Improved generation of anti-tumor immunity by antigen dose limitation
- 6.3k Downloads
The malignant cells of cutaneous T cell lymphoma (CTCL) display immunogenic peptides derived from the clonal T cell receptor (TCR) providing an attractive model for refinement of anti-tumor immunization methodology. To produce a clinically meaningful anti-tumor response, induction of cytotoxic anti-CTCL cells must be maximized while suppressive T regulatory cells (Treg) should be minimized. We have demonstrated that engulfment of apoptotic CTCL cells by dendritic cells (DC) can lead to either CD8 anti-CTCL responses or immunosuppressive Treg induction. Treg generation is favored when the number of apoptotic cells available for ingestion is high.
In this study, we sought to determine whether the balance between immunity and immunosuppression could be shifted towards a CD8 anti-CTCL response by lowering the ratio of apoptotic CTCL cells available for DC ingestion. CTCL cell apoptosis was produced by engagement of the TCR by anti-CD3 antibody affixed to magnetic beads.
The physical perturbation inherent in passage through a separation column induced monocytes to differentiate into DC, demonstrated by increased expression of class II and CD86 and decreased expression of the monocyte marker CD14. The immature DC internalized and processed apoptotic CTCL cells and could potentially present the tumor-derived peptides in the context of MHC class I and II. As the number of apoptotic cells increased, there was a dose-dependent increase in the expression of Treg markers CTLA-4, CD25, and FoxP3, with a ratio of apoptotic cell/DC loading of > 10:1 corresponding to the greatest Treg induction. These inducible phenotypic Treg also functionally inhibited CD8-mediated perforin expression in vitro. At lower levels of apoptotic cell/DC loading of < 5:1, there was an expansion of the CD8 T cell compartment with increased perforin expression and increased CTCL cell death, indicating anti-tumor activity.
These findings demonstrate that the ratio of apoptotic cells supplied to DC is an important determinant of whether CD8 anti-tumor immunity or immunosuppression is generated.
KeywordsDendritic Cell Apoptotic Cell Treg Cell Immature Dendritic Cell Autologous Dendritic Cell
Cutaneous T cell lymphoma (CTCL) is an umbrella designation that unifies a diverse group of clinical presentations on the basis of histopathologic and immunologic criteria. The malignancy is a clonal proliferation of epidermotropic T cells [1, 2, 3], that uniformly carry a common T cell receptor (TCR) and also display cell surface expression of a memory (CD45RO+), inducer (CD4+), and cutaneous homing leukocyte antigen (CLA+) phenotype. Initially the tumor cells localize in the skin of afflicted patients, surrounding Langerhans cells that contribute to the CTCL cell growth . As the disease progresses, the malignant cells become more poorly differentiated and often spread hematogenously throughout the body, as a leukemia forecasting a much poorer prognosis. It is the early epidermal focus of malignant T cells surrounding a central Langerhans cell , an immature member of the dendritic cell (DC) series , that is the diagnostic hallmark of the disease, the Pautrier microabscess . CTCL tumor cells lack the co-stimulatory molecules required to trigger an immune response contributing to their ability to evade induction of anti-tumor immunity and thus, persist and disseminate. Despite the role of DC in providing proliferative support for the malignancy, DC immunotherapy has demonstrated clinical benefit in this disease [7, 8].
We have recently found that passage of cells from CTCL patients through an anti-CD3 magnetic bead column can achieve the simultaneous induction of apoptosis in the malignant T cell population and differentiation in the monocyte population creating, after overnight culture, a population of tumor-loaded DC capable of initiating an immune response in vitro . This column-based method of generating tumor-loaded DC utilizes the mechanistic principles discovered in extracorporeal photopheresis (ECP), the first FDA-approved immunotherapy for cancer . Clinical trials of ECP have shown response rates of 73% in late stage CTCL patients with an extremely safe side effect profile . ECP has been recently modified by the addition of an overnight incubation step. The new procedure has been termed " Transimmunization", wherein malignant T cells were rendered apoptotic by ultraviolet A (UVA) light photo-activated 8-methoxypsoralen (8-MOP) and were avidly engulfed by immature DC, whose transition from monocytes was triggered by the physical perturbation produced when the cells were passed through the UVA exposure plate of the ECP apparatus. In Transimmunization, the transitioning DC and the apoptosing leukocytes are co-cultivated overnight to permit engulfment of the apoptotic cells and subsequent processing and presentation of the tumor-derived peptides prior to re-infusion into the host .
