hybridcrna supematants (80wl/well). The plates were washed and blocked with 0.5% BSA, 0.05% Tween-20 in PBS. An additional set of serum free hybridoma supematants (30/.1) was incubated separately with 125-I-IL-6-R (30«1, 30,000 cpm) for 2 hr at 37°C, and then the mixed samples (50pl) of antibody and antigen were transferred onto the antibocy coated plates for further incubation (overnight at 4°C). The plates were washed, individual wells
were
cut and counted in
a
gamma counter.
ELISA
Microtiter plates (Dynatech Laboratories, Alexandria, Virginia or NuncIlrrraxncplate Maxisorp, Denmark) were coated with anti IL-6-R monoclonal antibodies No. 34.4 (a subclone of 34) or No. 17.6 (a subclone of 17) (iirtrunoglobulin fraction, 120«l/well, 20¿_g/ml in PBS) overnight at 4°C. The plates were washed with PBS containing BSA (0.5%) and Tween-20 (0.05%) and blocked in the same solution for at least 2hrs at 37°C. The tested samples were diluted in the blocking solution and added to the wells (100«l/well) for 4 hrs at 37°C. The plates were than washed 3 times with PBS ccntaining Tween-20 (0.05%) followed by the addition of rabbit anti IL-6-R serum (1:1000, lOOwl/well) for a further incubation of 4hrs at 37°C. The plates were washed 3 times and a conjugate of goat-anti- rabbit-horseradish-peroxidase (HRP, BioMakor, Israel, 1:2000, 100«l/well ) was added for 2hrs at room temperature. The plates were washed 4 times and the color was developed
139
ABTS (2,2'-az_no-bis(3-et_hylben_±hiazoline-6- sulfàonic The plates were read by an automatic ELISA reader.
by
acid, Sigma) substrate.
Sandwich type solid phase RIA FVC microtiter plates (Dynatech) were coated, washed, blocked and the tested was added as described for the ELISA. The IL-6-R in the sample was detected antibody (subclone of 22, 100«l/well, 105 cpm) by the addition of instead of the anti IL-6-R rabbit serum. Following an incubation of 4hrs at 37°C or overnight at the plates were washed, cut and counted. IL-6-R preparation purified to homogeneity was used as a laboratory standard.
sample
125I-22.1
4°C,
Inhibition of binding of
-LZD-I-IL-6-R
to IL-6
by antibodies
were coated with __rira_r__affinity purified (5«g/ml, washed and blocked with 0.5% BSA, 0.05% _ween-20in PBS. dilutions made in the blocking solution (30«1) were iricubated separately Antibody with 12 -I-IL-6-R (30,000 cpm, 30«1) for 2 hr at rocm t_eirperature. Then the mixed samples (50«1) of antibody and antigen were transferred onto the IL-6 coated niicrotiter plate for a further incubation of 4 hr at rocm temperature. The plates were washed 3 times with PBS cxxitaining 0.05% Tween-20, individual wells were cut and counted in a gamma counter. Beu-kground counts were obtained using an excess of IL-6 (2#g/ml) instead of the antibody sample.
PVC microtiter
75«1). The plates
IL-Öq^-j
plates
were
Protein determination. Protein concentrations
crystalline bovine
serum
were determined by the fluorescarnine method albumin as a standard.
(21), using
RESULTS The __nra_ne response in mice injected with a pure preparation of a soluble IL-6-R was followed by the inverted sRIA, in which goat anti mouse antibodies were -I-IL-6-R for detection. used for coating of the niicrotiter plates and Following 5 injections, it was possible to dilute the serum 1:128,000 in this assay (3000 çpm above b_«_kground) and 1:1000 in a Western bloting analysis. The spleen of this mouse was used for fusion. Hybridoma supematants were screened for the presence of anti IL-6-R antibodies by the inverted sRIA. Those supematants that gave counts that were at least four times higher than the negative control value (250 cpm) were considered positive (Table 1). Thirty such anti IL-6-R hybridcrnas were selected from 800 hybridomas screened (Table 1). Seme of the antibodies were able to detect the soluble IL-6-R in a Western blotting analysis when the SDS-PAGE was run under reducing conditions (Fig.l) and seme (hybridcrnas No. 5, 22, 24, and 39) recognized the IL-6-R only when it was run under non reducing conditions (data not shewn). Hybridomas No. 17.6, 22.1, and 34.4 were further characterized. Affinity chrcinatography was performed using i_rirajr__aa_X)rbent pjrepared from antibody No. 34.4. Partially pjurified urinary proteins (25Cml) on a CMS column were loaded on the antibody column (0.5 ml) followed by extensive w___hings and elution at lew pH. Results of such purification procedure are summarized in Table 2. Purification of 7850 fold of the IL-6-R was achieved in this step, and the recovery was 87% acœrding to the ELISA rne___urinent. Silver stain analysis of SDS-PAGE, under non reducing conditions, of the eluted fractions revealed a broad band of a M.W. of 50,000 and several contaminants (Fig. 2). The high molecular weight contaminant (>92,000) was observed also when other proteins (e.g. soluble IFN-y-R) were purified from crude urine. This contaminant did not hybridize to anti IL-6-R MAbs upen Western blotting analysis (Fig. 1). Purification of the IL-6-R to homogeneity was obtained by RP-HPLC
(Fig. 3).
