DY determinants, possibly associated with novel class II ... · dy determinants, possibly...

DY DETERMINANTS, POSSIBLY ASSOCIATED WITH NOVELCLASS II MOLECULES, STIMULATE AUTOREACTIVE CD4+

T CELLS WITH SUPPRESSIVE ACTIVITY

BY GRAHAM PAWELEC,* NELSON FERNANDEZ,t THOMAS BROCKER,*E. MARION SCHNEIDER,* HILLIARD FESTENSTEIN,t AND

PETER WERNET*

From the *Immunology Laboratory, Medizinische Klinik, D-7400 Tiibingen,Federal Republic ofGermany, and the *Department ofImmunology, London Hospital Medical

College, London, United Kingdom El 2AD

The human MHC comprises loci encoding at least three main segregant seriesof polymorphic HLA class II products, HLA-DR, -DQ, and -DP. These appearto have broadly similar functions, at least in vitro, since all can serve as restrictionelements for T cell antigen recognition (1-3), as alloantigenic targets for T cell-mediated cytolysis (4-6), and as stimulatory antigens for lymphoproliferativeresponses (6) . Quite probably, not all class II products have even now beenidentified and characterized at the protein level (7). Further, the presence ofadditional class 11 genes, DOf and DZa, has also been reported (8, 9) . No surfaceproducts of these genes have yet been identified, although they are expressed atthe mRNA level (8, 9) . The existence of more class II molecules derived bytranscomplementation (10) or present as mixed isotypes (11, 12) is also to beexpected .mAbs reacting specifically with DR, DQ, or DP molecules have proven critical

in investigating the structure and function of class II moieties . Moreover, mAbsreacting with epitopes broadly distributed on products of more than one locusmay be informative in the definition of potential novel class II molecules (7) .Using mAb TU39, which appears to possess such a uniquely broad reactivity,leukemias of different lineages were often shown to contain a much lowerpercentage of DR', DP', and DQ+ cells than of TU39+ cells (13, 14). This extrareactivity could be interpreted as reflecting the reactivity of mAb TU39 withclass II molecules additional to the established DR, DQ, or DP series .

In a complementary approach, the identification of novel class II antigens maybe accomplished by using cloned lines from in vitro primings between allogeneicdonors thatched for established class II specificities, and by blocking theirstimulation with class II-specific mAbs. HLA-DP antigens were first demon-strated with such a primed lymphocyte typing (PLT)' technique (15) . Using aThis study was supported by the Deutsche Forschungsgemeinschaft grants SFB-120. AI, B5a, andF2, and by the Medical Research Council of the United Kingdom.

' Abbreviations used in this paper:

B-LCL, B-lymphoblastoid cell line ; LAD, lymphocyte-activatingdeterminant ; LP, lymphocyte proliferation ; PLT, primed lymphocyte test ; SA, suppressive activity ;SACl-RAM, Staphylococcus aureus Cowan 1 coupled to rabbit anti-mouse Ig; SC, suppressive cell ;TCC, T cell clone.

,J . Exp. MED. © The Rockefeller University Press - 0022-1007/88/01/0243/19 $2.00

243Volume 167 February 1988 243-261

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from

Published February 1, 1988

DY DETERMINANTS, POSSIBL Y ASSOCIA TED WITH NOVEL

CLASS 11 MOLECULES, STIMULATE AUTOREACTIVE CD4+

T CELLS WITH SUPPRESSIVE ACTIVITY

By GRAHAM PAWELEC,* NELSON FERNANDEZ,:I: THOMAS BROCKER,* E. MARI ON SCHNEIDER,* HILLIARD FESTENSTEIN,:I: AND

PETER WERNET*

From the *lmmunology Laboratory, Medizinische Klinik, D-7400 Tübingen, Federal Republic of Germany; and the :l:Department of Immunology, London Hospital Medical

College, London, United Kingdom EI 2AD

The human MHC comprises loci encoding at least three main segregant series of polymorphie HLA class II products, HLA-DR, -DQ, and -DP. These appear to have broadly similar functions, at least in vitro, since all can serve as restrietion elements for T cell antigen recognition (1-3), as alloantigenic targets for T cellmediated cytolysis (4-6), and as stimulatory antigens for Iymphoproliferative responses (6). Quite probably, not aB class II products have even now been identified and characterized at the protein level (7). Further, the presence of additional class II genes, DOß and DZa, has also been reported (8, 9). No surface products of these genes have yet been identified, although they are expressed at the mRNA level (8, 9). The existence of more class 11 molecules derived by transcomplementation (10) or present as mixed isotypes (11, 12) Is also to be expected.

mAbs reacting specifically with DR, DQ, or DP molecules have proven critical in investigating the structure and function of class II moieties. Moreover, mAbs reacting with epitopes broadly distributed on products of more than one locus may be informative in the definition of potential novel class 11 moleeules (7). Using mAb TÜ39, which appears to possess such a uniquely broad reactivity, leukemias of different lineages were often shown to contain a much lower percentage of DR+, DP+, and DQ+ cells than ofTÜ39+ cells (13,14). This extra reactivity could be interpreted as reflecting the reactivity of mAb TÜ39 with class Il molecules additional to the established DR, DQ, or DP series.

In a complementary approach, the identification of novel class II antigens may be accomplished by using cloned lines from in vitro primings between allogeneic donors matched for established class 11 specificities, and by blocking their stimulation with class II-specific mAbs. HLA-DP antigens were first demonstrated with such a primed Iymphocyte typing (PLT)I technique (15). Using a

This study was supported by the Deutsche Forschungsgemeinschaft grants SFB-120. AI, B5a, and F2, and by the Medical Research Council ofthe United Kingdom.

I Abbreviations used in this paper: B-LCL, B-lymphoblastoid cellline; LAD, lymphocyte-activating determinant; LP, lymphocyte proliferation; PL T, primed lymphocyte test; SA, suppressive activity; SACI-RAM, Staphylocvccus aureus Cowan I coupled to rabbit anti-mouse Ig;SC, suppressive cell; TeC, T cell clone.

J. Exp. MED. © The RockefeIler University Press· 0022-1007/88/01/0243/19 $2.00 243 Volume 167 February 1988 243-26 I

http://jem.rupress.org

244

AUTOREACTIVE SUPPRESSOR CELLS

similar rationale we previously reported (16) apparently novel lymphocyte-activating determinants (LADS) showing little polymorphism in the population,whose stimulation could be blocked by mAb TU39, but not by a number ofother less broadly reactive class II-specific mAbs. These LADS have beenoperationally designated DY (14) . The CD4' cloned lines responding to DYLADs exhibited weak and poorly biphasic proliferative responses not associatedwith the presence of any identifiable HLA specificity on the stimulating cells(16) . Such DY-reactive T cell clones (TCC) were also unusual in their ability toinhibit proliferative responses of other lymphocytes in an apparently HLA-unrestricted fashion (16) . In the present report it is shown that cells stimulatedby DY determinants display autoreactivity and could therefore constitute a self-maintaining circuit with suppressive activity for lymphocyte proliferation (LP) .Stimulation of LP and induction of suppressive activity via DY is shown not tobe blocked by mAbs specific for DR, DP, or DQ, but is blocked by mAb SG520and TU39. The latter mAb is shown to precipitate a putative novel class II-likenon-DR, -DQ, -DP molecule, which is therefore a good candidate for thestructure bearing DY determinants .

Materials and MethodsLymphocyte Priming.

Peripheral blood mononuclear cells (PBMC) from donors typedfor HLA-A, B, C, DR, Dw, DQw, and DPw antigens were isolated by density gradientcentrifugation . Stimulator cells were 7-irradiated at 20 Gy and mixed with unirradiatedresponding cells in equal amounts at 10' cells/ml . Priming cultures were performedbetween DR/Dw and DQ phenotypically matched donors in 16-mm-diameter cluster platewells (Costar, Cambridge, MA) or 25-cm2 tissue culture flasks (Falcon Labware, Oxnard,CA) in medium consisting of RPMI 1640 + 25 mM Hepes supplemented with 10% heat-inactivated pooled nontransfused male serum, with antibiotics . After 6 d, cultures weresupplemented with 20% vol/vol of conditioned medium (Lymphocult T; Biotest, Frank-furt, Federal Republic ofGermany) to a final concentration equivalent to 20 U/ml of IL-2 . After 10 d, the cultures were harvested, washed, and restimulated with twice theirnumber of cells from the original stimulating cell donor, in fresh medium supplementedwith Lymphocult T. After a further 4 d, PLT reagents were harvested and cryopreservedfor use in PLT tests .

Cloning and Cell Line Propagation. PLT cells were restimulated with the originalpriming cells in Costar wells in medium supplemented with 20% of Lymphocult T, using2 x 10' primed cells plus 4 x 10 5 stimulators in 2 ml medium with IL-2 at 20 U/ml . After4 d, cells were cloned at limiting dilution and cultured by intermittent feeding with freshmedium (every 2-4 d) and periodic restimulation with specific stimulator PBMC (every 7d) . Limiting dilution was performed in 1-mm diameter culture wells, seeding 0.3-0.45cells/well with 10 4 30 Gy irradiated PBMC stimulator cells . Control plates containing 4.5and 45 cells/well were set up to ensure single-hit characteristics of the limiting dilutioncurves (r > 0.9) . Contents of wells containing growing cells were transferred at 7-10 d to7-mm diameter culture wells with 10 5 stimulator cells and fresh medium . After a further3-5 d, cells were transferred to Costar wells, where they were maintained as above, usingpooled lymphocytes from at least 20 donors as stimulators .PLT Restimulation .

Primed cells, cultured cells, and cloned cells were restimulated inU-well microtiter plates, generally using 10 4 responders and 10 5 stimulators (PBMC) perwell . B-lymphoblastoid cell lines (B-LCL) and cloned TCC stimulators were used at lowercell numbers . PBMC and TCC were irradiated at 20 Gy and B-LCL at 80 Gy . Culturekinetic was varied . Medium was RPMI, 25 mM Hepes, 10% human serum, and antibiotics .37 kBq/well ["H]TdR (sp act, 185 GBq/mmol ; Amersham Corp., Arlington Heights, IL)was added 18 h before termination of the cultures .

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from


244 AUTOREACTIVE SUPPRESSOR CELLS

similar rationale we previously reported (16) apparently novel Iymphocyteactivating determinants (LADs) showing little po~ymorphism in the population, whose stimulation could be blocked by mAb TU39, but not by a number of other less broadly reactive c1ass II-specific mAbs. These LADs have been operationally designated DY (14). The CD4+ c10ned lines responding to DY LADs exhibited weak and poorly biphasic proliferative responses not associated with the presence of any identifiable HLA specificity on the stimulating cells (16). Such DY-reactive T cell clones (TCC) were also unusual in their ability to inhibit proliferative responses of other lymphocytes in an apparently HLAunrestricted fashion (16). In the present report it is shown that cells stimulated by DY determinants display autoreactivity and could therefore constitute a selfmaintaining circuit with suppressive activity for lymphocyte proliferation (LP). Stimulation of LP and induction of suppressive activity via DY is shown not to be blocked by mAbs specific for DR, DP, or DQ, but is blocked by mAb SG520 and TÜ39. The latter mAb is shown to precipitate a putative novel c1ass II-like non-DR, -DQ, -DP moleeule, which is therefore a good candidate for the structure bearing DY determinants.

Materials and Methods Lymphocyte Priming. Peripheral blood mononuclear cells (PBMC) from donors typed

for HLA-A, B, C, DR, Dw, DQw, and DPw antigens were isolated by density gradient centrifugation. Stimulator ceHs were -y-irradiated at 20 Gy and mixed with unirradiated responding cells in equal amounts at 106 cells/ml. Priming cultures were performed between DR/Dw and DQ phenotypicaHy matched donors in 16-mm-diameter cluster plate wells (Costar, Cambridge, MA) or 25-cm2 tissue culture flasks (Fa\con Labware, Oxnard, CA) in medium consisting of RPMI 1640 + 25 mM Hepes supplemented with 10% heatinactivated pooled nontransfused male serum, with antibiotics. After 6 d, cultures were supplemented with 20% vol/vol of conditioned medium (Lymphocult T; Biotest, Frankfurt, Federal Republic of Germany) to a final concentration equivalent to 20 U /ml of IL-2. After 10 d, the cultures were harvested, washed, and restimulated with twice their number of cells from the original stimulating cell donor, in fresh medium supplemented with Lymphocult T. After a further 4 d, PL T reagents were harvested and cryopreserved for use in PL T tests.

Cloning and Cell Line Propagation. PL T cells were restimulated with the original primin~ cells in Costar wells in medium supplemented with 20% of Lymphocult T, using 2 X 1 Ü" primed ceHs plus 4 X 105 stimulators in 2 ml medium with IL-2 at 20 U /ml. After 4 d, cells were cloned at limiting dilution and cultured by intermittent feeding with fresh medium (every 2-4 d) and periodic restimulation with specific stimulator PBMC (every 7 d). Limiting dilution was performed in I-mm diameter culture wells, seeding 0.3-0.45 cells/well with 104 30 Gy irradiated PBMC stimulator cells. Control plates containing 4.5 and 45 cells/well were set up to ensure single-hit characteristics of the limiting dilution curves (r > 0.9). Contents of wells containing growing ceHs were transferred at 7-10 d to 7-mm diameter culture wells with 105 stimulator cells and fresh medium. After a further 3-5 d, cells were transferred to Costar weHs, where they were maintained as above, using pooled lymphocytes from at least 20 donors as stimulators.

PLT Restimulation. Primed ceHs, cultured cells, and cloned cells were restimulated in U-well microtiter plates, generally using 104 responders and 105 stimulators (PBMC) per weIl. B-lymphoblastoid cell lines (B-LCL) and cloned TCC stimulators were used at lower cell numbers. PBMC and TCC were irradiated at 20 Gy and B-LCL at 80 Gy. Culture kinetic was varied. Medium was RPMI, 25 mM Hepes, 10% human serum, and antibiotics. 37 kBq/well [3HJTdR (sp act, 185 GBq/mmol; Amersham Corp., Arlington Heights, IL) was added 18 h be fore termination of the cultures.


