The in vitro induction of delayed-type hypersensitivity to allo-antigens of the mouse

Journal of Immunological Methods, 69 (1984) 149-163 Elsevier

149

JIM 03023

The In Vitro Induction of Delayed-Type Hypersensitivity to Allo-Antigens of the Mouse J. Barrington-Leigh Department of Immunology, University of AIberta, Edmonton, Alberta T6G 2H7, Canada (Received 20 September 1983, accepted 2 December 1983)

An in vitro procedure has been developed which allows the induction of delayed-type hypersensitivity (DTH) to major allo-antigens of the mouse. Murine spleen cells cultured with irradiated allogeneic cells develop allo-specific DTH reactivity (DTHR). The variables assessed to provide optimal induction of D T H R were (i) the culture system, (ii) the density of responders and the number of stimulator cells and (iii) the kinetics of induction. The reactivity of the allo-specific cells was assayed in 3 different ways in order to select the most sensitive: (i) by local footpad transfer into a mouse syngeneic with the responder cells together with the eliciting (irradiated) antigen-bearing cells; (ii) by local transfer into the footpad of a mouse syngeneic with the stimulator cells, in which the allo-antigens present in the subcutaneous tissues elicit the response, and (iii) by intravenous transfer into syngeneic or allogeneic mice which are challenged 24 h later in the footpad with spleen cells bearing the haplotype of the stimulator. The second assay is clearly the most sensitive (2 × that of the first and > 20 x that of the last). Observations are reported demonstrating the specificity of the swelling reaction. The kinetics of swelling and its T-cell dependence provide strong grounds for believing that the reaction is due to classical delayed-type hypersensitivity. Furthermore the T cells mediating the swelhng are of the phenotype Lytl + Lyt2 ± and I a - and are radiation resistant, whereas the ability to produce a swelling reaction is sensitive to 1000 rads. whole body irradiation. The system has been applied to determine both the specificity of T cells mediating DTH to major and minor allo-antigens and whether cytotoxic T cells (CTL) and DTH-active T cells are always induced under the same conditions. Key words: major histocompatibility complex - delayed hypersensitivity - allo- antigens

Introduction

The study of the roles played by products of the major histocompatibility complex (MHC) has in recent years acquired prominence in diverse areas of immunology, particularly since the discovery of the phenomenon of restriction in Abbreviations: MHC, major histocompatibility complex; CTL, cytolytic T-lymphocytes; HVG, hostversus-graft; GVH, graft-versus-host; MLC, mixed lymphocyte culture; CML, cell-mediated lympholysis; DTH, delayed-type hypersensitivity; LM, Leibovitz medium; MEM, minimal essential medium; FCS, fetal calf serum. 0022-1759/84/$03.00 © 1984 Elsevier Science Publishers B.V.

150 cell-mediated lympholysis (Zinkernagl and Doherty, 1974; Shearer et al., 1975). This phenomenon and the predominance of T-lymphocyte precursors with specificity for cell-surface allo-antigens encoded by the MHC (Nisbet et al., 1969) emphasize the importance of gaining an understanding of cellular allo-reactivity in each of its modes of expression. Many forms of cell-mediated immunity have been observed and studied in a variety of model systems, such as the mixed lymphocyte reaction (Dutton et al., 1965) and cell-mediated cytotoxicity (Brunner et al., 1970) in vitro, and the graft-versus-host reaction (Simonsen, 1962) and the host-versus-graft reaction (Billingham et al., 1954) in vivo. Whereas it is known that several forms of immunity can be involved in the process of graft rejection (Klein, 1976; Lafferty and Woolnough, 1977), it is probable that DTH is a major component. For example, it was found that footpad tuberculin-type delayed inflammatory responses could be elicited following skin-graft sensitization, by cells or fragments bearing major or minor histocompatibility antigens (Brent et al., 1962; Kon et al., 1976). DTH has also been described in other all0-graft systems (A1-Askari and Lawrence, 1969; Phillips et al., 1972; Smith and Miller, 1979 a, b; Van der Kwast et al., 1979; Greene et al., 1980; Liew and Simpson, 1980), and in certain tumor systems .(Halliday and Webb, 1969; Hoy and Nelson, 1969; Rouse et al., 1972; Paranjpe and Boone, 1974; Kon et al., 1976; Perry and Green, 1980). DTH can be induced against viruses (Mortensen et al., 1973; Zinkernagl and Doherty, 1974; Kraaijeveld et al., 1979) and has also been associated with GVH (Ruddle et al., 1974; Wolters and Benner, 1978). However, in each of these systems a state of DTH was induced in vivo, for which any analysis of regulatory and recognitive events must be complex. Recently, Bretscher (1979) and Ramshaw and Eidinger (1979) described the first in vitro system for inducing DTH to complex antigens on xenogeneic erythrocytes. In this paper an optimized in vitro system is described for the induction of H-2 specific DTH effector cells. Moreover, the in vitro system described herein allows the generation of considerably stronger DTH responses than those obtained via in vivo immunization. In addition, 3 different assays for measuring the allo-DTH reactivity obtained are compared in order to select the most sensitive. The sensitized cells can transfer DTH swelling activity either locally into a footpad or by intravenous injection. The ability to transfer in vitro induced DTH sensitivity systemically allows one to assess the importance of this DTH reactivity in various situations such as in the rejection of grafts, tumors and its efficacy in protection against viral infection, as well as its role in pathogenesis.

