0022-202X/83/8005-0398$02.00/0 TnE JouRNAL ·oF INV ESTIGATIVE DERMATOLOGY, 80:398-402, 1983 Copyright © 1983 by The Williams & Wi.lk ins Co.

Vol. 80, No.5 Prin.led in U.S.A.

Antigen-Specific T-Cell Lines in DNCB-Contact Sensitivity in Guinea Pigs

LADISLAV PoLAK, M.D. AND RIK J. ScHEPER, PH.D.

F. Hoffmann-La Roche & Co. Ltd. (LP), Central Research Units, Basle, Switzerland, and Pathological Institute (RJS), Free University Hospital, Amsterdam., The Netherlands

Expanded DNP-specific guinea pig T cells exhibit en­hanced in vitro and in vivo activity as demonstrated by DNA synthesis and systemic adoptive transfer of contact sensitivity. Moreover, expanded lymphocytes are capa­ble of accomplishing local passive transfer of contact sensitivity and producing interleukin 2, vascular perme­ability-increasing and macrophage migration-inhibiting factors. Expansion of hapten-specific lymphocytes offers a suitable model for studying eliciting mechanism of allergic contact dermatitis in humans.

During the 50-year history of research in experimental con­tact dermatitis several basic problems concerning its mecha­nism have been satisfactorily solved. The crucial role of T lymphocytes has been well established. The afferent and effer­ent pathways ofT-cell function as well as their proliferation and differentiation in the draining lymph nodes constitute a major part of the delayed-type hypersensitivity (DTH) char­acteristic of experimental contact dermatitis. Finally the in vivo and in vitro activity of effector T cells and their regu lation have been sufficiently documented [1]. However, a deeper insight into the eliciting mechanism of contact sensitivity requires preparation of lymphocyte populations highly enriched in T cells specific for the hapten used for immunization and contain­ing few, if any, lymphocytes of other specificities.

The possibility of obtaining such a lymphocyte population was described by Ben-Sasson et al [2]. They selected T lym­phocytes from hypersensitive guinea pigs on monolayers of antigen-pulsed stimulator cells. Fw-thermore, it has been shown that T lymphocytes in long-term bulk cultures survive and retain their specificity by repeated restimulations with the appropriate antigen . The increased reactivity toward the spe­cific antigen, thus achieved, was accompanied by a lowered or absent reactivity to other antigens to which the animals were simultaneously sensitized but which were not used for restim­ulations. This effect may be due to enrichment of specific effector cells as well as to their increased and accelerated activity. The technique of clonal expansion, sometimes pre­ceded by clonal selection, has been demonstrated for several proteins and hapten-protein conjugates as well as for histocom­patibility antigens [3].

In the present paper the clonal expansion technique was used for selecting lymphocyte populations highly enriched in hapten-

Manuscript received July 13, 1982; accepted for publication October 26, 1982.

Reprint requests to: Dr. L. Polak, F . Hoffmann-La Roche & Co Ltd., Central Research Units (68/252), CH-4002, Basle, Switzerland.

Abbreviations: BSS: balanced salt solution DNBSOa: 2,4-dinitrobenzene sulfonic acid sodium salt DNCB: 1-chloro-2,4-dinitrobenzene DNFB: 1-fluoro-2,4-dinitrobenzene DNP-M$: stimulator macrophages DTH: delayed type hypersensitivity FCA: Freund's complete adjuvant MIF: migration inhibitory factor N-M$: normal macrophages PEC: peritoneal exudate cells PHA: phytohemagglutinin

specific T cells. The functions of selected cell populations in the eliciting phase of contact sensitivity in guinea pigs is described.

MATERIALS AND METHODS

Animals

Inbred strain 2 guinea pigs of either sex weighing 350-400 g were used. The animals were bred at the Institute for Biomedical Research Fiillinsdorf, Switzerland. T hey were fed on pellet diet supplemented ad libitum with water containing vitamin C.

Immunization and Challenge

Animals were sensitized by injecting 500 Jig 1-fluoro-2,4-dinitroben­zene (DNFB; F luka AG, Buchs SG, Switzerland) in 0.5 ml Freund's complete adjuvant (FCA; DIFCO Lab., Detroit, Michigan) , divided into 5 doses, into the foot pads and the nape of the neck. Epicutaneous challenge was performed with 0.09%, 0.05%, and 0.03% solu tions of 1-ch loro-2,4-d.initrobenzene (DNCB; Merck, Darmstadt, FHG) in ace­tone. The individual reactions were evaluated 24 h later according to an arbitrary scale (pink, spots = 0.5; red confluent = 1; swollen = 2) and the degree of contact sensitivity was expressed as a total of all 3 individual reactions in each animal.