140
TABLE 1.
Screening of hybridomas bv 3 revert sRIA arxi inhibition of binding of I25-I-IL-6-R to IL-6
Hybridcrna
No.
a
Inhibition of binding to %
20,455 1,085 36,565 31,450 11,465 8,850 2,000 1,645 4,165 1,755 3,060 31,465 14,875 33,480 35,495 1,455 9,640 35,975 5,195 1,415 1,870 33,565 1,255 6,090 18,000 8,000 28,440 1,075 3,945 3,440
4 5 17 20 22 24 25 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 46 48 49 50 51 52 53
b
IsRIA cpm
IL-6a
Ig class
IgGl IgM
30 0 50 79 0 14
IgG2a IgGl IgG2 IgGl IgG2a IgGl
N.D.b N.D. N.D. N.D.
N.D.
IgM
N.D.
0 95 90 88 76
IgGl IgGl IgGl IgG3 IgM
N.D.
IgGl IgGl IgG2 IgGl IgGl IgGl IgGl
90 100 0 75 N.D.
100 N.D.
15 100
N.D.
N.D.
IgM IgGl IgGl IgM IgGl
IgGl
65 N.D.
42 N.D.
See Materials and Methods Not done
pair of antibodies suitable for a sandwich type sRIA and for an ELISA was by a competitive sRIA, in which the plates were coated with different -I-IL-6-R was competed by the same hybridcrna sup)ematants and the binding of battery of antibodies used fo coating. Nine antibodies were tested by such a box titration (4.4, 17.6 20.2, 22.1, 32.2, 34.4, 38.5, 42.4, 50.1; subclones of the MAbs listed in Table 1). Only antibody NO. 22.1 did not compete with the other 8 A
selected
antibodies
en
the
binding
to
-I-IL-6-R.
Thus it
was
concluded that this anti-
body binds the receptor through a different epitope than the other 8 antibodies and it can serve as a counter antibody to each of the other eight. The best results in terms of detection limit were obtained when the plates were coated with either antibody No. 17.6 or antibody No. 34.4, and the IL-6-R tested was detected by antibody No. 22.1. The detecting antibody could be either labeled with iodine and then used for sandwich typie sRIA, or coupled to an enzyme or biotinylated and then used for an ELISA ( not done ). The detection limit of the sandwich sRIA was 100
pg/ml
of IL-6-R
(Fig.4A).
ELISA was developed in which rnoncclonal antibody No. 17.6 was the capturing antibody while anti IL-6-R rabbit serum was used for detection. The ELISA was very
141
Western blotting analysis, lug of i-iinunoaffinity purified IL-6-R was run SDS-PAGE (under reducing conditions) and elecUüblotted. The nitrocellulose sheet was cut into 14 strips which were reacted with the different hybridcrna supematants. One strip was hybridized to mouse polyclonal serum and one was incubated with an irrelevant monoclonal antibody (117, anti lFN-j8).
Fig. 1.
on
reproducible and sensitive and was used for monitoring the purification steps of the IL-6-R (Table 2) and to detect the soluble IL-6-R in sera of patients. The
detection limit of such an ELISA was 60 pg/ml (Fig.4B). Antibodies capable of competing on the binding of the ligand to its receptor were selected by inhibition of the binding of labeled IL-6-R to IL-6 in a sRIA TABLE 2.
linrnunoaffinity purification of
IL-6-R from human urine ELISA
FLUORESCAMINE
Sample
Load Effluent El. fr. Ia El. fr. 2 El. fr. 3 El. fr. 4
Vol. ml
250Z
250t 1.2 1.2 1.2 1.2
Prot. Cone.
ug/ml 2200 2000 20 45 18 11
Prot. ug
Prot. Cone.
«g/ml 0.38 0.06 7.7
550,000 500,000 24 54 21 13
30 12 8
Total pxrot. eluted ®
b
Prot. ug
95 15 9 36 14 9 68
Elutian fracticn Equivalent to 250 L of crude urine
142
Purity Yield %
%
38 67 67 69 87
B 67 43
R-
30
-
Fig. 2.
SDS-PAGE of
affinity purified IL-6-R.
A. Eluticn fraction from antibody column NO. 34.4 indicated by R. B. Molecular weight markers (K).
(900ng).