Materials and Methods Mice

Two- to 4-month-old male or female mice of inbred strains, whose haplotypes are specified below, were obtained either from the stocks maintained at the University of Alberta, Edmonton, or from Jackson Laboratories (underlined): C B A / C a J (H-2 k), BALB/cCr (H-2d), C3H.SWSn (H-2b), C57BL/10J (H-2b), B10.D2 (H-2a), B10.A (2R) (kSd3b), B10.A (5R) (b3k2d4). The shorthand notion is used in which the 9

151 sub-regions of H-2 are referred to by haplotype in sequence, e.g., B10.A (2R) has (K, IA, IB, IJ, IE) k (IC, S) d, D d designated kSd3b. BALB/cCr is also abbreviated as BALB/c.

Preparation of cell suspension and M L C conditions Single cell suspensions from the spleens or thymi of normal mice were prepared by gently passing the tissue through a stainless steel wire mesh u~ader sterile conditions using Leibovitz medium (LM) containing 10% fetal calf serum (FCS). After settling in a tube for 2 min on ice, the supernatant was harvested to obtain cells free of clumps and fatty debris. After 1 wash, the cells were counted by trypan blue exclusion, centrifuged and resuspended in minimum essential medium (F-15, Grand Island Biol. Co., Grand Island, NY) supplemented with 10% FCS and 10 - 4 M mercaptoethanol (MEM). The cells were cultured at 37°C under 10% CO 2 either in (a) microtiter V-bottom trays (Baum and Pilarski, 1978) with 2-8 x 105 responders and 10 6 irradiated stimulators in 0.2 ml/well; (b) ac~ylamide rafts (Pilarski and Borshevsky, 1976) with 2-10 × 106 responders and 107 irradiated (1500 rads. from a 137Cs source) allogeneic stimulators in 1 ml of MEM; or (c) in Costar trays (Costar, Cambridge, MA) with 107 responders and 3 x 105 irradiated stimulators in 2 ml/well. The Costar plastic trays incorporate six 10 ml flat-bottomed cylindrical wells (area 10 mm2). Gentamicin was added to a final concentration of 50 btg/ml. Cells were harvested on the days of peak activity, namely day 5 or 6 of culture. Assay of DTH Cells were harvested, washed twice in LM, and counted using trypan blue exclusion. The dead and red cells (which presumably included most of the antigen) were removed by Isopaque-Ficoll (Parish and Hayward, 1974). Recovery of viable cells on day 6, when spleen cells were the responders, was about 50% from the rafts, 10% from the microtiter trays, and 75% from the Costar wells. In the first form of the assay for DTH activity, a standard dose of 107 sensitized cells was injected either alone (control) or with 5 × 106-107 irradiated syngeneic (control) or allogeneic spleen cells into a syngeneic hind footpad and the differential swelling between injected and uninjected footpads of each recipient mouse measured at 24 h post-injection using a micrometer (cf. Bretscher, 1979). In the second, more sensitive assay, the sensitized cells were injected alone into syngeneic (control) or allogeneic footpads and the 24 h swelling determined. Systemic transfer of in vitro induced allo-DTH was achieved by injecting 5 x 107 sensitized cells i.v. per recipient and challenging after 48 h with 2.5 x 107 fresh irradiated spleen cells of target or control strains. Swelling was recorded 24 h later. Note that a typical uninjected (normal) footpad measures approx. 160 U (1.6 mm) in thickness. Thus an indicated swelling of e.g., 80 U, as typically elicited with 107 sensitized cells, refers to ~an easily visible, 50% increase in thickness. Treatment of cells with antisera Initially an A K R anti-CBA anti-Thyl serum, whose specificity was characterized as described elsewhere (Baum and Pilarski, 1978), was used to determine whether the

152 swelling activity is due to T cells. 2 x 10 7 cells, in 0.75 ml of medium containing anti-Thyl serum or normal mouse serum at a dilution of 1/6, were incubated at 0°C for 20 min, washed, and resuspended in 1 ml of medium containing a non-toxic rabbit complement at a dilution of 1/6. After incubation at 37°C for 45 min, the cells were washed and the number of viable cells estimated by trypan blue exclusion. Comparable treatments were performed using monoclonal anti-Lytl.1, conventional anti-Ia k from (Dr. I.F.C McKenzie) as well as monoclonal anti-Thyl (New England Nuclear) at dilutions of 1/50, 1/16, and 1/1000, respectively, the latter confirming the AKR anti-CBA antiserum results.

Statistics Two-tailed t-tests were used in determining significance levels between the mean values for the variously treated groups. All means stated or implied to differ significantly satisfy 0.01 > P > 10 -6, n = 4 or 5. Each class of experiment has been performed 3 or more times. As an aid in interpreting the swelling data, a coding system has been added to the tables as follows: - , background or control level; barely significant (0.05 > P > 0.02); +, + +, + + + , all highly significant ( P as above), and ranging from 30% footpad increase to over 100%.

