- Email: [email protected]
Induction of antigen-specific suppressor cells in patients with hay fever receiving immunotherapy
Induction of antigen-specific suppressor in patients with hay fever receiving immunotherapy Hiroshi
Nagaya,
cells
M.D. Torrance, Calif.
In order to evaluate the effect of immunotherapy on subpopulations of lymphocytes, peripheral blood mononuclear cells were fractionated into T and non-T cells by incubating in anti-immunoglobulin-coated plates. T cells were further fractionated into rye grass antigen-adherent and nonadherent fractions and human serum albumin (HSA)-adherent and nonadherent fractions by incubating in rye grass antigen-coated and HSA-coated plates, respectively. The mean rye antigen-stimulated proltferative response of rye antigen-adherent fraction was significantly lower than that of rye antigen-nonadherent fraction in patients with rye grass hay fever receiving immunotherapy. There was no difference in the rye antigen-stimulated proltferative response between the HSA-adherent and nonadherent fractions. Both the rye antigen-adherent and nonadherent fractions proliferated similarly when they were stimulated by an unrelated antigen, Candida albicans. The rye antigen-stimulated proliferative response of rye antigen-nonadherent cells could be suppressed by coculturing with rye antigen-adherent cells from patients receiving immunotherapy. Furthermore, the treatment of cocultures with monoclonal antibody against suppressor-cytotoxic T cell (OKT8) subpopulation and guinea pig serum complement reversed the suppression. In patients receiving no immunotherapy, there was no difference in the rye grass antigen-stimulated proliferative response between the rye antigen-adherent and nonadherent fractions. These results suggest that immunotherapy induces a subpopulation of T cells that are adherent to antigen-coated plates and are capable of suppressing T cell proliferation stimulated by the specific antigen. (J ALLERGY CLIN IMMJNOL 75:388-94, 1985 .)
The atopy is characterized by hyperactivity of IgEproducing B cells against allergens. Manipulations that enhance suppressor T cell activity such as treatment with complete Freund’s adjuvant or repeated injections of native or altered antigen suppress IgE response.’ Immunotherapy has been reported to increase the levels of blocking antibody that suppresses histamine release, to decrease antigen-specific IgE antibody levels, to increase antigen-specific IgG antibody levels, and to suppress T cell responses to antigen as measured by 3H-thymidine incorporation. ‘-‘* In most of these studies, the observations were made in patients receiving immunotherapy for 1 yr or one From the Allergy-Immunology Section, Veterans Administration Medical Center, Long Beach, Calif., and the Department of Medicine, University of California College of Medicine, Irvine, Calif. Supported by the Medical Research Service of the Veterans Administration. Received for publication Jan. 17, 1983. Accepted for publication Aug. 1, 1984. Reprint requests: Hiroshi Nagaya, M.D., 4305 Torrance Blvd., Torrance, CA 90503. 388
Abbreviations used HSA: Human serum albumin MEM: Minimum essential medium
SRBC: Sheep red blood cell GPSC: Guinea pig serum complement
season. In the study in which there were good blocking IgG antibody responses to the first preseasonal course of immunotherapy, no additional rise in antibody titer was observed after the course of perennial treatment in the second year. ‘* We evaluated long-term effects of conventional immunotherapy on a variety of immunologic parameters in patients with rye grass hay fever for up to 3 yr. l3 The mean antigen-stimulated proliferative lymphocyte response decreased significantly after the first year, and the mean antigen-specific IgG antibody level increased significantly after the second year. Neither the mean antigen-specific IgE nor the mean total serum IgE level changed significantly at any point in time during the 3 yr of immunotherapy.
VOLUME NUMBER
75 3
Antigen-specific
suppressor
cells
389
Heparinized Venous Blood t Ficoll-Hypaque Gradient t Anti-lmmunoglobulin (lg) Plate Anti-19 Ad/herent MJ% E- Rosetting) I
/
Anti-$- Non-Adherent KO% E-Rosetting)
Rye Grass Anti&
L
Rye-Adherent
FIG. 1. Flow chart and Methods.
illustratina
\ Plate Human Serum Albumin (HSA) Plate
Rye-NkAdherent the method
/
HSA-Adherent
of cell separation
Since we found that the antigen-stimulated lymphocyte proliferation was the first in vitro test that changed significantly after immunotherapy was begun, an attempt was made to evaluate the effect of immunotherapy on subpopulations of lymphocytes. Our results suggest that immunotherapy induces a subpopulation of T cells that are adherent to antigencoated plates and are capable of suppressing T cell proliferation stimulated by the specific antigen. MATERiA#. AND METHODS Patient sebtion Patients with a history and physical findings consistent with the diagnosis of allergic rhinitis were selected. Each patient had our routine initial workup studies including prick skin tests with 60 common inhalant allergens. Patients with 4+ prick test reaction to rye grass antigen were selected for further studies. Patients receiving immunotherapy had received at lea&20,000 PNU of rye grass antigen in the preceding 1 yr. This study protocol was approved by the Human Studies Subcommittee of this institution. Antigen Lyophilized preservative-free rye grass antigen extracts were supplied by Hollister-Stier, Spokane, Wash. The antigen was freshly reconstituted before each procedure. Lyophilized preservation-free Can&da albicuns antigenextracts were obtained from Greer Laboratories,Lenoir, N. C.
\
HSA-Non-Adherent
as detailed
in the section
of Material
nightat 4” C with 4 ml of goatanti-humanimmunoglobulins (CappelLaboratories,Cochranville, Pa.) containingapproximately 2 mg/ml of antibody”. I5 (Fig. 1). The antiimmunoglobulin-coated plateswerewashedfive timeswith phosphate-buffered saline,pH 7.2, beforeaddingthe cells. After incubationthe plateswere swirled to remove nonadherentcells. Cellsnonadherent to the anti-immunoglobulin-coatedplateswereenrichedfor T cellsby 1%timesas evidencedby an increasein SRBCrosettingcellsfrom approximately60% to 90% andwill be called“T cells” for convenience.TheseT cellswerefurther incubatedfor 1 hr at room temperaturein Petri plates, 60 mm in diameter, which had beencoatedovernight at 4” C with either 2 ml of 1: 20 w/v of reconstitutedrye grassantigensolutionor 2 ml of HSA (1 mglml) in phosphate-buffered salineas control. After removingnonadherent cells, the plateswere incubatedon ice for 15 min to detachadherentcells for harvesting.Thus, we separated T cells into four fractions, namelyrye grass-adherent, rye grass-nonadherent, HSAadherent,and HSA-nonadherentfractions. Cellsadherent to anti-immunoglobuhn-coated plateswere depletedof T cellsandwerecalled“non-T cells.” The non-T cellswere addedbackto eachof the four fractionsproportionatelyso that the ratio of T cells to non-T cellsin eachof the four fractionswasequal. The viability of cellswas morethan 90% by trypan blue dye exclusion.‘6 Cell culture