Cell Suspensions

Lymphocyte suspensions were prepared from draining lymph nodes 14 days after sensitization. The cells were teased from the nodes a nd fLltered through nylon stocking mesh and through a 5-ml column packed with 1 m.l nylon wool (Leuko-Pack, Fenwal Lab., Deerfield, Illinois). ·

Stimulator macrophages (DNP-M$) were peritoneal exudate cells (PEC) harvested 4 days after an i.p. injection of 20 ml light paraffin oil (Drake Oil 6-VR White, Penna, Butler, Pennsylvania). They were irradiated with 3300 r and incubated in a 10 Jlg/ml DNFB solu tion in Hanks' balanced salt solution BSS (GIBCO Europe, Glasgow, Scotland) (pH 8.5, 30 min, 37°C, 1-2 x 107 M$/ml).

Long-term Culture

Lymph node lymphocytes (2 x 10n;n1l) were cultw·ed in RPMI 1640 (GIBCO) supplemented with 50 Jlg/ ml gentamycin (Essex Chemie AG, Lucerne, Switzerland), 100 Jig/ ml streptomycin and 100 IU/ml penicillin (GIBCO) , 10-G M 2-mercaptoethanol (Sigma Chemical and Co., St. Louis, Missouri), 2 mM glu tamin (GIBCO), 1 mM sodium pyruvate (GIBCO), 1 mM non-essential amino acids (GIBCO) a nd 5% heat­inactivated fetal calf serum (GIBCO) at 37°C in humified atmosphere of 5% CO" and 95% air. Twice a week about 50% of the medium was replaced by fresh supplemented RPM! containing 10% fetal calf serum. DNP-M$ (1 X 10°/ml) were added at the beginning of the cultw-e and then once a week.

Lymphocyte Proliferation Assay

Lymphocytes (1 X 10° /ml) from long-term cultlll'es were incubated for 96 h with 3 X 104 DNP-M<p in 75-mm round-bottom plastic tubes (Falcon, Oxnard, California) and pulsed for the last 18 h with 1 JiCi/ml of [3H]-thymidir1e (TRA 120 sp act 5 Ci/mM, Radiochemical Center, Arnersham, England). When primed nonexpanded lymphocytes were used, the concentration was 5 X 10° /ml and the incubation period 24 h longer. The DNA synthesis was determined on glass fiber ftlters (What­man Inc., Clifton, New York) by counting in a liquid scintillation spectrometer. To control cu ltures 3 X 10'' normal macrophages (N-M$) were added. On some occasions tuberculin PPD (100 Jig/ml, Statens Serum Institute, Copenhagen, Denmark) or PHA (purified phytohe­magglutinin, 1 Jlg/ml, Wellcome Research Laboratories, Beckenham, England) were added together with the normal macrophages.

Guinea pig interleukin 2 was prepared from normal guinea pig spleen cells (1 X 107/ml) incubated with 100 Jig concanavalin A (Con-A,

398

May 1983

Calbiochem, Behring Corp., La Jolla, California) for 48 h after a 24-h preincubation. lnterleukin 2 activity was assayed by the thymocyte test [ 4]. Guinea pig thymocytes (1 X 10,;/ ml) were cul tured with 1 f!g/ ml phytohemagglutinin (PHA) in the presence of 10% lymphokine-rich supernatant or control supernatant and the results expressed as cpm increment ± SE. No fmther purification of in te rleukin 2 was under­taken.

Transfer of Contact Sensitivity

Systemic adoptive transfer was performed by an i.v. injection of various numbers of primed or expanded lymphocytes.

Local passive transfer was performed either by an in tradermal injec­tion of 2.0 X 10" lymphocytes together with 2.0 X 10" DNP-M<P or N­M<P in 0.1 ml Hanks' BSS into the shaved skin of the flank or by an intradermal injection of I X 107 lymphocytes followed by an i.v. injec­tion of 500 mg/ kg 2,4-dinitrobenzene su lfonic ac id sodium salt (DNBSO", Eastman Kodak, Rochester, New York). The results are expressed as mean diameter of erythema in mm and increment of induration in 0.1 mm.

Lymphohines

Supernatants of 24-h cultures of primed or expanded lymphocytes (2.5 x 107

) with DNP-M<P (4 X 10") were examined for interleukin 2, vasculru· permeability-increasing, and macrophage migration-inhibiting factor activities. Interleukin 2 activity was assayed by the thymocyte test and the results expressed as cpm increment ± SE.

Vascular permeability- increasing factor activity was assessed accord­ing to Maillru·d et al [5]. Briefly, 0.1 ml of a 10 x concentrated lymphokine-containing or control supernatants were injected intrader­mally into the shaved skin of the flank followed by an i.v. injection of 0.5 ml of a 0.5% solution of Evans blue in 0.15 M saline. The diameter of the blue reaction was measured 4 h later.