Results

Culture conditions optimizing the induction of specific D T H Attempts to find conditions under which cells could be sensitized against allo-antigen in vitro began with cultures in microtiter trays similar to those known to induce allo-specific CTL, i.e., approximately equal numbers (10 6) of responder and stimulator spleen cells per well (0.2 ml) in a one-way MLC (Pilarski and Borshevsky, 1976). Cultures were subsequently optimized with respect to the following parameters: density of responders, number of stimulators, kinetics, percentage yields, ease of manipulation, specificity of swelling activity and differential induction of swelling versus cytotoxic activity. Three types of in vitro systems were examined for their efficacy in obtaining cells sensitized to produce a swelling reaction to allo-antigen (Table I). Highest yields were obtained on days 5 or 6 in acrylamide rafts or Costar wells (over 50%), the latter system being adopted since it is much easier to use. All the experiments described below were carried out using spleen cell responders. Thymocytes were found to be poor responders. It is apparent from Table I that Costar cultures are optimal for inducting D T H (107 responders (i.e., density of 106/cm 2) and 3 x 105 stimulators/well). Optimal specific swelling is induced in the range of 3-5 x 105 stimulator cells (Fig. 1); both the third-party and background swelling increase at higher doses. Responder cell densities in excess of 1.5 x 106/cm z give rise to a much reduced yield (e.g., < 10% at 107/cm 2, not shown) though not in activity per viable cell. Densities below the optimum show a marked drop in activity though not in yield. Specific swelling is defined here as that produced in the presence of both sensitized cells and relevant antigen minus the swelling produced by sensitized cells

153 TABLE I SYSTEMS F O R T H E IN VITRO I N D U C T I O N OF A L L O - D T H ACTIVITY Culture vessel a

Volume (ml)

Microtiter trays (V) Polyacrylamide rafts (V) Costar trays a b c d e

Optimal cell doses c ( X 10 6)

(%)

Footpad swelling e (1 U = 10/tm)

Yield d

R

S

0.2/well 1.0/raft

0.8 10.0

0.02 10.0

12 50

69 _+5 74 _+4

2.0/well

10.0

0.3

75

80 + 3

See Materials and Methods. V = V-bottomed; F = flat-bottomed, cylindrical well. Volume of medium containing cell doses indicated. R = responder population (spleen); S = irradiated stimulator population (spleen). Viability after 6 days in culture as a percentage of responder cell input. Swelling activity per 107 injected viable cells measured at 24 h.

injected with syngeneic cells. The yield of viable cells is almost independent of the number of stimulator cells up to a dose of 1 0 6 cells. However, the induction of specific swelling activity does not take place in the absence of antigen in culture (see Fig. 2). In the following sections the results pertaining to syngeneic recipients in the assay will be presented first, followed by those obtained using allogeneic recipients.

Syngeneic recipients: dependence of swelling activity on the presence of antigen both during culture and in the footpad Fig. 2 demonstrates the dependence of the swelling activity on the presence of antigen both in culture and in the footpad. CBA/CaJ spleen cells (107/well) were E o

- lO0--j~ ~

8o

m

o z

~

40

2o O

y -

° 0I'J

,

3xlO s

1 6

xlO

3 10

I

10z

STIMULATOR CELL DOSE

Fig. 1. Effect on induction of D T H activity of varying stimulator dose in culture. Fixed dose of 107 CBA/CaJ responder ceils per Costar well plus a range (0-107) of irradiated stimulator cells doses (BALB/c). 107 viable cells from harvest on day 6 assayed for their swelling activity in allogeneic (O, BALB/c), syngeneic (A, C B A / C a J ) or third-party ( O , C3H.SWSn) recipients. Four mice per point; bars denote 2 standard errors.

154 E

80 -

O

70 --

60 --

T

50

--

40

--

30

--

© Z

r-h 20-O O

10--

IA-

0GROUP

A

B

Fig. 2. Specific antigen dependence of sweiling response in syngeneic recipients. Costar cultures were set up at a dose of 107 CBA/CaJ responder spleen cells plus either zero (group A) or 3 x 105 (group B) irradiated BALB/c stimulator spleen cells per well. Cells harvested from either group on day 6 were assayed for their swelling activity in syngeneic (CBA/CaJ) recipients by injecting into their footpads 107 viable cells (of type A or B) together with 107 fresh, irradiated spleen cells of either allogeneic (~, BALB/c);"syngeneic (Q, CBA/CaJ), or third-party ([], C3H.SWSn) origin. Columns represent means of 5 recipients; bars denote 2 standard errors.

cultured without antigen (group A) or with 3 × 105 irradiated B A L B / c C r spleen cells as antigen (group B). The cells from each group were injected into the footpads of naive C B A / C a J mice together with l07 irradiated spleen cells that were (a) syngeneic with the stimulators, i.e., of B A L B / c C r origin, or (b) syngeneic with the responders, or (c) of a third-party (C3H.SWSn) strain, as a source of eliciting antigen. It can be seen that swelling of the order of 20 U is obtained in group A, and that swelling significantly above this level is obtained only when antigen is present both in culture and in the footpad. The level of 5 - 2 5 U appears to be the background level for the assays described herein since all cultured cells injected at a dose of 107 per footpad cause such a non-specific swelling unless the same antigen is present in the recipient footpad as was used for sensitization. Fig. 2 also suggests that the swelling activity is specific. Cells incubated either with or without antigen were injected into naive syngeneic animals together with 107 irradiated C 3 H / S W S n spleen cells as a source of third-party antigen. Significant swelling was not observed. Further experiments on the specificity of the swelling reaction are described below.