Thismethodhasbeendescribedpreviously.” Briefly, the cell concentrationof eachof thesefour fractionswas adjusted to 1 x 106per milliliter. Eachfraction wascultured Cell separation in triplicate in 0.5 ml of MEM supplemented with 10% Heparinizedvenousbloodwasdilutedwith an equalvolautologousplasma,100 U of penicillinper milliliter, 100 umeof Eagle’sMEM and wascentrifugedfor 35 min at pg of streptomycinper milliliter, and 2 mM/ml of L-gluroomtemperatureat 400 x g on a layer of Ficoll-Hypaque tamine.Rye grassantigenwasaddedto triplicate tubesat solutionwith the specificgravity of 1.077. Mononuclear the final concentrationsof 1: 500, 1:2500, and 1: 12,500 cellsrecoveredat the plasma-Ficollinterfacewere washed w/v. The final concentrations of Candida antigenwere 1: threetimesin MEM. Mononuclearcellswereresuspended 1000and 1: 10,000.The tubeswere incubatedat 37” C in in 2 ml of MEM at the approximateconcentrationof a humidifiedatmosphere of 5%CO, and95%air for 6 days. 20 X lo6 per milliliter and incubatedfor 30 min at room Before harvestingthe cells, 0.5 pCi of 3H-thymidinewas temperaturein polystyrenePetri plates(Falcon, Oxnard, addedto eachcell suspension for the final 4 hr of culture. Calif.), 100mmin diameter,which hadbeencoatedoverThe cell buttonswere precipitatedwith cold 5% trichlor-
399
.I. ALLERGY
Nagaya
TABLE I. Rye antigen-stimulated expressed as stimulation index Patient
R. S. w. 0. E. W. E. M. P. G. Mean 2 SEM
Unfrectionated
1.2 2.2 1.4 5.8 1.2 4.76 T 1.15
lymphocyte
proliferation
Rye-adherent
Rye-nonadherent
1.2 1.4 2.4 1.6 1.1 1.54 r 0.23
4.3 3.0 3.8 2.6 1.9 3.18 f 0.46
TABLE II. Rye antigen-stimulated lymphocyte proliferation immunotherapy expressed as stimulation index Patient
c. c. H. A. C. G. J. S. w. c. L. M. M. P. A. V. Mean ? SEM
Unfractionated
10.3 20.6 4.9 1.4 8.8 2.6 9.8 15.3 9.21 IL 2.13
Rye-adherent
Patient
R.S. 1 w. 0. E. W. E. M. Mean ? SEM
Rye-edhWMlt
8.57 9.28 2.87 9.36 7.52 t 1.35
in patients
Rye-nonadherent
10.8 6.0 1.6 1.5 3.5 2.4 7.1 2.5 4.42 +- 1.16
TABLE Ill. Candida antigen-stimulated proliferative lymphocyte response antigen-adherent and nonadherent patients receiving immunotherapy as stimulation index
in patients
of rye cells in expressed
Rye-nonadherent
8.91 7.88 4.36 10.06 7.80 ” 1.06
aceticacid,andthetrichloraceticacidinsolubleradioactivity wascountedin a liquid scintillationspectrometer.The T cell response wasexpressed asstimulationindexby dividing the radioactivity of antigen-stimulated cultureby the radioactivity of nonstimulated culture.Themaximumstimulation index wasconsideredasthe stimulationindex of the cells by theantigenregardless of antigenconcentration,although themaximumstimulationindex wasusuallyobservedat the rye grassantigenconcentrationof 1: 500 or 1: 2500. Coculture experiments For cocultureexperimentsT cellswereseparated by use of a modificationof the SRBC rosctting method described previously.” FreshSRBCssuspended in Alsever’ssolution were washedtwice in a large volumeof balanced salt so-
9.0 5.9 2.2 0.9 5.9 2.0 6.5 2.2 4.32 lr 1.01
receiving
CLIN. IMMUNOL. MARCH 1985
immunotherapy
HSA-adherent
4.3 4.8 2.1 \ 4.0 4.0 3.84 2 0.45
HSA-nonadherent
3.8 2.2 6.4 1.5 2.5 3.28 5 0.86
not receiving
HSA-adherent
10.9 1.4 2.3 1.6 8.6 5.3 4.4 5.3 5.72 t 1.10
HSA-nonadherent
9.7 10.4 3.4 0.8 18.1 3.1 8.7 11.9 8.26 k 1.99
lution and finally suspended in MEM containing 20% (by volume)of fetal calf serumthat hadbeenheat-inactivated at 56” C for 30 min andabsorbedwith one fifth volumeof packedfresh SRBCs.Equal volumes of SRBCs(4 x IO” cellsper milliliter) andperipheralbloodmononuclearcells (2 x IOhcells per milliliter) were mixed in plastic tubes and were centrifugedat 260 X g for 5 min at room temperature.After the centrifugationthe tubeswere kept at roomtemperaturefor 1 hr andthen at 4” C overnight.The cell buttonsweregently resuspended andwerecentrifuged on a layer of Ficoll-Hypaquesolutionat 400 x g at room temperaturefor 30 min. Rosettedcellsat the bottomof the tubesweretreatedwith O.OlM Tris-buffered0.84% NHCl to lyse SRBCs.The T cell populationcontainedmorethan 95% SRBC-rosettingcells and lessthan 1% surfaceimmunoglobulin-positive cellsby a directimmunofluorescence techniquedescribedpreviously.‘*T cellswereseparated into rye grassantigen-adherent, rye grassantigen-nonadherent, Candida antigen-adherent, and Candida antigen-nonadherent fractionsby incubatingfor 1 hr at roomtemperaturein Petri platesthat had been coatedovernight at 4” C with eitherrye grassantigenor Candida antigensolution(I: 20 w/v). Cells recoveredat the plasma-Ficollinterfacewere depletedof T cellsandwereaddedbackproportionatelyto eachof the four T cell fractions. Reconstituted2 x IO’ cells in 100 ~1 of culture mediumfrom eachof the four fractionswereplacedin wellsof microtiterplatesin triplicate and were culturedwith rye grassantigenfor 7 days. The cellswerepulsedwith 1 pCi per well of ‘H-thymidine for the final 18 hr of culture. Cultureswereharvestedby a
VOLUME NUMBER
75 3
Antigen-specific
suppressor
cells
391
TABLE IV. Rye antigen-stimulated
proliferative lymphocyte response of rye antigen-adherent and nonadherent fractions and their cocultures and Candida antigen-adherent and nonadherent fractions and their cocultures in patients receiving or not receiving immunotherapy expressed as stimulation index Mean
Cell fractions
Rye-adherent Rye-nonadherent Rye-adherentplusRye-nonadherent Candida-adherent Candida-nonadherent Candida-adherent plusCandidanonadherent
stimulation
Patients receiving immunotherapy
0.98 2.31 1.13 1.67
k t k k
0.10* (5)T 0.37*,* (5) 0.254 (5) 0.19 (3)
index
+ SEM No immunotherapy
1.80 -c 0.14 (3)f 2.09 5 0.26 (3) 1.80 rt 0.45 (3)
1.68 t 0.08 (3) 1.99 k 0.41 (3)
*p < 0.02. tNumberof patients in parentheses. $p< 0.05. cell harvesteron filter papers,andthe radioactivity incorporatedinto DNA was measuredin a liquid scintillation spectrometer. In orderto demonstrate the presenceof antigen-specific suppressor-cell activity in therye antigen-adherent fraction, 2 x 105cellsfrom the rye antigen-adherent fraction were coculturedwith 2 x 105cells from the rye antigen-nonadherentfraction. The rye antigen-stimulated proliferative response of thecoculture wascomparedto thoseof 2 X lo’ cellsfrom eitherthe rye antigen-adherent fractionor the rye antigen-nonadherent fraction culturedalone. The Can&h antigen-adherent fraction and the Candida antigen-nonadherentfraction werecoculturedas controls.The rye antigen-stimulated proliferative response of the coculturewas comparedto thoseof either the Candida antigen-adherent fraction or the Candidu antigen-nonadherent fraction culturedalone. In an attemptto characterizerye antigen-adherent and nonadherent T cell subpopulations from patientsreceiving immunotherapy in termsof cell surfacemarkersand cellular functions, coculture experimentswere performedin the presenceof monoclonalantibodyagainstinducer-helperT cell (OKT4) or suppressor-cytotoxic T cell (OKT8) (Ortho DiagnosticSystems,Inc., Raritan,N. J.). Ten microliters of freshlyreconstitutedOKT4 or OKT8 and 10 yl of GPSC wereaddedto eachmicrotiter well of triplicate cocultures containing2 X 10’rye antigen-adherent cellsand2 X lo5 rye antigen-nonadherent cellsin 200 ~1of culturemedium. The cellswereculturedwith rye grassantigenfor 7 days. The rye antigen-stimulated proliferative response of coculturescontainingOKT4 or OKT8 with GPSCwascompared to that of coculturescontainingGPSCalone. In order to testthe effect of OKT4 or OKT8 on the antigen-stimulated proliferative responseof rye antigen-nonadherent cells alone, 5’ ~1 of OKT4 or OKT8 and 5 ~1 of GPSCwere addedto wells containing2 x lo5 of rye antigen-nonadherentcells in 100 ~1 of culture medium.The cells were
culturedin triplicate with rye antigenfor 7 days. The rye antigen-stimulated proliferative responses were compared amongwells containingOKT4 with GPSC, OKT8 with GPSC,or GPSCalone.