Macrophage migration-inhibiting activity was measw·ed by the in­direct assay [1]. Normal PEC hru·vested as described for macrophages were packed into capillaries (microhematocrit tubes, Clay Adams, Parsippany, New York) and fixed in cultu1·e chambers (migration plate, Sterilin, Middlesex, England) with the help of a drop of silicon grease. The chambers were filled with 0.5 ml supplemented RPMI 1640 me­dium containing 15% isologous heat-inactivated serum and 10% of 10-fold concentrated lymphokine-rich or control superna tants. The cham­bers were covered with coverslips and incubated for 12-14 h at 37°C. The reactions were expressed as percent inhibi tion according to the formula:

migration ru·ea in lymphokine-rich supernatant 100- . . . I m1grat10n area 111 contro supernatant

Characterization of Lymphocyte Preparations

S urface lg-positive (8-Ig+) lymphocytes (B cells) were identified by rosette formation using rabbit anti-guinea pig lg-conjugated sheep red blood cells prepru·ed according to Gronowicz et al [6]. Rosette tests with rabbit red blood cells for the detection ofT cells were performed according to Wilson and Gtu·ner [7].

Nonspecific esterase-positive cells (mononuclear phagocytes) were determined cytochemically using a-naphthyl butyrate as a substrate [8].

Surface !a-positive cells were identified by immunofluorescence using an anti-guinea pig-la-monoclonal antibody prepru·ation 22C4, kindly provided by Dr. Shevach and fluorescein isothiocyanate- labeled rabbit antimouse immunoglobulin antiserum (Nordic Tilburg, The Nether­lands).

RESULTS

Surface Marhers of DNP-M~ Expanded Lymphocytes

In an attempt to characterize the cell population being se­lected by antigen priming and in vitro restimulation, we exam­ined the ability of these cells to form spontaneous E-rosettes or to manifest surface lg or Ia. From the data presented in Table I, it is evident that lymph node lymphocyte suspensions taken from in vivo primed animals contained a bout twice as many spontaneous E-rosette-forming cells (T cells) as B cells (lg+ cells). After repeated stimulations in vitro an almost complete disappearance of B cells was observed and the lymphocyte suspensions consisted almost exclusively of T cells. On some occasions the cell suspensions contained a small percentage of

DNP-SPECIFIC T-CELL LINES 399

macro phages, as judged by morphology and monocyte-esterase staining. However, these macrophages formed neither anti-Ig­conjugated sheep red blood cell rosettes nor spontaneous rabbit red blood cell (E) rosettes a nd were not included in t he Table. Furthermore, it has been shown by anti-Ia-monoclonal anti­body staining, that the number of ra+ lymphocytes increased after repeated in vitro stimulations from 43% to almost 100%.

Enhanced in Vitro DNA Synthesis of Expanded Lymphocytes

Lymph node lymphocytes primed in vivo with DNFB in FCA exhibited in vitro a significant although only moderate prolif­erative response to the antigenic preparation (DNP-M~) (cpm increment ± SE = 4900 ± 850). The response to tuberculin present in the adjuvant used for in vivo priming was signifi­cantly higher (13,983 ± 2,616) but about 10-fold lower than the response to the polyclonal T-cell mi togen PHA (142,788 ± 30,886).

However, when these lymphocytes were expanded in vitro by repeated stimulations with DNP-M~ (Fig 1) the specific prolif­erative response to this antigen was considerably enhanced (101,210 ± 24,047) whereas the increment in incorporation of the ra diola beled DNA precursor, [3H ]-thymidine, upon stimu­lation with tuberculin (PPD) fell to background values (1156 ± 67) . On the other hand, t h e response to the mitogen PHA remained unchanged, considering that lower numbers of ex­panded lymphocytes (1 X 105

) t han of nonexpanded ones (5 X 105

) were used. Furthermore, T lymphocytes kept in cul ture for more than 6

months under continuous restimulations with DNP-M~ re­tained their capacity of responding in vitro to this specific antigen (data not shown) . It should be mentioned that th ese cultures have been kept without adding interleukin 2.

Moreover, the capability to respond to DNP could be m ain­tained for 1 month, when the specific in vitro stimulations with DNP-M~ were replaced by adding interleukin 2 or PHA to the cultures. At the end of this period of nonspecific stimulation the cells responded to DNP-M~ at the same level as lympho­cytes specifically stimulated for 1 month with DNP-M~ alone.

