Spontaneous Autoimmunity Sufficiently Potent to Induce Diabetes Mellitus Is Insufficient to Protect against Insulinoma

This information is current as Kelley Zwicker, Christa Chatten, Kathy Gratton, Doug of September 25, 2021. Demetrick, Pau Serra, Afshin Shameli, Pere Santamaria and Oliver F. Bathe J Immunol 2009; 183:1705-1714; Prepublished online 1 July 2009;

doi: 10.4049/jimmunol.0800902 Downloaded from http://www.jimmunol.org/content/183/3/1705

Supplementary http://www.jimmunol.org/content/suppl/2009/07/07/jimmunol.080090 Material 2.DC1 http://www.jimmunol.org/ References This article cites 34 articles, 19 of which you can access for free at: http://www.jimmunol.org/content/183/3/1705.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on September 25, 2021

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2009 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Spontaneous Autoimmunity Sufficiently Potent to Induce Diabetes Mellitus Is Insufficient to Protect against Insulinoma1

Kelley Zwicker,2* Christa Chatten,2* Kathy Gratton,* Doug Demetrick,‡ Pau Serra,† Afshin Shameli,† Pere Santamaria,† and Oliver F. Bathe3*§

Intact tolerogenic mechanisms preclude effective immunity against tumors, as most tumor Ags differ little from normal host Ags. In contrast, when tolerance fails, the immune system becomes inappropriately activated against an autoantigen. We postulated that CD8؉ T cells activated during autoimmunity are capable of protecting against tumors that express the targeted autoantigen. To test this hypothesis, double-transgenic 8.3-NOD-RIPTAg mice were developed (where NOD is nonobese diabetic, RIP is rat insulin promoter, and TAg is large T Ag). In this model, individuals with the RIPTAg transgene develop insulinoma; those expressing a transgenic TCR (8.3-TCR) recognizing the islet-specific glucose 6 phosphatase cat- ؉ alytic subunit-related protein (IGRP) harbor a peripheral immune system dominated by diabetogenic CD8 T cells. Al- Downloaded from though tumor emergence was significantly slower in 8.3-NOD-RIPTAg mice compared with NOD-RIPTAg mice, all 8.3- NOD-RIPTAg mice eventually developed insulinoma. Tumor emergence was not secondary to clonal deletion or anergy. Ag loss and MHC down-regulation were not apparent. Endogenous 8.3-TCR CD8؉ T cells were recruited to the tumor site and proliferated upon arrival to the tumor, although they were notably absent from the central parts of more advanced tumors. These results demonstrate that a breakdown of tolerance capable of causing autoimmune disease is insufficient for effective /tumor immunity. Alterations in the tumor microenvironment may inhibit efficient and comprehensive delivery of CD8؉ T http://www.jimmunol.org cells to all regions of the tumor. These data suggest that any immunotherapeutic strategy for cancer must involve enhance- ment of a proinflammatory tumor microenvironment in addition to inhibition of tolerogenic mechanisms. The Journal of Immunology, 2009, 183: 1705–1714.

