| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Therapeutics |
Department of Pathology and Comprehensive Cancer Center, Ohio State University Medical Center, Columbus, Ohio 43210
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Because naïve T cells circulate primarily among the blood and lymphatics (1) , it is most likely that their activation initiates in the lymphoid organ. This is supported by the classic study of Barker and Billingham (2) , who reported that intact, afferent lymphatics were essential for the rejection of allografts. Moreover, direct injection of tumor antigen into spleens drastically increased the efficacy of CTL-priming (3) . Furthermore, priming of antiviral CTLs is severely depressed in mice lacking secondary lymphoid organs (4) . However, to our knowledge, it has not been demonstrated that CTL response is indeed initiated in the lymphoid organs. Although utilization of transgenic T cells in analyzing the T-cell activation (5) helps to overcome the difficulties associated with a low frequency of antigen-specific naïve T cells, systematic analyses on the sites of initial T-cell activation has thus far not been reported, perhaps because of technical issues related to analyzing the T-cell activation in nonlymphoid target organs, including tumors. Here we take advantage of transgenic mice expressing the TCR3 specific for the unmutated tumor antigen P1A and the fact that the tumor-infiltrating cells in a P1A-expressing tumor can be quantitated easily; and we carried out detailed kinetics analyses of the CD8-T cell activation in the tumor versus the lymphoid organ. Our results demonstrate that activation of T cells in the lymphoid organ precedes that in the tumors. Activated T cells then migrate into the tumor, where they receive additional stimulation and change cell surface phenotype.
The second issue to be addressed in this study is the mode of antigen presentation at early stage of T-cell response. Genetic analysis using bone marrow chimera revealed two modes of antigen presentation for priming CTLs in vivo. Direct-priming is mediated by engagement of T cells by the cells that synthesize the protein with antigenic epitope, whereas cross-priming, as originally described by Bevan (6) , was mediated by host antigen-presenting cells that take up antigens synthesized by other cells. Because of its profound implications on the development of tumor vaccine, the mechanism for priming tumor-specific T cells has been vigorously debated. Huang et al. (7) showed that MHC-deficient tumors expressing an influenza antigen primed CTLs, which indicates that CTLs can be activated by antigens synthesized in tumors without direct antigen presentation. In contrast, Kundig et al. (3) and Ochsenbein et al. (8) reported that an antigenic peptide from a glycoprotein of lymphocytic choriomeningitis virus expressed on tumor cells cannot induce cross-priming. However, all of these analyses were based on the accumulation of final products of the immune response, i.e., the increase of memory or effector CTLs. It remains unclear whether the mode of antigen presentation implicated was required for either initiation or amplification of a CTL response or even the survival of naïve T cells. Moreover, several recent reports indicate that in vivo CTL response can be primed by either direct- or cross-presentation, depending on antigenic epitopes and pathogens (9 , 10) . It is therefore valuable to use a natural tumor antigen to evaluate the mechanisms for priming antitumor CTL response. To address this critical issue, we adoptively transferred P1A-specific transgenic T cells (11) into mice that bore the P1A-expressing tumors. Our results demonstrated that both direct- and cross-presentations of the same antigenic epitope from P1A lead to activation of the P1A-specific T cells in the lymphoid organ.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cell Lines.
BALB/c Plasmocytoma J558 transfected with either vector alone (J558-Neo) or B7-1 (J558-B7) has been described (12)
. The mutant cell line ReB7 was derived from J558-B7, as described (13)
.
Adoptive Transfer of Purified Transgenic T Cells.
Pools of spleen and lymph node cells from the P1CTL-transgenic mice were incubated with a cocktail of mAbs (anti-CD4 mAb GK1.5, anti-FcR mAb 2.4G2, anti-CD11c mAb N418, and anti-MHC class II mAb M5/114). After the removal of unbound mAbs, the cells were incubated with anti-immunoglobulin-coated magnetic beads. The antibody-coated cells were removed by a magnet. The unbound cells consisted of >90% CD8 T cells, with no detectable CD4 T cells. The purified CD8 T cells were labeled with CFSE as described (14)
and adoptively transferred into tumor-bearing mice. The routes of injection were i.p., i.v., or intratumoral.