Adoptive Transfer of Contact Sensitivity with in. Vitro Expanded DNP-Specific T Cells

In order to achieve an adoptive transfer of DNCB-contact sensitivity in guinea pigs with in vivo primed lymph node lymphocytes an i.v. injection of about 4 X 108 viable cells is r equired (Fig 2). In contrast, the same degree of contact sensi­tivity was achieved with fewer than 1 X 107 expanded lympho­cytes. These results are considered as evidence for a high enrichment of DNP-specific T cells in the suspensions used for transfer. However, only a small proportion of these cells ap­peared to be systemically active effector cells of contact sensi­tivity. When epicutaneous challenge with DNCB is performed immediately after transfer, the elicited reaction is either of low

TABLE I. Marhers of hapten (DNP)-specific expanded lymphocytes

Thymocytes Lymph node cells

Expanded lymphocytes restimulated

2-3X 7-SX ---------------------------------------

% s-Ig+ % E-RFC % Ia+

1% 93 ± 5%

6%

26 ± 3% 62 ± 7%

43%

6±2% 0% 90% 90% 67% 100%

Lymph node lymphocytes from DNFB/ FCA-sensitized guinea pigs were cultured for several weeks in the presence of DNP-M0 (see Materials and Methods). Surface Ig was assessed by rosette formation using rabbit anti-guinea pig Ig-conjugated sheep red blood cells, rosette test forT-cell detection (E-RFC) was performed with rabbit red blood cells, and Ia+ lymphocytes were detected by anti-guinea pig la-mono­clonal antibody 22c4. The numbers represent results of 2 experiments, each consisting of a pool of lymphocytes fTom 4 guinea pigs, cultured 2 or 3 weeks and 7 or 8 weeks.

400 POLAK AND SCHEPER

Resoonelers Challenge In vitro

1Kl05

tn vttro exoandedLY

FIG 1. Increased in vitro DNA synthesis of in vivo primed or in vitro expanded lymph node lymphocytes from DNFB/FCA-sensitized guinea pigs. Lymphocytes were cultW'ed in the presence of DNP-M~ for 4 weeks. Results were expressed as cpm increment. The background cpm values of in vivo primed lymphocytes were 836 ± 79 and of expanded cells 4 965 ± 1062.

110 of

}~~~~- In vit ro expanded ce ll s

1 X 106 ffili3--< 3x106~

J X 107

:#:¥ffi:::f:%:f:f~:·::::::::::~;;;;~ .. ----. 2 X 107 B@i' . ~~

Ll x 107 ffl@}@.$JMWt&¥n@J.@@iP\@J..A .. ·. ~ .-.·· ···· J

I. x 108 N.O .

2 X 108 II .D.

; X 108 N.D.

N.D

N.D.

r1.n.

Degree 0 2 3 5 0

\!l VI VO or lmed

2 g~ntact '------'------'----'----l.----' '-----'------' sens i ti vitY

FIG 2. Degree of DNCB-contact sensitivity induced by transfer of different numbers of in vivo primed or in vitro expanded lymph node lymphocytes from DNFB/ FCA-sensitized guinea pigs. L~mphocytes were cultW'ed for 4 weeks in the presence of DNP-M~ and mJected l.V .

into syngeneic guinea pigs. The recipients were challenged eptcuta­neously 14 days after transfer and the reactions evaluated as stated m Materials and Methods. N.D.= not done.

intensity or not observed at all. However, when recipients injected witn lymphocytes of the same suspension were chal­lenged epicutaneously 4 or 14 days later, a considerable increase in the intensity of the resulting inflammatory reaction could be observed.

Furthermore, DNCB-contact sensitivity was transferred also with cells cultured in the presence of the antigen (DNP-M<jl) for about 6 months. The degree of contact sensitivity observed in guinea pigs adoptively sensitized with cells from these long­term cultures did not differ substantially from the degTee in­duced with cells of the same cell line but injected after a shorter period of culture (4 weeks) (data not shown) . These data correspond to the results obtained in vitro. Moreover, lympho­cytes expanded initially in the presence of DNP-M<jl but then cultured for another 4 weeks under nonspecific stimulations with either a mitogen (PHA) or interleukin 2 were still capable of transferring the same degree of DNCB-contact sensitivity as lymphocytes cultured in the presence of the specific antigen for the entire time (Fig 3).

The presence of high numbers of locally active effector cells in lymphocyte suspension was further shown by local passive transfer of delayed hypersensitivity to DNP compounds. Ex­panded lymphocytes (2.0 x 106), injected intradermally with an equal number of DNP-M<jl, ·elicited in normal guinea pigs an erythema and induration significantly exceeding the reaction induced by expanded cells injected with normal syngeneic mac­rophages (Fig 4). This tuberculin-like reaction could not be produced by lymphocytes from normal animals nor even by in vivo primed but not in vitro expanded cells injected together with the antigen. Moreover, similar delayed-type skin reactions were elicited in animals injected intradermally with expanded (but not with normal or in vivo primed) lymphocytes when the hapten (DNBSO~) was injected i.v. within 24 h after the intra­dermal injection (Fig 5) .