PAWELEC ET AL.

245

Quantification of IL-2 Secretion.

Supernatants of specifically stimulated T cell cloneswere prepared by incubating 4 x 10 6 cloned cells with 5 x 10 80 Gy-irradiated B-LCLin 2 ml of medium containing 1 % HS for 48 h. To quantitate IL-2 production, superna-tants were titrated onto IL-2-dependent T cell lines and ['H]TdR incorporation wasmeasured after 24 h. Highly purified natural IL-2 (Lymphocult T-HP; Biotest) was usedas a positive control, and Probit analysis was applied on an IBM-PC (program by Blaurock,M., G. Pawelec, and P. Wernet, submitted for publication) to calculate units of IL-2 withreference to the International Union of Immunological Societies-Biological ResponseModifiers Program (IUIS-BRMP) standard.mAbInhibition ofStimulation.

mAbs were added to restimulation assays at the initiationof culture and remained present for the duration of the assay. Stimulating cells wereplated first, followed by mAbs, and lastly, after a short pause, the responding cells wereplated . Previous experiments with the Tubingen series mAbs used here (TU22, anti-DQ;TU34, TU37, anti-DR; and TU59, binding the products of at least DR and DP) hadprovided no evidence of stimulation-inhibition mechanisms divorced from effects of themAbs binding to the stimulating cells (17, 18). Nonetheless, in certain experiments,stimulating cells were pretreated with mAbs by incubating in undiluted hybridomasupernatant at 4°C for 1 h, followed by washing three times in cold medium, andimmediate addition to culture wells already containing the responders . B-LCL used asstimulators had been shown by FACS analysis to bind strongly the mAbs used in theblocking experiments, although the level of expression of DQ and DP antigens was almostalways lower than that of DR antigens (data not shown). TO mAbs (13, 14, 17-20) wereused as tissue culture supernatants (25%, 1-3 ttg/ml), other mAbs as a 1 :100 dilution ofascites. Additional mAbs used were as follows: L243 (6), Q2/70 (18, 20), and SG157 (18)specific for HLA-DR (although Q2/70 may also bind DQ, reference 20); SPV-L3 (4) andLeu10 (18) specific for DQw and DQw1,3 molecules, respectively ; B7/21, specific forHLA-DP (5, 15); and PL5 (6), DA6.231 (6, 18), and SG520 (6, 18) "broadly" reactivewith multiple class II molecules. mAb B7/21 (anti-FA) ascites was a kind gift of F. Bach,Immunobiology Research Center, Minneapolis, MN, and the B7/21 hybridoma was agenerous gift of I. Trowbridge, Salk Institute, San Diego, CA; SPV-L3 was from J . deVries and H. Spits, UNICET, Dardilly, France; SG157 and SG520 came from S. Goyert,Hospital for Joint Diseases, New York; DA6.231 was from K. Guy, MRC Clinical andPopulation Cytogenetics Unit, Edinburgh, Scotland; Q2/70 was from S. Ferrone, NewYork Medical College, New York; PL5 was from R. Knowles, Memorial Sloan-KetteringCancer Center, New York ; and Leu10 and L243 were from Becton Dickinson & Co.,Mountain View, CA.

Sequential Immunodepletion� Procedures. For immunoprecipitation studies, mAbsL243, TU22, B7/21, and TU39 were purified by protein A-Sepharose 4B affinitychromatography according to Ey et al . (21) .TCC were biosynthetically labeled by resuspending 4 x 10' exponential growth phase

cells in 4 ml MEM without L-methionine . The medium was supplemented with 5%methionine-free (dialyzed three times) human serum and 20 U/ml IL-2, and, after 1 h at37°C, with 18,500 kBq of ['

nS]methionine (3 .7 x 10' ° Bq/mmol; Amersham Corp.) . Aftera further 4-h incubation, radiolabeling was terminated by washing the cells three timeswith cold saline 0.02% sodium azide, and 1 mg/ml L-methionine . Washed cells wereresuspended in 2 ml of NP-40 lysis buffer (10 mM Tris-HCl, pH 7.4, 150 mM NaCl,0.5% NP-40, 0.02% azide, and 1 mM PMSF), and incubated for 30 min on ice. Insolublematerial was removed by centrifuging at 10,000 g for 30 min . Cell extracts were loadedonto a Lens culinaris affinity column for glycoprotein purification according to Haymanand Crumpton 12). In a second series of experiments, 2 x 10' washed TCC were surfacelabeled with "~ I by the lactoperoxidase catalyzed method (23) . Briefly, TCC wereresuspended in 0 .5 ml of cold saline to which 18,000 kBq of carrier-free sodium `46 1(Amersham Corp.) was added. In rapid sequence, 30 U of lactoperoxidase (sp act, 90U/ml ; Calbiochem-Behring Corp., San Diego, CA), 15 ul of glucose oxidase (SigmaChemical Co.), and 75 ul of glucose (50 mg/ml, Sigma Chemical Co.) were added, andthe reaction was stopped after 30 min by repeated washing in cold saline .

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from


PA WELEC ET AL. 245

QuantiJication of IL-2 Secretion. Supernatants of specificallr stimulated T cell clones were prepared by incubating 4 X 10& cloned cells with 5 X 10 80 Gy-irradiated B-LCL in 2 ml of medium containing 1 % HS for 48 h. To quantitate IL-2 production. supernatants were titrated onto IL-2-dependent T cell lines and [SH]TdR incorporation was measured after 24 h. Highly purified natural IL-2 (Lymphocult T-HP; Biotest) was used as a positive control. and Probit analysis was applied on an IBM-PC (program by Blaurock. M .• G. Pawelec. and P. Wernet. submitted for publication) to calculate units of IL-2 with reference to the International Union of Immunological Societies-Biological Response Modifiers Program (IUIS-BRMP) standard.

mAb Inhibition of Stimulation. mAbs were added to restimulation assays at the initiation of culture and remained present for the duration of the assay. Stimulating cells were plated first. followed by mAbs. and lastly. after a short pause. the respondjng cells were pl~.ted. Prc:.vious experiments with"the Tübingen series mAbs used here (TU22. anti-DQ; TU34. TU37. anti-DR; and TU39. binding the products of at least DR and DP) had provided no evidence of stimulation-inhibition mechanisms divorced from effects of the mAbs binding to the stimulating cells (17. 18). Nonetheless. in certain experirrients. stimulating cells were pretreated with mAbs by incubating in undiluted hybridoma supernatant at 4°C for 1 h, followed by washing three times in cold medium, and immediate addition to culture weil, already containing the responders. B-LCL used as stimulators had been shown by F ACS analysis to bind strongly the mAbs used in the blocking experiments, although t~e level of expression of DQ and DP antigens was alm ost always lower than that ofDR antIgens (data not shown). TU mAbs (13.14,17-20) were used as tissue culture supernatants (25%. 1-3 "g/ml), other mAbs as a 1:100 dilution of ascites. Additional mAbs used were as folIows: L243 (6), Q2170 (18,20), and SG157 (18) specific for HLA-DR (although Q2170 mayaIso bind DQ, reference 20); SPV-L3 (4) and Leu10 (18) specific for DQw and DQw1,3 molecules. respectively; B7/21. specific for HLA-DP (5. 15); and PL5 (6). DA6.231 (6, 18). and SG520 (6, 18) "broadly" reactive with multiple class II molecules. mAb B7/21 (anti-FA) ascites was a kind gift of F. Bach. Immunobiology Research Center, Minneapolis. MN. and the B7/21 hybridoma was a generous gift of I. Trowbridge. Salk Institute. San Diego, CA; SPV-L3 was from J. de Vries and H. Spits. UNICET, Dardilly, France; SG157 and SG520 came from S. Goyert, Hospital for Joint Diseases, New York; DA6.231 was from K. Guy. MRC Clinical and Population Cytogenetics Unit, Edinburgh, Scotland; Q2/70 was from S. Ferrone. New York Medical College, New York; PL5 was from R. Knowles, Memorial Sloan-Kettering Cancer Center. New York; and Leu10 and L24S were from Becton Dickinson & Co., Mountain View, CA.

Sequ,n~~al Immunodepletion" Procedures. For immunoprecipitation studies, mAbs L243. TU22. B7/21. and TU39 were purified by protein A-Sepharose 4B affinity chromatography according to Ey et al. (21).

TCC were biosynthetically labeled by resuspending 4 X 107 exponential growth phase cells in 4 ml MEM without L-methionine. The medium was supplemented with 5% methionine-free (dialyzed three times) human serum and 20 U/ml IL-2, and. after 1 hat 37°C. with 18.500 kBq of[s5S]methionine (3.7 X 1010 Bq/mmol; Amersham Corp.). After a further 4-h incubation, radiolabeling was terminated by washing the cells three times with cold saline 0.02% sodium azide, and 1 mg/mI L-methionine. Washed cells were resuspended in 2 ml of NP-40 lysis buffer (10 mM Tris-HCI, pH 7.4, 150 mM NaCI, 0.5% NP-40, 0.02% azide, and 1 mM PMSF). and incubated for 30 min on ice. Insoluble material was removed by centrifuging at 10,000 g for 30 min. Cell extracts were loaded onto a Lens culinaris affinity column for glycoprotein purification according to Hayman and Crumpton &22). In a second series of experiments, 2 X 107 washed TCC were surface labeled Wlth 12 1 by the lactoperoxidase catalyzed method (23). Briefly, TCC were resuspended in 0.5 ml of cold saline to which 18.000 kBq of carrier-free sodium 1251 (Amersham Corp.) was added. In rapid sequence. 30 U of lactoperoxidase (sp act. 90 U/ml; Calbiochem-Behring Corp., San Diego. CA), 15 "I of glucose oxidase (Sigma Chemical Co.). and 75 "I of glucose (50 mg/mi, Sigma Chemical Co.) were added. and the reaction was stopped after 30 min by repeated washing in cold saline.


246


Imrnunoprecipitation studies were performed as previously described (24) . Briefly,labeled cell extracts were precleared with 100 jul of protein A-bearing Staphylococcusaureus Cowan I (SACI) and 50 yl of SACI coupled to rabbit anti-mouse globulin (SACI-RAM). 2 x 10 6 cell equivalents were then incubated with 1 tag of protein A-Sepharose-purified mAbs at 4°C for I h . Immunocompiexes were then incubated with 50 gel SACI-RAM to ensure identical binding capacity of the SACI to the various mAbs . Theprecipitates were then washed three times in lysis buffer without detergent. For sequentialinnnunoprecipitations, the supernatant was retained for serial transfer at each step totubes containing identical amounts (I Ag) of the different mAbs, followed by SACI-RAM,and repetition of the above . At each step, material was retained for gel analysis as follows :bound proteins were solubilized by boiling for 3 min in elution buffer containing SDSand DTT, and visualized by electrophoresis on 12 .5% polyacrylamide slab gel (25)followed by fixing and fluorography in Amplify (Amersham Corp.) and autoradiographyon Hyperfrlm MP (Amersham Corp.) at -70° .

(:ell-mediated Cytotoxicity Assay .

Standard 5 'Cr-release assays were performed as previ-ously described (26) . Briefly, 2-5 x 10' K562 line cells were incubated with 3 .7 x 10 6 Bqof sodium chromate (sp act, 22 .2 GBq/mg 5 'Cr ; Amersham-Buchler, Frankfurt, FRG) in0 .4 nil culture medium for 90 min, washed three times, and cocuitured for 4 h witheffector cells . Supernatants were removed by pipette thereafter for gamma spectroscopy .

Induction ofSuppressive Activity by T Cell Clones.

2 x 106 PBMC were incubated for 3d with 10 5 20 Gy-irradiated TCC in medium with 10% serum, without IL-2 . Harvestedcells were then cultured with 20 U/ml of IL-2 for at least 7 d before testing by titratinginto MLC . To investigate blockade of SA induction with mAbs, 3-d PBMC + TCCcultures were performed in the presence of 25% supernatant or 1 :100 diluted ascites .Cells were washed, cultured in IL 2 and then added to MLC . TCC cultured alone underthese conditions did not survive, and PBMC cultured alone did not cause suppression .

ResultsRestimulation of Clones Specific for DY Antigens Inhibited by mAb.

In threesensitization/cloning experiments with the same donors, ^-60% of all derivedclones showed autonomous proliferative capacity on alloantigen rechallenge . Ofthese, no more than 10% recognized LADs associated with the disparate DPwspecificity of the stimulator, whereas proliferative responses of the remainderdid not correlate with any DPw specificity . The majority of stimulators elicitedeither clearly positive or equivocal responses, as assessed by an objective clusterprogram . Even for the unequivocally positive responses, restimulation was notassociated with the presence of any particular established class I or class IIspecificity (16) . Surprisingly, the spontaneous [ sH]TdR incorporation of theclones incubated with medium alone was considerably higher than that forrepresentative PLT clones specific for DR-, DQ-, or DP-associated LADS. More-over, PBMC of the autologous donor from whom the clones were derived furtherstimulated these clones, although all responses were very weak compared withDR-, DQ-, or DP-specific responses . This is illustrated for four DY-reactiveclones designated 102-1, 102-3, 105-5, and 106-13 in Fig . 1 . Titrating thestimulating cells (Fig . 1), or altering the kinetics of restimulation (data not shown),indicated that not only did these response patterns remain stable, but thatautologous cells commonly stimulated more strongly than allogeneic cells . Similarresults were obtained when B-LCL instead of PBMC were used as stimulators(Table 1) .We attempted to characterize the stimulatory DY-LADs by blocking stimula-

tion with mAbs. A number of DR and DQ mAbs, and one DP mAb, failed to

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from



Immunoprecipitation studies were performed as previously described (24). Briefly, labeled cell extracts were precleared with I 00 ~I of protein A-bearing Staphylococcus aureus Cowan I (SACI) and 50 ~I of SACI coupled to rabbit anti-mouse globulin (SACIRAM). 2 X 106 cell equivalents were then incubated with 1 ~g of protein A-Sepharosepurified mAbs at 4°C for 1 h. Immunocomplexes were then ineubated with 50 ~I SACIRAM to ensure identieal binding eapaeity of the SACI to the various mAbs. The pl'eeipitates were then washed three times in lysis buffer without detergent. For sequential immunopreeipitations, the supernatant was retained for serial transfer at eaeh step to tubes containing identieal amounts (1 ~g) ofthe different mAbs, followed by SACI-RAM, and repetition of the above. At eaeh step, material was retained for gel analysis as folIows: bound proteins were solubilized by boiling for 3 min in elution buffer containing SDS and DTT, and visualized by eleetrophoresis on 12.5% polyacrylamide slab gel (25) followed by fixing and fluorography in Amplify (Amersham Corp.) and autoradiography on Hyperfilm MP (Amersham Corp.) at -70 0.