Flow Cytometry.
Both T-cell division and acquisition of cell surface-activation markers were determined by flow cytometry. The transgenic T cells were labeled by their expression of CD8 and TCR-
chain V8. Other mAbs used were anti-CD24 mAb (M1/69), anti-CD44 mAb (IM7), and anti-CD62L mAb (Mel14), all purchased from PharMingen (San Diego, CA). Briefly, spleens, lymph nodes, and tumors were harvested by surgery; single-cell suspensions were prepared by grinding on frosted slides. The cell debris was removed by centrifugation through a Ficoll-Hypaque solution. These viable cells were stained and analyzed.
Detection of Tumor Cells in the Spleen.
RAG-2(-/-) BALB/c mice were injected with 5 x 106 of either J558-Neo, J558-B7, or ReB7. At 2–3 weeks after tumor injection, the spleens from tumor-bearing mice were harvested, and the presence of tumor cells was evaluated in 96-well-plate microcultures. Briefly, single cell suspensions from the whole spleen were aliquoted into 96-well plates. At 2–3 weeks, the numbers of tumor-cell colonies were counted microscopically. Cloning efficiencies of each tumor type were determined by limiting dilution analysis of a known number of tumor cells in the presence of RAG-2(-/-) spleen cells and calculated based on the Poisson distribution. The number of tumor cells/spleen was calculated by the following formula:
Number of tumor cells/spleen = Number of tumor colonies/cloning efficiency.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, although a significant number of T cells were recovered from the spleen at 17, 36, 48, and 60 h after adoptive transfer, essentially no T-cell division was detected during the first 60 h. These results demonstrated that T-cell division within this period in the tumor-bearing mice is induced by tumors. Homeostatic T-cell division, however, can be found at 96 h after inoculation, and it can become quite substantial at 5–7 days after adoptive transfer (data not shown), as has been demonstrated with other transgenic mice (15
, 16)
.
|
, substantial numbers of T cells were detected in the spleen at 16 h after the adoptive transfer. At this point, no division of T cells had taken place, although a small proportion of T cells were enlarged. By 36 h, a small fraction of T cells had undergone one to two divisions. Twelve h later, a significant proportion of cells had undergone two divisions. By 60 h, most cells had undergone one to six divisions. A comparison between the T cells at 36 and 60 h revealed that the transgenic T cells can divide at a maximal rate of 4–5 h/division, which is consistent with previous estimates for viral and foreign antigen-specific T cells (17, 18, 19)
. Because no significant division of T cells was found in tumor-free mice during the same period (Fig. 1)
, the activation of T cells in the first 60 h studied here must have been induced by the tumor cells. Interestingly, tumor-specific T cells seemed to be trapped in the lymphoid organ early on, as very few T cells were detected within the tumors during the first 48 h of injection, and the few cells recovered can be easily accounted for by the contamination of blood. More importantly, because most of the T cells had divided several times when they appeared in the tumors, they are most likely the descendants of T cells activated in the secondary lymphoid organ.
|
, in the spleen, a significant proportion of T cells had divided one to three times at 36 h after the adoptive transfer. By the 60 h, the majority of T cells had undergone more than four divisions. In contrast, very few, if any, T cells were detected in the tumors before 60 h after adoptive transfer. Again, because the T cells had already undergone multiple divisions when they were first found in the tumors, they are likely the descendants of T cells activated in the lymphoid organ. Thus, expression of B7-1 on the tumor cells does not allow initiation of an antitumor CTL response in the tumor milieu.
|
, legends). As expected, a significant number of T cells had leaked into the blood stream and accumulated in the spleens. Nevertheless, no T cell division was detected within the tumors even at 42 h after adoptive transfer, when a substantial proportion of T cells in the spleen had divided two or more times (Fig. 4)
.
|
Cross Presentation Is Sufficient to Initiate T-Cell Clonal Expansion.