ost tumor Ags represent proteins derived from normal which CD8ϩ T cells that recognize an Ag expressed by spon- “self” proteins that are inappropriately expressed rel- taneously arising insulinoma and by normal ␤-islet cells are ative to the tissue of origin (1–3). Truly tumor-specific pathologically primed. Mice were bred on a nonobese diabetic M 4 Ags are rare. Therefore, with some exceptions, tumors are not (NOD) background, associated with a predisposition to auto- by guest on September 25, 2021 typically immunologically distinct. Recognizing that tumors are immunity due to multiple defects in various tolerogenic mech- derived from self, inhibiting tolerance currently represents the ma- anisms (6). The rat insulin promoter (RIP)-driven large T Ag jor strategic focus of cancer immunotherapists (4). Adoptive im- (TAg) transgene (RIPTAg) confers a propensity to develop spon- munotherapy and vaccination strategies each effectively bypass taneous insulinomas. The RIPTAg phenotype has been well char- tolerogenic mechanisms with occasional benefits in clinical trials. acterized, and all insulinomas appear to progress through similar Failures have largely been attributed to the emergence of Ag-loss histological stages (7–10). In individuals with the 8.3-TCR trans- variants (1, 2) and down-regulation of MHC class I (3, 5). To some gene, the vast majority of CD8ϩ T cells recognize residues 206– degree, these problems can be avoided by targeting numerous tu- 214 of the autoantigen IGRP (islet-specific glucose 6 phosphatase mor epitopes. If the autoreactive nature of the immune system is to catalytic subunit-related protein), which is present on normal be exploited against cancer, understanding the factors that limit ␤-cells and on insulinoma cells. As a result of enhanced recruit- effective tumor immunity despite a breakdown in tolerance would ment of CD8ϩ T cells to islets, mice with the 8.3-TCR transgene be useful. We considered that highly effective inhibition of toler- on the NOD background develop diabetes much earlier than NOD ance would be sufficient to eradicate a tumor. To investigate this mice, although the cumulative incidence of diabetes at 32 wk is hypothesis, we created a double-transgenic murine model in similar (11, 12). This model would enable the study of the inter- actions between pathologically primed diabetogenic CD8ϩ T cells and (normal or transformed) ␤-islet cells. † *Department of Surgery, Julia McFarlane Diabetes Research Centre and Department We demonstrate that a genetic predisposition to mount a vigor- of Microbiology and Infectious Diseases, ‡Department of Pathology, and §Depart- ment of Oncology, University of Calgary, Calgary, Alberta, Canada ous autoimmune response does indeed provide protection against Received for publication March 20, 2008. Accepted for publication May 25, 2009. an autologous tumor bearing the targeted autoantigen. Neverthe- The costs of publication of this article were defrayed in part by the payment of page less, this protection is incomplete, as the tumor eventually devel- charges. This article must therefore be hereby marked advertisement in accordance ops. The tumor microenvironment was thought to impede effective with 18 U.S.C. Section 1734 solely to indicate this fact. tumor immunity. These findings illustrate that even the best tumor 1 This research was supported through grants from the Cancer Research Society and vaccination strategies would not likely be sufficient to eradicate the American College of Surgeons. P. Serra was supported by a studentship from the Alberta Heritage Foundation for Medical Research (AHFMR). P. Santamaria was supported by the Canadian Diabetes Association and is a Scientist of the AHFMR. 4 2 Abbreviations used in this paper: NOD, nonobese diabetic; DM, diabetes mellitus; K.Z. and C.C. contributed equally to this work. IGRP, islet-specific glucose-6-phosphatase catalytic subunit-related protein; LN, 3 Address correspondence and request reprints to Dr. Oliver F. Bathe, Tom Baker lymph node; RIP, rat insulin promoter; TAg, large T antigen. Cancer Centre, Division of Surgical Oncology, 1331 29th Street Northwest, Calgary, Alberta, T2N 4N1 Canada. E-mail address: [email protected] Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00 www.jimmunol.org/cgi/doi/10.4049/jimmunol.0800902 1706 TUMOR ESCAPES IMPAIRED IMMUNE TOLERANCE tumors. Rather, approaches targeting the microenvironment may In vivo proliferation of 8.3-TCR CD8ϩ T cells was monitored by eval- be necessary to enhance successful tumor immunity. uating the reduction of CFSE (Invitrogen)-staining intensity. Purified 8.3- TCR CD8ϩ T cells were labeled with 5 ␮M CFSE. Naive 8.3-TCR CD8ϩ T cells (10 ϫ 106) in 0.5 ml of HBSS were injected via the tail vein. Six Materials and Methods days following adoptive transfer, recipients were sacrificed and donor cells Animals were identified using flow cytometry. Results were expressed as percentage of division calculated using the formula M2/(M2 ϩ M1) ϫ 100, where M2 8.3-NOD mice express the TCR␣␤ rearrangements of the H2Kd-restricted, ϩ represents a gate on cells that had diluted CFSE and M1 represents a gate ␤-islet cell-specific CD8 T cell clone NY8.3. These mice have an accel- on cells that had not undergone division. erated onset of diabetes mellitus (DM) compared with NOD mice due to ϩ In vivo proliferation was also measured by evaluating BrdU uptake. increased recruitment of diabetogenic CD8 T cells to pancreatic islets 8.3-NOD-RIPTAg mice received i.p. injections of 9.2 ␮g/ml BrdU in 200 (11). The NOD/Lt-TgN(RIP-TAg)1Lt-Prkdc-scid mouse (common name ␮l of PBS. Mice were sacrificed 24–31 h after injection, and lymphocytes NOD-scid-RIPTAg) was provided by Dr. D. Serreze (The Jackson Lab- were isolated from spleen, lymph node (LN), thymus, and tumor. Follow- oratory, Bar Harbor, ME). NOD-RIPTAg mice, which have the normal ing surface staining, an intracellular staining protocol (13) was used and complement of immune effectors, were generated by crossing NOD- cells were stained with either FITC-anti-BrdU or FITC-conjugated mouse scid-RIPTAG mice to NOD mice, eliminating the scid mutation. 8.3- IgG1 isotype control (BD Biosciences). BrdU uptake by T lymphocytes NOD-RIPTAg mice were produced by crossing 8.3-NOD and NOD- was evaluated by flow cytometry. RIPTAg mice. Such double-transgenic mice have tolerance to TAg (data not shown). The 8.3-TCR and RIPTAg transgenes were screened by PCR. Heterozygote scid mice were identified using PCR. Homozy- Isolation and analysis of RIPTAg tumor cells gotes were identified using the Ouchterlony assay. In some experi- RIPTAg tumors Ͼ2 mm in diameter were morselized in 3 ml of type V ments, hypoglycemic mice were fed sucrose-containing water to ensure collagenase (470U/ml; Sigma-Aldrich). Erythrocytes were lysed using an health in the later stages of insulinoma development. All mice were ammonium chloride lysis solution. The flasks were rinsed with PBS to Downloaded from housed under specific pathogen-free conditions at the University of Cal- remove nonadherent cells. Adherent cells were stained with 0.01 ␮g/ml gary (Calgary, Alberta, Canada), and experimental procedures were ap- 7-aminoactinomycin D and stained for MAC-1, CD11c, and H2Kd. proved by the University of Calgary Animal Care Committee in agree- ment with the Canadian Guidelines for Animal Research. Evaluation of tumor cell killing by CTL Monitoring diabetes and insulinoma CTL were developed by isolating splenocytes from 8.3-NOD mice and Blood glucose levels were measured 2–3 times weekly using a FastTake stimulating the cells with NRP-A7 (1 ␮g/ml) in complete medium with http://www.jimmunol.org/ glucose monitor (Johnson & Johnson). Animals were considered diabetic 10% FCS for 3 days. CTLs (20 ϫ 106) were injected i.v. into 8.3-NOD- after two readings Ն14 mM (hyperglycemia). The presence of insulinoma RIPTAg and NOD-RIPTAg mice. Blood glucose levels were followed was marked by two readings Յ4 mM (hypoglycemia). daily, and booster injections of 20 ϫ 106 CTLs were given weekly. Abs, cell purification, and flow cytometry Histopathology The following were purchased from BD Biosciences: PE-V␤8.1,8.2 TCR (MR5-2), Cy-chrome-CD8␣ (53-6.7), allophycocyanin-CD8␣ (53-6.7), bi- Freshly removed pancreata were cut longitudinally. If tumors were otin-CD25 (7D4), biotin-CD44 (IM7), biotin-CD69 (H1.F23), streptavidin- present they were also bivalved. The degree of inflammation was eval- Cy-chrome, FITC-IFN-␥ (XMG1.2), FITC-CD107␣ (1D4B), PE-CD11c uated on formalin-fixed paraffin-embedded samples after staining with (HL3) and PE-H2Kd (SF1-1.1). A FITC-conjugated rat anti-mouse IgG1 H&E (Sigma-Aldrich). Islets were scored for insulitis over five serial Ab was purchased from Cedarlane Laboratories. Anti-CD8 was derived sections, as previously described (12). by guest on September 25, 2021 from Lyt2.2 hybridoma and anti-CD4 was derived from GK1.5 hybridoma. IGRP, insulin, and cell differentiation markers were stained on 6-␮m Anti-F4/80 was derived from hybridoma supernatant. Anti-MAC-1 was sections from frozen blocks. Sections were fixed using equal volumes of purified from the M1/70 hybridoma and biotinylated. 100% methanol and acetone. Avidin and biotin were blocked (Vector Lab- CD8ϩ T cells were isolated using Abs against CD4, MAC-1, F4/80, and oratories). Serial sections were stained using the following dilutions: IGRP CD45R (derived from the B220 hybridoma) for negative selection with (1/800 dilution of polyclonal rabbit anti-mouse IGRP provided by J. Hutton Dynal beads. Cells were analyzed by flow cytometry using a FACScan (BD (University of Colorado, Denver, CO) (14); insulin (1/25 dilution guinea Biosciences) after conventional staining. Analysis software used included pig anti-swine polyclonal Ab; DakoCytomation); CD8 (neat Lyt2.2 super- CellQuest (BD Biosciences) and FlowJo (Tree Star). natant); CD4 (neat GK1.5 supernatant); and MAC-1 (1/5 dilution of M1/70 supernatant). Primary Abs were detected using 1/200 dilutions of biotin- Intracellular cytokine staining ylated rabbit anti-rat IgG for CD4, CD8, and MAC-1 and biotinylated goat ϩ anti-rabbit IgG for insulin, followed by incubation with HRP conjugated to To assess cytokine production by 8.3-TCR CD8 T cells, splenocytes or streptavidin (Zymed Laboratories). Endogenous peroxidase activity was peripancreatic lymphocytes from 8.3-NOD and 8.3-NOD-RIPTAg mice blocked using 3% H O /Tris-buffered saline. Color was developed using ϫ 6 2 2 were incubated in 24-well plates (2 10 cells/ml/well) with 100 nM diaminobenzidine tetrachloride, and counterstaining was done with NOD-related peptide (NRP-A7; KYNKANAFL) in RPMI 1640 containing ␮ hematoxylin. 5% FCS, glutamine (30 g/ml), penicillin (100 U/ml), streptomycin (100 Lesion size was evaluated using Openlab software (Improvision) by ␮ ϫ Ϫ5 g/ml), and 2-ME (5 10 M) (complete medium). Stimulation with 100 measuring the perimeter of lesions that had been stained for insulin. (All nM OVA peptide (SIINFEKL) served as a negative control. GolgiStop (BD IGRP-positive lesions were confirmed in serial sections to be insulin pos- Biosciences) was added for the last5hofculture. After 72 h, cells were itive.) Lesion size was evaluated using a ϫ4 objective. Those lesions that stained for surface expression of CD8 and the 8.3-TCR, and intracellular Ͻ ␥ were 10% of the field of view were defined as “normal” islets, whereas IFN- was stained using the Cytofix/Cytoperm kit (BD Biosciences) ac- those that were Ͼ10% of the field of view were “intermediate” lesions. cording to the manufacturer’s protocol. Those that exceeded the field of view were classified as “large” tumors. ϩ ϩ Proliferation assays Infiltration by CD8 T cells, CD4 T cells, or macrophages was evaluated using a scoring system similar to that used to score insulitis. The observer In vitro proliferation was measured using [3H]thymidine incorporation. was blinded to information on mouse age and genotype. Splenocytes or peripancreatic lymphocytes (1 ϫ 105) were incubated in flat-bottom, 96-well plates with the indicated concentrations of NRP-A7, Statistical analysis anti-CD3 (145-2C11 supernatant), or OVA for 48 h. Isolated 8.3-TCR CD8ϩ T cells or nontransgenic CD8ϩ T cells (1 ϫ 105) were incubated All values are expressed as mean Ϯ SEM. Deviations from Gaussian dis- with mitomycin C (Sigma-Aldrich)-treated tumor cells (1 ϫ 104) from tribution were tested using the Kolmogorov-Smirnov test. The differences 8.3-NOD-RIPTAg or NOD-RIPTAg mice for 72 h in flat-bottom, 96-well between two means were tested by the two-tailed Student t test for two plates. In conditions that used dendritic cells, bone marrow was harvested independent samples. The Mann-Whitney U test was used for nonpara- from NOD mice and cultured for 7 days with 10% IMDM (Invitrogen) and metric data. Categorical data were evaluated using the Fisher exact test. 20 ng/ml GM-CSF (from the F10.9 hybridoma provided by C. Brown, The time to hypoglycemia was evaluated using Kaplan-Meier analysis. The University of Pittsburgh, Pittsburgh, PA). [3H]Thymidine incorporation as- log-rank test was used to test for differences in incidence of hypoglycemia say was performed as previously described (1). At 16–18 h, cells were between groups. It was decided a priori that p Յ 0.05 would be considered harvested. statistically significant. The Journal of Immunology 1707 Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021 FIGURE 1. Incidence of autoimmunity and hypoglycemia in RIPTAg mice. A and B, Pancreata from 8.3-NOD-RIPTAg, NOD-RIPTAg, and NOD mice were evaluated at 3–4 wk and at 13–14 wk of age for inflammatory changes within islets. Data represent 50–259 islets from five to seven mice per group in two separate experiments. A, At 13–14 wk, 8.3-NOD-RIPTAg mice have a higher proportion of scores representing severe insulitis (i.e., 3 or 4) compared with NOD-RIPTAg mice. B, Infiltration by CD8ϩ T cells at 13–14 wk was significantly more severe in 8.3-NOD-RIPTAg lesions compared with NOD-RIPTAg lesions. C, Hypoglycemia was delayed in 8.3-NOD-RIPTAg mice. The curve for 8.3-NOD-RIPTAgs represents all mice that survived until hypoglycemia, independently of overt DM. D, Lifetime incidence of hypoglycemia in 8.3-NOD-RIPTAg mice as a function of overt DM before the onset of hypoglycemia. Data represent the same cohort of 8.3-NOD-RIPTAgs shown in C. There is no significant difference in the incidences of hypoglycemia between diabetic 8.3-NOD-RIPTAg mice and nondiabetic 8.3-NOD-RIPTAg mice.