IL-6-R band
(5ÓK)
(Table 1). Of 20 antibodies tested 10 inhibited extensively (more than 75%) the between the ligand and its receptor; 4 did not inhibit it at all, and the
binding
other 6 had inhibited such a binding to a different extent (14-65%). As expected, antibody No. 22.1 had no inhibiting capability, probably because it is directed to a different epitepe than the one binding to the ligand. Thus this antibody —i-1-1-1-1-1-1-1-r Purification of JL-6-Receptor on RP-300 HPLC
10
20
30
40 50 eo 70 Fraction number
80
90
Fig. 3. Purification of the urinary IL-6-R by RP-HPLC: Iirniuno-affinity purified urinary proteins were applied to the HPLC column, eluted at 0.5 ml/min by an acetonitrile gradient (dotted line), and the protein was rnonitored by a preparative fluorescamine system (solid line). 143
2345
IL-6-R
Fig. 4A.
Standard
curve
of
a
4B.
Standard
curve
of
an
could be used
as a
012345
[ng/ml] sandwich sRIA of ELISA of
a
pure
a
]?ure
IL-6-R [ng/ml] preparation of IL-6-R.
prreparation
of IL-6-R.
counter antibody to the capturing antibodies, which in a sandwich type sRIA and in an ELISA.
were
all
neutralizing antibodies,
DISCUSSION
The present study ck_rr__r__trates the convenience and success of ligand and irrmunoaffinity chrcrretography in purification of proteins that are present as traces in very crude protein ir_Lxtures. It also proves the existence of a soluble receptor for IL-6 in normal human urine. The observed prevalence of soluble cytokine receptors in urine {IL-2-R (22), IFN-ar-R, IL-6-R (14) and the two TNF-Rs (15, 23)} suggests that they may have an i-rrrtunoregulatory role, either by participation in the process of eliminating cytckines via the kidney or, if pxresent in the plasma as shewn for IL-2-R by modulating the availability of their correspcnding cytokines. The battery of anti IL-6-R antibodies described in this study will enable the approach to the ntechanism of action of this cytokine and its receptor. Seme of the antibodies did not inhibit the binding of IL-6 to its receptor. It rernains to be seen to which epitopes those antibodies are directed, and whether they affect the binding of the complex of IL-6 and its receptor to the gpl30, or influence seme of the various functions of IL-6. The antibodies capable of inhibiting the binding of IL-6 to its receptor are very pxecmising candidates to serve as neutralizing antibodies in biological assays and in vivo. The neutralizing capacity of these antibodies will enable to elucidate the role of IL-6 in regulation of the hematopoietic and iirrnune system both in normal and pathological situations, and the ELISA will serve as ferent diseases as well
a as
tool to tronitor the level of the soluble IL-6-R in difin different stages of the disease (work in progress). ACKN_*____DG_-_NFS
We thank Dr. C. Serafini of the Cesare Sereno Research Institute, Rome, for the supply of urinary proteins and Mrs. R. Chap from MonoYeda, The Weizmann Institute of Science, Israel for excellent technical assistance. M. Rubinstein has the Maurice and Edna Weiss Chair in Interferon research. This work was supported by a
grant from
InterPharm laboratories.
REE__R__NCES 1.
Nicola, N.A., Meteaff, D., Mabsumoto, M. and Johnson, G.R. (1983). Purification of a factor inducing differentiaticn in murine myelcmonocytic leukemia cells.
J. Biol. Chem.
258, 9017-9023. 144
Hirano, T., Taga, T., Nakano, N., Yasukawa, K., Kashiwamura, S., Shimizu, K., Nakajima, K., Pyun, K. H. and Kishinioto, K. (1985). Purification to hcmogeneity and characterization of human B cell differentiation factor (BCDF or BSF-p-2). Proc. Acad. Sei. USA 82, 5490-5494. Zilberstein, A., Ruggieri, R., Kbm, J.H. and Revel, M. (1986). Structure and expression of cDNA and genes for human interfercn-ySo, a distinct species inducible by growth sti_nulatory cytokines. EMBO J. 5, 2529-2537.
Hirano, T., Yasukawa, K., Harada, H., Taga, T., Watanabe, Y., Matsuda, T., Kashiwamura, S., Nakajima, K., Koyama, K., Iwamatu, A., Tsunasawa, S., Sakiyama, F., Matsui, H., Tatahara, Y., Taniguchi, T. and Kishimoto, T. (1986). Complementary DNA for a novel human interleukin (BSF-2) that induces B lymphocytes to produce __rrrur__globulin. Nature 324, 73-76. Lotz, M., Jirik, F., Kabouridis, R., Tsoukas, C., Hirano, T., Kisriirnoto,. T. and Carson, D.A. (1988). BSF-2/IL-6 is a costimulant for human thyrnocytes and T lyrrphocytes. J. Exp. Med. 167, 1253-1258. Pcucpart, P., Vandenabeele, P., Cayphas, S., Snick, J. V., Heageman, G., Kruys, V., Fiers, W., Content, J. (1987) B cell growth modulating and difer-
entiating activity of reccrnbinant human 26-kD protein (BSF-2, HPGF). EMBOJ. 6, 1219-1224.