Tumor antigens can be presented to T cells by the tumor cells (direct-presentation) or by the host APCs that take up the tumor antigen and present them in the MHC class I pathway (cross-presentation). To determine the role for cross-presentation in the T-cell response, we used a tumor mutant that lacks cell surface MHC class I. The cell line, which we called ReB7, expresses tumor antigen P1A but fails to express LMP2, LMP7, TAP1, and TAP2 (13)
. Data in Fig. 5, a and b
, confirmed that the cell line lacked cell-surface MHC class I H-2 Ld, the restricting MHC for P1A, and was not recognized by cytotoxic T cells. Therefore, the cell line was unable to present the antigen to T cells directly. As shown in Fig. 5c
, in mice that bore ReB7 tumors, T cells divided rapidly in the spleen. The rate of division was comparable with that in mice that bore either J558-B7 or J558-Neo tumors (Figs. 2
3
4
). Again, very few T cells were detected in the tumors before 48 h after adoptive transfer. Most T cells had finished multiple divisions before their arrival in the tumors at 60 h after adoptive transfer. The rapid T-cell division in the spleen of ReB7 tumor-bearing mice revealed that cross-priming was sufficient to initiate a T-cell response in the spleens but not in the tumors.
|
, can directly engage TCR on P1CTL to induce proliferative response. Although the cells that locally activated the P1CTL were not identified, we have reported that the plasmacytoma was infiltrated with large numbers of macrophages and dendritic cells (21
, 22)
. Because these cells are known to cross present tumor antigen, it is most likely that they presented the P1A to P1CTL to induce both recruitment and local proliferation of P1CTL within the MHC class I- tumor cells.
Direct Presentation by Tumor Cells Is Also Sufficient to Induce the P1A-specific CTL Response: Requirement for Both Antigen and Costimulator on the Tumor Cells.
To test whether the direct presentation of tumor antigens can also induce T-cell clonal expansion, we injected the J558-Neo and J558-B7 tumors into allogeneic C57BL/6.SCID mice lacking T and B cells. Because allogeneic NK cells confer resistance to the plasmacytoma (data not shown), we eliminated NK cells by monoclonal antibodies specific for the ß-chain of the IL-2 receptors (23)
. In the NK-depleted host, the J558-B7 and J558-Neo tumors grew at a comparable rate (data not shown).
Because the P1CTL TCRs recognize P1A presented by the H-2 Ld, the host APC (H-2b) cannot present the tumor antigen to P1CTL. To confirm this, we adoptively transferred T cells into mice that bore the MHC class I- tumor variant, ReB7. Because the T cells did not divide in the mice over a 4-day period (Fig. 6
, bottom), the host APC must have been unable to mediate cross-priming. Moreover, the T cells used were not contaminated with a sufficient number of H-2d APC to mediate cross-priming. In the C57BL6.SCID mice that bore the J558-B7 tumors, P1CTL had undergone rapid proliferation (Fig. 6
, middle). Over a 4-day period, P1CTL obtained from the peritoneal cavity (the site of injection), spleens, and tumors had undergone five or more divisions. Thus, direct presentation by B7-1+ tumor cells is sufficient to induce T-cell clonal expansion. Interestingly, in mice that bore J558-Neo tumors, essentially no T-cell division was observed (Fig. 6
, top). Although a significant number of T cells were present in the peritoneal cavity on day 4 after adoptive transfer, the T cells had not undergone any division. As expected, the numbers of T cells found in spleen and tumors were 13- to 40-fold lower than their counterparts in mice bearing the J558-B7 tumors. Thus, activation of T cells by tumor cells via direct antigen presentation requires the expression of both TCR ligand and costimulatory molecules. This is in striking contrast to what we observed in the syngeneic host, where expression of B7-1 on the tumor cells was not required for massive T-cell expansion.