RESULTS The results of rye grass antigen-stimulated lymphocyte proliferation in patients receiving and not receiving immunotherapy are illustrated in Tables 1 and
II, respectively. In patients receiving immunotherapy the mean stimulation index of rye grass antigen-adherent fraction (1.54 k 0.23) was significantly lower than that of rye grass antigen-nonadherent fraction (3.18 + 0.46, p < 0.02). The mean stimulation indices of HSA-adherent fraction and HSA-nonadherent fraction as controls were 3.84 +: 0.45 and 3.28 + 0.86, respectively. In patients not receiving immunotherapy the mean stimulation indices of rye grass antigen-adherent and nonadherent fractions were 4.42 it 1.16 and 4.32 + I .Ol, respectively, whereas those of HSA-adherent and nonadherent fractions were 5.72 + 1.10 and 8.26 + 1.99, respectively. The difference between the latter two values is not significant (p > 0.2). These results suggest that in patients receiving immunotherapy the antigen-adherent fraction is enriched for cells capable of suppressing antigen-stimulated lymphocyte proliferation. In patients not receiving immunotherapy no such emichment was observed by incubating the cells in antigencoated plates. Although the mean stimulation index of unfractionated cells cultured with rye antigen was lower in the group receiving immunotherapy (4.76 -+ 1.15) than in the group not receiving immunotherapy
J. ALLERGY CLIN. IMMUNOL.
392 Nagaya
MARCH1985
3-
t:
2-
:
-
Rye-Non-Adherent
C--O Rye-Adherent LL-dRye-Non-Adherent
I::00 RYE
ANTIGEN
+ Rye-Adherent
1:2&o CONCENTRATION
1: lzm
(w/v)
FIG. 2. A representative coculture experiment in patients receiving immunotherapy; 2 x IO5 rye antigen-adherent cells and 2 x IO5 rye antigen-nonadherent ceils were cultured either separately in 100 ~1 of culture medium or cocultured in 200 ~1 of culture medium in wells of microtiter plates for 7 days in triplicate in the presence of 1 : 500, 1: 2500, or 1: 12,500 rye antigen.
(9.21 + 2.13), the difference was not significant (0.05 < p < 0.1). However, the mean stimulation index of rye antigen-adherent fraction in the former group (1.54 ? 0.23) was significantly lower than that in the latter group (4.42 + 1.16, p < 0.05), suggesting that as a group mean value the antigen-stimulated proliferation of antigen-adherent fraction is a useful parameter in evaluating a response to immunotherapy. In order to demonstrate that the suppressor activity of rye grass antigen-adherent cells is antigen specific, rye antigen-adherent cells from four patients receiving immunotherapy were cultured with Candida antigen for 6 days in the final concentrations of 1: 1000 and 1: 10,000. The mean stimulation index of rye antigenadherent cells stimulated by Candida antigen was 7.52 * 1.35 @EM), whereas that of rye antigen-nonadherent cells stimulated by Cundidu antigen was 7.80 + 1.06 (Table III). Therefore, the suppressor activity of rye antigen-adherent cells was antigen specific . The results of coculture experiments are illustrated in Table IV. The rye antigen-stimulated lymphocyte proliferation of rye grass antigen-nonadherent cells was significantly reduced by coculturing with rye
grass antigen-adherent cells only in patients receiving immunotherapy, suggesting that rye grass antigen-adherent cells contained antigen-specific suppressor cells. Fig. 2 illustrates a representative dose-response curve of the coculture experiments in patients receiving immunotherapy. In control experiments there were no significant differences in the rye antigen-stimulated proliferative lymphocyte response among the Candida antigen-adherent fraction, Candidu antigen-nonadherent fraction, and their cocultures. In an attempt to characterize rye antigen-adherent and nonadherent cell subpopulations from patients receiving immunotherapy in terms of cell surface markers and cellular functions, coculture experiments were performed in the presence of monoclonal antibody against helper T cells (OKT4) or suppressor T cells (OKT8) with GPSC. Table V illustrates that the addition of OKT8 and GPSC to cocultures containing rye antigen-adherent and rye antigen-nonadherent subpopulations significantly increased the rye antigenstimulated proliferative response compared to that of cocultures to which GPSC alone was added. This result suggests that OKT8 plus GPSC abolished the suppressor cell activity of rye antigen-adherent cell subpopulation. In order to determine the effect of OKT4 or OKT8 on the rye antigen-stimulated proliferative response of rye grass antigen-nonadherent cell subpopulation alone, rye antigen-nonadherent cells were cultured with either OKT4 or OKT8 with GPSC. The addition of OKT4 and GPSC illustrated in the lower part of Table V abolished the proliferative response of rye antigen-nonadherent cells, whereas the addition of OKT8 and GPSC had no effect on the rye antigen-stimulated proliferative response of rye antigen-nonadherent cell subpopulation. This finding supports the interpretation that OKT8 and GPSC abolished the suppressor cell activity present in the rye antigen-adherent cell subpopulation of cocultures. DISCUSSION In order to evaluate the efficacy of immunotherapy objectively, we have measured a variety of immunologic parameters in our atopic patients receiving immunotherapy. In our studies the antigen-stimulated lymphocyte proliferation was the first in vitro test that changed significantly after immunotherapy was begun. I3 In an attempt to analyze immunologic events leading to a suppression of antigen-stimulated lymphocyte proliferation in patients receiving immunotherapy, we separated peripheral blood T lymphocytes into rye grass antigen-adherent and nonadherent cells according to the method of Taniguchi and Miller. I5 Although the mean stimulation index of unfractionated peripheral blood mononuclear cells cultured with
VOLUME NUMBER
75 3
Antigen-specific
suppressor
TABLE V. The effect of OKT4 or OKT8 with GPSC on the rye antigen-stimuleted proliferative lymphocyte response of cocultures of rye antigen-adherent and nonadherent subpopulations
ceils
3S3
from
patients receiving immunotherapy Cell culture
Mean
Rye-adherent plus rye-nonadherent plus complement alone Rye-adherent plus rye-nonadherent plus OKT4 plus complement Rye-adherent plus rye-nonadherent plus OKT8 plus complement Rye-nonadherent plus complement alone Rye-nonadherent plus OKT4 plus complement Rye-nonadherent plus OKT8 plus complement
sthulaekw,
0.67 0.98 1.78 2.24 0.76
-t t r t 4
indar
t SiW
0.18* (6)0.29 (6) 0.29” (6) 0.67$ (6) 0.13S.lj (h)
2.15 r 0.455 (6) ----_~-__
*p < 0.02. tNumber of patients in parentheses. tp < 0.01. Pp ‘c 0.02.