Results control. Blood glucose was measured beginning at 3 wk of age and The 8.3-TCR transgene confers incomplete protection against continued until the time of hypoglycemia. Of 52 8.3-NOD- insulinoma RIPTAg mice, 12 (23%) became diabetic and died before the end of the observation period; 11 (21%) became diabetic before be- In the absence of the 8.3-TCR transgene, the RIPTAg transgene coming hypoglycemic, and 29 (56%) were euglycemic before de- had little influence on the inflammatory response, as there was no difference between the degrees of inflammatory infiltrate in islets veloping hypoglycemia. 8.3-NOD-RIPTAg mice that did not die from NOD mice and age-matched NOD-RIPTAg mice. 8.3-NOD- before developing insulinoma had a delayed onset of hypoglyce- RIPTAg mice developed insulitis that was generally more severe mia compared with NOD-RIPTAg and NOD-scid-RIPTAG co- than that of age-matched NOD-RIPTAg mice (Fig. 1A). Infiltration horts (Fig. 1C). The median time to hypoglycemia was 125 days of normal islets by CD8ϩ T cells at 13–14 wk and at Ͼ14 wk was for 8.3-NOD-RIPTAg mice compared with 90 and 94 days for Ͻ significantly more severe in 8.3-NOD-RIPTAg islets compared NOD-RIPTAg and NOD-scid-RIPTAG mice, respectively ( p with NOD-RIPTAg islets (Fig. 1B). These trends were also ob- 0.0001). Despite the delay in tumor emergence in 8.3-NOD- served for CD4ϩ lymphocytes at these age groups (data not RIPTAg mice, ultimately 100% of individuals became hypogly- shown). This was consistent with the more rapid onset of CD8ϩ- cemic. Therefore, 8.3-NOD-RIPTAg mice were incompletely pro- mediated autoimmune DM seen with other models expressing the tected from a tumor. ϩ 8.3-TCR transgene (11). To investigate whether diabetogenic 8.3-TCR CD8 T cells To determine whether pathologically primed autoreactive CD8ϩ could inhibit tumors in the absence of T cells displaying endoge- T cells protect against a tumor expressing the targeted autoantigen, nous TCRs, we introduced the 8.3-TCR and RIPTAg transgenes tumor incidence was compared in 8.3-NOD-RIPTAg and NOD- into RAG2Ϫ/Ϫ mice, which cannot rearrange endogenous TCR or RIPTAg cohorts. NOD-scid-RIPTAg mice served as a baseline Ig genes. 8.3-NOD-RIPTAg.RAG2Ϫ/Ϫ (n ϭ 9) were not protected 1708 TUMOR ESCAPES IMPAIRED IMMUNE TOLERANCE against tumors; median time to hypoglycemia was 91 days, which was not significantly different from NOD-RIPTAg.RAG2Ϫ/Ϫ mice (n ϭ 13; median time to hypoglycemia was 89 days) or other mice not bearing the 8.3-TCR transgene (data not shown). This is con- sistent with the previous observation that RAG2Ϫ/Ϫ 8.3-NOD mice developed DM much less frequently and significantly later than RAG2ϩ 8.3-NOD mice (12). In that instance, the diabetes deceleration in RAG2Ϫ/Ϫ 8.3-NOD mice could be reversed by adoptive transfer of CD4ϩ T cells, suggesting the need for CD4ϩ help for the diabetogenicity of 8.3-TCR CD8ϩ T cells. We considered that overt DM represented a surrogate marker for a more exuberant autoimmune response. Therefore, we postulated that the emergence of overt DM before insulinoma provides en- hanced protection against insulinoma. However, there was no sig- nificant difference in the incidence of hypoglycemia in diabetic mice compared with nondiabetic mice (Fig. 1D). Therefore, the development of overt DM did not confer additional protection against insulinoma.