HuIFN-
fa,
Kawano, M., Hirano, T., Matsuda, T., Taga, T., Horii, Y., Iwato, K., Asaoko, H., Tang, b., Tanabe, 0., Tanaka, H., Kuramoto, A. and Kishimoto, T. (1988). Autocrine generation and requirement of BSF-2/IL-6 for human multiple myelomas.
Nature
332, 83-85.
Taga, T., Hibi, M., Hirata, Y., Yamasaki, K., Yasukawa, K., Matsuda, T., Hirano, T. and Kishimoto, T. (1989). Interleukin-6 triggers the association of its receptor with a possible signal transducer, gpl30. Cell 58, 573-581. Yamasaki, K., Taga, T., Hirata, Y., Yawata, H., Kawanishi, Y., Seed, B., Taniguchi, T., Hirano, T. and Kishimoto, T. (1988). Cloning and expression of the human Interleukin-6 (BSF-2/IFN-£2) receptor. Science. 241, 825-828.
Novick, D., Engelmann, H., Wallach, D., Leitner, O., Revel, M. and Rubinstein, M. (1990). Purification of soluble cytokine-receptors from human urine by ligand-affinity and ijirnuno-affinity
Monoclonal Antibodies to the Soluble Human IL-6 Receptor: Affinity Purification, ELISA, and Inhibition of Ligand Binding DANIELA NOVICK,1 HARTMUT ENGELMANN,1 M. REVEL,1 O. LEITNER2 and MENACHEM RUBINSTEIN1 'The Weizmann Institute
of Science, Dpi. of Molecular Genetics and Virology,
Rehovot 76100, Israel
2MonoYeda, Weizmann Institute, Rehovot, Israel
ABSTRACT
Soluble IL-6 receptor (IL-6-R) purified to homogeneity from normal human urine used for irrrnunization of mice and rabbits. Spleen cells derived from a mouse shewing a high binding titer to IL-6-R in an inverted solid phase rad-Loijrrnunoassay (IsRIA) and in a Western blotting analysis were fused to mouse myeloma cells. The hytaidemas were screened by the IsRIA, and 30 positive clones were isolated and characterized. They were suitable for affinity purification of the IL-6-R and for its detection by Western blot analysis, by ELISA and by sandwich type sRIA. Most of them inhibited the binding of labeled IL-6-R to IL-6 in a solid phase RIA. was
DJTTÎODUC-TON
Although IL-6 (1-7) was identified more than seven years ago, it is only within the past few years that the multiplicity of biologic activities of this molecule have begun to be recognized. IL-6 has been reported to be involved in (8, introduction): a. induction of ijiirajnoglobulin production in activated B cells; b. induction of proliferation of hybrid_rre/pl-__rracyTaxra/nTyelcma cells; c. induction of IL-2 production, cell growth and cytotoxic T cell differentiation; d. stimulation of multipotent colony formation in hematopoietic stem cells; e. regulation of acute phase response; f. growth inhibition and induction of differentiation into macrephages of myeloid leukemic cell lines and g. induction of neural differentiatien. A general question was raised how dees IL-6 with its multitude of biologic activities fit into the network of interactive cytokine that is so irnportant in the regulation of the hematopoietic and irnmune system. In order to address these issues, the relationship between this cytokine and its receptor is extensively studied. IL-6-R was cloned (9) and it was found that its ligand, the IL-6, triggers the association of the extracellular portion of the receptor with a r__n-ligand-binding membrane glyecprotein (gpl30, 8). Moreover, evidence was provided by us that the soluble part of the IL-6-R increases the HGF activity of IL-6 in mouse cells and also enhances IL-6's growth inhibitory effect on human breast carcinoma cells (10). A repertoire of antibodies directed against the extracellular portion of this receptor will enable to define the epitope responsible for this interaction and will help to shed light en whether the same mechanism is involved in the multifunction nature of IL-6. Two monoclonal antibodies against human IL-6-R prepared by iirrnunizaticn of mice with murine transfectant
137
cell line were described (11). They were used for identification and detection of the intact IL-6-R and the expression of IL-6-R on normal B cell T cell subsets. The present study describes the pxreparation of a battery of 30 anti IL-6-R monoclonal antibodies suitable for affinity purification, Western analysis and ELISA
of the IL-6-R and for inhibition of
binding
of IL-6 to its
receptor.
MATERIALS AND METHODS
Reccrnbinant IL-6 (rIL-6)
Affinity purified (5x10 units/mg, 12) rIL-6 from E. coli (5) or rIL-6 from was used t_hroughout this study.