Cell-mediated Cytotoxicity Assay. Standard 5lCr-release assays were performed as previously deseribed (26). Briefly, 2-5 X 106 K562 line eeJls were ineubated with 3.7 X 106 Bq of sodium chromate (sp aet, 22.2 GBq/mg slCr; Amersham-Buehler, Frankfurt, FRG) in 0.4 ml eulture medium for 90 min, washed three times, and eoeultured for 4 h with effeetor eells. Supernatants were removed by pipette thereafter for gamma spectroscopy.

Induction 01 Suppressive Activity by T Cell Clones. 2 X 106 PBMC were incubated for 3 d with 10" 20 Gy-irradiated TCC in medium with 10% serum, without IL-2. Harvested cells were then cultured with 20 U Iml of IL-2 for at least 7 d before testing by titrating into MLC. To investigate blockade of SA induction with mAbs, 3-d PBMC + TCC cultures were performed in the presence of 25% supernatant or 1: 1 00 diluted ascites. CeJls were washed, cultured in IL 2 and then added to MLC. TCC cultured alone under these conditions did not survive, and PBMC cultured alone did not cause suppression.

Results

Restimulation of Clones Specific JOT DY Antigens Inhibited by mAb. In three sensitization/cloning experiments with the same donors, -60% of all derived clones showed autonomous proliferative capacity on alloantigen rechallenge. Of these, no more than 10% recognized LADs associated with the disparate DPw specificity of the stimulator, whereas proliferative responses of the remainder did not correlate with any DPw specificity. The majority of stimulators elicited either c1early positive 01' equivocal responses, as assessed by an objective cluster program. Even for the unequivocally positive responses, restimulation was not associated with the presence of any particular established class I 01' c1ass 11 specificity (16). Surprisingly, the spontaneous [3H]TdR incorporation of the clones incubated with medium alone was considerably higher than that for representative PL T clones specific for DR-, DQ-, 01' DP-associated LADs. Moreover, PBMC ofthe autologous donor from whom the clones were derived further stimulated these clones, although all responses were very weak compared with DR-, DQ-, 01' DP-specific responses. This is illustrated for foul' DY-reactive clones designated 102-1, 102-3, 105-5, and 106-13 in Fig. 1. Titrating the stimulating cells (Fig. 1), 01' altering the kinetics of restimulation (da ta not shown), indicated that not only did these response patterns remain stable, but that autologous cells commonly stimulated more strongly than allogeneic cells. Similar results were obtained when B-LCL instead of PBMC were used as stimulators (Table I).

We attempted to characterize the stimulatory DY-LADs by blocking stimulation with mAbs. A number of DR and DQ mAbs, and one DP mAb, failed to


TABLE IDY Expression on B-LCL

Stimulator

Clone

Autologous

Allogeneic

FIGURE 1 . Autoreactivity ofDY-specific TCC. Constant104/well TCC were stimulatedby titrated amounts of irradi-ated PBMC from the same do-nor as the clones (open circles,autologous)or from the donorof the stimulating cells for theMLC from which the cloneswere derived (filled circles,specific) . Results are shown asmedian cpm of triplicate cul-tures (SEM was <12%).

Allogeneic

247

* Data presented as mean cpm t SEM of triplicates of 104 cloned cells per well stimulated with 10 5 PBMC or 2 .5 X 10 4 B-LCL cells .

block stimulation of DY-specific TCC by B-LCL, although each of the mAbs wascapable of blocking stimulation of other clones via the appropriate class II type .Thus, the left hand panel of Fig. 2 shows the relative responses of the DY-specificclones 102-1, 103-1, and 106-8 stimulated in the presence of a range of class II-specific mAbs . For comparison, the right hand panel of Fig. 2 shows stimulationinhibition patterns of the same mAb for anti-DP clone 64-2, anti-DR clone 249-13, and anti-DQ clone 233-7 . Essentially similar results were obtained by usingPBMC or stimulatory TCC instead of B-LCL as stimulators (data not shown) .Blocking of stimulation by a mixture of DR, DQ, and DP mAbs was also notseen (data not shown) . However, the broadly reactive class II-specific mAbsTU39 and SG520, but not PL5 or DA6.231, did block stimulation by DY (Fig .2) . Additionally, TU39 reduced the level of [3H]TdR incorporation to belowthat of clones incubated in medium alone. Inhibitory effects were not caused bynonspecific activity of the mAbs because in the presence of 20 U/ml of IL-2proliferation was not affected by TU39 (Table II). Furthermore, pretreatmentof the stimulating cells with mAb TU39 resulted in the retention of a degree ofinhibition of stimulation (Table II) .mAb TU39 Precipitates Class H-like Molecules Different from DR, DQ, or

DP. TCC were surface labeled with 1251 and 2 x 106 cell equivalents of solublepreparation were subjected to sequential immunoprecipitations designed todeplete molecules reacting with well-characterized mAbs specific for mono-

PBMC B-LCL

4,378 ± 327 5,911 ± 59010,102 t 991 9,628 t 8924,381 t 200 6,528 ± 5375,699 t 686 6,023 t 720

PBMC: B-LCL PBMC B-LCL

102-1 6,353 t 582* 8,972 t 549 4,316 t 335 5,291 t 938103-1 9,568 ± 720 7,623 t 836 7,399 t 877 8,443 t 562106-8 3,281 t 156 3,697 ± 110 2,926 t 126 3,468 t 479106-13 7,281 ± 881 9,428 ± 1,003 5,001 t 323 5,575 t 328

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from


10

o

E 10 0.

PA WELEC ET AL.

102-1

o I I I i I I I I I I I 1 40 20 10 5 2.5 0 40 20 10 5 2.5 0

Number 01 Stimulatory Cella x 1 ö4

TABLE

DY Expression on B-LCL

Stimulator

(:10m' AutologOllS Allogeneic

PBMC B-l.Cl. PBMC B-l.Cl.

102-1 6,353 ± 582' 8,972 ± 549 4,316 ± 335 5,291 ± 938 103-1 9,568 ± 720 7,623 ± 836 7,399 ± 877 8,443 ± 562 106-8 3,281 ± 156 3,697 ± 110 2,926 ± 126 3,468 ± 479 106-13 7,281 ± 881 9,428 ± 1.003 5,001 ± 323 5,575 ± 328

247

FIGURE 1. Autoreactivity of DY-specific TCC. Constant 104/well TCC were stimulated by titrated amounts of irradiated PBMC from the same donor as the clones (open circles, autologous) or from the donor of the stimulating cells for the MLC from which the clones were derived (filled circles, specific). Results are shown as median cpm of triplicate cultures (SEM was <12%).

Allogeneic

PBMC

4,378 ± 327 10,102 ± 991

4,381 ± 200 5,699 ± 686

B-LCL

5,911 ± 590 9,628 ± 892 6,528 ± 537 6,023 ± 720

* Dara presented as JJ1eall cpm ± SEM of triplicates of 10· c10ned cells per weil stimulated with 105 PBMC or 2.5 X 10· B-LCL cells.

block stimulation of DY-specific TCC by B-LCL, although each of the mAbs was capable of blocking stimulation of other clones via the appropriate c1ass 11 type_ Thus, the left hand panel of Fig. 2 shows the relative responses of the DY-specific clones 102-1, 103-1, and 106-8 stimulated in the presence of a range of c1ass 11-specific mAbs, For comparison, the right hand panel of Fig. 2 shows stimulation inhibition patterns of the same mAb for anti-DP clone 64-2, anti-DR clone 249-13, and anti-DQ clone 233-7. Essentially similar results were obtained by using PBMC or stimulatory TCC instead of B-LCL as stimulators (data not shown). Blocking of stimulation by a mixture of DR, DQ, and DP mAbs was also not seen (data not shown), However, the broadly reactive c1ass II-specific mAbs TÜ39 and SG520, but not PL5 or DA6.231, did block stimulation by DY (Fig. 2), Additionally, TÜ39 reduced the level of [3H]TdR incorporation to below that of clones incubated in medium alone. Inhibitory effects were not caused by nonspecific activity of the mAbs because in the presence of 20 U/ml of lL-2 proliferation was not affected by TÜ39 (Table II), Furthermore, pretreatment of the stimulating cells with mAb TÜ39 resulted in the retention of a degree of inhibition of stimulation (Table II).

mAb TÜ39 Precipitates Class II-like Molecules Different from DR, Dß or DP, TCC were surface labeled with 1251 and 2 X 106 cell equivalents of soluble preparation were subjected to sequential immunoprecipitations designed to deplete molecules reacting with well-characterized mAbs specific for mono-


248

102-1

OR

DO

DPBroad

Broad+DY

103-1,

249-13DR

00

OPBroad

Broed+DY

106-8DR


TABLE II

64-2DR

DO

OPBroad

Broad+DY

DR

00

OP

BroadBroad+OY

233-7OR

r .

0

60

100

11l0

0 -I 00 -1 100Percentage Relative Response

DY Stimulation Inhibited by mAb T039

DP

DPBroad

Broad ,.

BroedtDY

Broad+DY

FIGURE 2 .

Inhibition of stimulation by mAbs . (Left) DY-specific TCC; (rs'ght) representativeDP (64-2), DR (249-13) and DQ (233-7)-specific TCC . mAbs were as follows (from top tobottom): DR : T084, Tb37, L243, Q2/70, and SG157; DQ : T022, SPV-L3, Leu 10 ; DP :B7/21 ; broad : PL5, DA6.231 ; broad + DY : T039, SG520 . Results are shown as percentrelative response compared with the value in the presence of nonbinding control mAb(W6/32.HK) .

" Stimulating cells pretreated with mAb TU99.$ MAb'I'089 present for duration of coculture.1 104 TCC were stimulated with 2 .5 x 104 B-I.Cl, cells (Stimulator +) or in medium alone (stimulator -) . Where IL-2 was present

(I L-2 +), 20 U/ud was used . Where mAbTOM) was present (mAb TID39 +), 25% hybridonta culture supernatant was used.

morphic epitopes of DR, DQ, or DP molecules (Fig . 3 a, lanes 1-20) . Fivesequential precipitations with mAb L243 (lanes 1-S) sufficed to remove all DRmolecules (as shown by precipitation with SACI-RAM alone, lanes 6 and 7) . Thiswas followed on the same lysate by four precipitations with T1:J22 to remove DQmolecules (lanes 8-11, and SACI-RAM alone, lanes 12 and 13), and finally foursequential precipitations with 137/21 to remove DP molecules (lanes 14-17,SACI-RAM alone, lanes 18 and 19) . Finally, lane 20 shows that mAb T039 stillprecipitated heterodimeric molecules similar to class II products, even after DR,DQ, and DP molecules were depleted from the lysate .To show that the "extra" TU39+ molecules were indeed synthesized by the

StimulatormAh Tii991l,-2

Clone,108-I 1,869 t 290 6,492 t 690 14,287 t 1,501 2,496 t 948 758 t 114 13 .634 :t 1,271105-5 1,994 t 101 6,180 t 729 16,285 t 1,499 1,981 t 266 999 t 82 15 .200 :t 1,600106-8 1,995 t 204 9,456 t 888 22,187 t 9,009 9,059 t 422 1,055 t 296 21,864 * 2,149

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from



102-1 84-2 OR OR

00 00 OP OP

BroaCl Broad Broad+OY Broad+OY

103-1 249-13 OR DR -00 00

DP OP Broad Broad

Broad+OY - Broad+OY

108-8 233-7 OR DR

00 00 OP OP

Broad .road

Broad+OY =-- .raad+OY

0 110 100 150 0 50 100 110

P.ra.ntlgl Rllatlve Rllpan ..

FIGURE 2. Inhibition of stimulation by mAbl. (L.ft) DY·specific TCC; (",lat) reprelentative DP (64·2), DR (~49.1S)J._and DQ (2SS.7)·specific TCC. mAbl were as followl (from top to bottom): DR: TÜM, TuS7, L24S, Q2/70, and .SOl57; DQ: TÜ22, SPV·LS, Leu 10; DP: 87/21; broad: PL5, DA6.2S1; broad + DY: TÜS9, SG520. Results are shown al percent relative response compared with the value in the prelence of nonbinding control mAb (W6/S2.HK).

TABU 11 DY Stimulation Inhibit,d by mAb rÜJ9

Stimulator mAhTÜ39 11.-2

(;IOl1t"

103·1 105·~

106-H

I,H69 :I: 280 \,894:1: 101 1,995:1: 204

+

6,292:1: 6S0 6,180:1: 729 9,4&6:1: 888

• Stilllllilltinl! cell'l'retr.ated with .nAb TÜS9. * MAb "l'Ü89 pre •• nl for dur.tion ofcocultur •.

+ + +.