It was possible that the high blood glucose levels associated Downloaded from with DM masked the development of small insulinomas. To eval- uate this possibility, mice were sacrificed at 100 days and exam- ined for the appearance of gross tumor regardless of blood glucose level. 8.3-NOD-RIPTAg mice had a reduced prevalence of insu- linoma at 100 days of age compared with NOD-RIPTAg mice (69

vs 95%, respectively; p ϭ 0.002). This confirmed the initial ob- http://www.jimmunol.org/ servation that the 8.3-TCR transgene was protective. Moreover, there was no significant reduction in the prevalence of gross tumor in 8.3-NOD-RIPTAg mice that had developed overt DM before insulinoma (data not shown). Altogether, these data show that in- sulinoma develops despite an immune response that is sufficiently potent to induce the destruction of normal ␤-islet cells. Peripheral 8.3-TCR CD8ϩ T cells are not rendered anergic

It is possible that peripheral tolerogenic mechanisms, possibly re- by guest on September 25, 2021 lated to the tumor itself, impeded a sustained immune response against a tumor. In other experimental models, tumor-specific T cells become tolerant to tumor Ags due to clonal anergy (15, 16). Prolonged T cell anergy can lead to activation-induced cell death, resulting in peripheral deletion of tumor-specific T cell populations (17–19). Thus, 8.3-TCR CD8ϩ T cell anergy and clonal deletion were evaluated as potential mechanisms for incomplete protection FIGURE 2. Deletion and anergy do not affect the 8.3-TCR CD8ϩ T cell against insulinoma in 8.3-NOD-RIPTAg mice. response. A, The absolute number of 8.3-TCR CD8ϩ T cells in spleens and The absolute numbers of T cells present in secondary lymphoid peripancreatic LNs of 8.3-NOD and 8.3-NOD-RIPTAg mice was deter- tissue were determined in 8.3-NOD-RIPTAg mice as they aged. mined using flow cytometry. There is no evidence of clonal deletion of There was no significant difference in the absolute number of 8.3- 8.3-TCR CD8ϩ T cells in tumor-bearing 8.3-NOD-RIPTAg mice com- ϩ TCR CD8ϩ T cells derived from spleens or peripancreatic LNs of pared with age-matched 8.3-NOD controls. B, 8.3-TCR CD8 T cells from 8.3-NOD-RIPTAg mice compared with age-matched 8.3-NOD peripancreatic LNs derived from tumor-bearing mice proliferate in re- mice (Fig. 2A). To evaluate the possibility that thymic output sponse to NRP-A7 in a similar fashion to those from tumor-free 8.3-NOD- RIPTAg mice. Curves represent proliferative responses of lymphocytes masked deletion, we monitored in vivo persistence of 8.3-TCR from individuals, as measured by [3H]thymidine uptake assays. Lympho- CD8ϩ T cells following adoptive transfer. We injected 10 ϫ 106 ϩ cytes from 8.3-NOD mice served as a positive control. Data are represen- CFSE-labeled 8.3-TCR CD8 T cells into mice aged 13–14 wk. In tative of three independent assays with 1–3 mice per group. C, Proportion NOD-RIPTAg mice 6 days after adoptive transfer, an average of 8.3-TCR CD8ϩ T cells in spleen and peripancreatic LNs that secreted 5 ϩ 8.74 ϫ 10 8.3-TCR CD8 T cells were recovered from spleen, IFN-␥ in response to NRP-A7 as estimated by flow cytometry. There was peripancreatic LN, and peripheral LN. In NOD mice, 8.23 ϫ 105 no difference in the proportions of IFN-␥-positive cells in tumor-bearing 8.3-TCR CD8ϩ T cells were recovered, which was not signifi- 8.3-NOD-RIPTAg mice compared with 8.3-NOD counterparts. cantly different from the results seen in NOD-RIPTAg mice. Therefore, there is no evidence of CD8ϩ T cell clonal deletion secondary to tumor. magnitude as those derived from tumor-free 8.3-NOD-RIPTAg The functional capacity of the 8.3-TCR CD8ϩ T cell population mice and prediabetic 8.3-NOD controls (Fig. 2B). Similar results was then examined. The proliferative response against increasing were observed for splenic 8.3-TCR CD8ϩ T cells (data not doses of the 8.3-TCR peptide ligand NRP-A7 was assessed by shown). Thus, a tumor does not impair the proliferative capacity of measuring [3H]thymidine incorporation. 8.3-TCR CD8ϩ T cells 8.3-TCR CD8ϩ T cells systemically (i.e., in spleen) or regionally derived from peripancreatic LNs from tumor-bearing 8.3-NOD- (i.e., in peripancreatic LNs). Moreover, because proliferative re- RIPTAg mice responded to NRP-A7 with similar kinetics and sponses to Ag are seen in whole LN and spleen (data not shown), The Journal of Immunology 1709 Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021

FIGURE 3. Naive 8.3-TCR CD8ϩ T cells or nontransgenic CD8ϩ T cells were labeled with CFSE and adoptively transferred into age-matched NOD and NOD-RIPTAg recipients. A, Proliferation of donor cells was evaluated 6 days later by quantifying CFSE dilution. Proliferation was observed only in 8.3-TCR CD8ϩ T cells in peripancreatic LNs. B, The proportion of naive 8.3-TCR CD8ϩ T cells that divided was significantly higher in NOD-RIPTAg recipients (p ϭ 0.0066) and increased significantly with age in both groups (p ϭ 0.0013). Results represent 6–9 mice per group and were analyzed using two-way ANOVA.

it is unlikely that the responsiveness of 8.3-TCR CD8ϩ T cells is 8.3-NOD-RIPTAg mice with established tumor was comparable to due to the removal of a suppressive cell population during CD8ϩ that of T cells derived from tumor-free 8.3-NOD-RIPTAg mice, T cell isolation. and 8.3-NOD controls. ϩ Rather than inhibiting the 8.3-TCR CD8 T cell population as a ϩ whole, tumor progression may have inhibited individual T cells. Upon Naive 8.3-TCR CD8 T Cells are primed in tumor-draining stimulation with NRP-A7, there was no difference in the proportion of LNs IFN-␥ secreting 8.3-TCR CD8ϩ T cells derived from spleens or Tumors may have persisted secondary to immunological ignorance. peripancreatic LNs of tumor-bearing 8.3-NOD-RIPTAg mice com- Ignorance in the context of tumor immunity may occur because pared with those of their 8.3-NOD counterparts (Fig. 2C). Thus, the CD8ϩ T cells are exposed to insufficient levels of tumor Ag (20–22). proportion of CD8ϩ 8.3-TCR T cells that produce IFN-␥ in response Although recognition of untransformed ␤-cell Ag resulted in CD8ϩ T to peptide stimulation is not influenced by the presence of tumors. cell activation (as indicated by the emergence of DM), the elabo- Overall, these experiments demonstrate that the function of pe- ration of insulinoma Ags may have been insufficient for continued ripheral 8.3-TCR CD8ϩ T cells was not impaired as insulinoma cross-presentation and activation of CD8ϩ T cells in secondary developed. The in vitro response by CD8ϩ T cells derived from lymphoid tissue. 1710 TUMOR ESCAPES IMPAIRED IMMUNE TOLERANCE Downloaded from http://www.jimmunol.org/ FIGURE 4. RIPTAg tumor cells express H2Kd and IGRP. A, MHC class I (H2Kd) expression by insulinoma cells was evaluated by flow cytometry. Tumor cells from 8.3-NOD-RIPTAg mice retain H2Kd expression. Levels of expression are similar to those of NOD-RIPTAg tumor cells and NOD splenocytes. Data are representative of four independent experiments with 1–3 mice per group. B, Histological sections of pancreata from 13- to 14-wk-old 8.3-NOD-RIPTAg mice were assessed for IGRP by immunohistochemistry. All islets in 8.3-NOD-RIPTAg mice were IGRP positive. Data are representative of 52–75 islets and 6–13 intermediate size tumors using five mice per group. Data for NOD-scid-RIPTAG lesions (positive control) are not shown. C, Naive 8.3-TCR CD8ϩ T cells and nontransgenic CD8ϩ T cells were cultured with mitomycin C-treated tumor cells derived from 8.3-NOD-RIPTAg tumors. Proliferation was significantly higher in 8.3-TCR CD8ϩ T cells compared with nontransgenic CD8ϩ T cells when cultured with dendritic cells (DC). D, Blood glucose levels following CTL injection into 8.3-NOD-RIPTAg mice. Arrows indicate days when injections occurred. RIPTAg tumors remain targets to fresh CTL. Data are representative of three separate experiments with one or two mice per experiment. by guest on September 25, 2021