|
, whereas the distributions of the T-cell division numbers were comparable among the two groups, the amounts of T cells induced by cross-priming is about twice of that by direct-priming. Moreover, down-regulation of CD62L among the P1CTL in spleen is a late event, inasmuch as it was observed only among cells that had undergone more than six divisions, and because it was prominent on day 5 but barely so on day 3, as indicated in Figs. 8
and 9
. Therefore, the relative proportion of CD62Llow T cells can be used as an additional indicator for the efficiency of T-cell activation in vivo. As shown in Fig. 7
, over a 5-day period, cross-priming led to the acquisition of CD62Llow phenotype among 79% of P1CTL, whereas direct-priming lead to CD62L down-regulation among 52% of P1CTLs. Thus, cross priming is modestly more efficient than direct-priming.
|
|
|
, the majority of the T cells in the spleen had undergone one to five divisions, whereas most of those in the tumors had divided more than six times. Moreover, T cells in the spleen and in the tumors have distinct cell surface phenotypes. Thus, on day 3 of adoptive transfer, most of spleen T cells had high levels of CD62L, despite their extensive division. In contrast, the tumor-infiltrating T cells had down-regulated their CD62L. The HSA (also called CD24) is up-regulated transiently after T-cell activation (25 , 26) , and its expression is essential for the effector function of autoreactive T cells in the central nervous system (27) . Interestingly, whereas a substantial proportion of tumor-infiltrating T cells expressed high levels of CD24, T cells in the spleen of the same mice did not. The CD62L and CD24 phenotypes were not merely a reflection of cell division, as cells that have undergone the same divisions had distinct phenotypes depending on their anatomical location. As expected, all T cells had up-regulated the CD44 marker.
The accelerated activation of T cells within the tumor milieu can be attributable to either direct- or cross-presentation of tumor antigens. To test whether direct-presentation of tumor antigens is necessary for the accelerated T-cell division, we tested whether the MHC class I- variant of J558-B7, ReB7, can promote local T-cell activation. As shown in Fig. 9
, TIL from ReB7 had undergone substantially more division than the T cells in the spleen. Moreover, the T cells in the ReB7 tumors were CD44highCD62LlowCD24high/low, whereas the majority of those in the spleen were CD44highCD62LhighCD24low. Thus, local activation of T cells does not require direct antigen presentation by the tumor cells.
Quantitation of Tumor Cells in the Spleens of Mice with Large Tumor Burden.
We have reported that the J558-plasmocytoma is not metastatic in the immune-competent mice. We have also failed to detect tumor cells in the spleen of the RAG-2(-/-) mice with large tumors (
1 cm in diameter) by flow cytometry using an anti-PC1 antibody specific for plasma cells, by reverse transcription-PCR using primers specific for tumor antigen P1A, and by histology (data not shown). We therefore used microculture to detect the low number of tumor cells in the spleen.
We first established the cloning efficiency of the tumor cells by limiting dilution. RAG-2(-/-) spleen cells were added to the microculture to normalize against any potential effect of the spleen cells on tumor growth in vitro. As shown in Table 1
, all three lines of tumor cells used in the study exhibited high cloning efficiencies ranging from 22.5 to 42.9%, which indicates that a small number of tumor cells in the spleen can be detected by limiting dilution. The data from three to four individual spleens in each group were summarized in Table 1
. Although a significant number of tumor cells were found in the spleen of most mice bearing either J558-B7 or J558-Neo tumors, the majority of mice bearing ReB7 tumors had no tumor cells in their spleens.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Initiation of CTL Response in the Lymphoid Tissues.
We adoptively transferred CFSE-labeled transgenic T cells specific for tumor antigen P1A into syngeneic RAG-2 mice that bore P1A-expressing tumors and analyzed kinetically the T-cell response in the spleen, lymph nodes, and tumors. Interestingly, the T cells were restricted to the lymphoid system until 60 h after adoptive transfer. During this period, T cells in the spleen and lymph nodes underwent several rounds of division. A significant number of T cells appeared in the tumor at 60 h after adoptive transfer. Because these T cells had undergone multiple rounds of divisions when arriving at the tumor milieu, they must have been the descendants of the cells that had divided in the lymphoid organ. Thus, the initial activation of tumor-specific T cells takes place within the lymphoid organ.
