- Email: [email protected]
Role of macrophages as modulators but not as autonomous accessory cells in the proliferative response of immune T cells to soluble antigen
CELLULAR
IMMUNOLOGY
109, I- 11 ( 1987)
Role of Macrophages as Modulators but Not as Autonomous Accessory Cells in the Proliferative Response of Immune T Cells to Soluble Antigen’ JUN KAWAI,~ KAYO INABA, SHIN KOMATSUBARA, YOSHITAKA HIRAYAMA, KOJI NAITO, AND SHIGERU MURAMATSU~ Department of Zoology, Faculty of Science,Kyoto University, Kyoto 606, Japan Received September 4, 1986; acceptedMay 12. I987 The role of murine macrophages (M&) and that of splenic dendritic cells (DC) were investigated in the antigen-specific proliferative response of memory T cells of mice primed with keyhole limpet hemocyanin (KLH) 6 weeks or more before. Peritoneal Mq& whether expressing Ia antigens or not, did not function as autonomous accessorycells (A cells). A-cell activity of the spleen adherent cell population, which comprised M& in the majority and DC in the minority, was abolished by eliminating DC with a DC-specific monoclonal antibody and complement, and regained by the addition of a small number of DC. Though M@did not function as autonomous A cells, they augmented the proliferative response in the presence of a small number of DC. This occurred not only in the presence of free antigen, but also when DC and/or MI$ were pulsed with antigen. A culture supematant of M#Jhaving interleukin-1 activity was effective in enhancing the proliferation of T cells which responded to antigen-pulsed DC. On the other hand, interleukin-2 did not replace DC even in the presence of antigen-pulsed Ia+ M$J.We also investigated recently primed T cells, but no evidence was obtained in favor of the competence A Cells. 0 1987 Academic PWS, 1~. of M4 as aUtOnOtnOUS
INTRODUCTION Initiation of immune responsesis usually mediated by Ia-bearing (Ia’) accessory cells (A cells).4Dendritic cells (DC) are principal A cells in T-cell-dependent immune responses (l-8). We have so far demonstrated that Ia+ macrophages (Md) do not function as autonomous A cells but play a role as modulators in the primary mixedleukocyte reaction (MLR) and the primary antibody response to T-dependent antigens (6-8). Thus, these observations have been made solely in the primary immune response, so that it is necessaryfor us to check whether DC and M+ behave similarly ’ This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, and Culture of Japan, and the Shimizu Foundation Research Grant for 1984. 2 Present address:Shionogi Research Laboratories, Fukushima-ku, Osaka, 553 Japan. 3To whom correspondence should be addressed: Department of Zoology, Faculty of Science, Kyoto University, Kitashirakawa-Oiwakecho, Sakyo-ku, Kyoto 606, Japan. 4 Abbreviations used: A cells, accessory cells; C, complement; DC, dendritic cells; IL- 1, interleukin- I ; IL-2, interleukin-2; KLH, keyhole limpet hemocyanin; Mb, macrophages; MLR, mixed-leukocyte reaction; OVA, ovalbumin; SAC, spleen adherent cells.
0008-8749187$3.00 Copyright 0 1987 by Academic Press, Inc. Ail rights ofreproduction in any form reserved.
2
KAWAI
ET AL.
also in the secondary response. We investigated this in the antigen-specific proliferative response of memory T cells and also in that of recently primed T cells. The data to be presented below strongly argue that DC are competent, but are M@ virtually incompetent, to trigger the T cells, and that M@ modulate, augmentatively or suppressively, the level of the T-cell response triggered by DC. MATERIALS AND METHODS Mice. Male and female C3H/HeSlc (Shizuoka Agricultural Cooperative Association for Laboratory Animals, Shizuoka) and C3H/HeMs mice (Institute of Animal Experiments, Kyoto University) were used at the agesof 2-5 months. Mucrophuges. Pure populations of M4 were prepared from thioglycollate-elicited peritoneal exudate cells by the in vitro culture for 4 to 5 days as described (7). Ia molecules on M@ surfaces were induced by 100 IU/ml murine rIFN-7 (donated by Shionogi ResearchLaboratories, Osaka) or 20% CS (supernatant of spleen cell culture with concanavalin A) added to the culture medium, resulting in the Ia expression on more than 90% of Mr& In order to prepare spleen adherent cells (SAC), 3 X lo6 spleen cells were cultured in 24-well culture plates (A/S Nunc; Kamstrup, Roskilde) for 2 hr, and nonadherent cells were removed. The crude population of resident M$ containing a few percent of Ia+ cells was prepared as adherent cells after culturing resident peritoneal cells from normal mice for 5 hr. Dendritic cells. DC were prepared from collagenase-dispersed spleen cells as described (7). In the DC preparation, more than 90% of cells were Ia+, and no Fc receptor-bearing cells capable of forming rosettes with opsonized sheep erythrocytes were detectable. Memory T cells. Mice were immunized by injection into several subcutaneous sites of a total of 0.1-0.4 mg keyhole limpet hemocyanin (KLH; Calbiochem-Behring Corp., La Jolla, CA) emulsified in Freund’s complete adjuvant. After 6-10 weeks, draining lymph nodes were collected, treated with 0.5 mg/ml collagenase(Wako Pure Chemical Industries, Ltd., Osaka) and 30 pg/ml DNase I (Sigma Chemical Co., St. Louis, MO) for 30 min at 37°C and disrupted on stainless-steel sieves. Dispersed lymph node cells were incubated for 1 hr in plastic culture dishes (3003; Falcon, Oxnard, CA). T cells were enriched from the dish-nonadherent cells by a serial passagethrough a nylon-wool column (9) and a Sephadex G- 10 (Pharmacia Fine Chemicals, Uppsala) column (10). In some experiments, column-purified T-cell preparations were further treated with anti-Iak antiserum (A.TH anti-A.TL antiserum; Cedarlane Laboratories Ltd., Hornby, Ontario) plus complement (C) for 45 min at 37°C to completely eliminate Ia+ cells. Recently primed T cells. Lymph node cells were collected 12 days after KLH injection. Nylon-wool column-passed cells were prepared as the T-cell population and treated with anti-Iak plus C. Fractionation of the T-cell population was performed by centrifugation through step gradients of 50,60, and 70% Percoll (Pharmacia) at 1OOOg for 25 min at 5°C. The cells at each interface were collected with a Pasteur pipet, suspended in an excessamount of medium, and washed twice. Antigen pulsing. M4 on culture plates or DC in suspension were pulsed with 100 Kg/ml KLH for 30 min at 37°C and washed at least three times, unless otherwise described. Proliferative response o/T cells. Unless mentioned otherwise, 5 X lo5 purified T cells were cultured with varying numbers of M$ and/or DC in flat-bottomed 96-well
DC AND Mq3 IN THE PROLIFERATIVE
RESPONSE OF T CELLS
3
la%l+
la’M+ (xl0-4/well)
FIG. 1. Antigen-dependent proliferative responseofcolumn-purified T cells in the presenceof M$. KLHprimed memory T cells (5 X 105)prepared from lymph node cells by passagethrough nylon-wool and Sepbadex G- 10 columns were cultured in the presence (open) or absence (closed) of 50 rg/ml KLH with SAC or varying doses of Ia+ or Ia- peritoneal M$Jin modified Click’s medium containing 0.5% normal mouse serum on 96-well plates. Proliferation was measured by [3H]TdR incorporation during the last 18 hr ofa 5-day culture. Each symbol or column and vertical bar representsthe mean cpm oftriplicate cultures and the SEM, respectively.