To determine whether insulinoma escaped the CD8ϩ T cell re- ability as targets for CTL lysis. As shown in Fig. 4A, 8.3-NOD- sponse due to ignorance, naive 8.3-TCR CD8ϩ T cells were labeled RIPTAg tumor cells retain H2Kd expression. There was no sig- with CFSE and adoptively transferred into NOD-RIPTAg recipients. nificant difference in mean fluorescence intensity in 8.3-NOD- Donor cells were identified in spleen, peripancreatic LNs, and periph- RIPTAg tumors (n ϭ 5) compared with NOD-RIPTAg tumors eral LNs (i.e., axillary and inguinal LNs). No division was observed (n ϭ 6) (data not shown). Both islets and intermediate size tumors in cells that had trafficked to the spleen (data not shown) and periph- from 8.3-NOD-RIPTAg mice retain IGRP expression (Fig. 4B). In eral LNs (Fig. 3, A and B). Interestingly, CD8ϩ T cell proliferation in fact, 100% of lesions, including normal islets and intermediate size the peripancreatic LNs was significantly increased in NOD-RIPTAg lesions, were IGRP positive at 13–14 wk of age. recipients across all ages compared with NOD controls (Fig. 3B). To confirm that 8.3-TCR CD8ϩ T cells recognize tumors derived Nontransgenic CD8ϩ T cells were adoptively transferred in a similar from established 8.3-NOD-RIPTAg mice, tumors were cocultured fashion. These cells did not proliferate in peripancreatic LNs (Fig. with CD8ϩ T cells isolated from 8.3-NOD spleen or NOD spleen and 3A), demonstrating that the proliferation seen in adoptively transferred with dendritic cells grown from NOD bone marrow. [3H]Thymidine 8.3-TCR CD8ϩ T cells was Ag dependent. uptake was greater in tumors cocultured with 8.3-CD8ϩ T cells com- Ignorance is therefore not responsible for tumor progression pared with nontransgenic CD8ϩ T cells ( p ϭ 0.014; Fig. 4C). This is ϩ in this model. Moreover, naive 8.3-TCR CD8 T cells are consistent with our observation that adoptively transferred naive 8.3- cross-primed and are capable of proliferating in peripancreatic TCR CD8ϩ T cells divide in response to tumor in an Ag-dependent LNs in response to Ags derived from normal as well as trans- manner, even in older 8.3-NOD-RIPTAg mice. formed ␤-islet cells. We speculate that the enhanced prolifera- Finally, the ability of 8.3-TCR CD8ϩ T cells to eliminate tumor tion seen in NOD-RIPTAg mice is secondary to increased Ag cells from 8.3-NOD-RIPTAg mice with established tumors was load, although this does not completely exclude other variables evaluated. Activated 8.3-TCR CTLs were adoptively transferred such as the elaboration of tumor-derived factors that support T into hypoglycemic 8.3-NOD-RIPTAg mice. Following CTL admin- cell proliferation. istration, blood glucose rapidly increased before returning to hypo- glycemic levels a few days later (Fig. 4D). Upon receiving a booster RIPTAg tumor cells are suitable targets for CTL lysis injection of CTL, blood glucose levels in 8.3-NOD-RIPTAg mice Dysfunctional Ag presentation is well documented as a mechanism again increased to normoglycemic levels before decreasing. This of tumor escape, arising secondary to a reduction in MHC class I demonstrated that RIPTAg tumors, even in 8.3-NOD-RIPTAg mice, expression or to the generation of Ag loss variants (1, 23, 24). remained targets for freshly activated CD8ϩ T cells, although the Therefore, RIPTAg insulinoma cells were evaluated for their suit- inhibitory activity of these CTLs was temporary. The Journal of Immunology 1711 Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021

FIGURE 5. Leukocytes are recruited to the tumor site. A and B, Insulitis and CD8ϩ T cell infiltration was scored in islets, intermediate tumors, and large tumors of NOD-RIPTAg and 8.3-NOD-RIPTAg mice that had gross tumor at necropsy. A, The proportion of intermediate size lesions with scores representing severe insulitis (i.e., grades 3 and 4) was significantly increased in 8.3-NOD-RIPTAg mice compared with NOD-RIPTAg mice. Data are derived from the scoring of 4–46 lesions per mouse from five or six mice per group. B, In each genotype, CD8ϩ T cell infiltration was more severe in islets and intermediate tumors than in large tumors. Data are derived from the assessment of 5–31 lesions per mouse from five or six mice per group. C, CD8ϩ T cells infiltrated intermediate lesions but were restricted to the periphery of larger lesions. Similar patterns of infiltration were seen for CD4ϩ T cells. Results are representative of five or six large tumors per group. D, A high proportion of 8.3-TCR CD8ϩ T cells within insulinomas of 5- to 6-mo-old p ϭ 0.0028). The histogram ,ءء) p ϭ 0.0037) and lymph node ,ء) 8.3-NOD-RIPTAg mice incorporate more BrdU in tumors compared with spleen demonstrates fluorescence intensity of tumor-infiltrating 8.3-TCR CD8ϩ T cells stained with anti-BrdU or isotype control. The solid curve represents the isotype control, whereas the open curve corresponds to cells stained for BrdU.

Functional lymphocytes are recruited to the tumor site that this would illustrate differences in the inflammatory microen- In hypoglycemic 8.3-NOD-RIPTAg and NOD-RIPTAg mice vironments in normal islets vs tumors. There was no significant 16–31 wk of age, we analyzed the inflammatory infiltrate sepa- difference in the severity of inflammation in islets of NOD- rately in small, intermediate, and large lesions. It was considered RIPTAg vs 8.3-NOD-RIPTAg mice (Fig. 5A). However, a higher 1712 TUMOR ESCAPES IMPAIRED IMMUNE TOLERANCE Downloaded from