Chinese hamster ovary cells (CHO) (13)
Purification of soluble IL-6-R by ligand affinity chrcmatography (14) Concentrated crude urine (1000 fold) or partially purified urinary proteins on carboxy methyl sepharose (CMS, 15) were passed on an IL-6 (2.5 mg) coupled to an Affigel-10 column (1 ml, BioRad Laboratories, Richmond, CA). After extensive washings with PBS, bound proteins were eluted ty citric acid (25mM, pH 2.5) and immediately neutralized. Eluted fractions from the IL-6 column were further resolved by reversed phase HPLC.
lirtnunizaticrt of Mice and Cell Fusion
Balb/c mice were first injected into the foot pads with a pure preparation of a soluble IL-6-R (2.5/ig/mouse emulsified in complete Freund's adjuvant) and three weeks later subcutaneously in incomplete Freund's adjuvant. Three additional injections were given at 10 day intervals, subcutaneously in PBS. The mouse chosen for fusion received an intraperitoneal injection of this receptor and three days later lymphocytes from toe spleen and the inguinal lymph nodes (110x10 cells) were fused with 32x10-' NSO/1 myeloma variant (NSO cells, kindly provided by C. Milstein, MRC, Cambridge, U.K.). The fused cells were distributed into microculture plates (2.5x10 cells/well). Hyhridcinas that were found to secrete anti IL-6-R antibodies were cloned and recloned by the limiting dilution technique. -jnmunization of Rabbits The
except
pattern of iirmunizaticn was used for the rabbits as for the mice, for the dose, that in the case of the rabbits was 10«g/injection.
same
Labeling of IL-6-R and anti IL-6-R antibodies HPLC purified IL-6-R as well as antibody No. 22.1 were labeled by a modification of the Chlare-tiine-T method (16). Briefly 4 ug of IL-6-R in PBS was labeled with 1 mCi of iodine in the presence of 1 mg/ml of C3nloramine-T (30 sec, 4°C) to a specific activity of 8x10 cpm/wg. 100«g of antibody No. 22.1 (ascitic fluid irnmunoglbbulins obtained by 50% amonium sulfate precipitation) was labeled to specific activity of 3x10 cpm/«g.
Screening far anti JX-6-R Specific Hybridomas by
were tested for the presence of anti IL-6-R antibodies PVC microtiter plates were coated with affinity purified
Hybridcma supematants by
an
inverted sRIA.
IsRIA
138
goat anti mouse F(ab)o antibodies (BioMakor, 10«g/ml, 80«l/well). The plates were washed with PBS attaining BSA (0.5%) and Tween-20 (0.05%), and blocked with the
solution for 2 hrs at 37°C. Hybridcrna culture supematants (50«l/well) added and the plates were incubated for 4 hrs at 37°C. The plates were then washed and -I-IL-6-R (50«1, 105 cpm) was added for further incubation of 16 hrs at 4°C. The plates were washed and individual wells were cut and counted in a gamma counter.
washing were
Preparation of -jnnunoadsorbent and Affinity Chrcmatography
Iirir__r__glbbulin fraction of ascitic fluids (10 mg) was coupled to 1 ml of agarose-polyacryl-hydrazide (17). 250 ml of partially purified urinary proteins on carbox_>methyl sepharose (CMS, 15) were passed on a 0.5 ml column at a flow rate of 0.25 ml/min. Following washings with PBS bound proteins were eluted by citric acid (25mM pH 2.5) and immediately neutralized by IM HEPES pH 8.5. ,
Reversed phase HPLC Eluted fractions from the irrtnunoaffinity column were further resolved by reversed-phase HPLC (Aquapore RP-300, 4.6x30 mm, Brownlee Labs.Santa Clara, CA) using an acetonitrile gradient in 0.3% aq. trifluoroacetic acid (TFA). 0.5 ml fractions
were
collected.
SDS-PAGE and Western
blotting
Proteins resolved by SDS-PAGE (12%) (18) were visualized either by silver staining (19) or elecLrbblotted en nitrocellulose sheets as pjreviously described (20). Electroblotted proteins were reacted with MAbs followed by 125I-goat anti mouse antibodies (0.7x10 cpm/ml). The nitrocellulose sheets were then washed, dried and autoradiographed.
Competitive
sRIA
FVC microtiter plates were coated with serum free hybridcrna supematants (80wl/well). The plates were washed and blocked with 0.5% BSA, 0.05% Tween-20 in PBS. An additional set of serum free hybridoma supematants (30/.1) was incubated separately with 125-I-IL-6-R (30«1, 30,000 cpm) for 2 hr at 37°C, and then the mixed samples (50pl) of antibody and antigen were transferred onto the antibocy coated plates for further incubation (overnight at 4°C). The plates were washed, individual wells
were
cut and counted in
a
gamma counter.