plates in 0.2 ml of modified Click’s medium containing 0.5% normal mouse serum (7). M4 and DC were irradiated with 1500 R X-rays or treated with 40 pg/ml mitomytin C for 30 min at 37°C to prevent cell division prior to the culture. T-cell proliferation was evaluated by incorporation of tritiated thymidine ( [3H]TdR; New England Nuclear, Boston, MA) for 6-8 hr before cell harvest. IL-l-containing medium and recombinant IL-2. The culture supematant of M$ ingesting latex particles was used as the IL- 1-containing medium ( 11). Recombinant human IL-2 was donated by Dr. J. Hamuro (Ajinimoto Central Research Laboratories, Yokohama). RESULTS In this study, memory T cells and recently primed T cells from KLH-immunized mice were used in the experiments of Figs. 1 to 7, and in that of Fig. 8, respectively. In pilot experiments, we confirmed that these immune T cells proliferated specifically in response to KLH, but not to other unrelated antigens such as ovalbumin (OVA) under our conditions. Concentrations of KLH in the experiments described below were within the range of optimal concentrations estimated in preliminary experiments. Apparent A-Cell Activity of SAC, Ia+ M4, and la- M4 Nylon-wool and Sephadex G- 1O-passedcells (hereafter called T-cell preparation) were cultured with SAC, Ia+ Md, or Ia- M4 in the presence or absenceof KLH for 4 days. As shown in Fig. 1, the T-cell preparation was not responsive to KLH by itself, and supplementation of SAC or purified M$ was effective in supporting proliferation
4
KAWAI ET AL.
FIG.2. Memory T cells deprived of Ia+ accessorycells do not proliferate in the presence of antigen and Ia+ M$. Varying numbers of Ia+ peritoneal M4 were cultured in the presence (0) or absence (0) of 30 pg/ ml KLH with 5 X 10’ KLH-primed lymph node T cells which were pretreated with culture medium (A), with complement alone (B), or with anti-Ia antiserum plus complement (C). Cells were cultured for 4 days and [‘H]TdR incorporation proceeded for an additional 8 hr.
of T cells in response to KLH. A paradoxical result obtained in this experiment was that Ia- Mb appeared competent, though slightly inferior to Ia+ M4, as A cells in the proliferative response of memory T cells. The reason for this will be explained in the next experiment.
MacrophagesAre Not Autonomous A Cells In this experiment, the T-cell preparation was untreated or treated with anti-Ia antibody plus C or C alone, and cultured with Iaf M4 (Fig. 2). The T-cell preparation untreated or treated with C alone showed strong proliferation when cultured with Ia+ M$ (Figs. 2A and 2B), similar to the result in Fig. 1, showing that Ia+ M4 functioned as autonomous A cells. However, the T-cell preparation usually contained about 3% Ia+ cells which might play some role as A cells in the response. Indeed, the proliferative response of T cells was completely abrogated when the T-cell preparation was depleted of Iaf cells (Fig. 2C). These results strongly argue that Ia+ Mr$ do not function as autonomous A cells and that they induce T-cell proliferation in concert with a small number of Ia+ A cells in the T-cell preparation. This seemsto be the casealso with Ia- M4. SAC, comprising M4 in the majority and DC in the minority, exhibited A-cell activity for the proliferative response of T cells even after pretreatment of the T-cell preparation with anti-Ia antiserum plus C (Fig. 3). In order to identify the A cells in SAC, we prepared SAC in the presence of anti-Ia antiserum plus C or DC-specific 33Dl monoclonal antibody plus C during the process of adherent cell preparation. A-cell activity of SAC was almost completely abrogated after the above treatments (Fig. 3), indicating that Ia+ 33Dlf cells, i.e., DC, played an essential role in the response. Furthermore, A-cell activity was fully restored by supplementing the 33D 1+ cell-depleted SAC with a low dose DC which alone did not significantly induce a proliferative response.This indicates that splenic Mq5are not autonomous A cells but possessthe ability to enhance the DC-mediated proliferative response of T cells.
DC AND M&J IN THE PROLIFERATIVE
3H-TdR incorporation I 2 3 L
0 responder +SAC treated
RESPONSE OF T CELLS
5
( x10m4cpm 1
5 I
6
7
;I
with
C alo+C 33m+c 33Dl+C +1.3x104 DC +1.3x104m
FIG. 3. Contribution of DC to the manifestation of A-cell activity of SAC. T cells (2.5 X 106)deprived of Ia+ cells were cultured in the presence(open) or absence(closed) of 30 &ml KLH with DC and/or SAC which were untreated or pretreated with C alone, anti-Ia plus C, or anti-DC (33Dl) plus C. Cells were cultured on 16-mm dishes in 24-well plates. [3H]TdR incorporation proceeded for the last 6 hr of a 4-day culture.
DC Are Autonomous A Cells The experimental results in Figs. 2 and 3 argued that M$ were not autonomous A cells but collaborated with DC to manifest high A-cell activity. This, however, did not directly indicate the autonomy of DC as A cells. Thus, we performed an experiment in which the T-cell preparation treated with anti-Ia antibody plus C was cultured with KLH-pulsed or unpulsed DC or M@. Results shown in Fig. 4 clearly indicated that DC, in contrast to M+, manifested A-cell activity independently of the help of Mr$. The necessaryminimum dose of DC was 104to give a maximum response in 5 X lo5 T cells, or 2% of culture.
'
0
3 Number
6
IO
of cells
20
40
SO
(x~O-~)
FIG. 4. KLH-pulsed DC, but not Ia+ M+, do not induce proliferative response.T cells (5 X 10’) deprived of Ia+ cells were cultured with varying doses of DC or Me. DC (circles) and M+ (triangles) were pulsed with 100 &ml KLH (open) or unpulsed (closed) for 30 min at 37°C. [‘H]TdR incorporation proceeded for the last 18 hr of a 5-day culture.
6
KAWAI ET AL.
024802r,8024a unpulsed
pulsed at lOpg/ml la’M#/weU
pulsed at lOO~g/mt
(~10~~)
FIG. 5. Effect of antigen pulsing of DC and/or M+ on the enhancing activity of Me5in DC-mediated proliferative response. T cells (5 X 10’) deprived of la+ cells were cultured with unpulsed M4 (A), or with M4 puked with 10 (B) or 100 (C) &ml KLH in combination with unpulsed (a) or pulsed (100 pg/rnl KLH; 0) DC (5 X 103),or in the absence of DC (II). As a reference, T cells were cultured with unpulsed M$ and unpulsed DC in the presence of 30 pg/ml KLH (@). [3H]TdR was added 6 hr before the end of a 4-day culture.