+

14,287:1: 1,~01 2,496:1: 828 16,28~:I: 1,499 1,981 :I: 266 22,187:1: 8,008 8,0&8:1: 422

+ + +* +*

+

7&8:1: 114 18,684 :I: 1,271 998:1: 82 I &,200 :I: 1,600

1,0&&:1: 256 21,864:1: 2,149

'10' Tee wer. Slillluhtled with 2.5 X 10' B-Lel. c.lI. (Stil1lulator +) or in l1Iedium alon. (Stimulator -). Wh.r. IL-2 ws. present (11.-2 +). 20 U/11I1 "' •• u •• d. Wher. IIIAb TÜ39 11'.' preoent (mAb TÜ89 +), 2&% hybridoma culture .upernstsnt wa. u •• d.

morphic epitopes of DR, DQ, or DP moleeules (Fig. 3 a, lanes 1-20). Five sequential precipitations with mAb L243 (lanes 1-') sufficed to remove all DR molecules (as shown by precipitation with SACI-RAM alone, lanes 6 and 7). This was followed on the same lysate by four precipitations with TÜ22 to remove DQ molecules (lanes 8-11, and SACI·RAM alone, lanes 12 and 1 J), and finally four sequential precipitations with B7/21 to remove DP moleeules (lanes 14-17, SACI·RAM alone, lanes 18 and 19). Finally, lane 20 shows that mAb TÜ39 still precipitated heterodimeric moleeules similar to dass II products, even after DR, DQ, and DP molecules were depleted from the lysate.

To show that the "extra" TU39+ molecules were indeed synthesized by the


PAWELEC ET AL .

249

FIGURE 3. (a) mAb TU39detects molecules other thanHLA-DR, -DQ, and -DP. 2 x108 cell equivalents of 125I-lac-toperoxidase surface-labeledTCC lysates were subjected tosequential immunoprecipita-tion with class II-specificmAbs followed by SDS-PAGE :L243, lanes 1-5, and SACI-RAM alone, lanes 6 and 7;TU22, lanes 8-11, and SACI-RAM alone, lanes 12 and 13 ;B7/21, lanes 14-17, andSACI-RAM alone, lanes 18and 19 ; TU39, lane 20 . Rela-tive positions of the a and 0chains are indicated, as com-pared with SDS-PAGE molec-ular mass standards . (b) TCCthemselves synthesize TU39'"extra" class II-like molecules.Lysates of [ sBSlmethioninemetabolically labeled TCCwere subjected to sequentialimmunoprecipitations andSDS-PAGE analysis : L243,lanes 1-6, and SACI-RAMalone, lanes 7 and 8; TU22,lanes 9-11, and SACI-RAMalone, lanes 12 and 13; B7/21,lanes 14-16, and SACI-RAMalone, lanes 17 and 18 ; TU39,lane 19 .

TCC themselves, metabolic labeling with [s5S]methionine, followed by similarsequential immunoprecipitation procedures, was undertaken . Results ofone suchexperiment are shown in Fig. 3b . Six precipitations with L243 (lanes 1-6)removed DR molecules (as shown by precipitation with SACI-RAM alone, lanes7 and 8) ; next, repeated precipitations with TU22 removed DQ molecules (lanes9-11, SACI-RAM alone, lanes 12 and 13); after this, precipitations with B7/21(lanes 14-16) removed all DP molecules (SACI-RAM alone, lanes 17 and 18).Finally, lane 19 shows that TU39 still precipitated a large amount of class II-characteristic two-chain heterodimers also from metabolically labeled TCC.Essentially identical results were obtained also with B-LCL (data not shown) .

Suppressive but not Helper T Cell Clones Stimulate DY-speck PLTClones .

Sincecertain TCC as well as B-LCL expressed novel TU39+ non-DR, -DQ, -DPmolecules, they were tested for their expression of DY by using them as stimu-lators for PLT clones . It was found that only TCC that were suppressive for LPresponses in MLC were able to stimulate DY-specific clones . Helper TCC (definedby their ability to help B cells to secrete Ig, data not shown) that failed to suppressLP also failed to stimulate these reagents (Fig . 4) . Moreover, similar levels ofrestimulation responses were observed whether allogeneic or autologous stimu-lating cells were used . Thus, Fig. 4 shows the responses of the four DY-specific

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from


PAWELEC ET AL.

1 2 3 4 5 6 7 8 9 1011 121314 151617 18 19 20

b

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

249

FIGURE 3. (a) mAb TÜ39 detects molecules other than HLA-DR, -DQ, and -DP. 2 X 106 cell equivalents of 12~I_lac_ toperoxidase surface-Iabeled TCC Iysates were subjected to

a _ sequential immunoprecipita/3 _ tion with dass II-specific

mAbs followed by SDS-P AGE: L243, lanes 1-5, and SACIRAM alone, lanes 6 and 7; TÜ22, lanes 8-11, and SACIRAM alone, lanes 12 and 13; B 7 /21, lanes 14-17, and SACI-RAM alone, lanes 18 and 19; TÜ39, lane 20. Relative positions of the a and ß chains are indicated, as com-pared with SDS-PAGE molecular mass standards. (b) TCC themselves synthesize TÜ39T

"extra" dass II-like molecules. Lysates of [S~Slmethionine metabolically labeled TCC were subjected to sequential immunoprecipitations and

a SDS-PAGE analysis: L243, /3 lanes 1-6, and SACI-RAM

alone, lanes 7 and 8; TÜ22, lanes 9-11, and SACI-RAM alone,lanes 12 and 13; B7/21, lanes 14-16, and SACI-RAM alone,lanes 17 and 18; TÜ39, lane 19.

TCC themselves, metabolie labeling with [MS]methionine, followed by similar sequential immunoprecipitation proeedures, was undertaken. Results of one sueh experiment are shown in Fig. 3b. Six precipitations with L243 (Ianes 1-6) removed DR molecules (as shown hy precipitation with SACI-RAM alone, lanes 7 and 8); next, repeated precipitations with TÜ22 removed DQ molecules (Ianes 9-11, SACI-RAM alone, lanes 12 and 13); after this, precipitations with B7/21 (Ianes 14-16) removed all DP molecules (SACI-RAM alone, lanes 17 and 18). FinaIly, lane 19 shows that TÜ39 stiII precipitated a large amount of c1ass 11-eharaeteristie two-ehain heterodimers also from metabolieally labeled TCC. Essentially identieal results were obtained also with B-LCL (data not shown).

Suppressive but not Helper T Cell Clones Stimulate DY-specific PLT Clones. Sinee eertain TCC as weil as B-LCL expressed novel TÜ39+ non-DR, -DQ, -DP moleeules, they were tested for their expression of DY hy using them as stimulators for PL T clones. It was found that only TCC that were suppressive for LP responses in MLC were ahle to stimulate DY-speeifie clones. Helper TCC (defined by their ability to help B cells to seerete Ig, data not shown) that failed to suppress LP also failed to stimulate these reagents (Fig. 4). Moreover, similar levels of restimulation responses were observed whether allogeneic or autologous stimulating eells were used. Thus, Fig. 4 shows the responses of the four DY-speeifie


250


FIGURE 4.

Cloned suppressive but not helper T cells stimulate DY-specific PLT clones. Dataare presented as mean cpm of triplicate cultures ± SEM, after subtraction of background (cpmof clone cultured in medium alone) of 10' responders stimulated with 105 PBMC or 2.5 x 10'TCC or B-LCL for 66 h. The asterisk denotes that the stimulator cells were derived from thesame donor as the responding cells.

PLT clones 102-1, 103-1, 106-8, and 106-13 rechallenged with a range of TCC,as well as with autologous and allogeneic PBMC and autologous B-LCL cells .From these results, the expression of DY on activated TCC would seem tocorrespond to their functional status . Remarkably, this also applied to the DY-specific clones themselves, which were found to be autostimulatory, and couldrespond to one another (Fig . 4) . Such clones could be stimulated equally well by(irradiated) autochthonous cells as by cells from different clones.

Functional Activity of DY Antigens .

Since only those TCC with suppressiveactivity were capable of stimulating DY-specific PLT clones, and since TCCresponsive to DY were, unlike the majority of CD4+ PLT clones, themselvessuppressive (16), it seemed likely that DY could be a major regulator of suppres-sion . This possibility was further investigated in the following experiments . DY-specific clones 102-1, 103-1, 106-8, and 106-13 (Fig . 5, top) suppressed prolif-eration in allogeneic MLC practically as strongly as control suppressive clones29-31 and 38-15 (Fig . 5, bottom). Helper TCC, on the other hand (DR5- andDPw3-specific clones 248-3 and 64-2 in Fig. 5), generally failed to suppressunder these conditions . Moreover, other types of suppressive TCC, as well asthe DY-reactive PLT clones themselves, were able to induce SA in normalPBMC, which was blocked by TU39 but not by anti-DR, -DQ or -DP mAb (26),indicating the involvement of DY in the generation of suppression . When PBMCwere stimulated with DY' suppressive TCC (DY-specific PLT 106-8 or control

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from



S tl mul a l o r Re.ponde,

'02 - ' 103-. '06 ·8 '06 · ' 3

"' .. <:

24 -'9' !? u .. '03-" > '06-8' -;;

'" 29" 5 .. 0. 29-3' Cl

" 38, '5 " .. .,

256 ·7' c 0 U 248·3

Gi 257 -6 Cl 0; 260·' .<::

B ,LCC

PB MC'

P8MC

,03 .04 .'01 counlS pe, mlnUle

FIGURE 4. Cloned suppressive but not helper T cells stimulate DY-specific PL T clones. Data are presented as mean cpm of triplicate cultures ± SEM, after subtraction of background (cpm of clone cultured in medium alone) of 10' responders stimulated with 105 PBMC or 2.5 X 10' TCC or B-LCL for 66 h. The asterisk denotes that the stimulator cells were derived from the same donor as the responding cells.

PL T clones 102-1, 103-1, 106-8, and 106-13 rechallenged with a range of TCC, as weil as with autologous and allogeneic PBMC and autologous B-LCL cells. From these results, the expression of DY on activated TCC would see m to correspond to their functional status, Remarkably, this also applied to the DYspecific clones themselves, wh ich were found to be autostimulatory, and could respond to one another (Fig. 4). Such clones could be stimulated equally weil by (irradiated) autochthonous cells as by cells from different clones.

Functional Activity 01 DY Antigens. Since only those TCC with suppressive activity were capable of stimulating DY-specific PL T clones, and since TCC responsive to DY were, unlike the majority of CD4+ PL T clones, themse1ves suppressive (16), it seemed likely that DY could be a major regulator of suppression. This possibility was further investigated in the following experiments. DYspecific clones 102-1, 103-1, 106-8, and 106-13 (Fig. 5, top) suppressed proliferation in allogeneic MLC practically as strongly as control suppressive clones 29-31 and 38-15 (Fig. 5, bottom). Helper TCC, on the other hand (DR5- and DPw3-specific clones 248-3 and 64-2 in Fig. 5), generally failed to suppress under these conditions. Moreover, other types of suppressive TCC, as weil as the DY-reactive PL T clones themselves, were able to induce SA in normal PBMC, wh ich was blocked by TÜ39 but not by anti-DR, -DQ or -DP mAb (26), indicating the involvement of DY in the generation of suppression. When PBMC were stimulated with DY+ suppressive TCC (DY-specific PL T 106-8 or control


Responding coils per added call

PAWELEC ET AL . 251

FIGURE 5. Suppressive activity in MLC ofPLT clones specific for DY LADs . Cloned cellsas shown were irradiated and titrated at theratios shown directly into allo-MLC in whichthe responding cells were derived from a donormismatched for MHC class I and II specificitieswith the donor of the clones . Results are ex-pressed as percent suppression of the MLCperformed in the absence ofadded suppressivecells .

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from


-25

25

15

100

-25

25

15

'00

25 '25

PA WELEC ET AL. 251

··OY·.!.5PKitic

106-8 102-1 103-' 06-1)

625

29-)1 Nk/SC • 124-7 DRS

/'pJ.'15 SC I .

/ i / .. . ! /

/

FIGURE 5. Suppressive aCUvlty in MLC of PL T clones specific for DY LADs. Cloned cells as shown were irradiated and titrated at the ratios shown directly into allo-MLC in which the responding cells were derived from a donor mismatched for MHC c1ass land 11 specificities with the donor of the clones. Results are expressed as percent suppression of the MLC performed in the absence of added suppressive cells.

/ /

I /

)/ ./" /

If-·--Jf =4 ' .. "=r-" .. , 5 25

I '25

I 625


252

AUTOREACTIVE SUPPRESSOR CELLS-25-,

0-

25-.

a~mOmC

u St)-ma

75-,

Stimulator

toe-e anll-'Dr'124-7 anti-ORS

20-31 NKISC38-15 SC

248-3 anti-ORS

Realionding cells Per added tail

FIGURE 6.

Suppressive activity in MLC of PBMC stimulated with suppressive clones . Allo-geneic PBMC were cocultured for 3 d with the TCC shown (stimulator), followed by a further7-d culture in purified IL-2, before irradiation and titration at the ratios shown into ailo-MLC .Results are expressed as percent suppression of the MLC performed in the absence o£ addedsuppressive cells.

suppressive clones 29-31 and 38-15) for 3 d, followed by short-term culture inIL-2-supplemented medium, they were found to exert potent suppressive activityon LP responses in MLC (Fig . 6) . In contrast, cell lines derived from PBMCstimulated with DR5-specific nonsuppressive helper TCC 124-7 or 248-3 failedto suppress (Fig . 6) . In addition, to establish whether they recognized DYdeterminants, these lines were stimulated with PBMC and B-LCL in the absenceof IIA . Thus, PBMC were stimulated by irradiated suppressive clone 38-15 for3 d, followed by propagation with IL-2 for 10 d, and were then restimulatedwith a range of different cells . They were found to proliferate in the presenceof most stimulator cells, including PBMC from the autologous donor, GP (Fig .7), as well as allogeneic PBMC and B-LCL (KR), and suppressive (29-31, 38-15)but not helper (248-3) TCC. This is a pattern characteristic of "DY"-reactivecells . Moreover, this stimulation was preferentially blocked by mAb TU39, butnot TU22, TU34, or 137/21 (Fig . 7) . In contrast, PBMC cultured under thesame conditions with non-suppressive HLA-D-specific helper TCC respondedweakly to specific allogeneic but not to autologous PBMC and B-LCL. stimulators,

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from



e ; e .. C. ~ ., ~ .. • c .. u 0; ..