Vol. 80, No.5

Production of Lymphokines by DNP-Specific Expanded T­Cell Lines

Further evidence for the enrichment of DNP-speci.fic T cells in expanded cultures of primed lymphocytes is their capability to produce lymphokines (Table II). Supernatants from antigen­expanded lymphocyte cultures were assayed for interleukin 2, vascular permeability-increasing factor, and macrophage mig:l'a­tion inhibitory factor (MIF) activities. From the results pre­sented in Table II it is evident that expanded lymphocytes are capable of producing all 3 lymphokines investigated. However, in supernatants of nonexpanded lymphocytes only MIF activity was observed.

DISCUSSION

It is generally accepted that contact sensitivity is a form of DTH. The guinea pig model is particularly relevant for research in allergic contact dermatitis in humans [1).

In this study T -cell lines specific for DNCB contact sensitivity were selected and expanded in vitro. Repeated stimulation of in vivo primed unselected lymphocytes with hapten-modified macrophages (DNP-M<jl) yielded a selected T-cell population characterized by increased inCorporation of [3H]-thymidine in response to subsequent exposure to antigen (DNP-M<jl). The level of specific antigen-dependent in vitro DNA synthesis reached values equal to those observed after mitogenic stimu­lation with PH A. The disappearance of the in vitro response to

Stimulations in vitro

0 Degree of contact sensitivity± SE

1 2 3 4 5

DNP-Mcp l@tMi:::&&~Itt®l1NlM~WttlfM:H!M!f~J!;@t!t~~MMNllf--; 4.2 ± 0.2

3.2 ± 0.3

4.5 ± 0.3

IL 2

PHA

FIG 3. Maintenance of specific in vivo response of expanded lym­phocytes cultW'ed in the presence of interleukin 2 (IL 2) or mitogen (PHA ). Lymph node lymphocytes from DNFB/ FCA-sensitized guinea pigs expanded 4 weeks in the presence of DNP-M~ were cultured for another 4 weeks in the presence of IL2 or PHA before injection into syngeneic recipients. The recipients were challenged epicutaneously 14 days after transfer and the reaction evaluated 24 h later according to Materials and Methods.

Lymphocytes Antigen

Expanded DNP-Mcp

Expanded N-Mcp

Normal DNP-Mcp

Sensitized DNP-Mcp

Sensitized N-Mcp

Expanded DNP-Mcp

Expanded N-Mcp

Normal DNP-Mcp

Sensitized DNP-Mcp

Sensitized N-Mcp

0 I

2 I

Erythema (l'l mm)

6 8 I I

10 I

12 I

I n +< nt mr; ;;r ;;;; ;:m;;m;wmm uwM-< 122 ± o.4

nmummm:wm;w;;mmt8---; 8.o ± 12

Induration (increment in 0.1 mm)

0 2 4 6 8 10 12

7.5± 0.7

8.0± 1.4

7.0± 0.1

I I I I I I I I I

lmw:s:J:.1?::::;:m:m:w~r(iui:ti~~"li:'ii::ii:: i; ii·:·:··i;:·ii·::~·l "::":::::::~~ 9. 7 ± 3.9

@mm:e:;;:;J 3.6 ± 1.8

1.5 ± 0.1

2.5 ± 1.4

3.0 ± 0.7

FIG 4. Local passive transfer of DNCB-contact sensitivity with ex­panded lymphocytes. Expanded lymphocytes (2 X 106

) were injected intradermally into syngeneic recipients together with 2 X 106 haptenized (DNP-M~) or control (N-M~) macrophages. In vivo primed (sensitized) or normal lymph node lymphocytes served as controls. The reactions were evaluated 24 h later according to the size of erythema(~ mm) and degree of induration (increments in 0.1 mm).

May1983

Erythema (l'l mm)

0 2 4 6 8 10 12 Treatment

DNBS03 i.v.

L.¥mphocytes I I I I I I I I I I I I I

Expanded 1Wki.'Thk't¥t~"6'*&'%0,'%fu,'''k''4~ 11 .7 ± 0.8

7.5 ± 0.7 DNBS03 i.v.

DNBS03 i.v.

Sensitized

Nanna!

