Sep 1, 1989 - antigen receptor on T cells. Nevertheless, direct binding of such epitopes to MHC class II proteins on normal living antigen-presenting cells ...
The EMBO Journal vol.8 no.13 pp.4049-4052, 1989
Direct binding of a myasthenia gravis related epitope to MHC class 11 molecules on living murine antigenpresenting cells Edna Mozes, Molly Dayan, Einat Zisman, Stefan Brocke, Arieh Licht and Israel Pecht Department of Chemical Immunology, The Weizmann Institute of
Science, Rehovot 76100, Israel Communicated by I.Pecht
MHC gene products present antigenic epitopes to the antigen receptor on T cells. Nevertheless, direct binding of such epitopes to MHC class II proteins on normal living antigen-presenting cells (APCs) has not yet been demonstrated. We have previously shown a significant difference in the ability of T cells of myasthenia gravis (MG) patients to proliferate in response to the synthetic peptide p195-212 of the human acetylcholine receptor (AChR) a-subunit in comparison to healthy controls. The observed proliferative responses correlated significantly with HLA-DR5. Moreover, lymph node cells of various mouse strains that were primed with the T cell epitope, p195-212, were found to proliferate to different extents. To investigate these observations further, we designed an assay for direct binding of p195-212 to MHC class II proteins on the surface of freshly prepared splenic adherent cells. Binding of a biotinylated p195-212 was monitored using phycoerythrin - avidin by flow cytometry. Fifteen to sixty per cent of the cells were labeled following incubation with the biotinylated peptide. Binding was observed only to splenic adherent cells derived from mouse strains of which T cells were capable of proliferating in response to p195-212. The binding specificity, in terms of epitope structure and its site of interaction on the cells, was shown by its inhibition with an excess of the unlabeled peptide or with the relevant monoclonal anti-I-A antibodies. These results constitute the first direct evidence for the specific binding of a T cell epitope to live APC. Key words: fluorescence/human AChR/inbred mouse strains/ synthetic peptides/T cell epitope
ful. However, recent reports have shown the direct binding of a radiolabeled influenza matrix peptide 17-29 (Ceppellini et al., 1989) and of a biotinylated T cell determinant from influenza hemagglutinin (residue 307-319) (J.B.Rothbard et al., submitted) to products of the human class H MHC locus HLA-DR on HLA homozygous, Epstein -Barr virus (EBV) transformed, B cell lines. Nevertheless, direct binding of T cell epitopes to MHC class II proteins on antigenpresenting cells (APCs) has not yet been demonstrated. Myasthenia gravis (MG) is a well characterized autoimmune disorder, the symptoms of which are caused by an antibody mediated autoimmune response to the nicotinic acetylcholine receptor (AChR) (Lindstrom, 1985). We have previously shown a significant difference in the ability of T cells of MG patients, in comparison to healthy controls, to proliferate in response to the synthetic peptides p195 -212 and p257 -269 which represent sequences of the human AChR a-subunit. The proliferative responses to these peptides correlated significantly with HLA-DR5 and HLA-DR3 respectively (Brocke et al., 1988). Furthermore, various mouse strains that were primed with the T cell epitopes p195-212 and p257-269 of the human AChR responded to different extents to these peptides (unpublished results). Therefore, we designed direct binding experiments of fluorescently labeled derivatives of the p195 -212 peptide to MHC class II molecules on the surface of freshly prepared living splenic adherent cells.