We have previously shown that when CTCL cells encounter autologous DC loaded with high numbers of apoptotic cells, they adopt the phenotype and function of T regulatory (Treg) cells, expressing high levels of the Treg markers CTLA-4, CD25, and FoxP3, as well as secreting interleukin-10 (IL-10) and transforming growth factor-β (TBF-β) and suppressing normal T cell antigen driven secretion of interleukin-2 (IL-2) and interferon-γ(IFN-γ) . The induction of Treg from responding CTCL cells may be hindering the effectiveness of existing immunotherapies for CTCL, and understanding the mechanism of their induction is paramount to the generation of more effective immunotherapy.
In these experiments, we sought to determine if the level of apoptotic cell loading controlled the balance between the development of an anti-tumor immune response and immunosuppressive Treg cell generation. We found that when lower numbers of apoptotic cells were processed by DC, a CD8 T cell response could be stimulated while at higher levels of apoptotic cell up-take by DC we could induce CTCL cell Treg conversion and inhibit the CD8 T cell response.
Specimens were obtained from a leukapheresis harvest of CTCL patients (in accordance with the guidelines of the Yale human investigation committee) undergoing treatment with standard ECP. All patients had advanced disease with clonal CD4+ T cell populations present in the peripheral circulation as determined by immunophenotyping with antibodies to the clonotypic variable region of family-specific TCR or polymerase chain reaction to detect rearrangements of the beta or gamma chain of the TCR as well as CD8 T cell compartments > 10% of the circulating lymphocyte population.
On Day 0, peripheral blood mononuclear cells (MNL) were isolated from the leukapheresis harvest by centrifugation over a ficoll-hypaque gradient followed by two washes in RPMI 1640 (Gibco, Gaithersburg, MD) containing 10% AB serum and 2 mM EDTA. An aliquot of the MNL was then isolated, purified, and cultured for subsequent re-addition on Day 1. Within this aliquot, CD4+ and CD8+ cells were column-purified using 40 μl Macs α-human CD4 and CD8 microBeads (Miltenyi Bioteck, Auburn CA) according to manufacturer's instructions. The enriched CD4+ and CD8+ T cells (> 97% ± 0.5% CD4+ and > 98% ± 0.5% CD8+, respectively) were suspended in RPMI 1640 with 15% autologous serum and IL-2/IL-7 and cultured individually in separate wells of a 12 well tissue culture plate (Falcon). Another aliquot of the Day 0 isolated MNL was reserved to construct the apoptotic cell and DC dose response curves, and controls. MNL (2 × 107) from the leukapheresis harvest were incubated with 40 μl Macs α-human CD3 microBeads following the manufacturer's directions. As previously described , CTCL cell binding to CD3 antibody rendered the malignant T cells apoptotic. As a control, another portion of MNL (2 × 107) was incubated with 40 μl Macs α-human CD4 microBeads. Following the twenty minute incubation, the cells were passed through the magnetic bead column, allowing separation of the MNL into either CD3+ and CD3- fraction or CD4+ and CD4- fractions. The CD3-treated and CD4-treated cells were then cultured in 3 ml RPMI 1640 containing 15% AB serum. In some experiments, increasing doses of CD3-treated cells were added to 105 autologous DC obtained from the column eluate to create the dose response curves. Controls consisted of increasing doses of CD4-treated CTCL cells added to 105 autologous DC.
On Day 1, the individually cultured column-purified CD4 and CD8 cells (106) were added to the CD3-treated apoptotic cell-loaded DC as well as the CD4-treated control population. After a second overnight culture, the cells were harvested, counted, and immunophenotyped for markers of T cells, DC, Tregs, and apoptotic cells.
