Naive T cells recirculate mainly within the secondary lymphoid compartment, but once activated they can enter peripheral tissues and perform effector functions. To activate naive T cells, foreign antigens must traffic from the site of infection to the draining lymph nodes, where they can be presented by professional antigen presenting cells. For major histocompatibility complex class I–restricted presentation to CD8+ T cells, this can occur via the cross-presentation pathway. Here, we investigated the conditions allowing antigen access to this pathway. We show that the level of antigen expressed by peripheral tissues must be relatively high to facilitate cross-presentation to naive CD8+ T cells. Below this level, peripheral antigens did not stimulate by cross-presentation and were ignored by naive CD8+ T cells, although they could sensitize tissue cells for destruction by activated cytotoxic T lymphocytes (CTLs). Interestingly, CTL-mediated tissue destruction facilitated cross-presentation of low dose antigens for activation of naive CD8+ T cells. This represents the first in vivo evidence that cellular destruction can enhance access of exogenous antigens to the cross-presentation pathway. These data indicate that the cross-presentation pathway focuses on high dose antigens and those released during tissue destruction.

CD8+ CTLs, which recognize antigens presented by class I molecules encoded by the MHC, are important for immunity to viruses and other intracellular pathogens. Access to the class I–restricted presentation pathway generally requires that protein antigens are expressed within the cytoplasmic compartment of the presenting cell (1). This pathway has been described as the endogenous pathway, contrasting it with the class II–restricted pathway, which generally presents exogenous (extracellular) antigens.

Despite the apparent demarcation between the class I– and class II–restricted pathways, Bevan demonstrated over 20 years ago that under certain circumstances exogenous antigens can access the class I–restricted pathway (2). Such “cross-presentation” has now been observed in many tolerogenic as well as immunogenic responses (214). It has been proposed that the role of cross-presentation is to protect against tissue-tropic viruses that, in the absence of infection of professional APCs, would not access the class I–restricted pathway of these cells, and would therefore fail to prime naive CTLs (15, 16). In this model, the advantage of the cross-priming APC is that it can capture antigens from virus-infected tissues, process them into the class I pathway, and present them to naive CTLs in the draining LNs.

Recently, we used transgenic expression of the model antigen OVA to provide direct evidence that peripheral tissue antigens can be transported to draining LNs for class I–restricted presentation by a bone marrow–derived APC (10). In this earlier study, transgenic mice expressing membrane-bound OVA (mOVA) under the control of the rat insulin promoter (RIP) in the pancreas, kidney, and thymus were injected with class I–restricted, OVA-specific, CD8+ T cells from the OT-I transgenic line. When responses were assessed several days later, OT-I cells had proliferated and accumulated in nodes draining OVA-expressing tissues. Furthermore, antigen was presented by a bone marrow–derived cell, implying that tissue antigens were captured and then processed by a professional APC (the cross-presenting APC).

In this report, we examined the requirements for antigen entry into the cross-presentation pathway, using various transgenic mice expressing the model antigen OVA at different levels in the pancreas. Our studies show that only relatively high dose antigens constitutively gain effective entry into this pathway, but that cellular destruction may enhance access of low dose antigens. The implications of these findings with respect to immunity and tolerance are discussed.

Mice.

All mice were bred and maintained at the Walter and Eliza Hall Institute for Medical Research. OT-I, RIP-mOVA, and RIP-OVAlo transgenic mice have been described previously (10, 17). Note that RIP-OVAlo mice were referred to previously as RIP-OVA mice (17). RIP-OVAhi mice were generated using the same techniques.

Adoptive Transfer and FACS® Analysis.

Preparation and adoptive transfer of OT-I cells, 5,6-carboxy-succinimidyl-fluorescein-ester (CFSE)-labeling, and analysis on a FACScan® (Becton Dickinson, Mountain View, CA) were carried out as described previously (11, 18, 19). PE-conjugated anti-CD8 (YTS 169.4) was from Caltag Labs (South San Francisco, CA). Dead cells were excluded by propidium iodide staining.

Activation of OT-I Cells In Vitro.

OT-I cells were activated by culturing 107 OT-I spleen cells for 5 d in 30 ml of medium with 108 irradiated (1,500 cGy) B6 spleen cells previously coated for 30 min at 5 × 107 cell/ml with 0.2 μg/ml OVA257–264 peptide. Stimulator cells were washed once after peptide pulsing. Culture medium consisted of mouse tonicity RPMI1640, 10% fetal calf serum, 2 mM glutamine, 5 × 10−5 M 2-ME, and antibiotics.