rye grass antigen in this study was not significantly lower in the group receiving immunotherapy than in the group not receiving immunotherapy, the rye antigen-adherent subpopulation responded significantly less to rye antigen in the group receiving immunotherapy than in the group not receiving immunotherapy. More importantly, the mean rye antigenstimulated proliferation of rye antigen-adherent subpopulation was significantly lower than that of the rye antigen-nonadherent subpopulation only in the group receiving immunotherapy but not in the untreated group. In order to demonstrate the antigen specificity of this observation, T cells were also fractionated into the HSA-adherent and nonadherent fractions. There was no significant difference in the rye antigen-stimulated proliferative response between the HSA-adherent and nonadherent fractions both in the treated and untreated groups. In an attempt to demonstrate that the rye antigen-adherent subpopulation can respond to an indifferent antigen equally well as the rye antigen-nonadherent subpopulation can, the antigenadherent subpopulation was also cultured with Candidu antigen as an indifferent antigen. The Candida antigen-stimulated proliferation of rye antigen-adherent subpopulation was not different from that of rye antigen-nonadherent subpopulation. Therefore, the suppression of rye antigen-stimulated proliferation of rye antigen-adherent subpopulation in the patients receiving immunotherapy was antigen specific. In order to demonstrate that the rye antigen-adherent cell subpopulation from patients receiving immunotherapy has an antigen-specific suppressor cell activity, we performed coculture experiments. The rye antigen-stimulated lymphocyte proliferation of rye antigen-nonadherent cells was significantly suppressed by coculturing with rye antigen-adherent cells only in patients receiving immunotherapy but not in patients receiving no immunotherapy. There were no signifi-
cant differences in the rye antigen-stimulated proiiferative lymphocyte response between the Can&&r antigen-adherent cells and the Cundida antigen-nonadherent cells or their cocultures. In an attempt to characterize rye antigen-adherent and rye antigen-nonadherent subpopulations from patients receiving immunotherapy in terms of cell surface markers and cellular functions, coculture experiments were performed in the presence of monoclonal antibody against inducer-helper T cells 03KT4) or suppressor-cytotoxic T cells (OKT8) and GPSC. When OKT8 and GPSC were added to cocultures containing both rye antigen-adherent and rye antigennonadherent subpopulations, the rye antigen-stimulated proliferative response of the cocultures was significantly increased compared to the response of cocultures to which GPSC alone was added. This finding suggested that OKT8 plus GPSC abolished the suppressor cell activity of rye antigen-adherent cell subpopulation. Furthermore, when OKT4 and CPSC were added to cultures containing rye antigen-nonadherent cells alone, the rye antigen-stimulated proliferative response of rye antigen-nonadherent cells was abolished. In contrast, OKT8 and GPSC had no effect on the rye antigen-stimulated proliferative response of rye antigen-nonadherent cell subpopulation. These findings strongly suggest that the rye antigenadherent cell subpopulation contains cells capable of suppressing the rye antigen-stimulated proliferative response, and the treatment of the cells with OKT8 and GPSC can abrogate the suppressor function. It is most likely that the cells with the suppressor activity belong to the T cell subset identifiable hy OKT8. Antigen-stimulated lymphocyte proliferative responses have been reported to decrease in patients receiving immunotherapy.‘, “‘. ‘I’ ” immunotherapy has been demonstrated to generate antigen-specific suppressor T cells belonging to the subpopulation of
J. ALLERGY
394 Nagaya
lymphocytes bearing histamine receptors. I9 Although the histamine-activated suppressor cell system was markedly diminished in the atopic population, the concanavalin A-activated suppressor cell system was functionally equivalent in the atopic subjects and the normal control subjects.” The results of the present study demonstrated that rye antigen-adherent T cells obtained from patients with rye grass hay fever receiving immunotherapy can suppress the rye antigenstimulated proliferative lymphocyte response of rye antigen-nonadherent subpopulation. If we extrapolate our in vitro observations into the in vivo situation in patients receiving immunotherapy, the salutary effect of immunotherapy may depend on the activation of suppressor cell function of antigen-adherent T cells by the specific antigen injection. I am grateful to the entire staff of the Allergy-Immunology Section for their helpful contribution to this study. REFERENCES 1. Ishizaka K: Cellular events in the IgE antibody responses. Adv Immunol 23: 1, 1976 2. Lichtenstein LM, Norman PS, Winkenwerder WL, Osler AG: In vitro studies of human ragweed allergy: changes in cellular and humoral activity associated with specific desensitization. .I Clin Invest 45:1126, 1966 3. Pruzansky JJ, Patterson R: Histamine release from leukocytes of hypersensitive individuals. II. Reduced sensitivity of leukocytes after injection therapy. J ALLERGY 3944, 1967 4. Lichtenstein LM, Norman PS, Winkenwerder WL: A single year of immunotherapy for ragweed hay fever, immunologic and clinical studies. Ann Intern Med 75:553, 1971 5. Levy DA, Lichtenstein LM, Goldstein EO, Ishizaka K: Immunologic and cellular changes accompanying the therapy of pollen allergy. J Clin Invest 50:360, 1971 6. Lichtenstein LM, Ishizaka K, Norman PS, Sobotka AK, Hill BM: IgE antibody measurements in ragweed hay fever, relationship to clinical severity, and the results of immunotherapy. J Clin Invest 52:472, 1973 7. Anderson JA, Lane SR, Howard WA, Leiken S, Oppenheim
CLIN. IMMUNOL.
MARCH 1985
8.
9.
10. 11.
12.
JJ: The effect of hyposensitization on Alzernnria-induced lymphocyte blastogenesis. Cell Immunol 10442, 1974 Gaiten JG, Merler E, Colten HR: Allergy to ragweed antigen E: effect of specific immunotherapy on the reactivity of human T lymphocytes in vitro. Clin Immunol Immunopathol 4:32, 1975 Romagnani S, Biliotti G, Ricci M: Depression of grass polleninduced lymphocyte transformation by serum from hyposensitized patients. Clin Exp Immunol 19:83, 1975 Evans R, Pence H, Kaplan H, Rocklin RE: The effect of immunotherapy on humoral and cellular responses in ragweed hay fever. J Clin Invest 57:1378, 1976 Nagaya H, Lee SK, Reddy PM, Pascual H, Jerome D, Sadai J, Gupta S, Lauridsen J: Lymphocyte response to grass pollen antigens: a correlation with radioallergosorbent test and effect of immunotherapy. Ann Allergy 39:246, 1977 Norman PS, Lictenstein LM: The clinical and immunologic specificity of immunotherapy. J ALLERGY CLM IMMUNOL 61:370,
1978
13. Nagaya H: Long-term effects of conventional immunotherapy in southern California. Ann Allergy 44:193, 1980 14. Mage MG, McHugh LL, Rothstein TL: Mouse lymphocytes with and without surface immunoglobulin: preparative scale separation in polystyrene tissue culture dishes coated with specifically purified anti-immunoglobulin. J Immunol Methods 15:47, 1977 15. Taniguchi M, Miller JFAP: Enrichment of specific suppressor T cells and characterization of their surface markers. J Exp Med 146:1450, 1977 16. Nagaya H, Sieker HO: Lymphopenic effect of antilymphocyte serum. Proc Sot Exp Biol Med 121:722, 1966 17, Walter H, Nagaya H: Separation of human rosette- and nonrosette-forming lymphoid cells by countercurrent distribution in an aqueous two-phase system. Cell Immunol 19:158, 1975 18. Walter H, Moncla BJ, Weber TJ, Nagaya H: Membrane chargeassociated heterogeneity of B lymphocytes from human peripheral blood as reflected by cell partition in two-polymer aqueous phases. Exp Cell Res 122:380, 1979 19, Rocklin RE, Sheffer AL, Greineder DK, Melmon KL: Generation of antigen-specific suppressor cells during allergy desensitization. N Engl J Med 302: 1213. 1980 20. Beer DJ, Osband ME, McCaffrey RP, Soter NA, Rocklin RE: Abnormal histamine-induced suppressor-cell function in atopic subjects. N Engl J Med 306:454, 1982
Nagaya,
cells
M.D. Torrance, Calif.