Antigen Pulsing of Either DC or A44 Is Eflective in Inducing T-Cell Proliferation Results of an experiment in which 5 X lo5 T cells deprived of Ia+ cells were cultured with 5 X lo3 DC and/or 0 to 8 X lo4 Iaf M$, of which either or both were pulsed with IUH, are shown in Fig. 5. The result of culture with unpulsed DC and unpulsed M4 in the presence of free KLH is also shown in Fig. 5A as a reference. Iaf M$, even if KLH-pulsed, induced no significant T-cell proliferation in the absenceof DC, conforming to the result of preceding experiment (Fig. 4). A low dose (5 X 103)of DC pulsed with 100 pg/ml KLH induced only a marginal proliferative response. However, a substantial proliferative response was induced in the presence of pulsed DC in combination with unpulsed M$ (Fig. 5A). On the other hand, unpulsed DC induced a proliferative response in concert with Ia+ M@pulsed with 10 or 100 pg/ml KLH (Figs. 5B and 5C), and the level of responsewas dependent on the concentration of IUH for pulsing M$. The combination of pulsed DC with pulsed M$ was the most adequate for inducing a high response. The proliferative response induced by pulsed DC and M4 pulsed with 100 pg/ml KLH was the highest in this experiment (Fig. 5C) and comparable to that attained in the culture with unpulsed DC and unpulsed M$ in the presence of free antigens (Fig. 5A). Enhancing Activity of M$ Is Replaceable by Soluble Factor(s) Antigen-unpulsed Ia+ M+ augmented the proliferative responseof T cells mediated by pulsed DC (Fig. 5). Similar was the casewith unpulsed Ia- M$ (not shown). One of the implications of these results was the contribution of a antigen-nonspecific mechanism such as secretion of monokine(s) to enhance the response. The results shown in Fig. 6 supported this assumption. Latex-stimulated M4 culture supematant which had IL- 1 activity was employed as an monokine preparation. KLH-primed T
DC AND M$ IN THE PROLIFERATIVE
0
0
VI6 l/8 sup dilution
RESPONSE OF T CELLS
7
l/4
FIG. 6. The enhancing effect of IL- I on the DC-mediated proliferative response of T cells. T cells (5
X 105)deprived of Ia+ cells were cultured without (Cl) or with (2 X 103)DC which were unpulsed (A) or pulsed with 100 fig/ml KLH (0) or OVA (V). IL-l-containing medium was added to the culture at the final concentrations indicated on the abscissa.[3H]TdR was added 6 hr before the end of a 4-day culture.
cells deprived of Ia+ cells were cultured with KLH- or OVA-pulsed DC or unpulsed DC and IL- 1-containing M$ culture supernatant. The number of DC was fixed to be small (0.4% of the culture), so that both the KLH-specific response and syngeneic MLR were considerably low unless the monokine was added. The monokine itself did not trigger the KLH-primed T cells, but it augmented the KLH-specific proliferative response of T cells mediated by IUH-pulsed DC. Syngeneic MLR to unpulsed DC was also augmented, but the level it attained was approximately or less than one-third of that of the monokine-augmented IUH-specific response. The level of response to OVA-pulsed DC was similar to that of the response to unpulsed DC, indicating that IUH-primed T cells did not respond to OVA. Thus, the major part of the response in the presence of KLH-pulsed DC and monokine was attributed to the proliferation of IUH-specific T cells. Antigen-Pulsed Ia+ il44 Do Not Induce Prollyerative Response Even in the Presence of IL-2 Proliferative response of T cells occurred in the presence of IUH-pulsed Ia+ Md and a small number of unpulsed DC (Figs. 5B and 5C). At least two possible mechanisms were envisaged to explain this: (i) M$ transferred KLH to DC, and IUHspecific T cells responded to the IUH-carrying DC, or (ii) syngeneic MLR to DC resulted in the secretion of IL-2 from autoreactive T cells ( 12), whereby KLH-specific T cells became responsive to KLH-pulsed Ia+ M6. In order to check the latter possibility, we performed the experiment shown in Fig. 7. The results demonstrated that KLH-pulsed Ia+ M$ were unable to induce the proliferative response even in the presence of rIL-2 at lo-100 III/ml, though these concentrations of rIL-2 were sufficient to induce proliferation of CTLL, a murine IL-2-dependent cell line (data not shown).
KAWAI ET AL.
0
1
2
la’M#/well
4
8
(x10-4
FIG. 7. Inability of IL-2 to induce the proliferative responseof T cells in the presence of Ia+ Mb. T cells (5 X 10’) deprived of Ia+ cells were cultured with Ia+ Mb pulsed with 100 pp/rnl KLH in the presence of 10 (0), 33 (0), or 100 (0) IU/ml human rIL2. As a reference, T cells and pulsed Ia+ M& were cultured with 2 x lo3 unpulsed DC (Cl). [‘H]TdR incorporation proceeded for the last 6 hr of a 4&y culture.
Superiority of DC to Mq5as A Cells Even for Recently Primed T Cells The results presented hitherto in this paper strongly argue that M@ are much inferior to DC as A cells in the proliferative response of memory T cells of mice primed 6 weeks or more before. However, it remained to be investigated whether M$ would be competent as A cells if T cells are prepared from recently primed mice. Thus, lymph node T cells deprived of Ia+ cells were prepared from regional lymph nodes of mice immunized 12 days previously. As A-cell sources,we employed DC, Ia+ M4, and, in addition, peritoneal adherent cells containing resident M& The reason that we checked the resident M$ preparation, though somewhat crude, was that Kapsenberg et al. ( 13) reported A-cell activity of resident MI$ for recently primed T cells. Figure 8 shows that none of these cells served as potent A cells for the T-cell preparation. On the other hand, we fractionated the T-cell preparation by a discontinuous density gradient centrifugation into three fractions. Only the lightest and largest cell fraction, which was located at the interface between 50 and 60% Percoll, responded to IUH in the presence of A cells. T cells in the other two heavier cell fractions barely responded under the same conditions (data not shown). The A-cell activity was highest in DC, next highest in the resident M@ preparation, and actually inappreciable in Ia+ M4. DISCUSSION These experiments were conducted to investigate the role of DC and M$ in the antigen-specific proliferative response of immune T cells. We obtained no evidence that M$ function as autonomous A cells for memory T cells (Figs. 2 to 5, and 7).
DC AND MI$ IN THE PROLIFERATIVE
RESPONSE OF T CELLS
9
No. of DC or M4/well FIG. 8. The proliferative response of recently primed T cells supported by DC or a crude resident M@J population, but not by Ia+ Mqk Nylon-nonadherent lymph node T cells were prepared from mice primed 12 days previously, and unfractionated (closed symbols) or fractionated to obtain low-density large T cells (open symbols). T cells (5 X 10’) were cultured for 74 hr with varying numbers of DC (circles), crude resident Mg (triangles), or Ia+ peritoneal M$ (squares)in the presenceof 30 pglml KLH. [3H]TdR incorporation proceeded for the last 6 hr of culture.
This may also be the case for recently primed T cells, as will be discussed later. The apparent activity of Ia+ M4, and also Ia- M4, to induce proliferation of columnpurified T cells (Fig. 1) was caused by the synergy with Ia+ A cells which remained in the T-cell preparation. Thus, deprivation of Iaf cells from the T-cell preparation resulted in the abrogation of proliferative response of T cells even in the presence of IUH and Ia+ M&J(Fig. 2). On the other hand, it was not M@ but DC that played a key role in SAC to induce T-cell proliferation, since the A-cell activity of SAC was abrogated by selective depletion of DC and was restored by adding back a very small number of DC (Fig. 3). These results are consistent with the previous findings that M+ are not autonomous stimulators or A cells for unprimed lymphocytes in allogeneic and syngeneic MLR and the antibody response to T-dependent antigens, but they are able to augment the weak responsetriggered by a small number of DC (7,8, 14). There are several investigators who have argued that Ia+ M4 appear to be essential as autonomous A cells in the proliferative response of immune T cells (15- 17). They may have observed, however, the enhancing activity of Ia+ and Ia- M4, since contamination of a small number of DC in the Md and/or T cell preparations will result in the manifestation of synergy between DC and M4. In contrast to Mti, DC have been demonstrated to induce a strong proliferative response of antigen-primed T cells in the presence of soluble antigens (4, 16-18). IUH-pulsed DC also induce antigen-specific T-cell proliferation (Fig. 4). Pulsing of DC with KLH at 37°C was more efficient than that in the cold (data not shown). This may indicate that DC take up KLH through pinocytosis. Indeed, pinocytosis of M4 is diminished in the cold ( 19). On the other hand, commercially available IUH prep-
10
KAWAI
ET AL.