FIGURE 6. Further characterization of immune effectors that comprise the intratumoral microenvironment. The inflammatory infiltrate in established tumors from 8.3-NOD-RIPTAg and NOD-RIPTAg mice was characterized using flow cytometry. A, The increase in the number of CD4ϩ T cells in 8.3-NOD-RIPTAg tumors did not reach statistical significance compared with the number of CD4ϩ T cells on NOD-RIPTAg tumors (p ϭ 0.0553). B, There was no significant difference in the proportion of CD4ϩCD25ϩFoxp3ϩ T cells (p ϭ 0.43). C, The absolute number of macrophages was significantly higher /p ϭ 0.0431). Data are expressed as the absolute number of cells per milligram of tumor tissue and represent 5–11 mice http://www.jimmunol.org ,ءء) in 8.3-NOD-RIPTAg tumors per group. D, Macrophage infiltration of islets and tumor in 8.3-NOD-RIPTAg mice was visualized by immunohistochemistry. Macrophage infiltrate was .p ϭ 0.032). Data are derived from 1–12 lesions per mouse from three mice per group ,ء) significantly higher in large lesions compared with islets proportion of intermediate size lesions in 8.3-NOD-RIPTAg mice hypoglycemic while ingesting sucrose water (aged 20–24 wk). had scores reflecting severe inflammation (i.e., grade 3 or 4) than 8.3-TCR CD8ϩ and CD4ϩ T cells in these tumors readily incor- in NOD-RIPTAg mice ( p Ͻ 0.0001). Thus, the 8.3-TCR transgene porated BrdU but nontransgenic CD8ϩ T cells did not (Fig. 5D), was associated with increased leukocyte recruitment to intermedi- suggesting that the CD8ϩ T cells contained within tumors prolif- ate size lesions. Interestingly, this result was not observed in larger erated in an Ag-dependent manner. by guest on September 25, 2021 tumors. Large lesions in both 8.3-NOD-RIPTAg and NOD- Together, these data demonstrate a change in microenviron- RIPTAg mice typically had a paucity of infiltrating leukocytes, ment as tumors progress. That is, the islets and intermediate particularly in their central components. Therefore, the inflamma- lesions in mice with established tumors still harbored intense tory infiltrate is most pronounced in intermediate lesions, yet large inflammation, including CD8ϩ T cells. This was distinct from lesions contain inflammatory cells only in their periphery. ϩ ϩ the more restricted inflammatory response seen in large tumors Infiltration of islets and tumors by CD8 and CD4 lympho- within the same pancreata. As tumors enlarged, it appeared that cytes was also analyzed in mice with established tumors (16–31 the permissive microenvironment converted to an immunosup- wk of age). 8.3-NOD-RIPTAg mice had a higher proportion of ϩ pressive one. Although the proportion of tumor-infiltrating 8.3- intermediate size lesions with severe CD8 T cell infiltration (i.e., TCR CD8ϩ T cells that incorporated BrdU within large tumors ϭ grade 3 or 4) ( p 0.0013), whereas NOD-RIPTAg mice had a was similar to that seen in the thymus, the majority of cells did higher proportion of intermediate size lesions with mild infiltration not take up BrdU. It is these latter cells that may represent (i.e., grades 0–2) (Fig. 5B). CD4ϩ T cell infiltration was also dysfunctional cells. markedly increased in 8.3-NOD-RIPTAg mice (data not shown). Therefore, the 8.3-TCR transgene conferred a vigorous T cell-me- diated response against intermediate size adenomas. Remarkably, Factors in the tumor microenvironment may impair CD8ϩ in 8.3-NOD-RIPTAg mice that had more advanced insulinomas T cell function (i.e., 16–31 wk of age), normal islets and intermediate size lesions ϩ remained densely infiltrated by CD8ϩ T cells, yet CD8ϩ T cell in- We considered that the CD8 T cell response could be self-lim- filtration in large tumors from both 8.3-NOD-RIPTAg (Fig. 5C) and iting by recruiting other immune effectors that contribute to the NOD-RIPTAg (data not shown) mice was minimal. In fact, 100% of immunosuppressive tumor microenvironment. To evaluate other cel- large tumors in RIPTAg mice with and without the 8.3-TCR trans- lular constituents that might have influenced intratumoral trafficking, gene were Ͻ25% infiltrated. The same was noted with CD4ϩ T cell we sought to better characterize the cells contained in larger tumors. infiltrate (Fig. 5C). Importantly, in these larger lesions CD8ϩ and There was no difference between the numbers of B cells or dendritic CD4ϩ T cells were located primarily around the perimeter of the cells present in tumors from 8.3-NOD-RIPTAg and NOD-RIPTAg ϩ tumor, with a paucity of both T cell subsets in their middle. mice (data not shown). The increase in the number of CD4 T cells Although the above data suggest that tumors are recognizable by in 8.3-NOD-RIPTAg tumors did not reach statistical significance ϩ 8.3-TCR CD8ϩ T cells, it is possible that the T cells recruited to ( p ϭ 0.055) (Fig. 6A). Macrophages and CD4 T suppressor cells the lesion no longer recognized tumors and therefore ceased to were present in all established insulinomas. There was no significant proliferate at the tumor site. To test for this, BrdU uptake in tumor- difference in the proportions of CD4ϩCD25ϩFoxp3ϩ T cells between infiltrating T cells was studied in animals that had been persistently 8.3-NOD-RIPTAg and NOD-RIPTAg lesions (Fig. 6B). Interestingly, The Journal of Immunology 1713 there was a significant increase in the number of CD11bϩ cells iden- mors, CD8ϩ T cells were excluded from the centers of large tu- tified in 8.3-NOD-RIPTAg insulinomas compared with those derived mors, where the absence of any inflammatory response mediated from NOD-RIPTAg mice (Fig. 6C). by other immune effectors was striking. The CD11bϩ infiltrate was further characterized. The majority Indeed, in our model we argue that the immunosuppressive tu- (86.9 Ϯ 0.67%) consisted of F4/80ϩ macrophages. Of those, mor microenvironment limited CD8ϩ T cell-mediated tumor lysis, 85.6 Ϯ 2.9% expressed macrophage scavenger receptor 1, which is particularly near the center of the tumor. That immunosuppressive typically found on alternatively activated (M2) macrophages (25) microenvironment appeared not to be present in normal islets, as (supplemental Fig. 1).5 Myeloid-derived suppressor cells also have autoimmune DM frequently preceded tumor development. We immunosuppressive activity and are recognized as CD11bϩGr-1ϩ speculate that these microenvironmental factors were present early (26, 27). Only 1.6 Ϯ 1.1% of CD11bϩ cells coexpressed Gr-1 in tumor development, as one might surmise that even “early tu- within the tumor infiltrate (data not shown). mors” should not have developed in the presence of a sufficiently Macrophage infiltration was evaluated as a function of lesion size inhibitory antitumor immune response. We further speculate that and correlated with CD8ϩ T cell infiltration. In 8.3-NOD-RIPTAg this immunosuppressive microenvironment is a product of the mice, macrophage infiltrate increased steadily as lesion size increased other cell populations that comprise the tumor mass. (Fig. 6D). This was accompanied by augmented CD8ϩ T cell infil- The role of regulatory T cells in tumor immunity has recently tration in small and intermediate size lesions. However, in large tu- been extensively investigated, and it is well established that mors macrophage