Results Binding capacity of APCs of different inbred mouse strains to p195-212 The proliferative responses of primed lymph node cells from different mouse strains to peptide p195 -212 are summarized in Table I. As can be seen, significant responses were Table I. Proliferative responses of p195-212 primed lymph node cells of various mouse strains Strain
Introduction MHC presentation of T cell epitopes to the antigen receptor of T cells is a prerequisite for the specific activation of these cells (Schwartz, 1985; Buus et al., 1987). The interaction between immunologically active peptides and different MHC molecules has so far been demonstrated in binding studies of such peptides to solubilized, purified molecules of class II MHC and was shown to correlate with the ability of specific class II alleles to restrict the T cell response to specific peptides (Babbit et al., 1985; Buus et al., 1986; Guillet et al., 1987; Watts and McConnell, 1986; Luescher et al., 1988). In the past, experiments designed to demonstrate direct binding of peptides to MHC class II molecules on the surface of intact living cells were unsuccess© IRL Press
H-2 type
Optimal dose of peptide
Background values
Proliferative values
(Ag/well) C3H.SW C57BL/6 BALB/c AKR C3H/He SJL BIO.S
b b d k k s s
5 5 1 20 1 5 20
2368 4220 3133 15975 6432 1163 3471
+ ± ± ± ± ± ±
477 162 233 260 1865 366 115
4105 10296 15449 75602 108532 143568 46482
± ± + ± ± ± 4
341 253 305 4017 3785 3892 1852
Cells (5 x 105/well) were cultured in the presence of different concentrations of the peptide. At the end of 4 days' incubation period 0.5 AtCi of [3H]thymidine was added. Eighteen hours later cells were harvested and radioactivity was counted. Results are given as mean c.p.m. of triplicate cultures i SD in the presence of optimal peptide dose.
4049
E.Mozes et al. 30c
C3H SW
v1
:0t~
144.
~
SJL
~ ~ C7B/
8 0
I"Q A0 -
.. ,
..-.
60
149
W~~~~~~~~~~~ 2
E z
, ...
120 300 ,,,,
180
240
E
BALB/c
60
1;';I-'1
z 149-
120 180 Fluorescence Intensity
240
\
11
300F .
'4;
W-
-
.
.I
.
..
.
C57BL/6
.-
E Z
_.
60 n
ISO 120 Fluorescence Intensity
240
300
E
BALB/c
z
2-11
0
vz
$1, I 4~ Fluorescence Intensity
Fig. 1. Binding of the biotinylated analog of peptide p195-212 to the surface of living splenic adherent cells of various mouse strains. Splenic adherent cells (I x 106/100 /l) were incubated with the biotinylated p195-212 (50 1M) for 20 h and stained with phycoerythrin-streptavidin (0.5 Ag). -, staining of cells incubated with the biotinylated p195-212; . . . .. background fluorescence in the absence of biotinylated peptide.
obtained with the lymph node cells derived from most mouse strains examined. However, high (SJL, BlO.S, C3H/He), intermediate (AKR, BALB/c, C57BL) and low (C3H.SW) responder strains could be clearly resolved. It was therefore important to find out whether MHC class II molecules on APCs of the different mouse strains are capable of binding peptide p195 -212 and whether parallel differences could be observed in the binding capacity of such APCs. We took advantage of the fact that the N-terminal amine of the p195 -212 is not essential for its immunological function (unpublished observations) and modified it covalently so as to attach to it a fluorescent probe. Several procedures were examined and one effective way that we employed in the present study was the combination of biotin-phycoerythrin-streptavidin. Figure 1 illustrates binding measurements of the biotinylated p195 -212 peptide to freshly prepared splenic adherent cells of seven mouse strains. Binding was monitored by flow cytometric analysis (FACS-440) using phycoerythrin-streptavidin which forms a high affinity complex with biotin. As can be seen, 15 to about 60% of the cells of all the strains tested were labeled following 20 h incubation with the biotinylated peptide, except for splenic adherent cells of C3H.SW mice that did not bind. Preliminary time course measurements indicated that although significant binding of the peptide to the splenic adherent cells could be observed already following a 2.5 h incubation period with p195-212, optimal binding was
4050
60
120 100 Fluorescence Intensity
240
Fig. 2. Inhibition of binding of the biotinylated p195-212 by monoclonal anti-I-A antibodies. Splenic adherent cells (1 x 106/sample) were co-incubated with the biotinylated p195-212 (50 AM) and the different monoclonal anti-I-A antibodies (0.3 nM, see Materials and methods) for 20 h, and stained thereafter with phycoerythrin-streptavidin.-, staining of the cells incubated with the biotinylated p195 -212; --, staining of the cells incubated with both the biotinylated p195-212 and the monoclonal anti-I-A antibodies; . background fluorescence in the absence of biotinylated peptide. Table II. Inhibition of p195-212 binding to APC of different mouse strains using monoclonal anti-Ia antibodies
Strain
H-2
mAb used
% inhibition
C57BL/6
b
a I-Abd (34-5-3)
90 0 89 14 100 19
BALB/c
d
SJL
s
ai I-Ad (MKD6) a I-Ad (MKD6) a I-Ed (14-4-4) a I-As,b (Y3P) a I-Ad (MKD6)
The monoclonal anti-I-A antibodies were added (0.3 nM final concentration) to the APC together with the biotinylated peptide for 20 h of incubation. Phycoerythrin-streptavidin was added thereafter and the cells were subjected to flow cytometric analysis. All mAbs employed were of the IgG2a isotype.