arations comprise heterogeneous molecules of similar size. Therefore, it may be possible that small peptide fragments included in KLH preparations, which bind to DC, play a major role in inducing T-cell proliferation. To check this possibility, we dialyzed KLH to eliminate small fragments of molecular weight less than 10 kDa. The dialysis of KLH, however, did not affect the immunogenicity of KLH in the presence of DC (data not shown). Taken together, it seemslikely that DC can take up intact KLH through pinocytosis and present antigenic determinants to antigen-specific T cells; although DC are known to be nonphagocytic cells, their pinocytosis is not evident (20), and the mechanism of antigen processing of DC, if they do, is also not clear. We obtained results which imply the contribution of monokines to the enhancement of the DC-mediated proliferative response. First, unpulsed Iaf M4 augmented the response elicited by KLH-pulsed DC (Fig. 5). Second, Ia- M$ were also able to augment the response mediated by DC and free antigen (Fig. 1). M+ are well known to secrete IL- 1 when stimulated by various stimulants or adherence (2 1). The result that M4 culture supernatant containing IL-l activity augments the level of T-cell proliferation in response to a small number of KLH-pulsed DC (Fig. 6) strongly suggeststhat IL- 1 from M4 enhances the T-cell response directly or indirectly by amplifying the function of DC (22). Through interaction with DC, T cells begin to produce IL-2 which amplifies T-cell proliferation (12). The case in the experiments of Figs. 5B and 5C might be similar. Therefore, we examined the effectiveness of IL-2 in the culture comprising T cells and antigen-pulsed Ia+ M4. Human rIl-2 did not replace DC function to manifest synergy with KLH-pulsed Ia+ M$ (Fig. 6), though the molecular specimen and concentration of rIL-2 in our experiment were appropriate to induce proliferation of a murine IL-2-dependent cell line. This may indicate that memory T cells require not only antigenic stimulation in association with Ia molecules but also some additional signal from DC to become responsive to IL-2, and antigen-pulsed M+ are incapable of providing this kind of signal to T cells in the resting phase. Once activated, T cells probably no longer require such a signal. In fact, Inaba and co-workers have demonstrated that in allogeneic MLR, DC-activated T-cell blasts are able to proliferate by stimulation with allogeneic Ia+ M4 or B cells as well as DC, whereas resting virgin or memory T cells proliferate only by stimulation with DC (23-25). The results shown in Figs. 1 to 7 in this paper dealt with the A cell for memory T cells from mice primed 6 weeks or more before. In viewing the discussion above, however, it can be presumed that not only DC but also Ia+ M$ would serve as A cells if T cells are obtained from recently primed mice. On the other hand, Kapsenberg et al. (13) have reported that lymph node T cells collected from mice primed with OVA or KLH 10 to 16 days previously resumed proliferation with the aid of resident peritoneal M$. Their M&Jpreparation, however, contained more than 20% non-M4 cells, so that we suspect whether their results could be interpreted by simply assuming M$ to be autonomous A cells. Therefore, we compared the ability of DC, a crude resident M4 preparation, and a pure Ia+ M4 preparation to recall the proliferation of T cells collected 12 days after immunization of mice. The results in Fig. 8 present no evidence in support of the A-cell activity of M$ even for the recently primed T-cell population. A considerably high rate proliferation was observed in the activated large T-cell population in the presence of either DC or resident MI#Jpreparation, but not in that of Ia+ M+. This may suggest that the result of Kapsenberg et al. (13) was
DC AND Mb IN THE PROLIFERATIVE
RESPONSE OF T CELLS
11
caused by the contamination of some autonomous non-M4 A cells in their resident Mq5 preparation. One possibility, however, that cannot be ruled out is that their Tcell preparation might have contained a relatively large number of highly activated T cells which can proliferate well only with the aid of M4 in the absence of DC, similar to the casewith those in allogeneic MLR as mentioned above (23-25). If so, the discrepancy between their results and our’s in respect of the responsiveness of the unfractionated whole population of recently primed lymph node T cells can be interpreted. ACKNOWLEDGMENTS We gratefully acknowledge the advice of Dr. Ralph M. Steinman of the Rockefeller University in the course of the preparation of this manuscript. Thanks are also due to all other members of our laboratory for their fruitful discussionsand technical assistance.
REFERENCES 1. Steinman, R. M., and Witmer, M. D., Proc. Natl. Acad. Sci. USA 75,5132, 1978. 2. Nussenzweig, M. C., and Steinman, R. M., J. Exp. Med. 151, 1196, 1980. 3. Nussenzweig, M. C., Steinman, R. M., Gutchinov, B., and Cohn, Z. A., J. Exp. Med. 152, 1070, 1980. 4. Sunshine, G. H., Katz, D. R., and Feldmann, M., J. Exp. Med. 152, 1817, 1980. 5. Steinman, R. M., Nogueira, N., Witmer, M. D., Tydings, J. D., and Mellman, I. S., J. Exp. Med. 152, 1248, 1980. 6. Inaba, K., Steinman, R. M., Van Voorhis, W. C., and Muramatsu, S., Proc. Natl. Acad. Sci. USA 80, 6041,1983. 7. Naito, K., Komatsubara, S., Kawai, J., Mori, K., and Muramatsu, S., Cell. Immunol. 88,361, 1984. 8. Komatsubara, S., Hirayama, Y., Inaba, K., Naito, K., Yoshida, K., Kawai, J., and Muramatsu, S., Cell. Immunol. 95,288, 1985. 9. Julius, M. H., Simpson, E., and Herzenberg, L. A., Eur. J. Immunol. 3,645, 1973. 10. Ly, I. A., and Mishell, R. A., J. Immunol. Methods 5,239, 1974. 11. Gery, I., Davies, P., Derr, J., Krett, N., and Ban-anger,J. A., Cell. Immunol. 64,293, 1981. 12. Austyn, J. M., Steinman, R. M., Weinstein, D. E., Granelli-Pipemo, A., and Palladino, M. A., J. Exp. Med. 157, 1101, 1983. 13. Kapsenberg, M. L., Teunissen, M. B. M., Stiekema, F. E. M., and Keizer, H. G., Eur. J. Immunol. 16, 345,1986. 14. Inaba, K., Nakano, K., and Muramatsu, S., J. Immunol. 127,452, 1981. 15. Lee, K.-C., and Wong, M., J. Immunol. 128,2487,1982. 16. Miyazaki, H., andosawa, T., Eur. J. Immunol. 13,984, 1983. 17. Guidos, C., Wong, M., and Lee, K.-C., J. Immunol. 133, 1119, 1984. 18. Kaye, P. M., Chain, B. M., andFeldmann, M., J. Immunol. 134, 1930, 1985. 19. Steinman, R. M., Mellman, 1.S., Muller, W. A., and Cohn, Z. A., J. Cell. Biol. 96, 1, 1983. 20. Steinman, R. M., and Nussenzweig, M. C., Immunol. Rev. 53, 127, 1980. 2 1. Unanue, E. R., Kiely, J.-M., and Calderon, J., J. Exp. Med. 144, 155, 1976. 22. Koide, S. L., Inaba, K., and Steinman, R. M., J. Exp. Med. 165,5 15, 1987. 23. Inaba, K., and Steinman, R. M., J. Exp. Med. 160, 17 17, 1984. 24. Inaba, K., Koide, S., and Steinman, R. M., Proc. Natl. Acad. Sci. USA 82,7686, 1985. 25. Inaba, K., and Steinman, R. M., Science 229,475, 1985.