-25

o

25

50

75

'00 I 25

Reapondjng ceUa per added c::ell

I 125 F,25

Sttmul.tor

'08-8 .n'j-·OY· 124 -7 onti-OR6

2g-3' HK/SC

38-'6 SC

248-3 anti-DRS

FIGURE 6. Suppressive activity in MLC of PBMC stimulated with suppressive clones. Allogeneie PBMC were cocultured for 3 d with the TCC shown (stimulator). followed by a further 7-d culture in purified IL-2. before irradiation and titration at the ratios shown jnto aIlo-MLC. Results are expressed as percent suppression of the MLC performed in the absence of added suppressive cells.

suppressive clones 29-3 land 38-15) for 3 d, followed by short-term culture in I L-2-supplemented medium, they were found to exert potent suppressive activity on LP responses in MLC (Fig. 6). In contrast, cell lines derived from PBMC stimulated with DR5-specific nonsuppressive helper TCC 124-7 or 248-3 failed to suppress (Fig. 6). In addition, to establish whether they recognized DY determinants, these lines were stimulated with PBMC and B-LCL in the absence of lL-2. Thus, PBMC were stimulated by irradiated suppressive clone 38-15 for :-\ d, followed by propagation with IL-2 for 10 d, and were then restimulated witb a range of different cells. They were found to proliferate in the presence of most stimulator cells, inc\uding PBMC from the autologous donor, GP (Fig. 7), as weil as allogeneic PBMC and B-LCL (KR), and suppressive (29-31,38-15) but not helper (248-3) TCC. This is a pattern characteristic of "DY"-reactive cells. Moreover, this stimulation was preferentially blocked by mAb TÜ39, but not TÜ22, TÜ34, or B7/21 (Fig. 7). In contrast, PBMC cultured under the same conditions with non-suppressive HLA-D-specific helper TCC responded weakly to specific allogeneic but not to autologous PBMC and B-LCL stimulators.


PAWELEC ET AL.

253

FIGURE 7 .

Lymphocytes primed against suppressive clones recognize DY LADs. PBMC werecocultured with irradiated DY* clone 38-15 cells for 3 d, followed by a further 10-d culturein IL-2 . These cells were then restimulated for 66 h by the range of stimulating cells shown,in the presence of control mAb W6/32.HK (data presented in the left hand panel as meancpm t SEM, after subtraction ofthe background) or, in the rest of the Figure, in the presenceof the class II-specific mAb shown (data presented as percentage relative response comparedwith the value with W6/32.HK) .

were unable to respond to autologous T cell clones, and were inhibited by anti-DR mAb (data not shown) .No Further Functions of DY-specific Clones Found.

DY-specific PLT clonesexhibited a peculiarly weak autonomous proliferative capacity, although theygrew well when provided with exogenous IL-2 . Increasing the amount of stim-ulating antigen available by using larger numbers of stimulating cells did notresult in enhanced proliferative responses (Fig . 1) . Consistent with their weakproliferative capacity, and unlike DR-, DQ-, or DP-specific PLT helper cells,DY-reactive cells secreted very modest amounts of IL-2 into the culture mediumafter stimulation (Table III) . However, they were found to produce IL-2 withthe same kinetic as other class II-reactive PLT clones, i .e ., peaking at 36-48 h(data not shown) . Despite their CD4+, Leu-8- phenotype (16), these suppressiveTCC did not function as helpers for B cells (data not shown) . Cytolytic activityof DY-specific TCC on NK-susceptible targets or on B-LCL was not measurablein the "Cr-release assay (16) even in the presence of PHA and/or 20 U/ml ofIL-2 (Table IV).

DiscussionAfter priming between HLA-DR/Dw and DQ-matched homozygous typing

cells, a large proportion of derived clones was found to manifest weak autono-mous proliferative activity against LADs widely distributed in the population(16) . This peculiarly broad reactivity was not due to lack of monoclonality of thetest reagents, because Southern blotting of their DNA with TCR-S and -y chainprobes indicated monoclonal rearrangements of these genes (27) . Such clonesappeared to be similar to most DR-, DQ-, or DP-specific PLT clones in that they

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from


PAWELEC ET AL.

Responder : GP-PBMC and clone 38 - '5. lor 3days lollowed by 10 days culiu re wi lh I L - 2

St imulator, Antibody added 10 resllmulatlon ,

W6/32.HK

GP· P8MC

KR·PBMC 1--"""'---, KR · LCL I--r.-----J

38·'5

29 ·3'

248 ·3

c ,p .m.

Tü22 Tü34

11 perc entage relative response

Tü39

253

FIGURE 7. Lymphocytes primed against suppressive clones recognize DY LADs. PBMC were cocultured with irradiated DY+ clone 38-15 cells for 3 d, followed by a further 10-<1 culture in IL-2. These cells were then restimulated for 66 h by the range of stimulating cells shown, in the presence of control mAb W6/32.HK (data presented in the left hand panel as mean cpm ± SEM, after subtraction of the background) or, in the rest of the Figure, in the presence of the class II-specific mAb shown (data presented as percentage relative response compared with the value with W6/32.HK).

were unable to respond to autologous T cell clones, and were inhibited by antiDR mAb (data not shown).

No Further Functions 01 DY-specific Clones Found. DY-specific PL T clones exhibited a peculiarly weak autonomous proliferative capacity, although they grew weil when provided with exogenous IL-2. Increasing the amount of stimulating antigen available by using larger numbers of stimulating cells did not result in enhanced proliferative responses (Fig. 1). Consistent with their weak proliferative capacity, and unlike DR-, DQ-, or DP-specific PL T helper cells, DY-reactive cells secreted very modest amounts of IL-2 into the culture medium after stimulation (Table 111). However, they were found to produce IL-2 with the same kinetic as other c1ass II-reactive PL T clones, i.e., peaking at 36-48 h (data not shown). Despite their CD4+, Leu-8- phenotype (16), these suppressive TCC did not function as helpers for B cells (data not shown). Cytolytic activity of DY-specific TCC on NK-susceptible targets or on B-LCL was not measurable in the 5lCr-release assay (16) even in the presence of PHA and/or 20 V/mI of IL-2 (Table IV).

Discussion After priming between HLA-DR/Dw and DQ-matched homozygous typing

cells, a large proportion of derived clones was found to manifest weak autonomous proliferative activity against LADs widely distributed in the population (16). This peculiarly broad reactivity was not due to lack of monoclonality of the test reagents, because Southern blotting of their DNA with TCR-ß and -1' chain probes indicated monoclonal rearrangements of these genes (27). Such clones appeared to be similar to most DR-, DQ-, or DP-specific PL T clones in that they


25 4


TABLE III

Secretion of IL-2 by PLT Clones

* Clones were specifically stimulated for 48 h before collection of super-natant for IL-2 assay.Quantification of IL-2 by Probit analysis compared with the IUIS-BRMPIL-2 standard, concentration expressed in units per milliliter .

TABLE IV

Lack of Cytotoxicity ofDY-specific TCC

4 x 10` 5 'Cr-labeled K562 cells were incubated for 4 h in the presence of TCC or PBMC at the E/Tratios shown. Medium was supplemented with 1% PHA and/or 20 U/ml of IL-2 . Results areexpressed as percent specific "Cr release (mean of triplicates, SEM <18%).

were CD4+ and secreted IL-2, but they differed in that they exerted strongsuppressive activity on LP responses (14, 16). Although it is thought unlikelythat the suppressive activity could be due to cytotoxic effects, because these TCCshowed no killing of sensitive target cells even in the presence of PHA and IL-2("liable IV), it cannot be completely excluded that selective cytotoxicity on aminor population such as dendritic cells might contribute to inhibition of LPresponses . However, the ability of these clones to induce suppressive activity innormal PBMC could not be explained in the same way . Others have also describedCD4+ T cells with suppressor/inducer capability, in freshly isolated populations .(:I)4+,2H4+ (CD45R+) T cells were previously identified as suppressors/inducers(28) ; however, CD45R is rapidly lost from TCC during culture, regardless oftheir function (29) . Thus, the relationship between such CD45R+ cells, and thosedescribed here, is not clear . DY-reactive SA-inducers are also negative for

Effector E/TMedium

Target K562

1 % PHA

in the presence of:

20 U/ml IL-2 PHA + IL-2

103-1 50 :1 0.5 -1 .6 1 .5 0.217 :1 -2 .7 2 .1 0.3 -1.9

105-5 100:1 3 .1 -2.7 -2.3 0.633 :1 1 .1 3 .4 1 .8 3.9

106-8 50 :1 -6 .7 -4 .2 0.8 1.517 :1 -2 .2 -1 .3 1 .7 2.2

PBMC 100:1 66.5 95 .6 67 .3 88 .933 :1 48 .0 65 .1 43 .7 52.4

Clone* Specificity Secreted IL-2$

U/ml

102-1 DY 2.1102-3 DY 1 .8103-1 DY 4.3105-5 DY 3.6106-8 DY 5.064-2 DPw3 14.3

248-3 DR5 21 .6250-7 DR5 15.2257-6 DR5 14.9233-7 DQw3 16.4

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from


254

Effector

103-1

105-5

106-8

PBMC


TABLE 111 Secretion of IL-2 by PLT Clones

Clone* Specificity Secreted IL-2+

V/mi

102-1 DY 2.1 102-3 DY 1.8 103-1 DY 4.3 105-5 DY 3.6 106-8 DY 5.0 64-2 DPw3 14.3

248-3 DR5 21.6 250-7 DR5 15.2 257-6 DR5 14.9 233-7 DQw3 16.4

* Clones were specifically stimulated for 48 h before collection of supernatant for IL-2 assay.

+ Quantification of IL-2 by Probit analysis compared with the IUIS-BRMP IL-2 standard, concentration expressed in units per milliliter.

TABLE IV Lack of Cytotoxicity of DY-specific TCC

E/T Target K562 in the presence of:

Medium 1% PHA 20 U/ml IL-2 PHA + IL-2

50:1 0.5 -1.6 1.5 0.2 17:1 -2.7 2.1 0.3 -1.9

100:1 3.1 -2.7 -2.3 0.6 33:1 l.l 3.4 1.8 3.9 50:1 -6.7 -4.2 0.8 1.5 17:1 -2.2 -1.3 1.7 2.2

100:1 66.5 95.6 67.3 88.9 33:1 48.0 65.1 43.7 52.4

4 X 10" "Cr-Iabeled K562 cells were incubated for 4 h in the presence of TCC or PBMC at the E/T ratios shown. Medium was supplemented with 1% PHA and/or 20 U/ml of IL-2. Results are expressed as percent specific "Cr release (mean of triplicates, SEM <18%).

were CD4+ and secreted IL-2, but they differed in that they exerted strong suppressive activity on LP responses (14, 16). Although it is thought unlikely that the suppressive activity could be due to cytotoxic effects, because these TCC showed no killing of sensitive target cells even in the presence of PHA and IL-2 (Table IV), it cannot be completely excluded that selective cytotoxicity on a minor population such as dendritic cells might contribute to inhibition of LP responses. However, the ability of these clones to induce suppressive activity in normal PBMC could not be explained in the same way. Others have also described CD4+ T cells with suppressor/inducer capability, in freshly isolated populations. CD4+,2H4+ (CD45R+) T cells were previously identified as suppressors/inducers (28); however, CD45R is rapidly lost from TCC during culture, regardless of their function (29). Thus, the relationship between such CD45R+ cells, and those described here, is not c1ear. DY-reactive SA-inducers are also negative for


PAWELEC ET AL .

255

another marker, Leu-8 (16), which is said to identify CD4+ suppressor/inducercells (30) .

DY-specific TCC displayed an unusual type of autoreactivity, being restim-ulated not only by autologous PBMC or B-LCL, but also by HLA-mismatchedallogeneic cells . Using comparable priming conditions, similar reactivities inuncloned PLT cells have been noted previously (31), and, more recently, Martinet al . (32) have described possibly similar findings at the clonal level (32) . TheLADs involved in stimulating autoreactive lymphocyte populations have beenrepeatedly shown to be, or to be associated with, class 11 molecules. AlthoughFACS analysis showed that helper TCC expressed quantitatively similar amountsof DR, DQ, and DP antigens compared with the suppressive TCC (33), theydiffered in their ability to stimulate "DY"-specific clones (Fig . 4) . This associationof stimulatory activity with function of the clones surprisingly applied also to theautoreactive clones themselves, since these were mutually cross-stimulatory aswell as autostimulatory. This implies that such cells are able to present DY LADsto each other. It is somewhat puzzling, therefore, that stimulation by autologousirradiated PBMC or autochthonous clones led to greater proliferation thanobserved in medium alone, since DY-reactive cells by themselves should consti-tute an autostimulatory circuit. Indeed, [ sH]TdR incorporation by the clones inmedium alone was higher than seen for DR-, DQ- or DP-specific clones, but wasstill higher in the presence of irradiated cells, even when these came from thesame clone . In the case of autostimulatory PBMC or B-LCL, a more efficientantigen presentation, for example, by dendritic cells in the PBMC or by theLCL, than by the TCC themselves might help to explain this apparent anomaly.For self-stimulation by autochthonous clones, it is conceivable that the irradiatedstimulator TCC themselves would also respond to unirradiated cells in the culturewells by secreting IL-2 . This would amplify the response of the unirradiatedcells, since one of the characteristics of the DY response is the very limitedamounts of autocrine IL-2 production by stimulated cells (Table 111) .The broad class II-reactive mAb TC39, unlike exclusively or preferentially

DR-, DQ-, or DP-specific mAbs, blocked both LP stimulatory capacity and SA-inducing activity mediated by CD4+ suppressive, but not by CD4+ helper, TCC(26) . The same pattern of blocking applied to stimulation of the autoreactiveclones described here (Fig . 2) . Various mAbs against HLA-DR, -DQ, and -DPfailed to block stimulation of DY-specific autoreactive clones, whereas they couldblock stimulation of PLT clones by appropriate alloantigens . This is consistentwith, but does not prove, the presence of DY determinants on molecules otherthan DR, DQ, or DP. That stimulation of autoreactive clones was blocked bytwo of four broadly reactive mAbs, SG520 and TU39 (Fig . 2), however, impliesthat these clones do recognize structures bound by certain class II-specific mAbs .Since mixtures of anti-DR, -DQ, and -DP mAbs also failed to block stimulation(34, and data not shown), it is improbable that the clones were capable of reactingto shared determinants on all these molecules, and therefore were not blockedby mAbs against any single one. Blockade of DY-specific TCC stimulation byTU39 was abrogated by exogenous IL-2 (Table 11), suggesting that the mAbwas not mediating a nonspecific inhibitory effect . Despite this, [s H]TdR incor-poration was reduced by TU39 to below the level measured in medium alone.