Activation of naïve T cells requires both antigenic and costimulatory signals (28, 29, 30, 31) . Because the lymphoid system contains host APCs that express costimulatory molecules, one may suggest that the requirement for a lymphoid environment for activation of tumor-specific T cells reflects the need for costimulatory molecules. Here, we showed that expression of the costimulatory molecule B7-1 on the tumor cells does not change the sites of the initial activation. Moreover, naïve T cells tend to circulate within the lymphoid system (1 , 32) . Initial sites of T cell activation may therefore simply be caused by the localization of naïve T cells. This has also been disproved, as direct injection of antigen-specific T cells in the tumors failed to initiate T-cell activation in the tumor. Thus, in addition to the localization of T cells and costimulatory cells, the lymphoid organs have other properties that are uniquely suited for the initiation of the immune response. The exact nature of the lymphoid environment that promotes the initiation of T cell response is still unclear (33) . Because tumor cells can be easily manipulated genetically, our model may be valuable for dissecting the molecular basis for the initiation of T-cell response.
Although we are not aware of any direct evidence that T-cell response is initiated in the lymphoid organs, this concept offers the most logical interpretation of a long list of observations. (a) Naïve T cells normally circulate within the blood and lymphoid system, and the activated T cells acquire the ability to patrol the nonhematopoietic cells (1 , 32) . (b) Bone marrow-derived APC is the essential antigen-presenting cell for T-cell response to some viruses (34) or viral antigens expressed on the tumors (7) . (c) A sophisticated phagocytic and dendritic system allows the capture of antigens from various tissues and transportation of these antigens to the lymphoid system (35) . (d) Immune response is severely compromised in mice with disrupted spleen and lymph nodes (4) . Our direct kinetic analysis of T-cell activation in the lymphoid tissues and the target tissues (tumors) has substantially strengthened this notion.
Both Tumors and Host APC Can Initiate Antitumor CTL Response in the Lymphoid Organ.
Tumor antigens can be localized to the lymphoid tissues by at least two distinct mechanisms, including direct presentation by tumor cells that have metastasized into the lymphoid system or by cross-presentation by host APCs that have up-taken tumor antigen from the tumors. Our results, presented in the current studies, suggest that both of these mechanisms may contribute to the induction of a single clone of T cells specific for the P1A epitope. Thus, in syngeneic mice, a tumor variant that cannot present antigen directly to the T cells induces potent T-cell clonal expansion. Consistent with the notion that the host APCs are involved in antigen presentation, we found that expression of B7-1 on the tumor cells was not necessary for T-cell clonal expansion in the syngeneic mice. In contrast, in the allogeneic mice in which the host APC cannot present the correct antigenic epitope, tumor cells that expressed both MHC class I and costimulatory molecule B7-1 induced a potent proliferation. To our knowledge, this is the first demonstration that expression of B7-1 on the tumor cells enables them to prime directly the T-cell response in vivo, although the notion that expression of B7 converts tumor cells into professional APCs has been proposed for many years (36, 37, 38)
. Moreover, the requirement of B7-1 on the tumors for direct priming confirmed our previous in vitro study that presentation of both signal 1 and signal 2 by the same cells is required for optimal activation of T cells (39)
, because expression of B7-1 on host APC did not bypass the requirement for B7-1 on the tumor cells.
Because our models allow us to evaluate direct and indirect priming separately, we were able to compare the efficacy of direct and indirect priming using the same T cells and tumors with a similar level of nominal antigen (13 , 40) . Our data indicated that cross-priming is only modestly more efficient than direct priming. However, direct priming is subject to several limitations in tumor-bearing hosts. For instance, because most cancer cells are devoid of B7-1 and B7-2, they must be very inefficient in direct priming. In addition, our study involved no CD4 T cells that were known to enhance cross priming (41) . These two factors make it highly likely that cross priming is dominant under physiological conditions. Nevertheless, our finding that expression of costimulatory molecules on the tumor cells initiates direct priming, suggests that modification of tumor cells with costimulatory molecules can switch on a new mode of antigen presentation. Because host APCs are known to destroy certain tumor-antigenic epitopes (42) , direct priming may broaden the repertoire of antitumor T-cell response. Indeed, Johnson et al. reported that expression of B7-1 enlarged the spectrum of CTL epitopes (43) , although it is unclear whether these T cells were activated by direct priming.