In order to evaluate the effect of immunotherapy on subpopulations of lymphocytes, peripheral blood mononuclear cells were fractionated into T and non-T cells by incubating in anti-immunoglobulin-coated plates. T cells were further fractionated into rye grass antigen-adherent and nonadherent fractions and human serum albumin (HSA)-adherent and nonadherent fractions by incubating in rye grass antigen-coated and HSA-coated plates, respectively. The mean rye antigen-stimulated proltferative response of rye antigen-adherent fraction was significantly lower than that of rye antigen-nonadherent fraction in patients with rye grass hay fever receiving immunotherapy. There was no difference in the rye antigen-stimulated proltferative response between the HSA-adherent and nonadherent fractions. Both the rye antigen-adherent and nonadherent fractions proliferated similarly when they were stimulated by an unrelated antigen, Candida albicans. The rye antigen-stimulated proliferative response of rye antigen-nonadherent cells could be suppressed by coculturing with rye antigen-adherent cells from patients receiving immunotherapy. Furthermore, the treatment of cocultures with monoclonal antibody against suppressor-cytotoxic T cell (OKT8) subpopulation and guinea pig serum complement reversed the suppression. In patients receiving no immunotherapy, there was no difference in the rye grass antigen-stimulated proliferative response between the rye antigen-adherent and nonadherent fractions. These results suggest that immunotherapy induces a subpopulation of T cells that are adherent to antigen-coated plates and are capable of suppressing T cell proliferation stimulated by the specific antigen. (J ALLERGY CLIN IMMJNOL 75:388-94, 1985 .)
The atopy is characterized by hyperactivity of IgEproducing B cells against allergens. Manipulations that enhance suppressor T cell activity such as treatment with complete Freund’s adjuvant or repeated injections of native or altered antigen suppress IgE response.’ Immunotherapy has been reported to increase the levels of blocking antibody that suppresses histamine release, to decrease antigen-specific IgE antibody levels, to increase antigen-specific IgG antibody levels, and to suppress T cell responses to antigen as measured by 3H-thymidine incorporation. ‘-‘* In most of these studies, the observations were made in patients receiving immunotherapy for 1 yr or one From the Allergy-Immunology Section, Veterans Administration Medical Center, Long Beach, Calif., and the Department of Medicine, University of California College of Medicine, Irvine, Calif. Supported by the Medical Research Service of the Veterans Administration. Received for publication Jan. 17, 1983. Accepted for publication Aug. 1, 1984. Reprint requests: Hiroshi Nagaya, M.D., 4305 Torrance Blvd., Torrance, CA 90503. 388
Abbreviations used HSA: Human serum albumin MEM: Minimum essential medium
SRBC: Sheep red blood cell GPSC: Guinea pig serum complement
season. In the study in which there were good blocking IgG antibody responses to the first preseasonal course of immunotherapy, no additional rise in antibody titer was observed after the course of perennial treatment in the second year. ‘* We evaluated long-term effects of conventional immunotherapy on a variety of immunologic parameters in patients with rye grass hay fever for up to 3 yr. l3 The mean antigen-stimulated proliferative lymphocyte response decreased significantly after the first year, and the mean antigen-specific IgG antibody level increased significantly after the second year. Neither the mean antigen-specific IgE nor the mean total serum IgE level changed significantly at any point in time during the 3 yr of immunotherapy.
VOLUME NUMBER
75 3
Antigen-specific
suppressor
cells
389
Heparinized Venous Blood t Ficoll-Hypaque Gradient t Anti-lmmunoglobulin (lg) Plate Anti-19 Ad/herent MJ% E- Rosetting) I
/
Anti-$- Non-Adherent KO% E-Rosetting)
Rye Grass Anti&
L
Rye-Adherent
FIG. 1. Flow chart and Methods.
illustratina
\ Plate Human Serum Albumin (HSA) Plate
Rye-NkAdherent the method
/
HSA-Adherent
of cell separation
Since we found that the antigen-stimulated lymphocyte proliferation was the first in vitro test that changed significantly after immunotherapy was begun, an attempt was made to evaluate the effect of immunotherapy on subpopulations of lymphocytes. Our results suggest that immunotherapy induces a subpopulation of T cells that are adherent to antigencoated plates and are capable of suppressing T cell proliferation stimulated by the specific antigen. MATERiA#. AND METHODS Patient sebtion Patients with a history and physical findings consistent with the diagnosis of allergic rhinitis were selected. Each patient had our routine initial workup studies including prick skin tests with 60 common inhalant allergens. Patients with 4+ prick test reaction to rye grass antigen were selected for further studies. Patients receiving immunotherapy had received at lea&20,000 PNU of rye grass antigen in the preceding 1 yr. This study protocol was approved by the Human Studies Subcommittee of this institution. Antigen Lyophilized preservative-free rye grass antigen extracts were supplied by Hollister-Stier, Spokane, Wash. The antigen was freshly reconstituted before each procedure. Lyophilized preservation-free Can&da albicuns antigenextracts were obtained from Greer Laboratories,Lenoir, N. C.
\
HSA-Non-Adherent
as detailed
in the section
of Material
nightat 4” C with 4 ml of goatanti-humanimmunoglobulins (CappelLaboratories,Cochranville, Pa.) containingapproximately 2 mg/ml of antibody”. I5 (Fig. 1). The antiimmunoglobulin-coated plateswerewashedfive timeswith phosphate-buffered saline,pH 7.2, beforeaddingthe cells. After incubationthe plateswere swirled to remove nonadherentcells. Cellsnonadherent to the anti-immunoglobulin-coatedplateswereenrichedfor T cellsby 1%timesas evidencedby an increasein SRBCrosettingcellsfrom approximately60% to 90% andwill be called“T cells” for convenience.TheseT cellswerefurther incubatedfor 1 hr at room temperaturein Petri plates, 60 mm in diameter, which had beencoatedovernight at 4” C with either 2 ml of 1: 20 w/v of reconstitutedrye grassantigensolutionor 2 ml of HSA (1 mglml) in phosphate-buffered salineas control. After removingnonadherent cells, the plateswere incubatedon ice for 15 min to detachadherentcells for harvesting.Thus, we separated T cells into four fractions, namelyrye grass-adherent, rye grass-nonadherent, HSAadherent,and HSA-nonadherentfractions. Cellsadherent to anti-immunoglobuhn-coated plateswere depletedof T cellsandwerecalled“non-T cells.” The non-T cellswere addedbackto eachof the four fractionsproportionatelyso that the ratio of T cells to non-T cellsin eachof the four fractionswasequal. The viability of cellswas morethan 90% by trypan blue dye exclusion.‘6 Cell culture