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from


PA WELEe ET AL. 255

another marker, Leu-8 (16), which is said to identify CD4 + suppressor /inducer cells (30).

DY-specific TCC displayed an unusual type of autoreactivity, being restimulated not only by autologous PBMC or B-LCL, but also by HLA-mismatched allogeneic cells. Using comparable priming conditions, similar reactivities in uncloned PLT cells have been noted previously (31), and, more recently, Martin et al. (32) have described possibly similar findings at the clonal level (32). The LADs involved in stimulating autoreactive lymphocyte populations have been repeatedly shown to be, or to be associated with, class II molecules. Although F ACS analysis showed that helper TCC expressed quantitatively similar amounts of DR, DQ, and DP antigens compared with the suppressive TCC (33), they differed in their ability to stimulate "DY"-specific clones (Fig. 4). This association of stimulatory activity with function of the clones surprisingly applied also to the autoreactive clones themselves, since these were mutually cross-stimulatory as weil as autostimulatory. This implies that such cells are able to present DY LADs to each other. It is somewhat puzzling, therefore, that stimulation by autologous irradiated PBMC or autochthonous clones led to greater proliferation than observed in medium alone, since DY-reactive cells by themselves should constitute an autostimulatory circuit. Indeed, [3H]TdR incorporation by the clones in medium alone was higher than seen for DR-, DQ- or DP-specific clones, but was still higher in the presence of irradiated cells, even when these came from the same clone. In the case of autostimulatory PBMC or B-LCL, a more efficient antigen presentation, for example, by dendritic cells in the PBMC or by the LCL, than by the TCC themselves might help to explain this apparent anomaly. For self-stimulation by autochthonous clones, it is conceivable that the irradiated stimulator TCC themselves would also respond to unirradiated cells in the culture wells by secreting IL-2. This would amplify the response of the unirradiated cells, since one of the characteristics of the DY response is the very limited amounts of autocrine IL-2 production by stimulated cells (Table III).

The broad class II-reactive mAb TÜ39, unlike exclusively or preferentially DR-, DQ-, or DP-specific mAbs, blocked both LP stimulatory capacity and SAinducing activity mediated by CD4+ suppressive, but not by CD4+ helper, TCC (26). The same pattern of blocking applied to stimulation of the autoreactive clones described here (Fig. 2). Various mAbs against HLA-DR, -DQ, and -DP failed to block stimulation of DY-specific autoreactive clones, whereas they could block stimulation of PL T clones by appropriate alloantigens. This is consistent with, but does not prove, the presence of DY determinants on molecules other than DR, DQ, or DP. That stimulation of autoreactive clones was blocked by two of four broadly reactive mAbs, SG520 and TÜ39 (Fig. 2), however, implies that these clones do recognize structures bound by certain class II-specific mAbs. Since mixtures of anti-DR, -DQ, and -DP mAbs also failed to block stimulation (34, and da ta not shown), it is improbable that the clones were capable of reacting to shared determinants on all these molecules, and therefore were not blocked by mAbs against any single one. Blockade of DY-specific TCC stimulation by TÜ39 was abrogated by exogenous IL-2 (Table II), suggesting that the mAb was not mediating a nonspecific inhibitory effect. Despite this, [3H]TdR incorporation was reduced by TÜ39 to below the level measured in medium alone.


256


The most likely explanation for this finding is that an autostimulatory circuitwithin the clone was broken by the binding of the mAbto DY antigens, resultingin blockade of DY-directed self-stimulation . The existence of an autostimulatorycircuit within the clones would also help to explain why stimulation of DY-specific cells after pretreating only the stimulators with mAb TU39 was notreduced to background (Table II).There is evidence that TU39 binds determinants other than those carried by

DR, DQ, or DP molecules . First, studies of class II expression on leukemias andperipheral blood cells early after bone marrow transplantation demonstratedthat certain DR-,. DQ- , and DP' cells still reacted with mAb TU39, or that theproportion of T39-reactive cells was much higher than the sum of DR', DQ',and DP's cells (13, 14). Second, preliminary biochemical analyses providedevidence consistent with TU39 reactivity on non-DR, -DQ, or -DP class II-likemolecules of very similar molecular mass to DR molecules (Fig . 3) . Thus,sequential immunoprecipitations to remove DR (and DRw, reference 35), DQ,and DP molecules from lysates of surface-labeled TCC failed to deplete all ofthe class II-like molecules with which mAb TU39 could react. Furthermore,metabolic labeling showed that the putative novel class II-like TU3' non-DR,-DQ, -DP molecules actually were synthesized by the TCC themselves (Fig . 3b).In both surface-labeled and internally labeled cells, a relatively large amount ofT039-reactive material remained after depletion of the DR, DQ, and DPmolecules, emphasizing the potential novel nature of these moieties . Theirsimilarity to DR molecules may help to explain why they were not previouslydetected in two-dimensional gel analysis of whole lysates (24). Thepresent resultsare consistent with the proposal that DY determinants, defined in functionalassays, are carried by, or associated with, the, putative novel TU39+ non-DR,-DQ, -DP molecules demonstrated here . TheTU39+ novel class II-like moleculeswere, like the functionally defined DY determinants, also expressed on B-LCL(Table 1) . Metabolic labeling has confirmed that they were also synthesized byB-LCL (Fernandez, N ., unpublished results). It will be necessary to confirm thenovelty of DY by means of peptide mapping and amino acid sequencing, and toclarify the relationship between DY and established class II genes and productsat the structural level .DO# remains a possible candidate class II gene for which a protein product

has not yet been identified . Since it is differently regulated than other class IIgenes, it has been suggested that its function could also be different (9). It appearsto be a gene of a relatively low degree of polymorphism, which is consistent withthe findings pertaining to the DY LADs described here . However, DO(3 mRNAhas thus far not been found in T cells (9) . DZa (8) may also be a potentialcandidate for one of the DY chains, and this is known to be expressed in T cells(8). Another candidate may be the so-called "fourth Ia subset" described byCarra and Accolla (7), the nature of which is unknown, but which has also notyet been demonstrated in T cells. The retention of some LP stimulatory abilityand antigen presentation capacity by class I and class If loss-deletion mutant B-LCL (36) also suggests the existence of functionally relevant non-DR, -DQ, or-DP molecules, which may therefore have some relationship with DY .DY molecules clearly may not be the products of new class 11 genes, but rather

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from



The most likely explanation for this finding is that an autostimulatory circuit within the clone was broken by the binding of the mAb to DY antigens, resulting in blockade of DY -directed self-stimulation. The existence of an autostimulatory circuit within the clones would also help to explain why stimulation of DYspecific cells after pretreating only the stimulators with mAb TÜ39 was not reduced to background (Table II).

There is evidence that TÜ39 binds determinants other than those carried by DR, DQ, or DP molecules. First, studies of dass 11 expression on leukemias and peripheral blood cells early after bone marrow transplantation demonstrated that certain DR-, DQ-, and DP- cells still reacted with mAb TÜ39, or that the proportion of TÜ39-reactive cells was much higher than the sum of DR+, DQ+, and DP+ cells (13, 14). Second, preliminary biochemical analyses provided evidence eonsistent with TÜ39 reactivity on non-DR, -DQ, or -DP dass II-like molecules of very similar molecular mass to DR molecules (Fig. 3). Thus, sequential immunoprecipitations to remove DR (and DRw, reference 35), DQ, and DP molecules from lysates of surface-labeled TCC failed to deplete all of the dass II-like molecules with which mAb TÜ39 could react. Furthermore, metabolie labeling showed that the putative novel dass II-Iike TÜ39+ non-DR, -DQ, -DP molecules actually were synthesized by the TCC themselves (Fig. 3b). In both surface-labeled and internally labeled cells, a relatively large amount of TÜ39-reactive material remained after depletion of the DR, DQ, and DP molecules, emphasizing the potential novel nature of these moieties. Their similarity to DR moleeules may help to explain why they were not previously detected in two-dimensional gel analysis of whole lysates (20). The present results are consistent with the proposal that DY determinants, defined in funetional assays, are earried by, or associated with, the putative novel TÜ39+ non-DR, -DQ, -DP molecules demonstrated here. The TÜ39+ novel dass II-Iike molecules were, like the funetionally defined DY determinants, also expressed on B-LCL (Table I). Metabolie labeling has confirmed that they were also synthesized by B-LCL (Fernandez, N., unpublished results). It will be neeessary to confirm the novelty of DY by means of peptide mapping and amino acid sequencing, and to clarify the relationship between DY and established dass II genes and products at the struetural level.

DOß remains a possible eandidate class II gene for wh ich a protein product has not yet been identified. Sinee it is differently regulated than other dass 11 genes, it has been suggested that its funetion could also be different (9). It appears to be a gene of a relatively low degree of polymorphism, which is eonsistent with the findings pertaining to the DY LADs deseribed here. However, DOß mRNA has thus far not been found in T eells (9). DZa (8) mayaiso be a potential eandidate for one of the DY chains, and this is known to be expressed in T cells (8). Another candidate may be the so-ca lied "fourth Ia subset" described by Carra and Aceolla (7), the nature of which is unknown, but which has also not yet been demonstrated in T cells. The retention of some LP stimulatory ability and antigen presentation capacity by dass I and dass II loss-deletion mutant BLCL (36) also suggests the existence of functionally relevant non-DR, -DQ, or -DP molecules, which may therefore have some relationship with DY.

DY molecules dearly may not be the products of new dass 11 genes, but rather


PAWELEC ET AL .

257

novel structures formed between a and ft chains bf different known loci togenerate mixed isotypes. The existence of such mixed isotype class II moleculeshas been demonstrated in mouse L cells transfected with murine class II genes(37) . The relatively large amount of non-DR, -DQ, -DP TOW molecules onboth B and T cells, and the failure thus far to isolate functional genes for class11 molecules other than DR, DQ (and DX), DP, DO, and DZ, wouldbe consistentwith this possibility . Moreover, it has recently been reported that a variety ofDR-, DQ-, DP-, and DO-containing mixed isotypes can be expressed aftertransfection of genes encoding the appropriate chains into L cells (11) . Interest-ingly, these products were not detectable by DR- DQ-, or DP-specific mAbs, butwere, like DY, bound by broadly reactive class II-specific mAbs . Furthermore,the presence of such mixed isotypes has been implied, at least for DRa/DQ#chains, in a more physiological system on B-LCL (12) . Thus, a reasonable workinghypothesis is that DY is a mixed isotype, or a group of mixed isotypes, expressedon B-LCL and on activated T cells other than helper cells .A central role for DY LADs appears to be in the suppression of proliferative

responses, since all PLT clones specific for DY were highly suppressive, andbecause only suppressive, but not otherwise equally class II-positive helper TCCexpressed functionally defined DY on their surfaces. T039-inhibitable stimula-tion of LP responses in PBMC by suppressive clones is associated with inductionof suppression (38), and, as shown here (Fig . 7), such stimulated PBMC can alsorespond to DY with limited autonomous proliferation . Autoreactive DY-respon-sive T cells were unable to help B cell Ig production (data not shown), and werenot cytotoxic (16, Table IV). Thus far, the only effector function attributable tosuch cells is non-MHC-restricted potent suppression of LP responses stimulatedin MLC (16, Fig. 5) and by mitogens and antigens (PHA and PPD, data notshown), and, like other DY' clones, the induction of SA in PBMC (Fig . 6) . Sinceautoreactive clones both express and respond to DY, this would imply that theDY-associated suppressive circuit could be self maintaining by recruitment andby clonal expansion . However, the system presumably is kept in a steady state bythe relatively meager quantities of IL-2 produced by the autoreactive SC, so thatsuppression would not always predominate, although a background level ofsuppressive activity would be present .A degree of antigen specificity regulating the activity of the hypothesized DY-

suppressor circuit would be mediated by the local availability of higher concen-trations of IL-2 secreted by antigen-stimulated helper cells. This would be asimple strategy for upregulation of DY-stimulated suppression, which would thenfeed back to decrease helper function, thereby decreasing IL-2 availability,leading to its own downregulation . Further mechanisms for control of this typeof suppression would undoubtedly also have to exist, possibly involving contra-suppression (39) . Some preliminary evidence is consistent with this possibility(40) .