Using tumor cells or fibroblasts expressing a lymphocytic chloriomenigitis virus glycoprotein, Kundig et al. (3) and Ochsenbein et al. (8) have demonstrated that this viral antigen cannot induce cross-priming. In contrast, MHC class I+ tumors expressing an influenza viral antigen were found unable to induce direct priming even if B7-1 is expressed (44) . Recent studies using viral and bacterial pathogens indicate that the antigen used may dictate the mechanism of priming (9 , 10) . This conclusion makes it difficult to generalize the mechanism for priming antitumor CTL response with selected foreign antigens expressed on the tumors. Our study analyzed CTL response against a natural tumor antigen P1A and should therefore be more relevant to the mechanism of antitumor CTL response in vivo. In this regard, it is worthy to note that the P1A antigen seems to be the first that can activate T cells by both direct- and cross-priming, although direct-priming can be convincingly demonstrated only if the B7-1 is present on the tumor cells.
At this point, it is unclear why different antigens use different mechanisms to prime T cells. At least two contributing factors can be suggested. First, immune proteosome on professional APC was reported to destroy some antigenic epitopes (42) . It is predicted then that those epitopes that are sensitive to the immune proteosome would be unable to induce cross-presentation. Second, as shown in this study, immunogenicity of antigen-expressing cells determines whether an antigen can prime T cells directly. As such, the density and antigenic epitope and other immunological properties of tumors, such as costimulatory activity or expression of adhesion molecules, also contributes to their ability to prime T cells directly.
Implications for Antitumor CTL Response.
The obligatory lymphoid environment and the mode of antigen presentation have an important bearing on the timing and hence the efficacy of antitumor CTL response. Clearly, for the antigens that cannot be cross-presented, solid tumors must first migrate into the lymphoid organ to activate T cells, as has been demonstrated (8)
. Because J558-Neo and J558-B7 tumor cells were found in the spleen, they are likely responsible for direct activation of T cells. This, however, is not necessarily true for the antigens that induce cross-priming. Indeed, in the mice that bore the ReB7 tumors, only 1 of 4 spleens had any tumor cells, yet T-cell clonal expansion was observed in every one of >20 syngeneic mice that were analyzed. It is therefore highly unlikely that tumor metastasis is essential for the induction of cross-priming. The observed infiltration and maturation of dendritic cells in the tumors (45)
, and the ability of the dendritic cells to migrate to the lymphoid organ (35)
, should provide a mechanism for cross presentation of tumor antigens to the naïve T cells in the absence of tumor metastasis. Therefore, our conclusion that T-cell activation can take place in the absence of tumor metastasis would make it plausible to suggest that the antitumor CTL response can be initiated early during tumorigenesis.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 This work is supported by CA58033, AI32981, CA69091, and CA82355 from NIH. 
2 To whom requests for reprints should addressed, at Department of Pathology and Comprehensive Cancer Center, Ohio State University Medical Center, 129 Hamilton Hall, 1645 Neil Avenue, Columbus, OH 43210. Phone: (614) 292-3054; Fax: (614) 688-8152; E-mail: liu-3{at}medctr.osu.edu
3 The abbreviations used are: TCR, T-cell receptor; mAb, monoclonal antibody; CFSE, carboxylfluoresceindiacetate succinimidyl ester; APC, antigen-presenting cell; SCID, severe combined immunodeficient; NK, natural killer; HSA, heat-stable antigen; PE, phycoerythorin.