Antigen Dose Affects Cross-presentation.

To examine the role of antigen dose in cross-presentation of tissue antigens, two lines of transgenic mice were produced expressing different amounts of secreted OVA in the β cells of the pancreas, under the control of the RIP. These were termed RIP-OVAlo and RIP-OVAhi mice. RIP-OVAhi mice expressed sufficient OVA to show weak OVA-specific staining immunohistologically, whereas OVA expression by RIP-OVAlo mice could not be detected by this technique (Fig. 1). However, expression of OVA by the latter line has been shown previously using alternative approaches (reference 17 and see Fig. 3).

To investigate the effect of antigen dose on the access of peripheral tissue antigens to the cross-presentation pathway, naive CD8+ cells from the OT-I cell line were labeled with the fluorescent dye CFSE (11, 18, 19) and injected intravenously into both RIP-OVAlo and RIP-OVAhi mice. 3 d later, the pancreatic (draining) and inguinal (nondraining) LNs were examined for proliferating OT-I cells. When CFSE-labeled cells proliferate, their fluorescence is equally distributed between daughter cells resulting in a 2n-fold reduction in fluorescence intensity, where n is the number of cell divisions. Adoptive transfer of CFSE-labeled OT-I cells revealed that only RIP-OVAhi mice were able to supply sufficient OVA to the cross-presentation pathway for stimulation of naive OT-I cells in the pancreatic draining LN (Fig. 2). These findings indicated that antigen expression above a critical level was required for antigen access to the cross-presentation pathway. Neither line of mice became diabetic when 5 × 106 naive OT-I cells were adoptively transferred (data not shown).

Lower Levels of Tissue Antigen Expression that Do Not Facilitate Cross-presentation May Still Be of Physiological Relevance.

To determine whether the lower antigen dose expressed by RIP-OVAlo mice was of physiological relevance, we asked whether activated CTLs, which have the capacity to enter tissues, could recognize tissue cells expressing this low dose of OVA. RIP-OVAlo mice were injected with 107 activated OT-I cells (generated by stimulation with antigen for 5 d in vitro) and then examined for the onset of diabetes (Fig. 3). As a control, RIP-OVAhi mice were also injected with activated cells. Diabetes induction in the RIP-OVAlo mice indicated that the low level of OVA expressed by these mice was sufficient to target islet β cells for recognition and destruction by activated CTLs. Thus, although only high doses of OVA stimulated naive CTLs via the cross-presenting pathway, lower doses were able to target peripheral tissues for destruction by activated CTLs.

Tissue Destruction Facilitates Cross-presentation of Low Dose Antigens.

If, as proposed, the cross-presentation pathway is important for virus immunity, then it makes sense to focus on high dose antigens, since virus proteins are likely to be produced in relatively large amounts during replicative infections. An alternative way to bias the immune system towards detecting virus antigens would be to preferentially present antigens released during tissue destruction. With this in mind, we asked whether destruction of β cells expressing low doses of OVA would improve antigen access to the cross-presentation pathway. RIP-OVAlo mice were injected intravenously with in vitro–activated OT-I cells and then 4 d later with CFSE-labeled naive OT-I cells or B6 CD8+ T cells. The activated OT-I cells were used to specifically cause islet destruction (as shown by the induction of diabetes in Fig. 3), and the CFSE-labeled cells were used to measure the response of naive cells to antigen (OVA) cross-presented in the draining LNs. 3 d after the transfer of CFSE-labeled cells (that is, 7 d after the transfer of activated CTLs), the LNs of these mice were recovered and examined for proliferating cells (Fig. 4). As shown, naive CFSE-labeled OT-I cells proliferated in the pancreatic (draining) LNs of RIP-OVAlo mice that were previously injected with activated CTLs (Fig. 4, C and D), but not in untreated mice (Fig. 4, A and B), even if left for up to 8 d in vivo (data not shown). By contrast, CFSE-labeled B6 CD8+ T cells did not proliferate in the pancreatic LNs in either the presence (Fig. 4, E and F) or absence (data not shown) of activated CTLs. Thus, an antigen-specific response of OT-I cells was observed after the destruction of β cells by activated CTLs. When proliferation was examined in bm1→ RIP-mOVA.B6 bone marrow chimeras, where the bone marrow compartment expressed an inappropriate MHC haplotype for presentation of OVA to OT-I cells (10), proliferation was abolished (data not shown). This latter point indicates that a bone marrow– derived APC was responsible for capturing and cross-presenting OVA released by cellular destruction.