IMMUNOLOGY
109, I- 11 ( 1987)
Role of Macrophages as Modulators but Not as Autonomous Accessory Cells in the Proliferative Response of Immune T Cells to Soluble Antigen’ JUN KAWAI,~ KAYO INABA, SHIN KOMATSUBARA, YOSHITAKA HIRAYAMA, KOJI NAITO, AND SHIGERU MURAMATSU~ Department of Zoology, Faculty of Science,Kyoto University, Kyoto 606, Japan Received September 4, 1986; acceptedMay 12. I987 The role of murine macrophages (M&) and that of splenic dendritic cells (DC) were investigated in the antigen-specific proliferative response of memory T cells of mice primed with keyhole limpet hemocyanin (KLH) 6 weeks or more before. Peritoneal Mq& whether expressing Ia antigens or not, did not function as autonomous accessorycells (A cells). A-cell activity of the spleen adherent cell population, which comprised M& in the majority and DC in the minority, was abolished by eliminating DC with a DC-specific monoclonal antibody and complement, and regained by the addition of a small number of DC. Though M@did not function as autonomous A cells, they augmented the proliferative response in the presence of a small number of DC. This occurred not only in the presence of free antigen, but also when DC and/or MI$ were pulsed with antigen. A culture supematant of M#Jhaving interleukin-1 activity was effective in enhancing the proliferation of T cells which responded to antigen-pulsed DC. On the other hand, interleukin-2 did not replace DC even in the presence of antigen-pulsed Ia+ M$J.We also investigated recently primed T cells, but no evidence was obtained in favor of the competence A Cells. 0 1987 Academic PWS, 1~. of M4 as aUtOnOtnOUS
INTRODUCTION Initiation of immune responsesis usually mediated by Ia-bearing (Ia’) accessory cells (A cells).4Dendritic cells (DC) are principal A cells in T-cell-dependent immune responses (l-8). We have so far demonstrated that Ia+ macrophages (Md) do not function as autonomous A cells but play a role as modulators in the primary mixedleukocyte reaction (MLR) and the primary antibody response to T-dependent antigens (6-8). Thus, these observations have been made solely in the primary immune response, so that it is necessaryfor us to check whether DC and M+ behave similarly ’ This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, and Culture of Japan, and the Shimizu Foundation Research Grant for 1984. 2 Present address:Shionogi Research Laboratories, Fukushima-ku, Osaka, 553 Japan. 3To whom correspondence should be addressed: Department of Zoology, Faculty of Science, Kyoto University, Kitashirakawa-Oiwakecho, Sakyo-ku, Kyoto 606, Japan. 4 Abbreviations used: A cells, accessory cells; C, complement; DC, dendritic cells; IL- 1, interleukin- I ; IL-2, interleukin-2; KLH, keyhole limpet hemocyanin; Mb, macrophages; MLR, mixed-leukocyte reaction; OVA, ovalbumin; SAC, spleen adherent cells.
0008-8749187$3.00 Copyright 0 1987 by Academic Press, Inc. Ail rights ofreproduction in any form reserved.
2
KAWAI
ET AL.
also in the secondary response. We investigated this in the antigen-specific proliferative response of memory T cells and also in that of recently primed T cells. The data to be presented below strongly argue that DC are competent, but are M@ virtually incompetent, to trigger the T cells, and that M@ modulate, augmentatively or suppressively, the level of the T-cell response triggered by DC. MATERIALS AND METHODS Mice. Male and female C3H/HeSlc (Shizuoka Agricultural Cooperative Association for Laboratory Animals, Shizuoka) and C3H/HeMs mice (Institute of Animal Experiments, Kyoto University) were used at the agesof 2-5 months. Mucrophuges. Pure populations of M4 were prepared from thioglycollate-elicited peritoneal exudate cells by the in vitro culture for 4 to 5 days as described (7). Ia molecules on M@ surfaces were induced by 100 IU/ml murine rIFN-7 (donated by Shionogi ResearchLaboratories, Osaka) or 20% CS (supernatant of spleen cell culture with concanavalin A) added to the culture medium, resulting in the Ia expression on more than 90% of Mr& In order to prepare spleen adherent cells (SAC), 3 X lo6 spleen cells were cultured in 24-well culture plates (A/S Nunc; Kamstrup, Roskilde) for 2 hr, and nonadherent cells were removed. The crude population of resident M$ containing a few percent of Ia+ cells was prepared as adherent cells after culturing resident peritoneal cells from normal mice for 5 hr. Dendritic cells. DC were prepared from collagenase-dispersed spleen cells as described (7). In the DC preparation, more than 90% of cells were Ia+, and no Fc receptor-bearing cells capable of forming rosettes with opsonized sheep erythrocytes were detectable. Memory T cells. Mice were immunized by injection into several subcutaneous sites of a total of 0.1-0.4 mg keyhole limpet hemocyanin (KLH; Calbiochem-Behring Corp., La Jolla, CA) emulsified in Freund’s complete adjuvant. After 6-10 weeks, draining lymph nodes were collected, treated with 0.5 mg/ml collagenase(Wako Pure Chemical Industries, Ltd., Osaka) and 30 pg/ml DNase I (Sigma Chemical Co., St. Louis, MO) for 30 min at 37°C and disrupted on stainless-steel sieves. Dispersed lymph node cells were incubated for 1 hr in plastic culture dishes (3003; Falcon, Oxnard, CA). T cells were enriched from the dish-nonadherent cells by a serial passagethrough a nylon-wool column (9) and a Sephadex G- 10 (Pharmacia Fine Chemicals, Uppsala) column (10). In some experiments, column-purified T-cell preparations were further treated with anti-Iak antiserum (A.TH anti-A.TL antiserum; Cedarlane Laboratories Ltd., Hornby, Ontario) plus complement (C) for 45 min at 37°C to completely eliminate Ia+ cells. Recently primed T cells. Lymph node cells were collected 12 days after KLH injection. Nylon-wool column-passed cells were prepared as the T-cell population and treated with anti-Iak plus C. Fractionation of the T-cell population was performed by centrifugation through step gradients of 50,60, and 70% Percoll (Pharmacia) at 1OOOg for 25 min at 5°C. The cells at each interface were collected with a Pasteur pipet, suspended in an excessamount of medium, and washed twice. Antigen pulsing. M4 on culture plates or DC in suspension were pulsed with 100 Kg/ml KLH for 30 min at 37°C and washed at least three times, unless otherwise described. Proliferative response o/T cells. Unless mentioned otherwise, 5 X lo5 purified T cells were cultured with varying numbers of M$ and/or DC in flat-bottomed 96-well
DC AND Mq3 IN THE PROLIFERATIVE
RESPONSE OF T CELLS
3
la%l+
la’M+ (xl0-4/well)
FIG. 1. Antigen-dependent proliferative responseofcolumn-purified T cells in the presenceof M$. KLHprimed memory T cells (5 X 105)prepared from lymph node cells by passagethrough nylon-wool and Sepbadex G- 10 columns were cultured in the presence (open) or absence (closed) of 50 rg/ml KLH with SAC or varying doses of Ia+ or Ia- peritoneal M$Jin modified Click’s medium containing 0.5% normal mouse serum on 96-well plates. Proliferation was measured by [3H]TdR incorporation during the last 18 hr ofa 5-day culture. Each symbol or column and vertical bar representsthe mean cpm oftriplicate cultures and the SEM, respectively.