ELISA
Microtiter plates (Dynatech Laboratories, Alexandria, Virginia or NuncIlrrraxncplate Maxisorp, Denmark) were coated with anti IL-6-R monoclonal antibodies No. 34.4 (a subclone of 34) or No. 17.6 (a subclone of 17) (iirtrunoglobulin fraction, 120«l/well, 20¿_g/ml in PBS) overnight at 4°C. The plates were washed with PBS containing BSA (0.5%) and Tween-20 (0.05%) and blocked in the same solution for at least 2hrs at 37°C. The tested samples were diluted in the blocking solution and added to the wells (100«l/well) for 4 hrs at 37°C. The plates were than washed 3 times with PBS ccntaining Tween-20 (0.05%) followed by the addition of rabbit anti IL-6-R serum (1:1000, lOOwl/well) for a further incubation of 4hrs at 37°C. The plates were washed 3 times and a conjugate of goat-anti- rabbit-horseradish-peroxidase (HRP, BioMakor, Israel, 1:2000, 100«l/well ) was added for 2hrs at room temperature. The plates were washed 4 times and the color was developed
139
ABTS (2,2'-az_no-bis(3-et_hylben_±hiazoline-6- sulfàonic The plates were read by an automatic ELISA reader.
by
acid, Sigma) substrate.
Sandwich type solid phase RIA FVC microtiter plates (Dynatech) were coated, washed, blocked and the tested was added as described for the ELISA. The IL-6-R in the sample was detected antibody (subclone of 22, 100«l/well, 105 cpm) by the addition of instead of the anti IL-6-R rabbit serum. Following an incubation of 4hrs at 37°C or overnight at the plates were washed, cut and counted. IL-6-R preparation purified to homogeneity was used as a laboratory standard.
sample
125I-22.1
4°C,
Inhibition of binding of
-LZD-I-IL-6-R
to IL-6
by antibodies
were coated with __rira_r__affinity purified (5«g/ml, washed and blocked with 0.5% BSA, 0.05% _ween-20in PBS. dilutions made in the blocking solution (30«1) were iricubated separately Antibody with 12 -I-IL-6-R (30,000 cpm, 30«1) for 2 hr at rocm t_eirperature. Then the mixed samples (50«1) of antibody and antigen were transferred onto the IL-6 coated niicrotiter plate for a further incubation of 4 hr at rocm temperature. The plates were washed 3 times with PBS cxxitaining 0.05% Tween-20, individual wells were cut and counted in a gamma counter. Beu-kground counts were obtained using an excess of IL-6 (2#g/ml) instead of the antibody sample.
PVC microtiter
75«1). The plates
IL-Öq^-j
plates
were
Protein determination. Protein concentrations
crystalline bovine
serum
were determined by the fluorescarnine method albumin as a standard.
(21), using
RESULTS The __nra_ne response in mice injected with a pure preparation of a soluble IL-6-R was followed by the inverted sRIA, in which goat anti mouse antibodies were -I-IL-6-R for detection. used for coating of the niicrotiter plates and Following 5 injections, it was possible to dilute the serum 1:128,000 in this assay (3000 çpm above b_«_kground) and 1:1000 in a Western bloting analysis. The spleen of this mouse was used for fusion. Hybridoma supematants were screened for the presence of anti IL-6-R antibodies by the inverted sRIA. Those supematants that gave counts that were at least four times higher than the negative control value (250 cpm) were considered positive (Table 1). Thirty such anti IL-6-R hybridcrnas were selected from 800 hybridomas screened (Table 1). Seme of the antibodies were able to detect the soluble IL-6-R in a Western blotting analysis when the SDS-PAGE was run under reducing conditions (Fig.l) and seme (hybridcrnas No. 5, 22, 24, and 39) recognized the IL-6-R only when it was run under non reducing conditions (data not shewn). Hybridomas No. 17.6, 22.1, and 34.4 were further characterized. Affinity chrcinatography was performed using i_rirajr__aa_X)rbent pjrepared from antibody No. 34.4. Partially pjurified urinary proteins (25Cml) on a CMS column were loaded on the antibody column (0.5 ml) followed by extensive w___hings and elution at lew pH. Results of such purification procedure are summarized in Table 2. Purification of 7850 fold of the IL-6-R was achieved in this step, and the recovery was 87% acœrding to the ELISA rne___urinent. Silver stain analysis of SDS-PAGE, under non reducing conditions, of the eluted fractions revealed a broad band of a M.W. of 50,000 and several contaminants (Fig. 2). The high molecular weight contaminant (>92,000) was observed also when other proteins (e.g. soluble IFN-y-R) were purified from crude urine. This contaminant did not hybridize to anti IL-6-R MAbs upen Western blotting analysis (Fig. 1). Purification of the IL-6-R to homogeneity was obtained by RP-HPLC
(Fig. 3).
140
TABLE 1.
Screening of hybridomas bv 3 revert sRIA arxi inhibition of binding of I25-I-IL-6-R to IL-6
Hybridcrna
No.
a
Inhibition of binding to %
20,455 1,085 36,565 31,450 11,465 8,850 2,000 1,645 4,165 1,755 3,060 31,465 14,875 33,480 35,495 1,455 9,640 35,975 5,195 1,415 1,870 33,565 1,255 6,090 18,000 8,000 28,440 1,075 3,945 3,440
4 5 17 20 22 24 25 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 46 48 49 50 51 52 53
b
IsRIA cpm
IL-6a
Ig class
IgGl IgM
30 0 50 79 0 14
IgG2a IgGl IgG2 IgGl IgG2a IgGl
N.D.b N.D. N.D. N.D.