observed when the cells were incubated with the peptide for 20 h. Inhibition of the binding of p195-212 to APC In order to establish that the p195 -212 is indeed bound to MHC class II molecules on the surface of splenic adherent
MG-related peptide binding to MHC class 11 on cells
cells, monoclonal anti-I-A antibodies were co-incubated with the peptide in the binding experiments. Figure 2 demonstrates that 100% of the binding of p195 -212 to splenic adherent cells of SJL mice was inhibited by the anti-I-AS b antibody (Y3P). Similarly 90% of the peptide binding to C57BL/6 cells was inhibited by anti-I-Ab (34-5-3) and 89% of the peptide binding to BALB/c cells was inhibited by an antiI-Ad monoclonal antibody (MKD6). The inhibition caused by the various anti-I-A antibodies was specific since, as shown in Table II, anti-I-A antibodies of irrelevant specificities did not inhibit significantly the binding of p195-212 to the MHC class II molecules. As shown in Table II, p195 -212 bound to I-A molecules of the BALB/c APC since only an anti-I-Ad antibody was inhibitory whereas an anti-I-Ed monoclonal antibody did not cause significant inhibition. In order to ascertain that the biotinylated pl95-212 is bound to the same site as the unmodified peptide, the splenic adherent cells of BALB/c mice were incubated with an excess of p195-212 3 h prior to the addition of the biotinylated peptide to the incubation mixture. Figure 3 shows that the unmodified peptide inhibited 90% of the binding and hence competes with the biotinylated peptide for the MHC class II molecules of BALB/c cells. Table III demonstrates that the inhibition observed with the unmodified p195-212 is specific to peptides that are lad restricted, since a peptide with the sequence 12-26 of the Torpedo AChR a-subunit, the response to which is H-2d restricted, reduced the direct binding of the biotinylated p195-212 (51.4%) to the background level (9.6%). By contrast, a peptide with the sequence 90-101 of the myelin basic protein which is H-2s restricted did not affect the binding capacity of p195 -212 (49% binding, Table III) to BALB/c adherent cells. Peptide 90-101 did not bind directly to BALB/c APCs since following biotinylation, only 11% staining of the live APCs was observed, a value which is not significantly above background (8.9%, Table Ill). Thus, the data shown in Figure 3 and Table III also establish the binding specificity in terms of the requirements for the peptide structure.
Discussion In this study we have demonstrated for the first time the direct binding of a synthetic peptide which constitutes a stretch of the human AChR sequence to MHC class II molecules on living APCs derived from several different mouse strains. The observed binding capacity for the peptide correlated with the proliferative potential of the different mouse strains examined. T cells recognize antigens in association with MHC molecules on the APC surface (Unanue, 1984; Buus et al., 1987). This phenomenon became evident following the demonstration of antigenic peptides binding to purified MHC class II molecules (Babbit et al., 1985; Buus et al., 1986; Allen et al., 1987; Guillet et al., 1987). Very recently, direct binding to MHC class I molecules has been demonstrated as well (Bouillot et al., 1989; Chen and Parham, 1989). Most binding studies have been performed with antigenic epitopes derived from conventional antigens. However, it has been of great importance to extend them to autoantigens since the understanding of the binding of the latter to APCs and their handling by these cells could shed light on the mechanisms
L) E
-9o
zf C)14910 ~ 60 120
60
180
4 240
Fluorescence Intensity Fig. 3. Inhibition of binding of the biotinylated p195-212 in the presence of the unmodified peptide. Splenic adherent cells (1 x 106 cells/100 ul) were incubated with an excess (20-fold) of the unmodified p195-212 for 3 h prior to the addition of the biotinylated pl95-212 (50 AM). Phycoerythrin-streptavidin (0.5 ytg) was added thereafter. staining of the cells incubated with the biotinylated p195-212; ---* staining of the cells incubated with an excess of the . background fluorescence in the absence unmodified peptide; of biotinylated peptide. .