T-cell dependence and kinetics of the swelling reaction Cells displaying activity are sensitive to anti-Thyl treatment as recorded in Fig. 3. C B A / C a J spleen cells sensitized to B A L B / c C r were treated in various ways before

155

160-E

o

140 --

n

12o-

E ~-

120

~z

100 --

I

-

I00--

~

60

"-Z80---

©

z- n

I

60--

2

C3 40--

I---

© ~

C3

~ © u_

20OT

IT I-ii- "T-q- ~" GROUP NUMBER

20o

0

I

I

I

20

40

60

80

t

I

I

100

120

140

" TIME (hrs.)

Fig. 3. T-cell dependence of swelling activity in syngeneic recipients. Optimal cultures (107 CBA/CaJ spleen responder cells vs. 3x105 irradiated BALB/c spleen stimulator cells per Costar well) were harvested on day 6 and assayed for their swelling activity per 107 viable cells in syngeneic recipients after various treatments: I, 107 untreated cultured cells injected into the footpad together with 107 fresh irradiated BALB/c spleen cells as antigen, per CBA/CaJ (syngeneic) recipient. Passage of the sensitized cells through a nylon wool column prior to injection does not affect the swelling response (not shown); II, as in I but with fresh irradiated CBA/CaJ spleen cells as (control) antigen in place of BALB/c; III, as in I but with the sensitized cells treated with normal mouse serum and complement (17% lysis) before co-injection with the BALB/c (antigen-bearing) cells; IV, as in III but using anti-Thyl antiserum in place of normal mouse serum (76% lysis). The cell count was restored to its pretreatment value with fresh normal CBA/CaJ (syngeneic) spleen cell ('fillers') before injection with (allo-antigen-bearing) BALB/c cells; V, as in IV but without restoration with filler cells. Error bars denote 2 standard errors: 5 mice per group. Fig. 4. Kinetics of the swelling response. 10 7 sensitized cells (CBA/CaJ-anti-BALB/c) were injected into BALB/c recipients Q,), or into CBA/CaJ recipients together with 10 7 irradiated normal spleen cells or either allogeneic (O, BALB/c antigen) or syngeneic (O, CBA/CaJ, control) origin. Swelling activity was assayed as a function of time after injection. Four mice per point. Bars indicate 2 standard errors (not drawn if below _+2 U).

b e i n g i n j e c t e d i n t o r e c i p i e n t f o o t p a d s w i t h 10 7 i r r a d i a t e d B A L B / c s p l e e n cells as a s o u r c e of eliciting a n t i g e n ( g r o u p I) o r w i t h 10 7 i r r a d i a t e d s y n g e n e i c s p l e e n cells ( g r o u p II) as the c o n t r o l . T r e a t m e n t w i t h a n t i - T h y l a n t i s e r u m p l u s c o m p l e m e n t ( g r o u p s IV, V), b u t n o t w i t h c o m p l e m e n t a l o n e ( g r o u p III), r e d u c e s t h e s w e l l i n g a c t i v i t y o f the c o n t r o l level, e v e n w h e n s y n g e n e i c n o r m a l s p l e e n cells a r e a d d e d to c o m p e n s a t e for t h e loss b y lysis (86%) ( g r o u p IV). It is c l e a r t h a t a T cell is r e q u i r e d for the e l i c i t a t i o n o f the s w e l l i n g r e s p o n s e , t h a t is n o t p r e s e n t in large n u m b e r s in n o r m a l s p l e e n ( g r o u p I V

156 vs. group V), which is consistent with its being specific. The kinetics of the antigen-dependent swelling reaction in the footpad are characteristics of a D T H response, as shown in Fig. 4 (middle curve vs. lower). The peak swelling clearly occurs between 20 and 24 h. The absence of significant swelling before 16 h excludes the possibility of an immediate or Arthus response. The experiments described above provide reasonable grounds for believing that allo-specific D T H is induced in the Costar system, as assayed by swelling activity in normal syngeneic recipients in the presence of the relevant or control allo-antigen bearing spleen cells.

Allo-DTH in allogeneic recipients Cells sensitized in vitro were employed in a direct local transfer assay, in the hope that the presence of large amounts of allo-antigen in the subcutaneous tissue of the recipient would lead to increased sensitivity of the swelling response, and thus to a greater sensitivity in e x a m i n i n g responses to weaker allo-antigens. The following observations show this to be true.