To examine the kinetics of cross-presentation after cellular destruction, RIP-OVAlo mice were treated with activated CTLs to cause islet damage and then injected with CFSE-labeled naive OT-I cells either 1 or 4 d later. In previous studies we had shown that it takes 24 h for OT-I cells to begin proliferating after exposure to cross-presented antigen (11). Thus, we allowed the CFSE-labeled naive T cells to remain in these mice for a total of 3 d after transfer to ensure that there was enough time to adequately detect proliferation. Mice were killed and their pancreatic (draining) and inguinal (nondraining) LNs examined for proliferating OT-I cells (Fig. 4). No proliferation was seen in the pancreatic LNs from mice receiving CSFE-labeled cells on day 1 after initiation of β cell damage (Fig. 4, G and H), whereas considerable proliferation was evident in the day 4 group (Fig. 4, C and D). From these experiments, we can roughly estimate the minimum time necessary for antigen to appear in LNs after OT-I cell-induced β cell damage. Simply, since it takes at least 24 h for the cells to start proliferating after exposure to cross-presented antigen (11), then antigen must reach the LNs no earlier than day 3 after induction of tissue damage based on the lack of proliferation found in Fig. 4,G, and no later than day 6 given the proliferation shown in Fig. 4 C.

As previously reported, antigens expressed in peripheral tissues can be captured by bone marrow–derived APCs, processed via the cross-presentation pathway, and then presented in the draining LNs to naive CD8+ T cells (10). In this report, we show that peripheral tissue antigens must be expressed at relatively high levels to be cross-presented constitutively. When antigens were expressed at high doses, as in RIP-OVAhi mice (Fig. 2), they were cross-presented in the draining LNs and caused activation and proliferation of naive CD8+ T cells. Antigen expression levels below a certain threshold, as seen in RIP-OVAlo mice, were ignored by naive CD8+ T cells, which failed to proliferate in the draining LNs (Fig. 2).

Here, we also show that antigen access to the cross-presentation pathway can result after cellular destruction by CTL (Fig. 4). How CTL lysis allows antigen entry into this pathway is unknown, but it has recently been reported that apoptotic cells can be captured and presented by myeloid dendritic cells (20). This raises the possibility that apoptosis of target cells by activated CTLs may facilitate access to the cross-presentation pathway.

One of the critical observations of this report is that although lower dose antigens were unable to stimulate via the cross-presentation pathway, they still directly sensitized tissue (islet) cells, in which they were expressed, for recognition by activated CTLs. However, the direct recognition pathway is only available for recognition by activated CTLs, since only activated CTLs will have the capacity to leave the secondary lymphoid compartment and interact with these tissues. Thus, there appear to be three categories into which peripheral tissue antigens fall: category 1, those that are expressed at high levels and can be both cross-presented to naive CTLs and directly seen by activated CTLs; category 2, those that are expressed at lower concentrations and are nonstimulatory via the cross-presentation pathway, but can sensitize peripheral tissue cells for direct recognition by activated CTLs; and category 3, those antigens that can never be seen by CTLs.

The importance of these categories becomes apparent when we consider the physiological roles of the cross-presentation pathway. Recently, we provided evidence that cross-presentation of self-antigens leads to peripheral deletion of autoreactive CD8+ T cells (11). We showed that adoptively transferred OT-I cells were activated by cross-presented tissue antigens in RIP-mOVA mice, leading to proliferation and then deletion of the responding OT-I cells. If, as we have shown (11), self-tolerance is a consequence of cross-presentation of self-antigens, then only those antigens expressed at levels sufficiently high to be cross-presented will induce such tolerance. Thus, the idea that peripheral antigens are ignored (2123), may only apply to antigens that do not stimulate by cross-presentation, i.e., category 2 antigens. High dose self-antigens (category 1) would induce deletional tolerance via the cross-presentation pathway, whereas those lower dose antigens in category 2 would be ignored by naive CTLs but still able to sensitize peripheral tissues for recognition by activated CTLs. Perhaps this explains why some model self-antigens expressed by islet β cells were reported to induce tolerance (11, 24, 25), whereas others were ignored (2123). Analysis of the effect of antigen dose on peripheral tolerance will be the subject of a later report (Kurts, C., J.F.A.P. Miller, F.R. Carbone, and W.R. Heath, manuscript in preparation).