Table III. Binding of biotinylated p195-212 to BALB/c APCs in the presence of various unmodified peptides APCs
Biotinylated
Unmodified peptide
% binding
Human AChR 195-212 Torpedo AChR 12-26 Myelin basic protein 90- 101
8.9 51.4 11.7 9.6 49
p195 -212 (50 AM) +
-
+ + + +
+ + + +
Splenic adherent cells of BALB/c mice were incubated with an excess of the unmodified peptide (20-fold over the biotinylated p195-212) for 3 h prior to the addition of the biotinylated peptide. Phycoerythrin-streptavidin was added thereafter and the cells were subjected to flow cytometric analysis. The biotinylated peptide with sequence 90- 101 of the myelin basic protein yielded 11% binding to BALB/c APCs.
involved in autoimmunity. In the present study binding of a synthetic peptide (p195 -212), which constitutes a stretch of the ca-subunit sequence of the human AChR, has been examined. This peptide was designed to be a T cell epitope
(Rothbard and Taylor, 1988) and has indeed been shown by us previously to stimulate T cells of MG patients. The stimulatory capacity of p195-212 has been found to be associated with HLA-DR5 (Brocke et al., 1988). We have now further demonstrated (Table I) strain dependent differences in the ability of lymph node cells to proliferate
p195-212. It thus appears that p195-212 is a T cell epitope which may be relevant to MG and is therefore appropriate for the studies described here. Indeed, we have shown that the peptide binds to MHC molecules on the surface of adherent cells of the responding mouse strains. The interactions between fresh splenic adherent cells and the biotinylated p195-212 were examined. Binding of peptides to EBV transformed human B cells has recently been shown (Ceppellini et al., 1989; Rothbard et al., submitted). However, the present data constitute the first direct evidence
to
4051
E.Mozes et al.
for T cell epitopes binding to normal murine live APC. The binding of the biotinylated peptide was shown to be specific in terms of both binding site on the cells and epitope structure, since it could be inhibited with the relevant monoclonal anti-I-A antibodies (Figure 2 and Table II) and with excess of the free p195 -212 (Figure 3). Furthermore, an H-2d restricted peptide of the Torpedo AchR a-subunit (12-26) could also inhibit efficiently the binding of p195 -212 to BALB/c APCs (Table III). It is of interest that binding of p195 -212 to splenic adherent cells of BALB/c mice could be inhibited by anti-I-A but not by anti-I-E monoclonal antibodies since it has been previously reported that the ability to respond to AChR and disease susceptibility are mapped to I-A (Christadoss et al., 1979, 1985; Walder et al., 1983). The observed binding provides an important tool to screen for biological responsiveness and the results of such experiments might shed light on the role of p195-212 and other MG-related T cell epitopes in the pathogenesis of the disease. Furthermore, this approach is currently being employed in studies aimed at better understanding the mechanism of epitope -MHC interactions.
Materials and methods Synthetic peptides The human AChR ca-subunit peptide with sequence 195 -212 (DTPYLDITYHFVMQRLPL) was synthesized and characterized (Zamvil et al., 1986), and kindly provided by Dr Jonathan Rothbard (ICRF, London, UK). An H-2d restricted peptide with sequence 12-26 (LENYNKVIRPVEHHT) of the Torpedo AChR a-subunit was kindly provided by Dr Jean-Gerard Guillet (Institut Pasteur, Paris, France). A peptide with sequence 90- 101 (HFFKNIVTPRTP) of the myelin basic protein, restricted to H-2s was kindly provided by Dr Lawrence Steinman (Stanford University School of
Medicine, CA, USA).