N.D.
IgM
N.D.
0 95 90 88 76
IgGl IgGl IgGl IgG3 IgM
N.D.
IgGl IgGl IgG2 IgGl IgGl IgGl IgGl
90 100 0 75 N.D.
100 N.D.
15 100
N.D.
N.D.
IgM IgGl IgGl IgM IgGl
IgGl
65 N.D.
42 N.D.
See Materials and Methods Not done
pair of antibodies suitable for a sandwich type sRIA and for an ELISA was by a competitive sRIA, in which the plates were coated with different -I-IL-6-R was competed by the same hybridcrna sup)ematants and the binding of battery of antibodies used fo coating. Nine antibodies were tested by such a box titration (4.4, 17.6 20.2, 22.1, 32.2, 34.4, 38.5, 42.4, 50.1; subclones of the MAbs listed in Table 1). Only antibody NO. 22.1 did not compete with the other 8 A
selected
antibodies
en
the
binding
to
-I-IL-6-R.
Thus it
was
concluded that this anti-
body binds the receptor through a different epitope than the other 8 antibodies and it can serve as a counter antibody to each of the other eight. The best results in terms of detection limit were obtained when the plates were coated with either antibody No. 17.6 or antibody No. 34.4, and the IL-6-R tested was detected by antibody No. 22.1. The detecting antibody could be either labeled with iodine and then used for sandwich typie sRIA, or coupled to an enzyme or biotinylated and then used for an ELISA ( not done ). The detection limit of the sandwich sRIA was 100
pg/ml
of IL-6-R
(Fig.4A).
ELISA was developed in which rnoncclonal antibody No. 17.6 was the capturing antibody while anti IL-6-R rabbit serum was used for detection. The ELISA was very
141
Western blotting analysis, lug of i-iinunoaffinity purified IL-6-R was run SDS-PAGE (under reducing conditions) and elecUüblotted. The nitrocellulose sheet was cut into 14 strips which were reacted with the different hybridcrna supematants. One strip was hybridized to mouse polyclonal serum and one was incubated with an irrelevant monoclonal antibody (117, anti lFN-j8).
Fig. 1.
on
reproducible and sensitive and was used for monitoring the purification steps of the IL-6-R (Table 2) and to detect the soluble IL-6-R in sera of patients. The
detection limit of such an ELISA was 60 pg/ml (Fig.4B). Antibodies capable of competing on the binding of the ligand to its receptor were selected by inhibition of the binding of labeled IL-6-R to IL-6 in a sRIA TABLE 2.
linrnunoaffinity purification of
IL-6-R from human urine ELISA
FLUORESCAMINE
Sample
Load Effluent El. fr. Ia El. fr. 2 El. fr. 3 El. fr. 4
Vol. ml
250Z
250t 1.2 1.2 1.2 1.2
Prot. Cone.
ug/ml 2200 2000 20 45 18 11
Prot. ug
Prot. Cone.
«g/ml 0.38 0.06 7.7
550,000 500,000 24 54 21 13
30 12 8
Total pxrot. eluted ®
b
Prot. ug
95 15 9 36 14 9 68
Elutian fracticn Equivalent to 250 L of crude urine
142
Purity Yield %
%
38 67 67 69 87
B 67 43
R-
30
-
Fig. 2.
SDS-PAGE of
affinity purified IL-6-R.
A. Eluticn fraction from antibody column NO. 34.4 indicated by R. B. Molecular weight markers (K).
(900ng).
IL-6-R band
(5ÓK)
(Table 1). Of 20 antibodies tested 10 inhibited extensively (more than 75%) the between the ligand and its receptor; 4 did not inhibit it at all, and the
binding
other 6 had inhibited such a binding to a different extent (14-65%). As expected, antibody No. 22.1 had no inhibiting capability, probably because it is directed to a different epitepe than the one binding to the ligand. Thus this antibody —i-1-1-1-1-1-1-1-r Purification of JL-6-Receptor on RP-300 HPLC
10
20
30
40 50 eo 70 Fraction number
80
90
Fig. 3. Purification of the urinary IL-6-R by RP-HPLC: Iirniuno-affinity purified urinary proteins were applied to the HPLC column, eluted at 0.5 ml/min by an acetonitrile gradient (dotted line), and the protein was rnonitored by a preparative fluorescamine system (solid line). 143
2345
IL-6-R
Fig. 4A.
Standard
curve
of
a
4B.