The tolerance induced by cross-presented self-antigens contrasts with the immunity seen when foreign antigens were used to induce CTLs by cross-priming (25, 79, 13, 14). What determines whether tolerance or immunity is induced is unclear, but CD4+ T cell help appears to be important (12, 14, 26). Since the cross-presentation pathway is involved in both cross-tolerance and cross-priming, the mechanisms by which antigens gain access to this pathway is likely to affect both aspects. The focus of this pathway on high dose antigens may direct CTLs immunity towards cells producing large amounts of viral proteins, whereas the ability to recognize antigens associated with tissue destruction may be important for responses to cytopathic viruses.

In summary, this report provides the first in vivo evidence that cellular destruction can enhance access of peripheral antigens to the cross-presentation pathway. In addition, we show that high dose antigens are preferentially cross-presented, although lower doses may still be of physiological relevance by their ability to be directly presented to activated CTLs entering peripheral tissues. Together, these data lead to the conclusion that the cross-presentation pathway is focused on high dose antigens or those released during cellular destruction.

We thank Jenny Falso, Tatiana Banjanin, Freda Karamalis, and Paula Nathan for their technical assistance.

1
Germain
RN
,
Margulies
DH
The biochemistry and cell biology of antigen processing and presentation
Annu Rev Immunol
1993
11
403
450
[PubMed]
2
Bevan
MJ
Cross-priming for a secondary cytotoxic response to minor H antigens with H-2 congenic cells which do not cross-react in the cytotoxic assay
J Exp Med
1976
143
1283
1288
[PubMed]
3
Bevan
MJ
Minor H antigens introduced on H-2 different stimulating cells cross-react at the cytotoxic T cell level during in vivo priming
J Immunol
1976
117
2233
2238
[PubMed]
4
Gordon
RD
,
Mathieson
BJ
,
Samelson
LE
,
Boyse
EA
,
Simpson
E
The effect of allogeneic presensitization on H-Y graft survival and in vitro cell-mediated responses to H-y antigen
J Exp Med
1976
144
810
820
[PubMed]
5
Gooding
LR
,
Edwards
CB
H-2 antigen requirements in the in vitro induction of SV40-specific cytotoxic T lymphocytes
J Immunol
1980
124
1258
1262
[PubMed]
6
von Boehmer
H
,
Hafen
K
Minor but not major histocompatibility antigens of thymus epithelium tolerize precursors of cytolytic T cells
Nature
1986
320
626
628
[PubMed]
7
Carbone
FR
,
Bevan
MJ
Class I–restricted processing and presentation of exogenous cell-associated antigen in vivo
J Exp Med
1990
171
377
387
[PubMed]
8
Huang
AY
,
Golumbek
P
,
Ahmadzadeh
M
,
Jaffee
E
,
Pardoll
D
,
Levitsky
H
Role of bone marrow–derived cells in presenting MHC class I–restricted tumor antigens
Science
1994
264
961
965
[PubMed]
9
Arnold
D
,
Faath
S
,
Rammensee
H
,
Schild
H
Cross-priming of minor histocompatibility antigen-specific cytotoxic T cells upon immunization with the heat shock protein gp96
J Exp Med
1995
182
885
889
[PubMed]
10
Kurts
C
,
Heath
WR
,
Carbone
FR
,
Allison
J
,
Miller
JFAP
,
Kosaka
H
Constitutive class I–restricted exogenous presentation of self antigens in vivo
J Exp Med
1996
184
923
930
[PubMed]
11
Kurts
C
,
Kosaka
H
,
Carbone
FR
,
Miller
JFAP
,
Heath
WR
Class I–restricted cross-presentation of exogenous self antigens leads to deletion of autoreactive CD8+T cells
J Exp Med
1997
186
239
245
[PubMed]
12
Kurts
C
,
Carbone
FR
,
Barnden
M