SummaryA set of T cell clones (TCC) isolated from HLA-DR-, Dw-, DQ-matched

allogeneic MLCs was found to proliferate autonomously when stimulated withcells carrying a wide range of class I or 11 specificities . This apparently unre-

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from


PA WELEC ET AL. 257

novel structures formed between a and ß chains bf different known loci to generate mixed isotypes. The existence of such mixed isotype class II molecules has been demonstrated in mouse L cells transfected with murine class 11 genes (37). The relatively large amount of non-DR, -DQ, -DP TÜ39+ molecules on both Band T cells, and the failure thus far to isolate functional genes for class II molecules other than DR, DQ (and DX), DP, DO, and DZ, would be consistent with this possibility. Moreover, it has recently been reported that a variety of DR-, DQ-, DP-, and DO-containing mixed isotypes can be expressed after transfection of genes encoding the appropriate chains into L cells (11). Interestingly, these products were not detectable by DR- DQ-, or DP-specific mAbs, but were, Iike DY, bound by broadly reactive class II-specific mAbs. Furthermore, the presence of such mixed isotypes has been implied, at least for DRa/DQß chains. in a more physiological system on B-LCL (12). Thus, a reasonable working hypothesis is that DY is a mixed isotype, or a group of mixed isotypes, expressed on B-LCL and on activated T cells other than helper cells.

A central role for DY LADs appears to be in the suppression of proliferative responses, since all PL T clones specific for DY were highly suppressive, and because only suppressive, but not otherwise equally class li-positive helper TCC expressed functionally defined DY on their surfaces. TÜ39-inhibitable stimulation of LP responses in PBMC by suppressive clones is associated with induction of suppression (38), and, as shown here (Fig. 7), such stimulated PBMC can also respond to DY with limited autonomous proliferation. Autoreactive DY-responsive T cells were unable to help B cell Ig production (data not shown), and were not cytotoxic (16, Table IV). Thus far, the only effector function attributable to such cells is non-MHC-restricted potent suppression of LP responses stimulated in MLC (16, Fig. 5) and by mitogens and antigens (PHA and PPD. da ta not shown), and, Iike other DY+ clones. the induction of SA in PBMC (Fig. 6). Since autoreactive clones both express and respond to DY, this would imply that the DY-associated suppressive circuit could be self maintaining by recruitment and by c10nal expansion. However, the system presumably is kept in a steady state by the relatively meager quantities of IL-2 produced by the autoreactive SC, so that suppression would not always predominate, although a background level of suppressive activity would be present.

A degree of antigen specificity regulating the activity of the hypothesized DYsuppressor circuit would be mediated by the local availability of higher concentrations of IL-2 secreted by antigen-stimulated helper cells. This would be a simple strategy for upregulation ofDY-stimulated suppression, which would then feed back to decrease helper function, thereby decreasing IL-2 availability, leading to its own downregulation. Further mechanisms for control of this type of suppression would undoubtedly also have to exist, possibly involving contrasuppression (39). Some preliminary evidence is consistent with this possibility (40).

Summary A set of T cell clones (TCC) isolated from HLA-DR-, Dw-, DQ-matched

allogeneic MLCs was found to proliferate autonomously when stimulated with cells carrying a wide range of c1ass I or II specificities. This apparently unre-


25 8


stricted proliferation was relatively weak, andonly low levels of IL-2 were presentin the supernatants of stimulated cells. Autologous as well as allogeneic PBMCand B lymphoblastoid cell lines (B-LCL) were capable of stimulating such clones,which were also restimulated by suppressive, but not by helper, TCC. Moreover,such clones displayed the unusual property of autostimulation . mAb inhibitionexperiments suggested that class II- or class II-restricted antigens were involvedin stimulation . Thus, certain "broad" mAbs (TU39, SG520) reacting with mul-tiple locus products inhibited activation of these reagents, but none of thosereacting more specifically with DR (TU34, TU37, L243, Q2/70, SG157), DQ(TID22, SPV-L3, LeuIO), or DP (137/21), or mixtures of these mAbs, were ableto do so . Evidence from sequential immunoprecipitation experiments suggestedthat mAb TU39 bound class II-like molecules other than DR, DQ, and DP onTCC and B-LCL, and it is therefore proposed that such putative novel class A-like molecules may carry the stimulating determinants for these autoreactiveclones . DY-reactive clones lacked helper activity for B cells but mediated potentsuppressive activity on T cell proliferative responses that was not restricted bythe HLA type of the responding cells. Suppressive activity was induced in normalPBMC by such clones, as well as by independent suppressive clones, which wasalso inhibited only by mAb TU39 . These findings lead to the proposal that DY-reactive autostimulatory cells may constitute a self-maintaining suppressive cir-cuit, the level of activity of which would be regulated primarily by the availabilityof IL-2 in the microenvironment .

We would like to thank A. Rehbein, 1. Balko, and C. Schmidt for their expert assistance,and C. Miller for artwork.

Receivedfor publication 13 March 1987 and in revisedform 6 August 1987 .

References

1 . Lamb, J . R ., J . N . Woody, R . J . Hartzman, and D . D . Eckels . 1982 . I n vitro influenzavirus-specific antibody production in man : antigen-specific and HLA-restricted in-duction of helper activity mediated by cloned human T lymphocytes . J. Immunol.129:1465 .

2 . Qvigstad, E ., T . Moen, and E . Thorsby . 1984 . T cell clones with similar antigenspecificity may be restricted by DR, MT (DC) or SB class II HLA molecules .Immunogenetics . 19:455 .

3 . Pawelec, G., E . M . Schneider, C . Miifer and P . Wernet . 1985 . HLA-DR, MB- andnovel DC-related determinants restrict purified protein derivative of tuberculin(PPD)-stimulated human T cell proliferation . Eur. J. Immunol. 15 :12 .

4 . Spits, H ., J . Borst, M. Giphart, J . Coligan, C . Terhorst, and J . E . de Vries . 1984 .HLA-DC can serve as restriction elements for human cytotoxic T lymphocytes . Eur.J. Immunol. 14:299 .

5 . Ohta, N ., A . Anichini, N . L . Reinsmoen, G . Strassmann, P . Wernet, and F . H . Bach .1985 . Analysi s of human class II antigens by cloned cytolytic T cell reagents : a studyusing loss mutant lymphoblastoid cell lines and monoclonal antibodies detectingHLA-DP product(s) . Hum. Immunol. 13:21 .

6 . Eckels, D. D., A . Zeevi, P . G . Beatty, N . Flomenberg, S . Goyert, R . W. Knowles, E .Mickelson, G . Nepom, P . Parham, G . Pawelec, and N . L . Reinsmoen . 1986 . ASH I

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from



stricted proliferation was relatively weak, and only low levels of IL-2 were present in the supernatants of stimulated cells. Autologous as weil as aIIogeneic PBMC and B Iymphoblastoid celilines (B-LCL) were capable of stimulating such clones, which were also restimulated by suppressive, but not by helper, TCC. Moreover, such clones displayed the unusual property of autostimulation. mAb inhibition experiments suggested that c1ass 11- or c1ass II-restricted antigens were involved in stimulation. Thus, certain "broad" mAbs (TÜ39, SG520) reacting with multiple locus products inhibited activation of these reagents, but none of those reacting more specificaIIy with DR (TÜ34, TÜ37, L243, Q2/70, SG157), DQ (TÜ22, SPV-L3, LeulO), or DP (B7/21), or mixtures ofthese mAbs, were able to do so. Evidence from sequential immunoprecipitation experiments suggested that mAb TÜ39 bound c1ass II-like molecules other than DR, DQ, and DP on TCC and B-LCL, and it is therefore proposed that such putative novel c1ass 11-like molecules may carry the stimulating determinants for these autoreactive clones. DY-reactive clones lacked helper activity for B ceIIs but media ted potent suppressive activity on T cell proliferative responses that was not restricted by the HLA type of the responding ceIIs. Suppressive activity was induced in normal PBMC by such clones, as weII as by independent suppressive clones, wh ich was also inhibited only by mAb TÜ39. These findings lead to the proposal that DYreactive autostimulatory ceIIs may constitute a self-maintaining suppressive circuit, the level of activity of which would be regulated primarily by the availability of IL-2 in the microenvironment.

We would like to thank A. Rehbein, I. Balko, and C. Schmidt for their expert assistance, and C. Miller for artwork.

Received for publication 13 March 1987 and in revised form 6 August 1987.

References l. Lamb,j. R., j. N. Woody, R. j. Hartzman, and D. D. EckeIs. 1982. In vitro influenza

virus-specific antibody production in man: antigen-specific and HLA-restricted induction of helper activity mediated by cloned human T lymphocytes. J. Immunol. 129:1465.

2. Qvigstad, E., T. Moen, and E. Thorsby. 1984. T cell clones with similar antigen specificity may be restricted by DR, MT (DC) or SB class II HLA molecules. 1mmunogenetics. 19:455.

3. Pawelec, G., E. M. Schneider, C. Müller and P. Wernet. 1985. HLA-DR, MB- and novel DC-related determinants restrict purified protein derivative of tuberculin (PPD)-stimulated human T cell proliferation. Eur. J. 1mmunol. 15: 12.

4. Spits, H., j. Borst, M. Giphart, j. Coligan, C. Terhorst, and j. E. de Vries. 1984. HLA-DC can serve as restriction elements for human cytotoxic T lymphocytes. Eur. J. 1mmunol. 14:299.

5. Ohta, N., A. Anichini, N. L. Reinsmoen, G. Strassmann, P. Wernet, and F. H. Bach. 1985. Analysis of human class II antigens by cloned cytolytic T cell reagents: a study using loss mutant lymphoblastoid cell lines and monoclonal antibodies detecting HLA-DP product(s). Hum. Immunol. 13:21.

6. EckeIs, D. D., A. Zeevi, P. G. Beatty, N. Flomenberg, S. Goyert, R. W. Knowles, E. Mickelson, G. Nepom, P. Parham, G. Pawelec, and N. L. Reinsmoen. 1986. ASHI


PAWELEC ET AL .

259

workshop summary report of the Science and Education subcommittee : structuraland functional relationship of human class Il MHC molecules . Hum. Immunol . 15 :68 .

7 . Carra, G., and R . S. Accolla . 1987 . Structura l analysis of human la antigens revealsthe existence of a fourth molecular subset distinct from DP, DQ and DR molecules .J . Exp . Med. 165:47 .

8 . Trowsciale, J ., and A. Kelly . 1985 . The human H LA class II alpha chain gene DZais distinct from genes in the DP, DQ and DR subregions . EMBO (Eur. Mol . Biol.Organ.)J. 4:2231 .

9 . Tonelle, C ., R . DeMars, and E. Long . 1985 . DO# : a new a chain gene in HLA-Dwith distinct regulation of expression . EMBO (Eur. Mol . Biol . Organ.) J . 4:2839 .

10 . Charron, D . J ., V . Lotteau, and P . Turmal. 1984 . Hybrid HLA-DC antigens providemolecular evidence for gene trans-complementation . Nature (Lond .) . 312:157 .

11 . Long, E . O., S . Rosen-Bronson, S . Jacobson, and E . P . Sekaly . 1987 . Molecularanalysis of the diversity and function of MHC class II antigens (Abstract) . Immuno-biology. 3(Suppl .):204 .

12 . Lotteau, V ., L . Teyton, P . Hermans, and D. Charron . 1987 . HLA-DRa-DQ/3 interi-sotypic heterodimers : influence of allelic polymorphism of DW-chain on the a-#-cross-pairing (Abstract) . Immunobiology. 3(Suppl .):205 .

13 . Wernet, P ., C . P . Muller, and P . Ostendorf. 1984 . Reactivity of lymphocytic andmyelocytic leukaemia blasts with monoclonal antibodies specific for the differenthuman class 11 molecules. Disease Markers . 2 :449 .

14 . Wernet, P., G . Pawelec, and E . M . Schneider . 1986 . Cellula r detection of humanclass II antigens : delineation of a novel HLA-DP-like suppressor restriction system,DY, the sequential expression of class II antigens and a pronounced functionalflexibility of class II alloproliferative T cell clones . In Human Class II Antigens . AComprehensive Review of Structure and Function . B . Solheim, E . Muller, and S .Ferrone, editors. Springer-Verlag, Heidelberg . 281-298.

15 . Sanchez-Perez, M., and S . Shaw . 1986 . HLA-DP: current status . In Human Class IIAntigens . A Comprehensive Review ofStructure and Function . B . Solheim, E. Muller,and S . Ferrone, editors . Springer Verlag, Heidelberg . 83-108 .

16 . Pawelec, G., P . Wernet, R . Rosenlund, M. Blaurock, and E . M . Schneider . 1984 .Strong lymphoproliferative suppressive function of PLT clones specific for SB-likeantigens . Hum . Immunol . 9:145 .

17 . Pawelec, G . P ., S . Shaw, A . Ziegler, C . Muller, and P . Wernet . 1982 . Differentialinhibition of HLA-D or SB-directed secondary lymphoproliferative responses withmonoclonal antibodies detecting human la-like determinants . J. Immunol. 129:1070 .

18 . Steel, M ., editor . 1984 . Human MHC class II antigens : genetics, structure andfunction . Proceedings of the 1st International Workshop on monoclonal antibodiesto human MHC class II antigens, Edinburgh . Disease Markers . Special issue . 2 :1-369 .

19 . Ziegler, A ., J . Heinig, C . Muller, H . G6tz, F . P . Thinnes, B . Uchanska-Ziegler, andP . Wernet . Analysis by sequential immunoprecipitations of the specificities of themonoclonal antibodies TID22, 34, 35, 36, 37, 39, 43, 58, and YD1/63 .HLK directedagainst human HLA class 11 antigens . Immunobiology. 171 :77 .

20 . Crumpton, M. J ., J . G . Bodmer, W. F . Bodmer,J . M . Heyes,J . Lindsay, and C. Rudd .1984 . Biochemistry of class II antigens : Workshop report . In HistocompatibilityTesting 1984 . E . D . Albert, M . P . Baur, and W. R . Mayr, editors . Springer-Verlag,Berlin . 29-37 .

21 .