Expanded

DNBS03 i.v. Expanded

DNBS03 i.v. Sensitized

DNBS03 i.v. Normal

Expanded

IB'tillU%1l.Dk%&\%-.-,

mm&L'"thmt4•~ ltm;Mi&1j\h1W&\U$1-o

Induration (increment in 0.1 mm)

0 2 4 6 8 10 12

5.7 ± 0.8

7.5 ± 0.7

~~ii~'~ii~~~' ii' iiii'ii'~~~~~~~ 1&1JW1i®&'tkt"illih-a"\'tCmtX1Jitl I 9. 7 ± 3. 1

• IWiEM I .... 1.0 ± 0.2

1.6 ± 1.3

1.0 ± 0.4

FIG 5. Local passive transfer of DNCB-contact sensit ivity with ex­panded lymphocytes. Same protocol as in Fig. 4 but the hapten (DNBSOa) was injected i.v. within 1 h after the intradermal injection of lymphocytes instead of addi tion of haptenized or control macro­phages. The results are expressed as in Fig 4.

TABLE II. Lymplwhin.e activities of supernatants of in vivo primed or in vitro expanded lymphocytes from DNFB! FCA -sen.sitized

g uinea pigs

Vascular Macrophage t d f 1 permeabiJ- migration

Lyrn)hk~y es usl t • ym- In terleukin 2 (cmp ± SE) ity-increas- inhi bition P 10 me proc uc 10 11 ing factor (%facto r (% in-

increase) hibi tion)

In vivo primed only 1,157 ± 211 10 43 In vitro stimulated

l-2X 5,455 ± 97 60 63 4X 40,998 ± 559 54 45 16X 38,459 ± 2,703 45 41

Thymocytes 212 Thymocytes + PHA 1,316

an unrelated antigen (tuberculin) used together with DNFB during priming is considered as evidence for in vitro selection and expansion of DNP-specific T cells. These expanded cells maintained specific antigen responsiveness, which could be demonstrated both in vitro and in vivo over an extended period of time. Moreover, since addition of interleukin 2 was not required, we believe that sufficient amounts of this lymphokine is pl'Oduced by the cultW'ed cells themselves.

The crucial proof that the selected lymphocytes are t ruly cells involved in contact sensitivity was established in vivo by transfer experiments. Systemic transfer of contact sensitivity with unselected lymphoid cells from in vivo primed allogeneic [9] or syngeneic (10) donors was considered as the final evidence for the existence of a non-antibody-mediated hypersensitivity reaction. For this type of transfer high numbers of lymph node lymphocytes were required, although syngeneic inbred donors and recipients were used and challenges performed 14 days after the cell injection (adoptive transfer). If expanded lympho­cytes were used the same degree of contact sensitivity was achieved with transfer of 50- to 100-fold fewer lymphocytes. However, this relatively small number of lymphocytes induced contact sensitivity in recipients only after a certain resting period. Guinea pigs challenged shortly after transfer of ex­panded lymphocytes exhibited only a moderate or even nega­tive response to the epicuta neously applied hapten. There exist at least two ways to explain this phenomenon. It is possible that by in vitro cul turing, the lymphocyte suspensions were emiched mainly in specific memory cells (blasts) capa ble of proliferating and differentiating but not of mediating directly an inflamma­t ory reaction. For this pmpose transformation into lymphokine­releasing effector cells is necessary which occurs in syngeneic recipients. Effector cells, on the other hand, are terminal cells

DNP-SPECIFIC T-CELL LINES 401

incapable of fmther proliferation and multiplication (1]. Their frequency in primed or expanded lymphocyte suspensions may be too low to manifest an immediate effect in passively sensi­tized guinea pigs. Another explanation may be the impaired migratory capacity of expanded cells shortly after transfer (Scheper and Polak, unpublished observation [ll]) . The fact that local passive transfer and lymphokine production, which did not require intact cell migration, were not altered, supports this second explanation.

. It is noteworthy that in our experiments the cells expanded initially in the presence of specific hapten-modified stimulator cells without the use of conditioned media, and maintained their specificity even when t he antigen was omitted and re­placed by the addition of guinea pig interleukin 2 or the mitogen PHA to the culture media.

The method of systemic adoptive transfer of contact sensit iv­ity was later supplemen ted by the technique of local passive transfer (12,13). By this technique lymphocytes from sensit ized guinea pigs were injected intradermally, either together with the antigen or the antigen was injected i.v. shortly after the intradermal injection of sensitized cells. However, this transfer was accompHshed only with PEC but not with lymph node lymphocytes from guinea pigs sensitized to protein antigens. Local passive transfer was never achieved with cells from ani­mals exhibiting contact sensitivity to haptens. This could be explained by a low frequency of mature effector cells in lym­phocyte suspensions from contact-sensit ive guinea pigs, provid­ing that amounts of lymphokine per antigen-specific cell is the same.