,
Blanas
E
,
Allison
J
,
Heath
WR
,
Miller
JFAP
CD4+ T cell help impairs CD8+T cell deletion induced by cross-presentation of self-antigens and favors autoimmunity
J Exp Med
1997
186
2057
2062
[PubMed]
13
Pulaski
BA
,
Yeh
KY
,
Shastri
N
,
Maltby
KM
,
Penney
DP
,
Lord
EM
,
Frelinger
JG
Interleukin 3 enhances cytotoxic T lymphocyte development and class I major histocompatibility complex “re-presentation” of exogenous antigen by tumor-infiltrating antigen-presenting cells
Proc Natl Acad Sci USA
1996
93
3669
3674
[PubMed]
14
Bennett
SR
,
Carbone
FR
,
Karamalis
F
,
Miller
JFAP
,
Heath
WR
Induction of a CD8 cytotoxic T lymphocyte response by cross-priming requires cognate CD4 help
J Exp Med
1997
186
65
70
[PubMed]
15
Bevan
MJ
Antigen recognition. Class discrimination in the world of immunology
Nature
1987
325
192
194
[PubMed]
16
Rock
KL
A new foreign policy: MHC class I molecules monitor the outside world
Immunol Today
1996
17
131
137
[PubMed]
17
Blanas
E
,
Carbone
FR
,
Allison
J
,
Miller
JFAP
,
Heath
WR
Induction of autoimmune diabetes by oral administration of autoantigen
Science
1996
274
1707
1709
[PubMed]
18
Lyons
AB
,
Parish
CR
Determination of lymphocyte division by flow cytometry
J Immunol Methods
1994
171
131
137
[PubMed]
19
Hodgkin
PD
,
Lee
JH
,
Lyons
AB
B cell differentiation and isotype switching is related to division cycle number
J Exp Med
1996
184
277
281
[PubMed]
20
Albert
ML
,
Sauter
B
,
Bhardwaj
N
Dendritic cells acquire antigen from apoptotic cells and induce class I–restricted CTLs
Nature
1998
392
88
91
21
Ohashi
PS
,
Oehen
S
,
Buerki
K
,
Pircher
H
,
Ohashi
CT
,
Odermatt
B
,
Malissen
B
,
Zinkernagel
RM
,
Hengartner
H
Ablation of “tolerance” and induction of diabetes by virus infection in viral antigen transgenic mice
Cell
1991
65
305
317
[PubMed]
22
Oldstone
MB
,
Nerenberg
M
,
Southern
P
,
Price
J
,
Lewicki
H
Virus infection triggers insulin-dependent diabetes mellitus in a transgenic model: role of anti-self (virus) immune response
Cell
1991
65
319
331
[PubMed]
23
Heath
WR
,
Karamalis
F
,
Donoghue
J
,
Miller
JF
Autoimmunity caused by ignorant CD8+ T cells is transient and depends on avidity
J Immunol
1995
155
2339
2349
[PubMed]
24
Lo
D
,
Freedman
J
,
Hesse
S
,
Palmiter
RD
,
Brinster
RL
,
Sherman
LA
Peripheral tolerance to an islet cell-specific hemagglutinin transgene affects both CD4+ and CD8+ T cells
Eur J Immunol
1992
22
1013
1022
[PubMed]
25
Adams
TE
,
Alpert
S
,
Hanahan
D
Non-tolerance and autoantibodies to a transgenic self antigen expressed in pancreatic beta cells
Nature
1987
325
223
228
[PubMed]
26
Guerder
S
,
Matzinger
P
A fail-safe mechanism for maintaining self-tolerance
J Exp Med
1992
176
553
564
[PubMed]

C. Kurts is supported by a fellowship from the Deutsche Forschungsgemeinschaft (Grant Ku1063/1-2). This work was funded by National Institutes of Health grant AI-29385 and grants from the National Health and Medical Research Council of Australia and the Australian Research Council.

Author notes

Address correspondence to William R. Heath, Immunology Division, The Walter and Eliza Hall Institute, P.O. Royal Melbourne Hospital, Parkville 3050, Victoria, Australia. Phone: 61-3-9345-2555; Fax: 61-3-9347-0852; E-mail: heath@wehi.edu.au or to Francis R. Carbone, The Department of Pathology and Immunology, Monash Medical School, Commercial Road, Prahran 3181, Victoria, Australia. Phone: 61-3-9276-2744; Fax: 61-3-9529-6484; E-mail: carbone@cobra.path.monash.edu.au