Ey, P. L ., S . J . Prowse, and C . R . Jenskin . 1978 . Isolation of pure IgGI, IgG2a andIgG2b immunoglobulins from mouse serum protein using protein A-Sepharose .Immunochemistry . 15 :429 .

22 . Hayman, M . S ., and M. J . Crumpton . 1972 . Isolatio n of glycoproteins from pig

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from


PA WELEC ET AL. 259

workshop summary report of the Science and Education subcommittee: structural and functional relationship of human class II MHC molecules. Hum. Immunol. 15:68.

7. Carra, G., and R. S. Accolla. 1987. Structural analysis of human Ia antigens reveals the existence of a fourth molecular subset distinct from DP, DQ and DR molecules. J. Exp. Med. 165:47.

8. Trowsdale, j., and A. Kelly. 1985. The human HLA class 11 alpha chain gene DZa is distinct from genes in the DP, DQ and DR subregions. EMBO (Eur. Mol. Biol. Organ.)]. 4:2231.

9. Tonelle, c., R. DeMars, and E. Long. 1985. DOß: a new ß chain gene in HLA-D with distinct regulation of expression. EMBO (Eur. Mol. Biol. Organ.)]. 4:2839.

10. Charron, D. j., V. Lotteau, and P. Turmal. 1984. Hybrid HLA-DC antigens provide molecular evidence for gene trans-complementation. Nature (Lond.). 312:157.

11. Long, E. 0., S. Rosen-Bronson, S. Jacobson, and E. P. Sekaly. 1987. Molecular analysis of the diversity and function of MHC class 11 antigens (Abstract). Immunobiology. 3(Suppl.):204.

12. Lotteau, V., L. Teyton, P. Hermans,and D. Charron. 1987. HLA-DRa-DQßinterisotypic heterodimers: influence of allelic polymorphism of DQß-chain on the a-ßcross-pairing (Abstract). Immunobiology. 3(Suppl.):205.

13. Wernet, P., C. P. Muller, and P. Ostendorf. 1984. Reactivity of Iymphocytic and myelocytic leukaemia blasts with monoclonal antibodies specific for the different human class II molecules. Disease Markers. 2:449.

14. Wernet, P., G. Pawelec, and E. M. Schneider. 1986. Cellular detection of human class II antigens: delineation of a novel HLA-DP-like suppressor restriction system, DY, the sequential expression of class II antigens and a pronounced functional flexibility of class II alloproliferative T cell clones. In Human Class 11 Antigens. A Comprehensive Review of Structure and Function. B. Solheim, E. Möller, and S. Ferrone, editors. Springer-Verlag, Heidelberg. 281-298.

15. Sanchez-Perez, M., and S. Shaw. 1986. HLA-DP: current status. In Human Class 11 Antigens. A Comprehensive Review ofStructure and Function. B. Solheim, E. Möller, and S. Ferrone, editors. Springer Verlag, Heidelberg. 83-108.

16. Pawelec, G., P. Wernet, R. Rosenlund, M. Blaurock, and E. M. Schneider. 1984. Strong lymphoproliferative suppressive function of PL T clones specific for SB-like antigens. Hum. Immunol. 9: 145.

17. Pawelec, G. P., S. Shaw, A. Ziegler, C. Müller, and P. Wernet. 1982. Differential inhibition of HLA-D or SB-directed secondary Iymphoproliferative responses with monoclonal antibodies detecting human la-like determinants. J. Immunol. 129: 1 070.

18. Steel, M., editor. 1984. Human MHC class II antigens: genetics, structure and function. Proceedings of the 1st International Workshop on monoclonal antibodies to human MHC class II antigens, Edinburgh. Disease Markers. Special issue. 2: 1-369.

19. Ziegler, A., j. Heinig, C. Müller, H. Götz, F. P. Thinnes, B. Uchanska-Ziegler, and P. Wernet. Analysis by sequential immunoprecipitations of the specificities of the monoclonal antibodies TÜ22, 34, 35, 36, 37, 39,43,58, and YDlj63.HLK directed against human H LA class II antigens. Immunobiology. 171 :77.

20. Crumpton, M.j.,j. G. Bodmer, W. F. Bodmer,j. M. Heyes,j. Lindsay, and C. Rudd. 1984. Biochemistry of class II antigens: Workshop re port. In Histocompatibility Testing 1984. E. D. Albert, M. P. Baur, and W. R. Mayr, editors. Springer-Verlag, Berlin.29-37.

21. Ey, P. L., S. j. Prowse, and C. R. Jenskin. 1978. Isolation of pure IgG I, IgG2a and IgG2b illllllunogiobulins from mouse serum protein using protein A-Sepharose. Immunochemistry. 15:429.

22. HaYlllan, M. S., and M. j. Crumpton. 1972. Isolation of glycoproteins from pig


260


lymphocyte membrane using Lens culinaris phytohemagglutinin . Biochem . Biophys .Res. Commun . 47:923 .

23 . Hubbard, A., and Z . Cohn . 1972 . The enzymatic iodination of the red cell membrane .J. Cell Biol. 55:390 .

24 . Fernandez, N ., M . Labeta, M . Kurpisz, R. Solana, and H. Festenstein . 1986 . HLA-DQ molecular heterogeneity in HLA-DR4-Dw4 consanguineous cell lines . DiseaseMarkers . 4:191 .

25 . Laemmli, U. K . 1970 . Cleavage of structural proteins during assembly of the headof bacteriophage T4 . Nature (Lond .) . 227 :68 .

26 . Pawelec, G., E . M . Schneider, and P . Wernet . 1986 . Acquisition of suppressiveactivity and natural killer-like cytotoxicity by human alloproliferative "helper" T cellclones . J. Immunol . 136:402 .

27 . Wernet, P., E . M . Schneider, F . Kalthoff, and G. Pawelec . 1987 . Delineation of anovel HLA class II-like restriction system "DY" for the induction of autoreactivesuppressor cells in man: analysis by T cell clones and their antigen receptor generearrangements . Transplant. Proc. In press .

28 . Morimoto, C., N . L . Letvin, J . A . Distaso, W. R . Aldrich, and S . F . Schlossman . 1985 .The isolation and characterisation of the human suppressor inducer T cell subset . J.Immunol . 134:1508 .

29 . Pawelec, G., H .J . Buhring, E . M . Schneider, M . Ernst, C . Schmidt, A . Rehbein, I .Balko, and P . Wernet . 1987 . Expression and functional relevance of leukocytecommon antigens on T cell clones. In Leukocyte Typing III . A . J. McMichael, editor .Oxford University Press, Oxford . 818-822 .

30 . Damle, N. K., N . Mohaghegpour, G. K . Kansas, D . M. Fishwild, and E. G. Engleman .1985 . Immunoregulatory T cell circuits in man . J. Immunol . 134:235 .

31 . Wank, R., D . J . Schendel, M . E . Blanco, and B. Dupont . 1979 . Secondary MLCresponses of primed lymphocytes after selective sensitisation to non-HLA-D deter-minants . Scand . J . Immunol . 9 :499 .

32 . Martin, A., T . H . Eiermann, L . Tricas, and S . F . Goldmann . 1986 . An autoreactiveT cell clone derived after allostimulation in vitro : evidence for recognition of apolymorphic self MHC class II gene product . Transplant. Proc. 19:875 .

33, Wernet, P ., E . M . Schneider, F . Kalthoff, and G . Pawelec . 1987 . Constitutivefunctional heterogeneity without detectable somatic mutation of the antigen receptorgenes in helper T cell clones : possible regulation by novel HLA-class II "DY"determinants . lmmunol . Rev . 96:109 .

34 . Pawelec, G., E . M . Schneider, A . Rehbein, I . Balko, and P . Wernet . 1986 . Dissectionof suppressor cell generation in vitro . Hum . lmmunol . 17 :343 .

35 . Shackelford, D . A ., L . A . Lampson, and J . L . Strominger . 1983 . Separation of threeclass II antigens from a homozygous human B cell line .J. lmmunol . 130:289 .

36 . Chen, B . P ., R . DeMars, and P . Sondel . 1987 . Presentation of soluble antigen tohuman T cells by products of multiple HLA-linked loci : analysis of antigen presen-tation by a panel ofcloned, autologous, HLA-mutant Epstein-Barr virus-transformedlymphoblastoid cell lines . Hum . lmmunol. 18:75 .

37 . Germain, R . N ., and H. Quill . 1986 . Unexpected expression of a unique mixedisotype class lI molecule by transfected L cells . Nature (Lond .) . 320:72 .

38 . Pawelec, G., E . M. Schneider, and P . Wernet . 1984 . Cloned human T lymphocyteswith lymphostimulatory activity preferentially activate suppressor cells . Eur. J. lm-munol . 14:335 .

39 . Gershon, R . K ., D . D . Eardley, S . Durum, D. R . Green, R. W. Shen, K. Yamauchi,H. Cantor, and D . B . Murphy . 1981 . Contrasuppression : a novel immunoregulatoryactivity . J . Fxp . Med . 153:1533 .

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from



Iymphocyte membrane using Lens culinaris phytohemagglutinin. Biochem. Biophys. Res. Commun. 47:923.

23. Hubbard, A., and Z. Cohn. 1972. The enzymatic iodination ofthe red cell membrane. J. Cell Biol. 55:390.

24. Fernandez, N., M. Labeta, M. Kurpisz, R. Solana, and H. Festenstein. 1986. HLADQ molecular heterogeneity in HLA-DR4-Dw4 consanguineous cell Iines. Disease Markers. 4: 191.

25. Laemmli, U. K. 1970. Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature (Lond.). 227:68.

26. Pawelec, G., E. M. Schneider, and P. Wernet. 1986. Acquisition of suppressive activity and natural killer-Iike cytotoxicity by human alloproliferative "helper" T cell clones. J. Immunol. 136:402.

27. Wernet, P., E. M. Schneider, F. Kalthoff, and G. Pawelec. 1987. Delineation of a novel HLA c1ass II-Iike restrietion system "DY" for the induction of autoreactive suppressor cells in man: analysis by T cell clones and their antigen receptor gene rearrangements. Transplant. Proc. In press.

28. Morimoto, C., N. L. Letvin,J. A. Distaso, W. R. Aldrieh, and S. F. Schlossman. 1985. The isolation and characterisation of the human suppressor inducer T cell subset. J. Immunol. 134: 1508.

29. Pawelec, G., H.-J. Bühring, E. M. Schneider, M. Ernst, C. Schmidt, A. Rehbein, I. Balko, and P. Wernet. 1987. Expression and functional relevance of leukocyte common antigens on T cell clones. In Leukocyte Typing 111. A.J. McMiehael, editor. Oxford University Press, Oxford. 818-822.

30. Damle, N. K., N. Mohaghegpour, G. K. Kansas, D. M. Fishwild, and E. G. Engleman. 1985. Immunoregulatory T cell circuits in man. J. Immunol. 134:235.

31. Wank, R., D. J. Schendel, M. E. Blanco, and B. Dupont. 1979. Secondary MLC responses of primed Iymphocytes after selective sensitisation to non-HLA-D determinants. Scand. J. Immunol. 9:499.

32. Martin, A., T. H. Eiermann, L. Trieas, and S. F. Goldmann. 1986. An autoreactive T cell clone derived after allostimulation in vitro: evidence for recognition of a polymorphie self MHC c1ass II gene product. Transplant. Proc. 19:875.

33. Wernet, P., E. M. Schneider, F. Kalthoff, and G. Pawelec. 1987. Constitutive functional heterogeneity without detectable somatic mutation ofthe antigen receptor genes in helper T cell clones: possible regulation by novel HLA-c1ass II "DY" determinants. Immunol. Rev. 96: 109.

34. Pawelec, G., E. M. Schneider, A. Rehbein, I. Balko, and P. Wernet. 1986. Dissection of suppressor cell generation in vitro. Rum. Immunol. 17:343.

35. Shackelford, D. A., L. A. Lampson, and J. L. Strominger. 1983. Separation of three c1ass II antigens from a homozygous human B cellline.J. Immunol. 130:289.

36. Chen, B. P., R. DeMars, and P. Sondel. 1987. Presentation of soluble antigen to human T cells by products of multiple HLA-Iinked loci: analysis of antigen presentation by a panel of c1oned, autologous, HLA-mutant Epstein-Barr virus-transformed Iymphoblastoid celllines. Rum. Immunol. 18:75.

37. Germain, R. N., and H. Quill. 1986. Unexpected expression of a unique mixed isotype c1ass II molecule by transfected L cells. Nature (Lond.). 320:72.

38. Pawelec, G., E. M. Schneider, and P. Wernet. 1984. Cloned human T Iymphocytes with Iymphostimulatory activity preferentially activate suppressor cells. Eur. J. Immunol. 14:335.

39. Gershon, R. K., D. D. Eardley, S. Durum, D. R. Green, R. W. Shen, K. Yamauchi, H. Cantor, and D. B. Murphy. 1981. Contrasuppression: a novel immunoregulatory activity. J. Exp. Med. 153: 1533.


PAWELEC ET AL.

26 1

40 . Schneider, E . M., G . Pawelec, L. Shi, H.J . Buhring, and P . Wernet . 1987 . Generationof CD4-positive suppressor T cells from mixed lymphocyte cultures in the presenceof interleukin 2 receptor antibody TD69: an in vitro model for transplantationtolerance induction . Transplantation (Baltimore) . 44 :295 .

on Novem

ber 20, 2008 jem

.rupress.orgD

ownloaded from


PAWELEC ET AL. 261

40. Schneider, E. M., G. Pawelec, L. Shi, H.]. Bühring, and P. Wernet. 1987. Generation of CD4-positive suppressor T cells from mixed lymphocyte cultures in the presence of interleukin 2 receptor antibody TÜ69: an in vitro model for transplantation tolerance induction. Transplantation (Baitimore). 44:295.


DY determinants, possibly associated with novel class II ... · dy determinants, possibly...

Documents

Transcript of DY determinants, possibly associated with novel class II ... · dy determinants, possibly...