The successful local passive transfer of tuberculin type DTH with expanded but not with in vivo primed lymph node lym­phocytes in om experiments is considered as evidence of con­siderable enrichment of hapten-specific T cells in these suspen­sions. Furthermore, it indicates that the frequency of mature effector cells in suspensions of expanded lymphocytes may be sufficient to induce contact sensitivity upon challenge inlmedi­ately after adoptive transfer. The fail m e of immediate systemic transfer may be caused by temporary impaired migration ca­pacity of expanded cells [11).

The postulated effector mechanism, nam ely release of inflam­matory lymphokines, which has been sufficiently documented for tuberculin type reactions (DTH) to protein antigens was never convincingly demonstrated for contact sensitivity to hap­tens [14). This fa ilure may also be due to a relatively low frequency of antigen-specific T-effector cell in lymphocyte suspensions from hapten (in contrast to protein)-sensit ive ani­mals. The other possibility, namely, that the effector mecha­nism of contact sensit ivity might be different from the mecha­nism of classic delayed hypersensitivity as indicated by the finding that contact sensitivity reactions differ from the classical DTH by the presence of large (about 20%) numbers of baso­philic granulocytes (15), is difficult to confirm since the role of basophils and their products (histamine, etc.) in con tact sensi­t ivity skin lesions remains controversial [16].

OW' experiments with expanded hapten-specific cell lines seem to support the view of the essential identity of effector mechanisms of DTH and contact sensit ivity. This is evidenced by the release of several lymphokines into culture supematants upon specific in vitro stimulation. Thus, interleukin 2, vascular permeability-increasing a nd macrophage migration-inhibiting factors were demonstrated in supernatants of expanded lym­phocytes cultmed for 24 h with specific hapten-modified mac­rophages (DNP-M$) . Moreover, the lymphocyte density re­quired for lymphokine production was 5- to 10-fold lower than the density required for induction of these lymphokines in protein antigen-primed nonexpanded cells. It should be stre sed that nonexpanded primed cells from contact-sensitive guinea pigs did not produce interleukin 2 and vascular permeabili ty­increasing factor upon a ntigenic stimulation. MIF production was, however, observed.

The results presented in this study are consistent with the

402 LAN~ ET AL

hypothesis that contact sensitivity in guinea pigs is mediated by T cells and their products, inflammatory lymphokines. These mediators could be induced in detectable amounts from an expanded population of hapten-specific T cells.

Thus, our experiments suggest similarity if not identity in the eliciting mechanisms of contact sensitivity and tuberculin-like DTH to protein antigens. This view is further supported by the fact that antilymphokine antibodies are capable of inhibiting both contact sensitivity reactions to hap tens and DTH reactions to proteins [17).

Clones of hapten-specific T cells should be useful for a better understanding of the eliciting mechanisms of contact sensitivity.

The excellent technical assistance by Mrs. V. Weiss and Miss A. Stierlin, and the skillful secretarial help of Miss M. Jenni and Miss E. Moll is gratefully acknowledged.

REFERENCES 1. Polak L: Immunological aspects of contact sensitivity. Edited by P

Dukor, P Kall6s, Z Trnka, BH Waksman, AL de Week. Basel, S Karger, 1980

2. Ben-Sasson SZ, Paul WE, Shevach EM, Green 1: In vitro selection and extended cu lture of antigen-specific T lymphocytes. I. De­scription of selection culture procedure and ini tial characteriza­tion of selected cells. J Exp Med 142:90-105, 1975

3. Ben-Sasson SZ, Paul WE: Selection and enrichment of antigen­specific T lymphocyte populations. Isr J Med Sci 12:414-424, 1976

4. Paetkan V, Mills G, Gerhar t S, Monticone V: Proliferation of murine thymic lymphocytes in vitro is mediated by the concan­avalin A-induced re lease of a lymphokine (costimu lator). J Im­munol 117:1320-1324, 1976

5. Maillard JL, Pick E, Turk JL: Interaction between "sensitized lymphocytes" and antigen in vitro. V. Vascula1· permeability induced by skin-reactive factor. Int Arch Allergy 42:50-68, 1972

0022-202X/83/8005-0402$02.00/0 THE JOURNAL OF I NVESTIGATIV E DERMATOLOGY, 80:402-405, 1983 Copyright © 1983 by The Williams & Wilkins Co.