T-cell dependence and kinetics of swelling The uppermost curve in Fig. 4 records the time course of swelling activity in allogeneic recipients. It can be seen that the peak swelling is greater in this form of assay, typically by a factor of 2 above that obtained in the syngeneic system employed initially. Table II demonstrates that for C B A / C a J reactivity against B A L B / c C r antigens a T-cell population bearing L y t l , but not Ia, is required for transfer of activity which is not detectably present in normal syngeneic spleen. The case of the Lyt2 marker is complex. Experiments using conventional and monoclonal anti-Lyt2.1 antisera have consistently indicated a moderate dependence of the swelling reaction on Lyt2-bearing cells. This contrasts with a clear evidence for a Ly2.2 bearing cell in the case of B A L B / c C r D T H responses against major and m i n o r allo-antigens (data not shown), and with H-Y specific D T H effectors which are L y t l ÷2 ÷ (Barrington-Leigh, 1982). TABLE II T-CELL DEPENDENCE OF SWELLING ACTIVITY IN ALLOGENEIC RECIPIENTS Treatment of sensitized cells

% Lysis

NMS+C' Anti-Thyl + C' Anti-Lytl + C' Anti-lak + C'

30 82 66 79

Footpad swelling ( + SE) (1 U = 10/~m) 77+3 87+7 11 +_2 30 _ 3 79 + 4

CBA/CaJ spleen optimally sensitized against BALB/c, and injected either untreated, or pre-incubated with normal mouse serum or anti-Thyl, anti-Lytl or anti Ia antiserum and complement (C') (see Materials and Methods), with indicated % loss by lysis. DTH elicited by injecting a total of 107 cells per normal BALB/c (allogeneic) recipient. Normal syngeneic spleen cells (CBA/CaJ) added as 'fillers' to replace numerically the corresponding loss by lysis. Means of 5 mice per group. Background swelling from 10 7 unsensitized cells was 25 + 2 U in BALB/c recipients, 15 + 3 in syngeneic (CBA/CaJ) animals.

157 160 E

0

~-

140

-

120

-

I.--

100 -

© z

8O-

~

6O-

C3

~

4o-

© ~

20O-

I

Bo
Ba(,S

• SENSITIZED CELLS

Fig. 5. Specificity of the swelling response in allogeneic recipients. 107 BALB/c spleen cells, sensitized ('B a C' or 'B a S') in optimal Costar cultures to either CBA/CaJ or C3H.SWSn spleen stimulators, were injected alone into syngeneic (BALB/c) or allogeneic (CBA/CaJ or C3H.SWSn) recipient footpads. Five mice per groups; bars indicate 2 standard errors. B = BALB/c; C = CBA/CaJ; S = C3H.SWSn. Letters within columns denote the recipients' strain.

m i n o r allo-antigens (data not shown), and with H - Y specific D T H effectors which are L y t l + 2 ÷ (Barrington-Leigh, 1982).

Specificity In a 2-way (criss-cross) experiment 107 B A L B / c C r spleen cells sensitized against C B A / C a J or C 3 H . S W S n were injected into C B A / C a J , B A L B / c C r or C 3 H / S W S n recipients and the 24 h swelling c o m p a r e d (Fig. 5). A similar pattern of specificity is seen as in the syngeneic assay (Fig. 2), again showing the dependence of the swelling reaction on an interaction between sensitized cells and sensitizing antigen. In this case the antigen is already present in the subcutaneous tissues of recipients of the appropriate strain.

Dose-response of swelling activity Fig. 6 shows that the specific swelling elicited in allogeneic recipients depends in a straightforward m a n n e r on the n u m b e r of sensitized cells injected.

Absence of allogeneic effects To ensure that the D T H swelling was not affected by an allogeneic effect in the f o r m of a host response to the injected cells, ' A anti-B' cells ( C B A / C a J a n t i - B A L B / c ) were injected into either (A × B)F 1 or B footpads. C o m p a r a b l e swelling was elicited

158 E

o i-.

120-':"

2

T

_

o z

-.J

5 O

0 0

........ i ........ ........ i ........ I 2xlO 6 4xlO 6 6xlO 6 8 x106 10 x106 DOSAGE OF CELLS

P

Fig. 6. Dose-response curve the swelling activity in allogeneic recipients. Varying numbers of CBA/Ca.I spleen cells sensitized in optimal Cos'tar cultures towards BALB/c spleen cells were injected into normal BALB/c recipients (11), and the swelling recorded (at 24 h). The effect of adding CBA/CaJ spleen filler ceils, to a total of 107 injected cells, is indicated at 2 points (D). The 'dose-response' curve for non-sensitized (normal) CBA/CaJ spleen cells (e) is also given. For comparison the responses in syngeneic recipients (CBA/CaJ) in the same experiment are shown: 10 7 sensitized cells injected together with 107 irradiated BALB/c spleen cells as antigen (-), or without antigen (zx), per CBA/CaJ recipient. Bars indicate 2 standard errors. Five mice per point.

in both strains of mouse, cf. groups I and II of Fig. 7. In the first case, none (or very little) of a hypothetical allogeneic effect is expected to occur in contrast to the second case. A anti-B sensitized cells gave very little swelling (group VI) when injected into footpads of ' C ' (third-party) strain mice (C3H.SWSn), showing that an allogeneic environment per se does not cause a swelling reaction, but that swelling requires, and can be obtained in, the presence of the relevant antigen(s) (group VII). In group II, A anti-B cells were injected into (B × C)F 1 recipients. The observations confirm that a third-party environmental factor does not affect the size of the swelling. Fig. 7 also illustrates the greater sensitivity of the allogeneic as c o m p a r e d to the syngeneic form of the assay (group I vs. V, VII). In addition, a specific D T H reaction dependent on the presence of both activated responder cells ( ' A ' - a n t i - ' B ' ) and target antigen ('B') can be elicited in a third-party host ('C'), i.e., is not restricted to hosts of either ' A ' (responder) or ' B ' (stimulator) strains, (group VII vs. VI, c o m p a r e d with V vs. IV). This result is expected if the allogeneic environment per se does not influence the reaction.