infiltration continued to rise, and suddenly CD8ϩ T CD4ϩCD25ϩ T cells can influence T cell function in the tumor cells were constrained to the tumor periphery. microenvironment (28, 29). For example, CD4ϩCD25ϩ T cells ϩ Together, these data demonstrate that the enhanced CD8 T cell were found to inhibit intratumoral T cell function in a TAg trans- Downloaded from response in 8.3-NOD-RIPTAg mice was accompanied by the re- genic model of prostate dysplasia (29). We were unable to identify cruitment of other immune effector populations, including macro- any alterations in the numbers of CD4ϩCD25ϩFoxP3ϩ cells in phages and CD4ϩ T cells. However, the increase in CD4ϩ T cells insulinomas derived from 8.3-TCR transgenic mice as compared within the tumor compartment of 8.3-NOD-RIPTAg mice was not with individuals without this transgene. However, this does not significantly larger than that seen in NOD-RIPTAg animals. We exclude the possibility that CD4ϩCD25ϩ suppressor T cells con- ϩ postulate that the inflammatory response secondary to the influx of tribute to the dysfunction of intratumoral CD8 T cells. http://www.jimmunol.org/ CD8ϩ T cells altered the tumor microenvironment to an immuno- Although tumor-associated macrophages have the potential to suppressive one and ultimately may have altered the capability of induce a protective innate antitumor immune response, they have CD8ϩ T cells to effectively lyse tumors. also been shown to inhibit T cell function directly at the tumor site (30). Their polarization to an inhibitory phenotype may be a prod- Discussion uct of the tumor microenvironment, including cytokines produced Most immunotherapeutic strategies for generating immunity by other inflammatory cell populations (31). We identified an im- against tumors rely on the inhibition of tolerogenic mechanisms. pressive number of macrophages in both 8.3-NOD-RIPTAg and Such strategies have dominated cancer immunotherapeutics be- NOD-RIPTAg insulinomas. The fact that they were particularly by guest on September 25, 2021 cause most tumor Ags reflect the antigenic repertoire of “self” and numerous in individuals with the 8.3-TCR transgene suggests that the homing of macrophages into a tumor is a function of diabeto- are therefore not recognized by the immune system as foreign. We ϩ have created a spontaneous tumor model that fully tests the capa- genic CD8 T cells, although this must be investigated further. It bility of this strategy to inhibit tumor development. The model is is also possible that the prominent macrophage population is sim- important for the following reasons: 1) the tumor develops spon- ply secondary to the diabetogenic inflammatory response, as insu- taneously; 2) the Ag recognized by CD8ϩ T cells is a naturally litis in NOD mice is similarly accompanied by macrophage infiltra- occurring endogenous Ag that is expressed in normal ␤-islet cells tion (32). The potentially immunosuppressive role of intratumoral as well as in transformed ␤-islet cells; 3) diabetogenic CD8ϩ T macrophages must be investigated further in this model. cells are primed spontaneously due to the breakdown of tolerance Bone marrow-derived myofibroblasts may also influence the na- mediated by multiple mechanisms (12); and 4) the relative mag- ture of the intratumoral inflammatory response. Myofibroblasts re- nitudes of autoimmunity and tumor immunity can be monitored. semble muscle cells and fibroblasts and have been identified within Even in the presence of overt autoimmunity, despite the initial RIPTAg insulinomas (33). They serve not only as structural com- inhibition of tumor development, tumors ultimately emerged. This ponents in the tumor microenvironment but also as key regulators effectively demonstrates that immunotherapeutic strategies that of the intratumoral immune response (34). The distribution of T rely solely on bypassing tolerance are insufficient for cancer cells and macrophages has been shown to be associated with large therapeutics. sheaths of myofibroblasts at the tumor margin of both rat colorec- In a TCR transgenic model in which a single Ag is targeted, a tal tumors and RIPTAg insulinomas (33, 35). Evidence suggests number of factors could have contributed to the ultimate emer- that tumor-associated myofibroblasts not only prevent T cell infil- gence of tumors. Such factors included clonal deletion, clonal an- tration in established tumors in vivo, but also have the ability to ergy, poor delivery of tumor-specific immune effectors to tumor, shield tumor cells from T cell-mediated lysis (33). The paucity of Ag loss, or MHC down-regulation. None of these factors were T cells at the center of large tumors in our model may therefore be found to contribute to tumor progression. Rather, CD8ϩ T cells secondary to the presence of insulinoma-associated myofibro- efficiently homed to early tumors, as well as to normal islets. Tu- blasts, although this was not specifically evaluated. mors appeared capable of priming CD8ϩ T cells in draining In summary, we have demonstrated that even an immune re- peripancreatic LNs even at later stages of development. Moreover, sponse sufficiently robust to induce autoimmunity is not suffi- ϩ ciently robust to retard tumor progression. Although functional intratumoral CD8 T cells retained at least some of their capability ϩ to respond to larger tumors, as evidenced by their intratumoral CD8 T cells were delivered to tumors, initially inhibiting tumor proliferation. However, unlike in normal islets and in smaller tu- progression, the tumor microenvironment differed from that of normal islets. These microenvironmental perturbations resulted in the exclusion of inflammatory cells from the central portions of 5 The online version of this article contains supplemental material. larger tumors, ultimately allowing unfettered tumor growth. Our 1714 TUMOR ESCAPES IMPAIRED IMMUNE TOLERANCE observations demonstrate that immunotherapeutic strategies that bone marrow-derived antigen-presenting cells is the dominant mechanism in the rely solely on breaking down tolerance are insufficient for cancer induction of T-cell tolerance during B-cell lymphoma progression. Blood 98: 1070–1077. therapeutics. Rather, success will depend on the identification and 17. Deeths, M. J., R. M. Kedl, and M. F. Mescher. 1999. CD8ϩ T cells become inhibition of factors that contribute to an immunosuppressive in- nonresponsive (anergic) following activation in the presence of costimulation. J. Immunol. 163: 102–110. tratumoral microenvironment. 18. Hernandez, J., S. Aung, W. L. Redmond, and L. A. Sherman. 2001. Phenotypic and functional analysis of CD8ϩ T cells undergoing peripheral deletion in re- Acknowledgments sponse to cross-presentation of self-antigen. J. Exp. Med. 194: 707–717. 19. Staveley-O’Carroll, K., E. Sotomayor, J. Montgomery, I. Borrello, L. Hwang, We acknowledge Dr. Chris Mody for helpful advice during this work. We S. Fein, D. Pardoll, and H. Levitsky. 1998. Induction of antigen-specific T cell also thank Sandy Eyton-Jones for technical assistance. anergy: An early event in the course of tumor progression. Proc. Natl. Acad. Sci. USA 95: 1178–1183. Disclosures 20. Dalyot-Herman, N., O. F. Bathe, and T. R. Malek. 