plates in 0.2 ml of modified Click’s medium containing 0.5% normal mouse serum (7). M4 and DC were irradiated with 1500 R X-rays or treated with 40 pg/ml mitomytin C for 30 min at 37°C to prevent cell division prior to the culture. T-cell proliferation was evaluated by incorporation of tritiated thymidine ( [3H]TdR; New England Nuclear, Boston, MA) for 6-8 hr before cell harvest. IL-l-containing medium and recombinant IL-2. The culture supematant of M$ ingesting latex particles was used as the IL- 1-containing medium ( 11). Recombinant human IL-2 was donated by Dr. J. Hamuro (Ajinimoto Central Research Laboratories, Yokohama). RESULTS In this study, memory T cells and recently primed T cells from KLH-immunized mice were used in the experiments of Figs. 1 to 7, and in that of Fig. 8, respectively. In pilot experiments, we confirmed that these immune T cells proliferated specifically in response to KLH, but not to other unrelated antigens such as ovalbumin (OVA) under our conditions. Concentrations of KLH in the experiments described below were within the range of optimal concentrations estimated in preliminary experiments. Apparent A-Cell Activity of SAC, Ia+ M4, and la- M4 Nylon-wool and Sephadex G- 1O-passedcells (hereafter called T-cell preparation) were cultured with SAC, Ia+ Md, or Ia- M4 in the presence or absenceof KLH for 4 days. As shown in Fig. 1, the T-cell preparation was not responsive to KLH by itself, and supplementation of SAC or purified M$ was effective in supporting proliferation
4
KAWAI ET AL.
FIG.2. Memory T cells deprived of Ia+ accessorycells do not proliferate in the presence of antigen and Ia+ M$. Varying numbers of Ia+ peritoneal M4 were cultured in the presence (0) or absence (0) of 30 pg/ ml KLH with 5 X 10’ KLH-primed lymph node T cells which were pretreated with culture medium (A), with complement alone (B), or with anti-Ia antiserum plus complement (C). Cells were cultured for 4 days and [‘H]TdR incorporation proceeded for an additional 8 hr.
of T cells in response to KLH. A paradoxical result obtained in this experiment was that Ia- Mb appeared competent, though slightly inferior to Ia+ M4, as A cells in the proliferative response of memory T cells. The reason for this will be explained in the next experiment.
MacrophagesAre Not Autonomous A Cells In this experiment, the T-cell preparation was untreated or treated with anti-Ia antibody plus C or C alone, and cultured with Iaf M4 (Fig. 2). The T-cell preparation untreated or treated with C alone showed strong proliferation when cultured with Ia+ M$ (Figs. 2A and 2B), similar to the result in Fig. 1, showing that Ia+ M4 functioned as autonomous A cells. However, the T-cell preparation usually contained about 3% Ia+ cells which might play some role as A cells in the response. Indeed, the proliferative response of T cells was completely abrogated when the T-cell preparation was depleted of Iaf cells (Fig. 2C). These results strongly argue that Ia+ Mr$ do not function as autonomous A cells and that they induce T-cell proliferation in concert with a small number of Ia+ A cells in the T-cell preparation. This seemsto be the casealso with Ia- M4. SAC, comprising M4 in the majority and DC in the minority, exhibited A-cell activity for the proliferative response of T cells even after pretreatment of the T-cell preparation with anti-Ia antiserum plus C (Fig. 3). In order to identify the A cells in SAC, we prepared SAC in the presence of anti-Ia antiserum plus C or DC-specific 33Dl monoclonal antibody plus C during the process of adherent cell preparation. A-cell activity of SAC was almost completely abrogated after the above treatments (Fig. 3), indicating that Ia+ 33Dlf cells, i.e., DC, played an essential role in the response. Furthermore, A-cell activity was fully restored by supplementing the 33D 1+ cell-depleted SAC with a low dose DC which alone did not significantly induce a proliferative response.This indicates that splenic Mq5are not autonomous A cells but possessthe ability to enhance the DC-mediated proliferative response of T cells.
DC AND M&J IN THE PROLIFERATIVE
3H-TdR incorporation I 2 3 L
0 responder +SAC treated
RESPONSE OF T CELLS
5
( x10m4cpm 1
5 I
6
7
;I
with
C alo+C 33m+c 33Dl+C +1.3x104 DC +1.3x104m
FIG. 3. Contribution of DC to the manifestation of A-cell activity of SAC. T cells (2.5 X 106)deprived of Ia+ cells were cultured in the presence(open) or absence(closed) of 30 &ml KLH with DC and/or SAC which were untreated or pretreated with C alone, anti-Ia plus C, or anti-DC (33Dl) plus C. Cells were cultured on 16-mm dishes in 24-well plates. [3H]TdR incorporation proceeded for the last 6 hr of a 4-day culture.
DC Are Autonomous A Cells The experimental results in Figs. 2 and 3 argued that M$ were not autonomous A cells but collaborated with DC to manifest high A-cell activity. This, however, did not directly indicate the autonomy of DC as A cells. Thus, we performed an experiment in which the T-cell preparation treated with anti-Ia antibody plus C was cultured with KLH-pulsed or unpulsed DC or M@. Results shown in Fig. 4 clearly indicated that DC, in contrast to M+, manifested A-cell activity independently of the help of Mr$. The necessaryminimum dose of DC was 104to give a maximum response in 5 X lo5 T cells, or 2% of culture.
'
0
3 Number
6
IO
of cells
20
40
SO
(x~O-~)
FIG. 4. KLH-pulsed DC, but not Ia+ M+, do not induce proliferative response.T cells (5 X 10’) deprived of Ia+ cells were cultured with varying doses of DC or Me. DC (circles) and M+ (triangles) were pulsed with 100 &ml KLH (open) or unpulsed (closed) for 30 min at 37°C. [‘H]TdR incorporation proceeded for the last 18 hr of a 5-day culture.
6
KAWAI ET AL.
024802r,8024a unpulsed
pulsed at lOpg/ml la’M#/weU
pulsed at lOO~g/mt
(~10~~)
FIG. 5. Effect of antigen pulsing of DC and/or M+ on the enhancing activity of Me5in DC-mediated proliferative response. T cells (5 X 10’) deprived of la+ cells were cultured with unpulsed M4 (A), or with M4 puked with 10 (B) or 100 (C) &ml KLH in combination with unpulsed (a) or pulsed (100 pg/rnl KLH; 0) DC (5 X 103),or in the absence of DC (II). As a reference, T cells were cultured with unpulsed M$ and unpulsed DC in the presence of 30 pg/ml KLH (@). [3H]TdR was added 6 hr before the end of a 4-day culture.