Monoclonal anti-I-A antibodies The following monoclonal anti-Ia antibodies were used in the study: Anti I-Asb (Y3P, Janeway et al., 1984); Anti I-Ab.d (34-5-3, Ozato et al., 1982); Anti I-Ad (MKD6, Kappler and Marrack, 1981); Anti I-Ed (14-4-4, Sachs and Ozato, 1980). All monoclonal antibodies are of the IgG2a subclass.
N-Terminal biotinylation of the peptide p195-212 1. 1 mg of p195 -212 (or an equivalent amount of the other peptides) was dissolved in 0.5 ml of aqueous 0.1 M NaHCO3. When necessary 30 ytl 0.1 M NaOH was added in order to achieve complete solubilization. The solution was cooled in ice. Biotin-N-hydroxy-succinimide (BNHS) was added to 30% molar excess [133 jg/133 Il a, N-dimethylformamide (DMF)]. The reaction was allowed to proceed at 0°C for 2 h (Wilchek and Bayer, 1988). Since excess of unreacted BNHS is hydrolyzed in the reaction mixture, it does not interfere with binding to the cells and therefore no further purifica-
tion of the conjugate was performed. Mice Mice of the different inbred strains C3H.SW, C57BL/6, BALB/c, AKR, C3H/He, SJL and B10.S were obtained from the Jackson Laboratory, Bar Harbor, USA, or from Olac, Blackthorn, Bicester, Oxon, UK. All mice were used at the age of 8-12 weeks.
Proliferative responses Lymph node cells of mice taken 10- 14 days after primary immunization were prepared in a single cell suspension and cultured in the presence of different concentrations of p195 -212. The assays were performed in flat bottom microtiter plates (Nunc, Roskilde, Denmark) at a cell concentration of S x 105/well in RPMI medium supplemented with 1% autologous normal mouse serum (Axelrod et al., 1986) 2 mM glutamine, 1 mM sodium pyruvate, nonessential amino acids, 100 U/ml penicillin, 100 ,tg/ml streptomycin, 0.25 yg/ml fungizone (Bio Lab, Jerusalem, Israel), 5 x l0-5 M 2-mercaptoethanol (Fluka Ab, Buchs, Switzerland) and 10 mM HEPES buffer (Raught Ltd, Essex, UK). Concanavalin A (Con A, Bio-Yeda, Israel) was used as a control for culture conditions. After 4 days of incubation [3H]thymidine (0.5 ACi of S Ci/mmol, Nuclear Research Center, Negev, Israel) was added to the cultures for 18 h. Thereafter cells were harvested and radioactivity was counted. Results are expressed as mean counts per minute (c.p.m.) of triplicate cultures iSD.
4052
Binding of p195-212 to splenic adherent cells Spleen cells (1 x 108) suspended in RPMI-1640 and supplemented with 10% FCS were incubated in Petri dishes (100 x 20 mm, Falcon Plastic)
at 37°C for 1 h. Thereafter non-adherent cells were removed and the plates were washed three times with RPMI and placed on ice. Adherent cells were collected using a rubber policeman. Splenic adherent cells (1 x 106/sample) were incubated with the biotinylated peptide p195 -212 (50 AM) in phosphate-buffered saline (PBS) containing 0.1% bovine serum albumin (BSA), in a 37°C incubator containing 5% CO2 for 20 h. The cells were washed twice at 4°C with the above solution. Phycoerythrin-streptavidin (Biomeda Corporation, CA) was then added (0.5 Ag/100 Al cell suspension) and the reaction mixture was incubated for 30 min at 4°C. After washing ( x2 with PBS containing 0.1 % BSA) the cells were analyzed by flow cytometry using a FACS 440
(Becton-Dickinson). Phycoerythrin was excited using 488 nm (blue line) and emission was monitored at 575 nm. In each analysis minimally 104 cells were examined. For inhibition of binding, monoclonal anti-Ia antibodies (0.3 nM final concentration) were added to the first incubation mixture together with the biotinylated peptide. In order to inhibit the binding, an excess (20-fold) of the unmodified p195 -212 was added to the cells 3 h prior to the addition of the biotinylated p195-212.