Vol. 80, No. 5

6. Gronowicz E, Coutinho A, Melchers F: A plaque assay for all cells secretmg lg ot a given type or class. Eur J lmmunol 6:588-590 1976 '

7. Wilson AB, Gumer BW: Increased affinity of guinea pig thymocytes and thymus-derived lymphocytes for papain-treated rabbit erythrocytes as compared to untreated erythrocytes. J Immunol Methods 7:163-168, 1975

8. Mullink H, von Blomberg. M, Wilders ~M •. Drexhage HA, Alons CL: A Simple cytochemical method for distmgUishing EAC ro­settes formed by lymphocytes and monocytes. J Inununol Methods 29:133-137, 1975

9. Landste!n.er K, q hase MW: Experiments of transfer of cutaneous sensitivi ty to sunple compounds. Proc Soc Exp Bioi Med 49:688-690, 1942

10. Chase MW: Cellula1: transfer between homozygous and heterozy­gous gumea pigs with regard to contact-type hypersensitivity and antibody production. Fed Proc 19:214, 1960

11. Dailey MO, Fa~hman CG, Butcher EC, Pillemer E, Weissman I: Abnormal migratiOn of T lymphocyte clones. Immunology 128:2134-2136, 1982

12. Metaxas MN, Metaxas-Buhler M: Passive transfer of local cuta­neous hypersensitivity to tubercu lin. Proc Soc Exp Bioi M ed 69:163-165, 1948

13. Blazkovec AA, Sorkin E, Turk JL: A study of the passive cellular transfer of local cutaneous hypersensitivity. lnt Arch Allergy 27:289-303, 1965

14. von Blombe1:g M,.RijlaarsdaJ? U, Scheper RJ : Direct macrophage migratiOn mhibitiOn test m DNCB contact sensitized guinea pigs Int Arch Allergy 50:593-512, 1976 ·

15. Dvorak HF, Dvorak AM, Simpson BA, Richerson HB, Leskowitz S, Karnovsky MJ: Cutaneous basophil hypersensitivity. II. A light and electron microscopic description. J Exp Med 132:558-582, 1970

16. Polak L, Rufli T: Vaso~ctive media.tors in contact sensitivity, New Trends 111 Allergy. Ed1ted by W Rmg, G Burg. Berlin/Heidelberg/ New York, Springer-Verlag, 1981, pp 187-199

17. Geczy CL •. Geczy AF, de Week .AL: Antibodies to guinea pig lymphokmes. II. SuppressiOn of delayed hypersensitivity reac­t ions by a 'second generation' goat antibody against guinea pig lymphokines. J Immunol 117:66-72, 1976

Vol. 80, No.5 Printed in U.S.A.

Class-Specific Antibodies to Gluten in Dermatitis Herpetiformis

ALFRED T. LANE, M.D., J . CLARK HUFF, M.D., JOHN J. ZONE, M.D., AND WILLIAM L. WESTON, M.D.

Department of Medicine and Pediatrics, University of Rochester School of Medicine (A TL), Rocheste1 ~ N ew York; Department of Dermatology, University of Colorado School of Medicine (JC!i, WL W), Boulder, Colorado; and Department of Medicine, University of Utah

School of Medicine (JJZ), Salt Lal<e City, Utah, U.S.A .

An immune reaction to wheat protein has been previ­ously proposed to explain the pathogenesis of dermatitis h e rpetiformis. In order to detect and characterize anti­bodies to gluten in human sera, we developed an enzyme immunoassay for class-specific antibodies. Results of

Manuscript received April 26, 1982; a~cepted for publication Novem-be r 4, 1982. ·

This investigation was supported in part by Public Health Services Research Grant RR-64 from the Division of Research ResoUJ"ces and by Grant 1R01AM 28412-01 from the National Institutes of Health.

Reprint requests to: Alfred T. Lane, M.D., University of Rochester School of Medicine, 601 Elmwood Avenue, P . 0. Box 697, Rochester, New York 14642.

Abbreviations: BSS: borate-buffered saline CD: celiac disease DH: dermatitis herpetiformis EIA: enzyme immunoassay SBB: smaU-bowel bypass

this assay in 49 patients with dermatitis herpetiformis were compared with those of 38 normal control subjects, 11 patients with celiac disease, and 6 small-bowel bypass patients. IgA antibodies to gluten were significantly more frequent in dermatitis herpetiformis sera (28/49) than in normal control sera (4/38). lgG antibodies to gluten were significantly more frequent in both celiac disease (10/11) and dermatitis herpetiformis (16/49) sera than in control (5/38) sera. Dermatitis herpetiformis sera also had an increased prevalence of lgM antibodies to gluten (19/49). Small-bowel bypass patients demon­strated no antibody to gluten. Antibodies to gluten in dermatitis herpetiformis objectively mark a state of im­mune reactivity to wheat protein and may be involved in the genesis of the cutaneous lgA immune deposits and the skin disease.

Dermatitis herpetiformis (DH) is a chronic pruritic skin condition characterized by subepidermal blister formation and