Antigen Pulsing of Either DC or A44 Is Eflective in Inducing T-Cell Proliferation Results of an experiment in which 5 X lo5 T cells deprived of Ia+ cells were cultured with 5 X lo3 DC and/or 0 to 8 X lo4 Iaf M$, of which either or both were pulsed with IUH, are shown in Fig. 5. The result of culture with unpulsed DC and unpulsed M4 in the presence of free KLH is also shown in Fig. 5A as a reference. Iaf M$, even if KLH-pulsed, induced no significant T-cell proliferation in the absenceof DC, conforming to the result of preceding experiment (Fig. 4). A low dose (5 X 103)of DC pulsed with 100 pg/ml KLH induced only a marginal proliferative response. However, a substantial proliferative response was induced in the presence of pulsed DC in combination with unpulsed M$ (Fig. 5A). On the other hand, unpulsed DC induced a proliferative response in concert with Ia+ M@pulsed with 10 or 100 pg/ml KLH (Figs. 5B and 5C), and the level of responsewas dependent on the concentration of IUH for pulsing M$. The combination of pulsed DC with pulsed M$ was the most adequate for inducing a high response. The proliferative response induced by pulsed DC and M4 pulsed with 100 pg/ml KLH was the highest in this experiment (Fig. 5C) and comparable to that attained in the culture with unpulsed DC and unpulsed M$ in the presence of free antigens (Fig. 5A). Enhancing Activity of M$ Is Replaceable by Soluble Factor(s) Antigen-unpulsed Ia+ M+ augmented the proliferative responseof T cells mediated by pulsed DC (Fig. 5). Similar was the casewith unpulsed Ia- M$ (not shown). One of the implications of these results was the contribution of a antigen-nonspecific mechanism such as secretion of monokine(s) to enhance the response. The results shown in Fig. 6 supported this assumption. Latex-stimulated M4 culture supematant which had IL- 1 activity was employed as an monokine preparation. KLH-primed T
DC AND M$ IN THE PROLIFERATIVE
0
0
VI6 l/8 sup dilution
RESPONSE OF T CELLS
7
l/4
FIG. 6. The enhancing effect of IL- I on the DC-mediated proliferative response of T cells. T cells (5
X 105)deprived of Ia+ cells were cultured without (Cl) or with (2 X 103)DC which were unpulsed (A) or pulsed with 100 fig/ml KLH (0) or OVA (V). IL-l-containing medium was added to the culture at the final concentrations indicated on the abscissa.[3H]TdR was added 6 hr before the end of a 4-day culture.
cells deprived of Ia+ cells were cultured with KLH- or OVA-pulsed DC or unpulsed DC and IL- 1-containing M$ culture supernatant. The number of DC was fixed to be small (0.4% of the culture), so that both the KLH-specific response and syngeneic MLR were considerably low unless the monokine was added. The monokine itself did not trigger the KLH-primed T cells, but it augmented the KLH-specific proliferative response of T cells mediated by IUH-pulsed DC. Syngeneic MLR to unpulsed DC was also augmented, but the level it attained was approximately or less than one-third of that of the monokine-augmented IUH-specific response. The level of response to OVA-pulsed DC was similar to that of the response to unpulsed DC, indicating that IUH-primed T cells did not respond to OVA. Thus, the major part of the response in the presence of KLH-pulsed DC and monokine was attributed to the proliferation of IUH-specific T cells. Antigen-Pulsed Ia+ il44 Do Not Induce Prollyerative Response Even in the Presence of IL-2 Proliferative response of T cells occurred in the presence of IUH-pulsed Ia+ Md and a small number of unpulsed DC (Figs. 5B and 5C). At least two possible mechanisms were envisaged to explain this: (i) M$ transferred KLH to DC, and IUHspecific T cells responded to the IUH-carrying DC, or (ii) syngeneic MLR to DC resulted in the secretion of IL-2 from autoreactive T cells ( 12), whereby KLH-specific T cells became responsive to KLH-pulsed Ia+ M6. In order to check the latter possibility, we performed the experiment shown in Fig. 7. The results demonstrated that KLH-pulsed Ia+ M$ were unable to induce the proliferative response even in the presence of rIL-2 at lo-100 III/ml, though these concentrations of rIL-2 were sufficient to induce proliferation of CTLL, a murine IL-2-dependent cell line (data not shown).
KAWAI ET AL.
0
1
2
la’M#/well
4
8
(x10-4
FIG. 7. Inability of IL-2 to induce the proliferative responseof T cells in the presence of Ia+ Mb. T cells (5 X 10’) deprived of Ia+ cells were cultured with Ia+ Mb pulsed with 100 pp/rnl KLH in the presence of 10 (0), 33 (0), or 100 (0) IU/ml human rIL2. As a reference, T cells and pulsed Ia+ M& were cultured with 2 x lo3 unpulsed DC (Cl). [‘H]TdR incorporation proceeded for the last 6 hr of a 4&y culture.
Superiority of DC to Mq5as A Cells Even for Recently Primed T Cells The results presented hitherto in this paper strongly argue that M@ are much inferior to DC as A cells in the proliferative response of memory T cells of mice primed 6 weeks or more before. However, it remained to be investigated whether M$ would be competent as A cells if T cells are prepared from recently primed mice. Thus, lymph node T cells deprived of Ia+ cells were prepared from regional lymph nodes of mice immunized 12 days previously. As A-cell sources,we employed DC, Ia+ M4, and, in addition, peritoneal adherent cells containing resident M& The reason that we checked the resident M$ preparation, though somewhat crude, was that Kapsenberg et al. ( 13) reported A-cell activity of resident MI$ for recently primed T cells. Figure 8 shows that none of these cells served as potent A cells for the T-cell preparation. On the other hand, we fractionated the T-cell preparation by a discontinuous density gradient centrifugation into three fractions. Only the lightest and largest cell fraction, which was located at the interface between 50 and 60% Percoll, responded to IUH in the presence of A cells. T cells in the other two heavier cell fractions barely responded under the same conditions (data not shown). The A-cell activity was highest in DC, next highest in the resident M@ preparation, and actually inappreciable in Ia+ M4. DISCUSSION These experiments were conducted to investigate the role of DC and M$ in the antigen-specific proliferative response of immune T cells. We obtained no evidence that M$ function as autonomous A cells for memory T cells (Figs. 2 to 5, and 7).
DC AND MI$ IN THE PROLIFERATIVE
RESPONSE OF T CELLS
9
No. of DC or M4/well FIG. 8. The proliferative response of recently primed T cells supported by DC or a crude resident M@J population, but not by Ia+ Mqk Nylon-nonadherent lymph node T cells were prepared from mice primed 12 days previously, and unfractionated (closed symbols) or fractionated to obtain low-density large T cells (open symbols). T cells (5 X 10’) were cultured for 74 hr with varying numbers of DC (circles), crude resident Mg (triangles), or Ia+ peritoneal M$ (squares)in the presenceof 30 pglml KLH. [3H]TdR incorporation proceeded for the last 6 hr of culture.
This may also be the case for recently primed T cells, as will be discussed later. The apparent activity of Ia+ M4, and also Ia- M4, to induce proliferation of columnpurified T cells (Fig. 1) was caused by the synergy with Ia+ A cells which remained in the T-cell preparation. Thus, deprivation of Iaf cells from the T-cell preparation resulted in the abrogation of proliferative response of T cells even in the presence of IUH and Ia+ M&J(Fig. 2). On the other hand, it was not M@ but DC that played a key role in SAC to induce T-cell proliferation, since the A-cell activity of SAC was abrogated by selective depletion of DC and was restored by adding back a very small number of DC (Fig. 3). These results are consistent with the previous findings that M+ are not autonomous stimulators or A cells for unprimed lymphocytes in allogeneic and syngeneic MLR and the antibody response to T-dependent antigens, but they are able to augment the weak responsetriggered by a small number of DC (7,8, 14). There are several investigators who have argued that Ia+ M4 appear to be essential as autonomous A cells in the proliferative response of immune T cells (15- 17). They may have observed, however, the enhancing activity of Ia+ and Ia- M4, since contamination of a small number of DC in the Md and/or T cell preparations will result in the manifestation of synergy between DC and M4. In contrast to Mti, DC have been demonstrated to induce a strong proliferative response of antigen-primed T cells in the presence of soluble antigens (4, 16-18). IUH-pulsed DC also induce antigen-specific T-cell proliferation (Fig. 4). Pulsing of DC with KLH at 37°C was more efficient than that in the cold (data not shown). This may indicate that DC take up KLH through pinocytosis. Indeed, pinocytosis of M4 is diminished in the cold ( 19). On the other hand, commercially available IUH prep-
10
KAWAI
ET AL.