Standard
curve
of
an
could be used
as a
012345
[ng/ml] sandwich sRIA of ELISA of
a
pure
a
]?ure
IL-6-R [ng/ml] preparation of IL-6-R.
prreparation
of IL-6-R.
counter antibody to the capturing antibodies, which in a sandwich type sRIA and in an ELISA.
were
all
neutralizing antibodies,
DISCUSSION
The present study ck_rr__r__trates the convenience and success of ligand and irrmunoaffinity chrcrretography in purification of proteins that are present as traces in very crude protein ir_Lxtures. It also proves the existence of a soluble receptor for IL-6 in normal human urine. The observed prevalence of soluble cytokine receptors in urine {IL-2-R (22), IFN-ar-R, IL-6-R (14) and the two TNF-Rs (15, 23)} suggests that they may have an i-rrrtunoregulatory role, either by participation in the process of eliminating cytckines via the kidney or, if pxresent in the plasma as shewn for IL-2-R by modulating the availability of their correspcnding cytokines. The battery of anti IL-6-R antibodies described in this study will enable the approach to the ntechanism of action of this cytokine and its receptor. Seme of the antibodies did not inhibit the binding of IL-6 to its receptor. It rernains to be seen to which epitopes those antibodies are directed, and whether they affect the binding of the complex of IL-6 and its receptor to the gpl30, or influence seme of the various functions of IL-6. The antibodies capable of inhibiting the binding of IL-6 to its receptor are very pxecmising candidates to serve as neutralizing antibodies in biological assays and in vivo. The neutralizing capacity of these antibodies will enable to elucidate the role of IL-6 in regulation of the hematopoietic and iirrnune system both in normal and pathological situations, and the ELISA will serve as ferent diseases as well
a as
tool to tronitor the level of the soluble IL-6-R in difin different stages of the disease (work in progress). ACKN_*____DG_-_NFS
We thank Dr. C. Serafini of the Cesare Sereno Research Institute, Rome, for the supply of urinary proteins and Mrs. R. Chap from MonoYeda, The Weizmann Institute of Science, Israel for excellent technical assistance. M. Rubinstein has the Maurice and Edna Weiss Chair in Interferon research. This work was supported by a
grant from
InterPharm laboratories.
REE__R__NCES 1.
Nicola, N.A., Meteaff, D., Mabsumoto, M. and Johnson, G.R. (1983). Purification of a factor inducing differentiaticn in murine myelcmonocytic leukemia cells.
J. Biol. Chem.
258, 9017-9023. 144
Hirano, T., Taga, T., Nakano, N., Yasukawa, K., Kashiwamura, S., Shimizu, K., Nakajima, K., Pyun, K. H. and Kishinioto, K. (1985). Purification to hcmogeneity and characterization of human B cell differentiation factor (BCDF or BSF-p-2). Proc. Acad. Sei. USA 82, 5490-5494. Zilberstein, A., Ruggieri, R., Kbm, J.H. and Revel, M. (1986). Structure and expression of cDNA and genes for human interfercn-ySo, a distinct species inducible by growth sti_nulatory cytokines. EMBO J. 5, 2529-2537.
Hirano, T., Yasukawa, K., Harada, H., Taga, T., Watanabe, Y., Matsuda, T., Kashiwamura, S., Nakajima, K., Koyama, K., Iwamatu, A., Tsunasawa, S., Sakiyama, F., Matsui, H., Tatahara, Y., Taniguchi, T. and Kishimoto, T. (1986). Complementary DNA for a novel human interleukin (BSF-2) that induces B lymphocytes to produce __rrrur__globulin. Nature 324, 73-76. Lotz, M., Jirik, F., Kabouridis, R., Tsoukas, C., Hirano, T., Kisriirnoto,. T. and Carson, D.A. (1988). BSF-2/IL-6 is a costimulant for human thyrnocytes and T lyrrphocytes. J. Exp. Med. 167, 1253-1258. Pcucpart, P., Vandenabeele, P., Cayphas, S., Snick, J. V., Heageman, G., Kruys, V., Fiers, W., Content, J. (1987) B cell growth modulating and difer-
entiating activity of reccrnbinant human 26-kD protein (BSF-2, HPGF). EMBOJ. 6, 1219-1224.
HuIFN-
fa,
Kawano, M., Hirano, T., Matsuda, T., Taga, T., Horii, Y., Iwato, K., Asaoko, H., Tang, b., Tanabe, 0., Tanaka, H., Kuramoto, A. and Kishimoto, T. (1988). Autocrine generation and requirement of BSF-2/IL-6 for human multiple myelomas.
Nature
332, 83-85.
Taga, T., Hibi, M., Hirata, Y., Yamasaki, K., Yasukawa, K., Matsuda, T., Hirano, T. and Kishimoto, T. (1989). Interleukin-6 triggers the association of its receptor with a possible signal transducer, gpl30. Cell 58, 573-581. Yamasaki, K., Taga, T., Hirata, Y., Yawata, H., Kawanishi, Y., Seed, B., Taniguchi, T., Hirano, T. and Kishimoto, T. (1988). Cloning and expression of the human Interleukin-6 (BSF-2/IFN-£2) receptor. Science. 241, 825-828.
Novick, D., Engelmann, H., Wallach, D., Leitner, O., Revel, M. and Rubinstein, M. (1990). Purification of soluble cytokine-receptors from human urine by ligand-affinity and ijirnuno-affinity