Received 5/ 1/01. Accepted 7/16/01.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-expressing tumor: linkage between inflammation and specific immunity. Cancer Res., 59: 5549-5553, 1999.. J. Immunol., 159: 360-368, 1997.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  P. Yu, Y. Lee, Y. Wang, X. Liu, S. Auh, T. F. Gajewski, H. Schreiber, Z. You, C. Kaynor, X. Wang, et al. Targeting the Primary Tumor to Generate CTL for the Effective Eradication of Spontaneous Metastases J. Immunol., August 1, 2007; 179(3): 1960 - 1968. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. F. May Jr, K. Lute, E. Kocak, S. Abdessalam, L. Yin, O. Li, Z. Guan, G. Philips, P. Zheng, and Y. Liu Immune competence of cancer-reactive T cells generated de novo in adult tumor-bearing mice Blood, January 1, 2007; 109(1): 253 - 258. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. M. Hargadon, C. C. Brinkman, S. L. Sheasley-O'Neill, L. A. Nichols, T. N. J. Bullock, and V. H. Engelhard Incomplete Differentiation of Antigen-Specific CD8 T Cells in Tumor-Draining Lymph Nodes J. Immunol., November 1, 2006; 177(9): 6081 - 6090. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X.-F. Bai, J.-Q. Liu, P. S. Joshi, L. Wang, L. Yin, J. Labanowska, N. Heerema, P. Zheng, and Y. Liu Different Lineages of P1A-Expressing Cancer Cells Use Divergent Modes of Immune Evasion for T-Cell Adoptive Therapy Cancer Res., August 15, 2006; 66(16): 8241 - 8249. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Li, X. Chang, H. Zhang, E. Kocak, C. Ding, P. Zheng, and Y. Liu Massive and destructive T cell response to homeostatic cue in CD24-deficient lymphopenic hosts J. Exp. Med., July 10, 2006; 203(7): 1713 - 1720. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Liu, J. Idoyaga, A. Charalambous, S.-i. Fujii, A. Bonito, J. Mordoh, R. Wainstok, X.-F. Bai, Y. Liu, and R. M. Steinman Innate NKT lymphocytes confer superior adaptive immunity via tumor-capturing dendritic cells J. Exp. Med., December 5, 2005; 202(11): 1507 - 1516. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. R. Jenkinson, N. A. Williams, and D. J. Morgan The Role of Intercellular Adhesion Molecule-1/LFA-1 Interactions in the Generation of Tumor-Specific CD8+ T Cell Responses J. Immunol., March 15, 2005; 174(6): 3401 - 3407. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Roychowdhury, K. F. May Jr., K. S. Tzou, T. Lin, D. Bhatt, A. G. Freud, M. Guimond, A. K. Ferketich, Y. Liu, and M. A. Caligiuri Failed Adoptive Immunotherapy with Tumor-Specific T Cells: Reversal with Low-Dose Interleukin 15 but not Low-Dose Interleukin 2 Cancer Res., November 1, 2004; 64(21): 8062 - 8067. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-X. Gao, X. Chang, X. Zheng, J. Wen, L. Yin, P. Du, P. Zheng, and Y. Liu A new role for CD28 in the survival of autoreactive T cells in the periphery after chronic exposure to autoantigen Int. Immunol., October 1, 2004; 16(10): 1403 - 1409. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. S. Yang, C. E. Monken, and E. C. Lattime Intratumoral Vaccination with Vaccinia-Expressed Tumor Antigen and Granulocyte Macrophage Colony-Stimulating Factor Overcomes Immunological Ignorance to Tumor Antigen Cancer Res., October 15, 2003; 63(20): 6956 - 6961. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Liu, J. X. Gao, J. Wen, L. Yin, O. Li, T. Zuo, T. F. Gajewski, Y.-X. Fu, P. Zheng, and Y. Liu B7DC/PDL2 Promotes Tumor Immunity by a PD-1-independent Mechanism J. Exp. Med., June 16, 2003; 197(12): 1721 - 1730. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Yu, M. T. Spiotto, Y. Lee, H. Schreiber, and Y.-X. Fu Complementary Role of CD4+ T Cells and Secondary Lymphoid Tissues for Cross-presentation of Tumor Antigen to CD8+ T Cells J. Exp. Med., April 21, 2003; 197(8): 985 - 995. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X.-F. Bai, J. Liu, K. F. May Jr, Y. Guo, P. Zheng, and Y. Liu B7-CTLA4 interaction promotes cognate destruction of tumor cells by cytotoxic T lymphocytes in vivo Blood, April 15, 2002; 99(8): 2880 - 2889. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