arations comprise heterogeneous molecules of similar size. Therefore, it may be possible that small peptide fragments included in KLH preparations, which bind to DC, play a major role in inducing T-cell proliferation. To check this possibility, we dialyzed KLH to eliminate small fragments of molecular weight less than 10 kDa. The dialysis of KLH, however, did not affect the immunogenicity of KLH in the presence of DC (data not shown). Taken together, it seemslikely that DC can take up intact KLH through pinocytosis and present antigenic determinants to antigen-specific T cells; although DC are known to be nonphagocytic cells, their pinocytosis is not evident (20), and the mechanism of antigen processing of DC, if they do, is also not clear. We obtained results which imply the contribution of monokines to the enhancement of the DC-mediated proliferative response. First, unpulsed Iaf M4 augmented the response elicited by KLH-pulsed DC (Fig. 5). Second, Ia- M$ were also able to augment the response mediated by DC and free antigen (Fig. 1). M+ are well known to secrete IL- 1 when stimulated by various stimulants or adherence (2 1). The result that M4 culture supernatant containing IL-l activity augments the level of T-cell proliferation in response to a small number of KLH-pulsed DC (Fig. 6) strongly suggeststhat IL- 1 from M4 enhances the T-cell response directly or indirectly by amplifying the function of DC (22). Through interaction with DC, T cells begin to produce IL-2 which amplifies T-cell proliferation (12). The case in the experiments of Figs. 5B and 5C might be similar. Therefore, we examined the effectiveness of IL-2 in the culture comprising T cells and antigen-pulsed Ia+ M4. Human rIl-2 did not replace DC function to manifest synergy with KLH-pulsed Ia+ M$ (Fig. 6), though the molecular specimen and concentration of rIL-2 in our experiment were appropriate to induce proliferation of a murine IL-2-dependent cell line. This may indicate that memory T cells require not only antigenic stimulation in association with Ia molecules but also some additional signal from DC to become responsive to IL-2, and antigen-pulsed M+ are incapable of providing this kind of signal to T cells in the resting phase. Once activated, T cells probably no longer require such a signal. In fact, Inaba and co-workers have demonstrated that in allogeneic MLR, DC-activated T-cell blasts are able to proliferate by stimulation with allogeneic Ia+ M4 or B cells as well as DC, whereas resting virgin or memory T cells proliferate only by stimulation with DC (23-25). The results shown in Figs. 1 to 7 in this paper dealt with the A cell for memory T cells from mice primed 6 weeks or more before. In viewing the discussion above, however, it can be presumed that not only DC but also Ia+ M$ would serve as A cells if T cells are obtained from recently primed mice. On the other hand, Kapsenberg et al. (13) have reported that lymph node T cells collected from mice primed with OVA or KLH 10 to 16 days previously resumed proliferation with the aid of resident peritoneal M$. Their M&Jpreparation, however, contained more than 20% non-M4 cells, so that we suspect whether their results could be interpreted by simply assuming M$ to be autonomous A cells. Therefore, we compared the ability of DC, a crude resident M4 preparation, and a pure Ia+ M4 preparation to recall the proliferation of T cells collected 12 days after immunization of mice. The results in Fig. 8 present no evidence in support of the A-cell activity of M$ even for the recently primed T-cell population. A considerably high rate proliferation was observed in the activated large T-cell population in the presence of either DC or resident MI#Jpreparation, but not in that of Ia+ M+. This may suggest that the result of Kapsenberg et al. (13) was
DC AND Mb IN THE PROLIFERATIVE
RESPONSE OF T CELLS
11
caused by the contamination of some autonomous non-M4 A cells in their resident Mq5 preparation. One possibility, however, that cannot be ruled out is that their Tcell preparation might have contained a relatively large number of highly activated T cells which can proliferate well only with the aid of M4 in the absence of DC, similar to the casewith those in allogeneic MLR as mentioned above (23-25). If so, the discrepancy between their results and our’s in respect of the responsiveness of the unfractionated whole population of recently primed lymph node T cells can be interpreted. ACKNOWLEDGMENTS We gratefully acknowledge the advice of Dr. Ralph M. Steinman of the Rockefeller University in the course of the preparation of this manuscript. Thanks are also due to all other members of our laboratory for their fruitful discussionsand technical assistance.
REFERENCES 1. Steinman, R. M., and Witmer, M. D., Proc. Natl. Acad. Sci. USA 75,5132, 1978. 2. Nussenzweig, M. C., and Steinman, R. M., J. Exp. Med. 151, 1196, 1980. 3. Nussenzweig, M. C., Steinman, R. M., Gutchinov, B., and Cohn, Z. A., J. Exp. Med. 152, 1070, 1980. 4. Sunshine, G. H., Katz, D. R., and Feldmann, M., J. Exp. Med. 152, 1817, 1980. 5. Steinman, R. M., Nogueira, N., Witmer, M. D., Tydings, J. D., and Mellman, I. S., J. Exp. Med. 152, 1248, 1980. 6. Inaba, K., Steinman, R. M., Van Voorhis, W. C., and Muramatsu, S., Proc. Natl. Acad. Sci. USA 80, 6041,1983. 7. Naito, K., Komatsubara, S., Kawai, J., Mori, K., and Muramatsu, S., Cell. Immunol. 88,361, 1984. 8. Komatsubara, S., Hirayama, Y., Inaba, K., Naito, K., Yoshida, K., Kawai, J., and Muramatsu, S., Cell. Immunol. 95,288, 1985. 9. Julius, M. H., Simpson, E., and Herzenberg, L. A., Eur. J. Immunol. 3,645, 1973. 10. Ly, I. A., and Mishell, R. A., J. Immunol. Methods 5,239, 1974. 11. Gery, I., Davies, P., Derr, J., Krett, N., and Ban-anger,J. A., Cell. Immunol. 64,293, 1981. 12. Austyn, J. M., Steinman, R. M., Weinstein, D. E., Granelli-Pipemo, A., and Palladino, M. A., J. Exp. Med. 157, 1101, 1983. 13. Kapsenberg, M. L., Teunissen, M. B. M., Stiekema, F. E. M., and Keizer, H. G., Eur. J. Immunol. 16, 345,1986. 14. Inaba, K., Nakano, K., and Muramatsu, S., J. Immunol. 127,452, 1981. 15. Lee, K.-C., and Wong, M., J. Immunol. 128,2487,1982. 16. Miyazaki, H., andosawa, T., Eur. J. Immunol. 13,984, 1983. 17. Guidos, C., Wong, M., and Lee, K.-C., J. Immunol. 133, 1119, 1984. 18. Kaye, P. M., Chain, B. M., andFeldmann, M., J. Immunol. 134, 1930, 1985. 19. Steinman, R. M., Mellman, 1.S., Muller, W. A., and Cohn, Z. A., J. Cell. Biol. 96, 1, 1983. 20. Steinman, R. M., and Nussenzweig, M. C., Immunol. Rev. 53, 127, 1980. 2 1. Unanue, E. R., Kiely, J.-M., and Calderon, J., J. Exp. Med. 144, 155, 1976. 22. Koide, S. L., Inaba, K., and Steinman, R. M., J. Exp. Med. 165,5 15, 1987. 23. Inaba, K., and Steinman, R. M., J. Exp. Med. 160, 17 17, 1984. 24. Inaba, K., Koide, S., and Steinman, R. M., Proc. Natl. Acad. Sci. USA 82,7686, 1985. 25. Inaba, K., and Steinman, R. M., Science 229,475, 1985.