Thismethodhasbeendescribedpreviously.” Briefly, the cell concentrationof eachof thesefour fractionswas adjusted to 1 x 106per milliliter. Eachfraction wascultured Cell separation in triplicate in 0.5 ml of MEM supplemented with 10% Heparinizedvenousbloodwasdilutedwith an equalvolautologousplasma,100 U of penicillinper milliliter, 100 umeof Eagle’sMEM and wascentrifugedfor 35 min at pg of streptomycinper milliliter, and 2 mM/ml of L-gluroomtemperatureat 400 x g on a layer of Ficoll-Hypaque tamine.Rye grassantigenwasaddedto triplicate tubesat solutionwith the specificgravity of 1.077. Mononuclear the final concentrationsof 1: 500, 1:2500, and 1: 12,500 cellsrecoveredat the plasma-Ficollinterfacewere washed w/v. The final concentrations of Candida antigenwere 1: threetimesin MEM. Mononuclearcellswereresuspended 1000and 1: 10,000.The tubeswere incubatedat 37” C in in 2 ml of MEM at the approximateconcentrationof a humidifiedatmosphere of 5%CO, and95%air for 6 days. 20 X lo6 per milliliter and incubatedfor 30 min at room Before harvestingthe cells, 0.5 pCi of 3H-thymidinewas temperaturein polystyrenePetri plates(Falcon, Oxnard, addedto eachcell suspension for the final 4 hr of culture. Calif.), 100mmin diameter,which hadbeencoatedoverThe cell buttonswere precipitatedwith cold 5% trichlor-
399
.I. ALLERGY
Nagaya
TABLE I. Rye antigen-stimulated expressed as stimulation index Patient
R. S. w. 0. E. W. E. M. P. G. Mean 2 SEM
Unfrectionated
1.2 2.2 1.4 5.8 1.2 4.76 T 1.15
lymphocyte
proliferation
Rye-adherent
Rye-nonadherent
1.2 1.4 2.4 1.6 1.1 1.54 r 0.23
4.3 3.0 3.8 2.6 1.9 3.18 f 0.46
TABLE II. Rye antigen-stimulated lymphocyte proliferation immunotherapy expressed as stimulation index Patient
c. c. H. A. C. G. J. S. w. c. L. M. M. P. A. V. Mean ? SEM
Unfractionated
10.3 20.6 4.9 1.4 8.8 2.6 9.8 15.3 9.21 IL 2.13
Rye-adherent
Patient
R.S. 1 w. 0. E. W. E. M. Mean ? SEM
Rye-edhWMlt
8.57 9.28 2.87 9.36 7.52 t 1.35
in patients
Rye-nonadherent
10.8 6.0 1.6 1.5 3.5 2.4 7.1 2.5 4.42 +- 1.16
TABLE Ill. Candida antigen-stimulated proliferative lymphocyte response antigen-adherent and nonadherent patients receiving immunotherapy as stimulation index
in patients
of rye cells in expressed
Rye-nonadherent
8.91 7.88 4.36 10.06 7.80 ” 1.06
aceticacid,andthetrichloraceticacidinsolubleradioactivity wascountedin a liquid scintillationspectrometer.The T cell response wasexpressed asstimulationindexby dividing the radioactivity of antigen-stimulated cultureby the radioactivity of nonstimulated culture.Themaximumstimulation index wasconsideredasthe stimulationindex of the cells by theantigenregardless of antigenconcentration,although themaximumstimulationindex wasusuallyobservedat the rye grassantigenconcentrationof 1: 500 or 1: 2500. Coculture experiments For cocultureexperimentsT cellswereseparated by use of a modificationof the SRBC rosctting method described previously.” FreshSRBCssuspended in Alsever’ssolution were washedtwice in a large volumeof balanced salt so-
9.0 5.9 2.2 0.9 5.9 2.0 6.5 2.2 4.32 lr 1.01
receiving
CLIN. IMMUNOL. MARCH 1985
immunotherapy
HSA-adherent
4.3 4.8 2.1 \ 4.0 4.0 3.84 2 0.45
HSA-nonadherent
3.8 2.2 6.4 1.5 2.5 3.28 5 0.86
not receiving
HSA-adherent
10.9 1.4 2.3 1.6 8.6 5.3 4.4 5.3 5.72 t 1.10
HSA-nonadherent
9.7 10.4 3.4 0.8 18.1 3.1 8.7 11.9 8.26 k 1.99
lution and finally suspended in MEM containing 20% (by volume)of fetal calf serumthat hadbeenheat-inactivated at 56” C for 30 min andabsorbedwith one fifth volumeof packedfresh SRBCs.Equal volumes of SRBCs(4 x IO” cellsper milliliter) andperipheralbloodmononuclearcells (2 x IOhcells per milliliter) were mixed in plastic tubes and were centrifugedat 260 X g for 5 min at room temperature.After the centrifugationthe tubeswere kept at roomtemperaturefor 1 hr andthen at 4” C overnight.The cell buttonsweregently resuspended andwerecentrifuged on a layer of Ficoll-Hypaquesolutionat 400 x g at room temperaturefor 30 min. Rosettedcellsat the bottomof the tubesweretreatedwith O.OlM Tris-buffered0.84% NHCl to lyse SRBCs.The T cell populationcontainedmorethan 95% SRBC-rosettingcells and lessthan 1% surfaceimmunoglobulin-positive cellsby a directimmunofluorescence techniquedescribedpreviously.‘*T cellswereseparated into rye grassantigen-adherent, rye grassantigen-nonadherent, Candida antigen-adherent, and Candida antigen-nonadherent fractionsby incubatingfor 1 hr at roomtemperaturein Petri platesthat had been coatedovernight at 4” C with eitherrye grassantigenor Candida antigensolution(I: 20 w/v). Cells recoveredat the plasma-Ficollinterfacewere depletedof T cellsandwereaddedbackproportionatelyto eachof the four T cell fractions. Reconstituted2 x IO’ cells in 100 ~1 of culture mediumfrom eachof the four fractionswereplacedin wellsof microtiterplatesin triplicate and were culturedwith rye grassantigenfor 7 days. The cellswerepulsedwith 1 pCi per well of ‘H-thymidine for the final 18 hr of culture. Cultureswereharvestedby a
VOLUME NUMBER
75 3
Antigen-specific
suppressor
cells
391
TABLE IV. Rye antigen-stimulated
proliferative lymphocyte response of rye antigen-adherent and nonadherent fractions and their cocultures and Candida antigen-adherent and nonadherent fractions and their cocultures in patients receiving or not receiving immunotherapy expressed as stimulation index Mean
Cell fractions
Rye-adherent Rye-nonadherent Rye-adherentplusRye-nonadherent Candida-adherent Candida-nonadherent Candida-adherent plusCandidanonadherent
stimulation
Patients receiving immunotherapy
0.98 2.31 1.13 1.67
k t k k
0.10* (5)T 0.37*,* (5) 0.254 (5) 0.19 (3)
index
+ SEM No immunotherapy
1.80 -c 0.14 (3)f 2.09 5 0.26 (3) 1.80 rt 0.45 (3)
1.68 t 0.08 (3) 1.99 k 0.41 (3)
*p < 0.02. tNumberof patients in parentheses. $p< 0.05. cell harvesteron filter papers,andthe radioactivity incorporatedinto DNA was measuredin a liquid scintillation spectrometer. In orderto demonstrate the presenceof antigen-specific suppressor-cell activity in therye antigen-adherent fraction, 2 x 105cellsfrom the rye antigen-adherent fraction were coculturedwith 2 x 105cells from the rye antigen-nonadherentfraction. The rye antigen-stimulated proliferative response of thecoculture wascomparedto thoseof 2 X lo’ cellsfrom eitherthe rye antigen-adherent fractionor the rye antigen-nonadherent fraction culturedalone. The Can&h antigen-adherent fraction and the Candida antigen-nonadherentfraction werecoculturedas controls.The rye antigen-stimulated proliferative response of the coculturewas comparedto thoseof either the Candida antigen-adherent fraction or the Candidu antigen-nonadherent fraction culturedalone. In an attemptto characterizerye antigen-adherent and nonadherent T cell subpopulations from patientsreceiving immunotherapy in termsof cell surfacemarkersand cellular functions, coculture experimentswere performedin the presenceof monoclonalantibodyagainstinducer-helperT cell (OKT4) or suppressor-cytotoxic T cell (OKT8) (Ortho DiagnosticSystems,Inc., Raritan,N. J.). Ten microliters of freshlyreconstitutedOKT4 or OKT8 and 10 yl of GPSC wereaddedto eachmicrotiter well of triplicate cocultures containing2 X 10’rye antigen-adherent cellsand2 X lo5 rye antigen-nonadherent cellsin 200 ~1of culturemedium. The cellswereculturedwith rye grassantigenfor 7 days. The rye antigen-stimulated proliferative response of coculturescontainingOKT4 or OKT8 with GPSCwascompared to that of coculturescontainingGPSCalone. In order to testthe effect of OKT4 or OKT8 on the antigen-stimulated proliferative responseof rye antigen-nonadherent cells alone, 5’ ~1 of OKT4 or OKT8 and 5 ~1 of GPSCwere addedto wells containing2 x lo5 of rye antigen-nonadherentcells in 100 ~1 of culture medium.The cells were
culturedin triplicate with rye antigenfor 7 days. The rye antigen-stimulated proliferative responses were compared amongwells containingOKT4 with GPSC, OKT8 with GPSC,or GPSCalone.