Radiation sensitivity The specific cells able to elicit a swelling reaction in allogeneic footpads are relatively X-ray resistant, whereas the recipient mouse is X-ray sensitive. Thus C B A

159

160

T --

E

~

14o -

I----

~_

~2o -

~©

I00

I -

-

80--

Z

1

_..i ill

~

6o-

c¢'3

C3

~

4o-

I-.-

©

0u _

200-

-1RECIPIENT: BB ANTIGEN:

.

-V

AB .

.

BC .

AA

AA ~'B

CC

CC

--

Fig. 7. Absence of allogeneic effects on the swelling response. 107 cells (designated A-anti-B, see key below) of strain C B A / C a J (here 'A') sensitized in Costar trays to irradiated B A L B / c spleen cells (here 'B') were injected into homozygous (AA, BB or CC) or F l (A × B = AB, B x C = BC) recipients, with (groups V, VII) or with (groups I-IV, VI) 107 irradiated B A L B / c ('B') spleen cells (as antigen), as indicated. A = C B A / C a J ; B = B A L B / c ; C = C3H.SWSn. COlumns BB, AB and BC (groups I-III) show responses in allogeneic or semi-allogeneic recipients. AA columns (IV, V) represent the response in syngeneic recipients (with or without antigen); CC columns (VI, VII) show the B-antigen dependence of an A-anti-B response in third-party (C) recipients. Five mice per column with bars of 2 standard errors.

anti-BALB/c cells, injected into B A L B / c feet, produce a large swelling (line 1 of Table III) which is comparable to that observed when the sensitized cells are given 1500 rads. before injection (line 3). Irradiating the recipients with 1000 rads. 24 h

T A B L E Ili R A D I A T I O N SENSITIVITY OF SWELLING R E A C T I O N C B A / C a J spleen cells sensitized against B A L B / c spleen for 6 days in Costar trays. Irradiated cells given 1500 rads. prior to injection into recipient footpad ( B A L B / c ) When irradiated, the recipients were given 1000 rads. whole body irradiation 24 h prior to injection of 107 sensitized cells into the footpad. Means of 5 mice per group. Treatment of sensitized cells -

1500 rads.

Treatment of recipient -

1000 rads. -

Footpad swelling ( + SE) (1 U = 10 # m) + 12 46 + 4 140 + 7

156

160 TABLE IV SYSTEMIC TRANSFER OF IN VITRO INDUCED ALLO-DTH DTH reactivity in CBA/CaJ spleen cells, induced in vitro (6 days) against BALB/cCr, was assayed systemically by i.v. transfer (5>(10 7 sensitized cells/mouse) into normal, syngeneic or third-party (allogeneic) recipients, which were challenged in the footpad after 48 h with 2.5 x 107 fresh, irradiated spleen cells of the indicated target strains. Swelling activity was recorded 24 h later, n = 4. Responder

Stimulator

Recipient

Antigen

Target

Footpad swelling

Code

CBA/CaJ

BALB/c

CBA/CaJ

BALB/c CBA/CaJ

H-2 -

57 + 3 6 _+2

+ -

CBA/CaJ

BALB/c

B10/J

B10.D2 B10.A(2R)

H-2 -

49 + 4 4+ 2

+ -

CBA/CaJ

BALB/c

B10.A(2R)

B10.D2 B10.A(2R)

H-2 -

65 + 7 10 + 3

+ -

before injection considerably diminishes the swelling reaction (line 2), as observed previously ( U h r and Scharrf, 1960).

Systemic transfer of in vitro induced allo-DTH Cells sensitized to allo-antigen in vitro might exhibit altered h o m i n g or recirculation properties when re-injected into mice. To test for this possibility, 5 x 107 C B A / C a J a n t i - B A L B / c C r sensitized cells from primary culture were injected i.v. into syngeneic mice ( C B A / C a J ) as well as 2 B10 congeneics, B 1 0 / J differing fully at H-2 from both responder and stimulator strains, and B10.A (2R) differing at the K, IA, and D sub-regions only (Table IV). W h e n challenged in the footpad with a high dose of target-bearing and control normal spleen cells, 48 h post i.v. injection, moderately large responses were obtained as would be expected if there were no H-2 restriction on this transfer. Clearly there is no impediment to recirculation and specific homing in this system, though the sensitivity of the assay is considerably lower than that of the local transfer into a footpad.