2000. Reversal of CD8ϩ T cell ignorance and induction of anti-tumor immunity by peptide-pulsed APC. J. Im- The authors have no financial conflict of interest. munol. 165: 6731–6737. 21. Spiotto, M. T., P. Yu, D. A. Rowley, M. I. Nishimura, S. C. Meredith, References T. F. Gajewski, Y. X. Fu, and H. Schreiber. 2002. Increasing tumor antigen expression overcomes “ignorance” to solid tumors via crosspresentation by bone 1. Bathe, O. F., N. Dalyot-Herman, and T. R. Malek. 2003. Therapeutic limitations ϩ marrow-derived stromal cells. Immunity 17: 737–747. in tumor-specific CD8 memory T cell engraftment. BMC Cancer 3: 21. 2. Jager, E., M. Ringhoffer, J. Karbach, M. Arand, F. Oesch, and A. Knuth. 1996. 22. Wilcox, R. A., D. B. Flies, G. Zhu, A. J. Johnson, K. Tamada, A. I. Chapoval, Inverse relationship of melanocyte differentiation antigen expression in mela- S. E. Strome, L. R. Pease, and L. Chen. 2002. Provision of antigen and CD137 noma tissues and CD8ϩ cytotoxic-T-cell responses: evidence for immunoselec- signaling breaks immunological ignorance, promoting regression of poorly im- J. Clin. Invest. tion of antigen-loss variants in vivo. Int. J. Cancer 66: 470–476. munogenic tumors. 109: 651–659. 23. Bai, X. F., J. Liu, O. Li, P. Zheng, and Y. Liu. 2003. Antigenic drift as a mech- 3. Seliger, B., A. Hohne, D. Jung, M. Kallfelz, A. Knuth, E. Jaeger, H. Bernhard, Downloaded from F. Momburg, R. Tampe, and C. Huber. 1997. Expression and function of the anism for tumor evasion of destruction by cytolytic T lymphocytes. J. Clin. peptide transporters in escape variants of human renal cell carcinomas. Exp. Invest. 111: 1487–1496. Hematol. 25: 608–614. 24. Garcia-Lora, A., M. Martinez, I. Algarra, J. J. Gaforio, and F. Garrido. 2003. 4. Zwicker, K. A., and O. F. Bathe. 2004. Exploring the relationship between tumor MHC class I-deficient metastatic tumor variants immunoselected by T lympho- immunity and autoimmunity. In Recent Research Developments in Cancer. cytes originate from the coordinated downregulation of APM components. Int. J. S. G. Pandalai, ed. Transworld Research Network, Kerala, India, p. 343–378. Cancer 106: 521–527. 5. Smith, M. E., S. G. Marsh, J. G. Bodmer, K. Gelsthorpe, and W. F. Bodmer. 25. Mantovani, A., A. Sica, S. Sozzani, P. Allavena, A. Vecchi, and M. Locati. 2004. 1989. Loss of HLA-A,B,C allele products and lymphocyte function-associated The chemokine system in diverse forms of macrophage activation and polariza- antigen 3 in colorectal neoplasia. Proc. Natl. Acad. Sci. USA 86: 5557–5561. tion. Trends Immunol. 25: 677–686. http://www.jimmunol.org/ 6. Zucchelli, S., P. Holler, T. Yamagata, M. Roy, C. Benoist, and D. Mathis. 2005. 26. Gabrilovich, D. I., V. Bronte, S. H. Chen, M. P. Colombo, A. Ochoa, Defective central tolerance induction in NOD mice: genomics and genetics. Im- S. Ostrand-Rosenberg, and H. Schreiber. 2007. The terminology issue for my- munity 22: 385–396. eloid-derived suppressor cells. Cancer Res. 67: 425. 27. Gabrilovich, D. I., M. P. Velders, E. M. Sotomayor, and W. M. Kast. 2001. 7. Folkman, J., K. Watson, D. Ingber, and D. Hanahan. 1989. Induction of angio- ϩ genesis during the transition from hyperplasia to neoplasia. Nature 339: 58–61. Mechanism of immune dysfunction in cancer mediated by immature Gr-1 my- 8. Grant, S. G., I. Seidman, D. Hanahan, and V. L. Bautch. 1991. Early invasiveness eloid cells. J. Immunol. 166: 5398–5406. characterizes metastatic carcinoid tumors in transgenic mice. Cancer Res. 51: 28. Antony, P. A., C. A. Piccirillo, A. Akpinarli, S. E. Finkelstein, P. J. Speiss, D. R. Surman, D. C. Palmer, C. C. Chan, C. A. Klebanoff, W. W. Overwijk, et al 4917–4923. ϩ ϩ 9. Hanahan, D., G. Christofori, P. Naik, and J. Arbeit. 1996. Transgenic mouse 2005. CD8 T cell immunity against a tumor/self-antigen is augmented by CD4 models of tumour angiogenesis: the angiogenic switch, its molecular controls, T helper cells and hindered by naturally occurring T regulatory cells. J. Immunol. and prospects for preclinical therapeutic models. Eur. J. Cancer 32A: 174: 2591–2601. by guest on September 25, 2021 2386–2393. 29. Tien, A. H., L. Xu, and C. D. Helgason. 2005. Altered immunity accompanies 10. Naik, P., J. Karrim, and D. Hanahan. 1996. The rise and fall of apoptosis during disease progression in a mouse model of prostate dysplasia. Cancer Res. 65: multistage tumorigenesis: down-modulation contributes to tumor progression 2947–2955. from angiogenic progenitors. Genes Dev. 10: 2105–2116. 30. Saio, M., S. Radoja, M. Marino, and A. B. Frey. 2001. Tumor-infiltrating mac- ϩ 11. Verdaguer, J., J. W. Yoon, B. Anderson, N. Averill, T. Utsugi, B. J. Park, and rophages induce apoptosis in activated CD8 T cells by a mechanism requiring P. Santamaria. 1996. Acceleration of spontaneous diabetes in TCR-␤-transgenic cell contact and mediated by both the cell-associated form of TNF and nitric nonobese diabetic mice by ␤-cell cytotoxic CD8ϩ T cells expressing identical oxide. J. Immunol. 167: 5583–5593. endogenous TCR-␣ chains. J. Immunol. 157: 4726–4735. 31. Gallina, G., L. Dolcetti, P. Serafini, C. De Santo, I. Marigo, M. P. Colombo, 12. Verdaguer, J., D. Schmidt, A. Amrani, B. Anderson, N. Averill, and G. Basso, F. Brombacher, I. Borrello, P. Zanovello, et al. 2006. Tumors induce ϩ P. Santamaria. 1997. Spontaneous autoimmune diabetes in monoclonal T cell a subset of inflammatory monocytes with immunosuppressive activity on CD8 nonobese diabetic mice. J. Exp. Med. 186: 1663–1676. T cells. J. Clin. Invest. 116: 2777–2790. 13. Tough, D. F., and J. Sprent. 1994. Turnover of naive- and memory-phenotype T 32. Yoon, J. W., H. S. Jun, and P. Santamaria. 1998. Cellular and molecular mech- cells. J. Exp. Med. 179: 1127–1135. anisms for the initiation and progression of ␤ cell destruction resulting from the 14. Martin, C. C., L. J. Bischof, B. Bergman, L. A. Hornbuckle, C. Hilliker, collaboration between macrophages and T cells. Autoimmunity 27: 109–122. C. Frigeri, D. Wahl, C. A. Svitek, R. Wong, J. K. Goldman, et al. 2001. Cloning 33. Direkze, N. C., K. Hodivala-Dilke, R. Jeffery, T. Hunt, R. Poulsom, D. Oukrif, and characterization of the human and rat islet-specific glucose-6-phosphatase M. R. Alison, and N. A. Wright. 2004. Bone marrow contribution to tumor- catalytic subunit-related protein (IGRP) genes. J. Biol. Chem. 276: 25197–25207. associated myofibroblasts and fibroblasts. Cancer Res. 64: 8492–8495. 15. Sotomayor, E. M., I. Borrello, E. Tubb, J. P. Allison, and H. I. Levitsky. 1999. 34. Silzle, T., G. J. Randolph, M. Kreutz, and L. A. Kunz-Schughart. 2004. The In vivo blockade of CTLA-4 enhances the priming of responsive T cells but fails fibroblast: sentinel cell and local immune modulator in tumor tissue. Int. J. Can- to prevent the induction of tumor antigen-specific tolerance. Proc. Natl. Acad. cer 108: 173–180. Sci. USA 96: 11476–11481. 35. Lieubeau, B., M. F. Heymann, F. Henry, I. Barbieux, K. Meflah, and 16. Sotomayor, E. M., I. Borrello, F. M. Rattis, A. G. Cuenca, J. Abrams, K. Stave- M. Gregoire. 1999. Immunomodulatory effects of tumor-associated fibroblasts in ley-O’Carroll, and H. I. Levitsky. 2001. Cross-presentation of tumor antigens by colorectal-tumor development. Int. J. Cancer 81: 629–636.