RESULTS The results of rye grass antigen-stimulated lymphocyte proliferation in patients receiving and not receiving immunotherapy are illustrated in Tables 1 and
II, respectively. In patients receiving immunotherapy the mean stimulation index of rye grass antigen-adherent fraction (1.54 k 0.23) was significantly lower than that of rye grass antigen-nonadherent fraction (3.18 + 0.46, p < 0.02). The mean stimulation indices of HSA-adherent fraction and HSA-nonadherent fraction as controls were 3.84 +: 0.45 and 3.28 + 0.86, respectively. In patients not receiving immunotherapy the mean stimulation indices of rye grass antigen-adherent and nonadherent fractions were 4.42 it 1.16 and 4.32 + I .Ol, respectively, whereas those of HSA-adherent and nonadherent fractions were 5.72 + 1.10 and 8.26 + 1.99, respectively. The difference between the latter two values is not significant (p > 0.2). These results suggest that in patients receiving immunotherapy the antigen-adherent fraction is enriched for cells capable of suppressing antigen-stimulated lymphocyte proliferation. In patients not receiving immunotherapy no such emichment was observed by incubating the cells in antigencoated plates. Although the mean stimulation index of unfractionated cells cultured with rye antigen was lower in the group receiving immunotherapy (4.76 -+ 1.15) than in the group not receiving immunotherapy
J. ALLERGY CLIN. IMMUNOL.
392 Nagaya
MARCH1985
3-
t:
2-
:
-
Rye-Non-Adherent
C--O Rye-Adherent LL-dRye-Non-Adherent
I::00 RYE
ANTIGEN
+ Rye-Adherent
1:2&o CONCENTRATION
1: lzm
(w/v)
FIG. 2. A representative coculture experiment in patients receiving immunotherapy; 2 x IO5 rye antigen-adherent cells and 2 x IO5 rye antigen-nonadherent ceils were cultured either separately in 100 ~1 of culture medium or cocultured in 200 ~1 of culture medium in wells of microtiter plates for 7 days in triplicate in the presence of 1 : 500, 1: 2500, or 1: 12,500 rye antigen.
(9.21 + 2.13), the difference was not significant (0.05 < p < 0.1). However, the mean stimulation index of rye antigen-adherent fraction in the former group (1.54 ? 0.23) was significantly lower than that in the latter group (4.42 + 1.16, p < 0.05), suggesting that as a group mean value the antigen-stimulated proliferation of antigen-adherent fraction is a useful parameter in evaluating a response to immunotherapy. In order to demonstrate that the suppressor activity of rye grass antigen-adherent cells is antigen specific, rye antigen-adherent cells from four patients receiving immunotherapy were cultured with Candida antigen for 6 days in the final concentrations of 1: 1000 and 1: 10,000. The mean stimulation index of rye antigenadherent cells stimulated by Candida antigen was 7.52 * 1.35 @EM), whereas that of rye antigen-nonadherent cells stimulated by Cundidu antigen was 7.80 + 1.06 (Table III). Therefore, the suppressor activity of rye antigen-adherent cells was antigen specific . The results of coculture experiments are illustrated in Table IV. The rye antigen-stimulated lymphocyte proliferation of rye grass antigen-nonadherent cells was significantly reduced by coculturing with rye
grass antigen-adherent cells only in patients receiving immunotherapy, suggesting that rye grass antigen-adherent cells contained antigen-specific suppressor cells. Fig. 2 illustrates a representative dose-response curve of the coculture experiments in patients receiving immunotherapy. In control experiments there were no significant differences in the rye antigen-stimulated proliferative lymphocyte response among the Candida antigen-adherent fraction, Candidu antigen-nonadherent fraction, and their cocultures. In an attempt to characterize rye antigen-adherent and nonadherent cell subpopulations from patients receiving immunotherapy in terms of cell surface markers and cellular functions, coculture experiments were performed in the presence of monoclonal antibody against helper T cells (OKT4) or suppressor T cells (OKT8) with GPSC. Table V illustrates that the addition of OKT8 and GPSC to cocultures containing rye antigen-adherent and rye antigen-nonadherent subpopulations significantly increased the rye antigenstimulated proliferative response compared to that of cocultures to which GPSC alone was added. This result suggests that OKT8 plus GPSC abolished the suppressor cell activity of rye antigen-adherent cell subpopulation. In order to determine the effect of OKT4 or OKT8 on the rye antigen-stimulated proliferative response of rye grass antigen-nonadherent cell subpopulation alone, rye antigen-nonadherent cells were cultured with either OKT4 or OKT8 with GPSC. The addition of OKT4 and GPSC illustrated in the lower part of Table V abolished the proliferative response of rye antigen-nonadherent cells, whereas the addition of OKT8 and GPSC had no effect on the rye antigen-stimulated proliferative response of rye antigen-nonadherent cell subpopulation. This finding supports the interpretation that OKT8 and GPSC abolished the suppressor cell activity present in the rye antigen-adherent cell subpopulation of cocultures. DISCUSSION In order to evaluate the efficacy of immunotherapy objectively, we have measured a variety of immunologic parameters in our atopic patients receiving immunotherapy. In our studies the antigen-stimulated lymphocyte proliferation was the first in vitro test that changed significantly after immunotherapy was begun. I3 In an attempt to analyze immunologic events leading to a suppression of antigen-stimulated lymphocyte proliferation in patients receiving immunotherapy, we separated peripheral blood T lymphocytes into rye grass antigen-adherent and nonadherent cells according to the method of Taniguchi and Miller. I5 Although the mean stimulation index of unfractionated peripheral blood mononuclear cells cultured with
VOLUME NUMBER
75 3
Antigen-specific
suppressor
TABLE V. The effect of OKT4 or OKT8 with GPSC on the rye antigen-stimuleted proliferative lymphocyte response of cocultures of rye antigen-adherent and nonadherent subpopulations
ceils
3S3
from
patients receiving immunotherapy Cell culture
Mean
Rye-adherent plus rye-nonadherent plus complement alone Rye-adherent plus rye-nonadherent plus OKT4 plus complement Rye-adherent plus rye-nonadherent plus OKT8 plus complement Rye-nonadherent plus complement alone Rye-nonadherent plus OKT4 plus complement Rye-nonadherent plus OKT8 plus complement
sthulaekw,
0.67 0.98 1.78 2.24 0.76
-t t r t 4
indar
t SiW
0.18* (6)0.29 (6) 0.29” (6) 0.67$ (6) 0.13S.lj (h)
2.15 r 0.455 (6) ----_~-__
*p < 0.02. tNumber of patients in parentheses. tp < 0.01. Pp ‘c 0.02.