Discussion The development and characterization of an in vitro system for the induction of D T H to allo-antigens was undertaken for several reasons: (i) to provide a means for optimizing and analyzing the inductive process and its regulation; (ii) to determine the principal allo-antigens against which D T H reactivity is induced; (iii) to investigate the role of such antigens in restricting responses to minor allo-antigens; (iv) to find characteristics of allo-specific D T H effectors distinguishing them from cytotoxic T cells induced under comparable (or distinct) conditions; and (v) to provide a general procedure for the induction of D T H to weak allo- or related antigens (such as H - Y or tumor specific) whereby in vitro restimulation leads to greatly magnified activity as assayed in vitro.

161 The experiments reported may be taken to define an allo-antigen specific DTH response which for major antigens can be induced in vitro in a one-way MLC between allogeneic spleen cells. The swelling response elicited in the footpads of normal recipients by the injection of cultured cells is specific for those allo-antigens as shown by its dependence on the presence of specific antigen both in the sensitizing culture and in the test recipient (Fig. 2). Cells sensitized against the allo-antigens of a given allogeneic strain gave rise to a much greater swelling when tested by co-injection into the footpad with cells bearing that same set of antigens, than with third-party allogeneic cells (Fig. 2). In a substantially more sensitive assay, the sensitized cells are injected directly into the footpad of normal allogeneic mice of target or third-party strains. This method generates a 2-3-fold greater specific swelling, with no increase in background level, than that obtained in the syngeneic assay (Figs. 2, 6, 7). It demonstrates the presence of appropriate antigens in the sub-cutaneous tissue of the target-bearing allo-recipients (as well as on spleen cells), and that the mechanism of the delayed inflammatory response proceeds undeterred in the allogeneic environment. F~ and third-party controls have ruled out the possibility that an 'allogeneic' effect, either enhancing or inhibitory, contributes significantly to the swelling. Thus sensitized cells injected alone into allogeneic or semi-allogeneic recipients elicit an equally strong footpad swelling reaction, which is absent in syngeneic, third-party or semi-allogeneic third-party F 1 mice (Fig. 7). Evidence that the observed swelling activity is indeed an expression of DTH is as follows. Firstly, the swelling activity can be transferred by allo-sensitized but not unsensitized or normal cells (Fig. 2). Secondly, the specific swelling in either syngeneic or allogeneic recipients is sensitive to treatment with anti-Thyl antiserum and complement (Fig. 3 and Table II). Moreover, none of the swelling activity is removed by passage of the in vitro sensitized cells through a nylon column prior to injection (data not shown). Thus a specific nylon wool non-adherent T lymphocyte is required to elicit the swelling. Third, the time course of the swelling, which reaches a peak by 20-24 h, is characteristic of a DTH and not of an Arthus reaction (Fig. 4). Fourth, the specific swelling response was abolished by irradiating recipients with 1000 rads. 24 h prior to the injection of the sensitized cells, but was undiminished if the cells were irradiated (1500 rads.) and injected into normal recipients (Table III). Thus the function of the effector population is radiation resistant reflecting their end-differentiated state but the swelling depends on a radiation-sensitive host reaction, presumably a differentiation and proliferation of macrophages (Beller and Unanue, 1982). Finally, it was found that the effector cell population bears Lytl, in some cases Lyt2, but in no case Ia (Table II), which is consistent with previous determinations of the Ly phenotype of other DTH effector T cells (Dennert and Hatlen, 1975; Vadas et al., 1976; Smith and Miller, 1979a, b). The absence of Ly7 on the in vitro induced effectors (Shaw et al., 1981) also distinguishes them from T helper cells specific for allo-antigens (A1-Adra et al., 1980). Delayed responses to DNCB and to protein antigens such as F G G have been shown to be restricted by the K, I or D and by the I region(s) respectively (Miller and Vadas, 1976). In an in vivo system (Smith and Miller, 1979b), DTH reactivity against non-H-2 antigens was found to be restricted on transfer by the K end, but

162 not the D sub-region of H-2. In these instances in vivo sensitized cells were transferred i.v. to a normal recipient which was subsequently challenged with antigen," or antigen-bearing cells, at a peripheral site. In the present system no restriction in local transfer is observed for anti-H-2 responses (Fig. 7) nor is there a restriction for systemic transfer (Table IV). These observations confirm the lack of restriction for anti-MHC responses observed in the in vivo systems but also show that recirculation of allogeneically activated cells is possible in a third-party host. This latter result may be of value for immunoregulatory studies conducted in vivo. The conditions for the induction in vitro of DTH to allo-antigens in the present system clearly resemble those for the induction of CTL (Engers and Macdonald, 1976). Two main questions arise from this. Firstly, is the swelling activity in fact a result of the induction of CTL and not of a separate DTH-specific class of cells? Secondly, if these 2 classes of effector cells do exist, what factors govern the induction of 1 class versus the other? Studies aimed at answering these questions, employing the system described here, show that D T H and CTL to major alloantigens are not always coordinately induced, and that potent DTH, both to major and minor H antigens can be induced to some of those antigens against which it is not possible to raise potent cytotoxicity (Barrington-Leigh, 1982; and JBL in preparation). These observations, as well as the surface phenotype of the cell-mediated effector cells show that D T H and cytotoxic cells belong to different subclasses of T cell. The system developed here should be useful for analyzing the cellular interactions that determine whether a D T H or CTL response is induced•

Acknowledgements I am grateful to Mr. A. Murji for his enthusiastic and competent technical assistance. I thank Drs. P.A. Bretscher and L.M. Pilarski for many invaluable discussions and Drs. J.F.A.P. Miller and E. Diener for their comments on an earlier version of the manuscript. This work was supported by the National Cancer Institute and the Medical Research Council of Canada.