Acknowledgements The authors are grateful to Drs Jonathan Rothbard, Jean-Gerard Guillet and Lawrence Steinman for providing synthetic peptides. The research was supported in part by the Los Angeles Chapter of the Myasthenia Gravis Foundation and by the Crown Foundation. I.P. acknowledges the support of the J.M.Heinemann Foundation.
References Allen,P.M., Matsueda,G.R., Evans,R.J., Dunbar,J.B., Marshall,G.R. and Unanue,E.R. (1987) Nature, 327, 713-715. Axelrod,O. and Mozes,E. (1987) Immunobiology, 172, 99-109. Babbit,B.P., Allen,P.M., Matsueda,G., Haber,E. and Unanue,E.R. (1985) Nature, 317, 359-361. Bouillot,M., Choppin,J., Cornille,F., Martinon,F., Papo,T., Gomard,E., Fournie-Zaluski,M.-C. and Levy,J.-P. (1989) Nature, 339, 473-475. Brocke,S., Brautbar,C., Steinman,L., Abramsky,O., Rothbard,J., Neumann,D., Fuchs,S. and Mozes,E. (1988) J. Clin. Invest., 82, 1894-1900. Buus,S., Settle,A., Colon,S.M., Jevis,D.M. and Grey,H.M. (1986) Cell, 47, 1071-1077. Buus,S., Sette,A. and Grey,H.M. (1987) Immunol. Rev., 98, 115-141. Ceppellini,R., Frumeuto,G., Ferrara,G.B., Tosi,R., Chersi,A. and Pernis,B. (1989) Nature, 339, 392-394. Christadoss,P., Lennon,V.A. and David,C. (1979) J. Immunol., 123, 2540-2543. Christadoss,P., Lindstrom,J., Melvold,R. and Talal,N. (1985) Immunogenetics, 21, 33-39. Cohen,B.P. and Parham,P. (1989) Nature, 337, 743-745. Guillet,J.G., Lai,M.Z., Briner,T.J., Smith,J.A. and Gefter,M.L. (1987) Nature, 324, 260-262. Janeway,C.A., Jr, Courand,P.J., Lerner,E.A., Babich,J., Wettstein,P. and Murphy,D.B. (1984) J. Immunol., 132, 662-667. Kappler,J. and Marrack,P. (1981) J. Exp. Med., 153, 1198-1214. Lindstrom,J. (1985) Annu. Rev. Immunol., 3, 109-131. Luescher,I.F., Allen,P.M. and Unanue,E.R. (1988) Proc. Natl. Acad. Sci. USA, 85, 871-874. Ozato,K., Mayer,N.M. and Sachs,D.H. (1982) Transplantation, 34,
113-119. Rothbard,J.B. and Taylor,W.R. (1988) EMBO J., 7, 93-100. Sachs,D.H. and Ozato,K. (1980) J. Immunol., 124, 533-540. Schwartz,R.H. (1985) Annu. Rev. Immunol., 3, 237-261. Unanue,E.R. (1984) Rev. Immunol., 2, 395-428. Waldor,M.K., Sriram,S., McDavitt,H.O. and Steinman,L. (1983) Proc. Natl. Acad. Sci. USA, 80, 2713-2717. Watts,T.H. and McConnell,H.M. (1986) Proc. Natl. Acad. Sci. USA, 83, 9660-9664. Wilchek,M. and Bayer,E.A. (1988) Anal. Biochem., 171, 1-32. Zamvil,S., Mitchel,D., Moore,A., Kitamura,K., Steinman,L. and Rothbard,J. (1986) Nature, 324, 258-260. Received
on
July 31, 1989; revised on September 1, 1989