rye grass antigen in this study was not significantly lower in the group receiving immunotherapy than in the group not receiving immunotherapy, the rye antigen-adherent subpopulation responded significantly less to rye antigen in the group receiving immunotherapy than in the group not receiving immunotherapy. More importantly, the mean rye antigenstimulated proliferation of rye antigen-adherent subpopulation was significantly lower than that of the rye antigen-nonadherent subpopulation only in the group receiving immunotherapy but not in the untreated group. In order to demonstrate the antigen specificity of this observation, T cells were also fractionated into the HSA-adherent and nonadherent fractions. There was no significant difference in the rye antigen-stimulated proliferative response between the HSA-adherent and nonadherent fractions both in the treated and untreated groups. In an attempt to demonstrate that the rye antigen-adherent subpopulation can respond to an indifferent antigen equally well as the rye antigen-nonadherent subpopulation can, the antigenadherent subpopulation was also cultured with Candidu antigen as an indifferent antigen. The Candida antigen-stimulated proliferation of rye antigen-adherent subpopulation was not different from that of rye antigen-nonadherent subpopulation. Therefore, the suppression of rye antigen-stimulated proliferation of rye antigen-adherent subpopulation in the patients receiving immunotherapy was antigen specific. In order to demonstrate that the rye antigen-adherent cell subpopulation from patients receiving immunotherapy has an antigen-specific suppressor cell activity, we performed coculture experiments. The rye antigen-stimulated lymphocyte proliferation of rye antigen-nonadherent cells was significantly suppressed by coculturing with rye antigen-adherent cells only in patients receiving immunotherapy but not in patients receiving no immunotherapy. There were no signifi-
cant differences in the rye antigen-stimulated proiiferative lymphocyte response between the Can&&r antigen-adherent cells and the Cundida antigen-nonadherent cells or their cocultures. In an attempt to characterize rye antigen-adherent and rye antigen-nonadherent subpopulations from patients receiving immunotherapy in terms of cell surface markers and cellular functions, coculture experiments were performed in the presence of monoclonal antibody against inducer-helper T cells 03KT4) or suppressor-cytotoxic T cells (OKT8) and GPSC. When OKT8 and GPSC were added to cocultures containing both rye antigen-adherent and rye antigennonadherent subpopulations, the rye antigen-stimulated proliferative response of the cocultures was significantly increased compared to the response of cocultures to which GPSC alone was added. This finding suggested that OKT8 plus GPSC abolished the suppressor cell activity of rye antigen-adherent cell subpopulation. Furthermore, when OKT4 and CPSC were added to cultures containing rye antigen-nonadherent cells alone, the rye antigen-stimulated proliferative response of rye antigen-nonadherent cells was abolished. In contrast, OKT8 and GPSC had no effect on the rye antigen-stimulated proliferative response of rye antigen-nonadherent cell subpopulation. These findings strongly suggest that the rye antigenadherent cell subpopulation contains cells capable of suppressing the rye antigen-stimulated proliferative response, and the treatment of the cells with OKT8 and GPSC can abrogate the suppressor function. It is most likely that the cells with the suppressor activity belong to the T cell subset identifiable hy OKT8. Antigen-stimulated lymphocyte proliferative responses have been reported to decrease in patients receiving immunotherapy.‘, “‘. ‘I’ ” immunotherapy has been demonstrated to generate antigen-specific suppressor T cells belonging to the subpopulation of
J. ALLERGY
394 Nagaya
lymphocytes bearing histamine receptors. I9 Although the histamine-activated suppressor cell system was markedly diminished in the atopic population, the concanavalin A-activated suppressor cell system was functionally equivalent in the atopic subjects and the normal control subjects.” The results of the present study demonstrated that rye antigen-adherent T cells obtained from patients with rye grass hay fever receiving immunotherapy can suppress the rye antigenstimulated proliferative lymphocyte response of rye antigen-nonadherent subpopulation. If we extrapolate our in vitro observations into the in vivo situation in patients receiving immunotherapy, the salutary effect of immunotherapy may depend on the activation of suppressor cell function of antigen-adherent T cells by the specific antigen injection. I am grateful to the entire staff of the Allergy-Immunology Section for their helpful contribution to this study. REFERENCES 1. Ishizaka K: Cellular events in the IgE antibody responses. Adv Immunol 23: 1, 1976 2. Lichtenstein LM, Norman PS, Winkenwerder WL, Osler AG: In vitro studies of human ragweed allergy: changes in cellular and humoral activity associated with specific desensitization. .I Clin Invest 45:1126, 1966 3. Pruzansky JJ, Patterson R: Histamine release from leukocytes of hypersensitive individuals. II. Reduced sensitivity of leukocytes after injection therapy. J ALLERGY 3944, 1967 4. Lichtenstein LM, Norman PS, Winkenwerder WL: A single year of immunotherapy for ragweed hay fever, immunologic and clinical studies. Ann Intern Med 75:553, 1971 5. Levy DA, Lichtenstein LM, Goldstein EO, Ishizaka K: Immunologic and cellular changes accompanying the therapy of pollen allergy. J Clin Invest 50:360, 1971 6. Lichtenstein LM, Ishizaka K, Norman PS, Sobotka AK, Hill BM: IgE antibody measurements in ragweed hay fever, relationship to clinical severity, and the results of immunotherapy. J Clin Invest 52:472, 1973 7. Anderson JA, Lane SR, Howard WA, Leiken S, Oppenheim
CLIN. IMMUNOL.
MARCH 1985
8.
9.
10. 11.
12.
JJ: The effect of hyposensitization on Alzernnria-induced lymphocyte blastogenesis. Cell Immunol 10442, 1974 Gaiten JG, Merler E, Colten HR: Allergy to ragweed antigen E: effect of specific immunotherapy on the reactivity of human T lymphocytes in vitro. Clin Immunol Immunopathol 4:32, 1975 Romagnani S, Biliotti G, Ricci M: Depression of grass polleninduced lymphocyte transformation by serum from hyposensitized patients. Clin Exp Immunol 19:83, 1975 Evans R, Pence H, Kaplan H, Rocklin RE: The effect of immunotherapy on humoral and cellular responses in ragweed hay fever. J Clin Invest 57:1378, 1976 Nagaya H, Lee SK, Reddy PM, Pascual H, Jerome D, Sadai J, Gupta S, Lauridsen J: Lymphocyte response to grass pollen antigens: a correlation with radioallergosorbent test and effect of immunotherapy. Ann Allergy 39:246, 1977 Norman PS, Lictenstein LM: The clinical and immunologic specificity of immunotherapy. J ALLERGY CLM IMMUNOL 61:370,
1978
13. Nagaya H: Long-term effects of conventional immunotherapy in southern California. Ann Allergy 44:193, 1980 14. Mage MG, McHugh LL, Rothstein TL: Mouse lymphocytes with and without surface immunoglobulin: preparative scale separation in polystyrene tissue culture dishes coated with specifically purified anti-immunoglobulin. J Immunol Methods 15:47, 1977 15. Taniguchi M, Miller JFAP: Enrichment of specific suppressor T cells and characterization of their surface markers. J Exp Med 146:1450, 1977 16. Nagaya H, Sieker HO: Lymphopenic effect of antilymphocyte serum. Proc Sot Exp Biol Med 121:722, 1966 17, Walter H, Nagaya H: Separation of human rosette- and nonrosette-forming lymphoid cells by countercurrent distribution in an aqueous two-phase system. Cell Immunol 19:158, 1975 18. Walter H, Moncla BJ, Weber TJ, Nagaya H: Membrane chargeassociated heterogeneity of B lymphocytes from human peripheral blood as reflected by cell partition in two-polymer aqueous phases. Exp Cell Res 122:380, 1979 19, Rocklin RE, Sheffer AL, Greineder DK, Melmon KL: Generation of antigen-specific suppressor cells during allergy desensitization. N Engl J Med 302: 1213. 1980 20. Beer DJ, Osband ME, McCaffrey RP, Soter NA, Rocklin RE: Abnormal histamine-induced suppressor-cell function in atopic subjects. N Engl J Med 306:454, 1982