References Al-Adra, A.R., L.M. Pilarski and I.F.C. McKenzie, 1980, Immunogenetics10, 521. AI-Askari, S. and H.S. Lawrence, 1969, Transplant. Proc. 1,400. Barrington-Leigh, J., 1982, Transplantation 33, 642. Baum, L.L. and L.M. Pilarski, 1978, J. Exp. Med. 148, 1579. Beller, D.I. and E.R. Unanue, 1982, in: Lymphokines,Vol. 6, ed. S.B. Mizel (AcademicPress, New York) p. 25. Billingham, E.R., L. Brent and P.B. Medawar, 1954, Proc. R. Soc. (London) Ser. B. 143, 58. Brent, L., J.B. Brown and P.B• Medawar, 1962, Proc. R. Soc. (London) Ser. B. 156, 187. Bretscher, P.A., 1979, Eur. J. Immunol. 9, 311.

163 Brunner, K.T., J. Manual, H. Rudolf and B. Chapuis, 1970, Immunology 18, 50. Dennert, G. and L.E. Hatlen, 1975, Nature (London) 257, 486. Dutton, R.W., 1965, J. Exp. Med. 122, 759. Engers, H.D. and F.W. Fitch, 1979, J. Immunol. Methods 25, 13. Engers, H.D. and H.R. Macdonald, 1976, in: Contemporary Topics in Immunobiology, Vol. 5, ed. W.O. Weigle (Plenum, New York) p. 145 Greene, M.I., B. Benacerraf and M.E. Dorf, 1980, Immunogenetics 11,267. Halliday, W.J. and M. Webb, 1969, J. Natl. Cancer Inst. 43, 141. Hoy, W.E. and D.S. Nelson, 1969, Nature (London) 222, 1001. Klein, G., 1976, in: Contemporary Topics in Immunobiology, Vol. 5, ed. W.O. Weigle (Plenum, New York) p. 316. Kon, N.D. and P.A. Klein, 1976, J. Immunol. 117, 413. Kon, N.D., J.T. Forbes and P.A. Klein, 1976, Int. J. Cancer 17, 613. Kraaijeveld, C.A., M. Harmsen and B. Khader Boutahar-Trouw, 1979, Infect. Immun. 23, 119. Lafferty, K.J. and J. Woolnough, 1977, Immunol. Rev. 35, 245. Liew, F.Y. and E. Simpson, 1980, Immunogenetics 11,225. Miller, J.F.A.P. and M.A. Vadas, 1976, Cold Spring Harbor Symp. Quant. Biol. XLI, 579. Mortensen, R.F., W.S. Ceglowski and H. Friedman, 1973, J. Immunol. 111,657. Nisbet, N.W., M. Simonsen and M. Zaleski, 1969, J. Exp. Med. 129, 459. Paranjpe, M.S. and C.W. Boone, 1974, Int. J. Cancer 13, 179. Parish, C.R. and J.A. Hayward, 1974, Proc. R. Soc. (London) Ser. B. 187, 65. Perry, L.L. and M.I. Greene, 1980, J. Immunol. 125, 738. Phillips, S.M., C.B. Carpenter and J.P. Merrill, 1972, Cell. Immunol. 5, 249. Pilarski, L.M. and L.V. Borshevsky, 1976, Eur. J. lmmunol. 6, 906. Ramshaw, I.A. and D. Eidinger, 1979, Immunology 42, 42. Rouse, B., T. Wagner and A.W. Harris, 1972, J. Immunol. 108, 1353. Ruddle, N.H., M.Y.K. Armstrong and F.F. Richards, 1974, J. Immunol. 112, 706. Shaw, J., L.M. Pilarski, A.R. A1-Adra, J. Barrington-Leigh, J. Wilkins, P.M. Hogarth, I.F.C. McKenzie and V. Paetkau, 1981, Transplantation 31, 56. Shearer, G.M., T.G. Rehn and C.A. Carbarino, 1975, J. Exp. Med. 141, 1348. Simonsen, M., 1962, Prog. Allergy 6, 349. Smith, F. and J.F.A.P. Miller, 1979a, Int. Arch. Allergy Appl. lmmunol. 58, 285. Smith, F.I. and J.F.A.P. Miller, 1979b, J. Exp. Med. 150, 965. Uhr, J.W. and M. Scharrf, 1960, J. Exp. Med. 112, 65. Vadas, M.A., J.F.A.P. Miller, I.F.C. McKenzie, S.E. Chism, F.W. Shen, E.A. Boyse, J.R. Gamble and A.M. Whitelaw, 1976, J. Exp. Med. 144, 10. Van der Kwast, T.H., J.G. Olthof and R. Benner, 1979, Cell. Immunol. 47, 182. Wolters, E.A.J. and R. Benner, 1978, Transplantation 26, 40. Zinkernagl, R.M. and P.C. Doherty, 1974, Nature (London) 248, 701.