In order to monitor enrichment of the purified CD4+ and CD8+ cell populations and to measure the effect of loading DC with differing number of apoptotic malignant T cells, the cells were stained by two-color immunofluorescence with a panel of antibodies to monocytes, DC, Tregs, and cytotoxic T cells. Cells (1 × 106) were incubated with 10–20 μl of fluorochrome conjugated monoclonal antibody for 30 minutes in the dark at 4°C. The antibodies were directly conjugated to fluorescein (FITC) or phycoerythrin (PE) and included: CD3-FITC (pan T cell); CD4-FITC (inducer T cell); CD8-FITC (cytotoxic T cell); CD25-FITC (IL2 receptor); CD14-FITC (monocytes); HLA-DR-FITC (anti-class II MHC molecule) and CD86-PE (B7.2 co-stimulatory molecule) and their isotype controls. Cells were washed once and suspended in PBS and read on a FC500 flow cytometer (Beckman Coulter) within 24 hours.
Combined membrane and cytoplasmic staining was performed following manufacturers instructions for cell fixation and permeabilization (Intraprep kit, Beckman Coulter). Antibody combinations included: membrane CD4-FITC/cytoplasmic CTLA4-PE (inhibitory member of the co-stimulatory family); membrane CD3-FITC/cytoplasmic CTLA4-PE; membrane CD4-FITC/cytoplasmic FoxP3 (inhibitory transcription factor); membrane CD3-FITC/cytoplasmic Apo2.7-PE (apoptotic cells); membrane CD8-FITC/cytoplasmic perforin-PE (cytotoxic granule); and isotype controls (Beckman Coulter). Data was analyzed using the CXP software (Beckman Coulter).
The expression of DC markers and the induction of a Treg or cytotoxic T cell response were evaluated statistically in 3 to 5 replicate cultures by the student's t test or if the data was not normally distributed the Mann-Whitney Rank Sum Test using the Sigma Stat analysis program.
CTCL Cells Rendered Apoptotic by CD3 Antibody Binding
Passage of Monocytes through Magnetic Column Induces DC differentiation
In addition, expression of class II and CD86 increased significantly (P ≤ 0.017–P ≤ 0.029 respectively) upon co-culture with column-differentiated DC fed apoptotic CTCL cells as compared to Day 0 controls. Class II expression increased from the level found on Day 0 monocytes to an intermediate increase found on the transitioning immature DC that had been co-cultured with viable CTCL cells (P ≤ 0.011) to the maximal increase seen when apoptotic CTCL cells (Fig. 2B) were co-cultured with the immature DC. A 5.5-fold increase in class II expression was found when the DC were fed apoptotic CTCL cells and compared to the monocytes isolated on Day 0. Expression of the co-stimulatory molecule CD86 also increased from day 0 to the intermediate increase found on transitioning immature DC that had been co-cultured with viable CTCL cells (P ≤ 0.029) to the maximal increase in DC fed apoptotic CTCL cells (Fig. 2C). A 25.8-fold increase in CD86 expression was found when DC that had engulfed apoptotic cells were compared to monocytes tested on primary isolation. Therefore, passage of CTCL cells through an anti-CD3 magnetic bead column and subsequent co-culture of newly apoptotic CTCL cells with column activated monocytes generated significantly enhanced DC differentiation. Co-culture of column-activated monocytes with viable CTCL cells was also effective in generating immature DC but the addition of apoptotic CTCL cells enhanced the adoption and stabilization of this phenotype.
Induction of a Treg phenotype
Treg CTCL cells suppress CD8-mediated perforin expression
Expansion of CD8 T cell compartment following co-culture with autologous DC that have ingested low numbers of apoptotic cells
Increased perforin activity and increased T cell death following loading of autologous DC with low numbers of apoptotic cells
The development of more effective DC-based immunotherapy has been hampered by a number of complicating factors, ranging from the length of culture necessary for DC development in vitro as well as a lack of knowledge of the most appropriate stage of DC maturation required for effective anti-tumor vaccination . In addition, the most appropriate antigen source for DC loading remains to be determined, with a variety of existing strategies including: purified peptides ; tumor-targeted viral vectors ; tumor/DC hybrids; apoptotic whole tumor cells; and tumor lysates having been attempted and resulted in substantial variability in long-term clinical responses .
Utilizing a magnetic bead column, we were able to generate potentially immunogenic DC in a fashion that offers several advantages over existing methodologies: 1) DC generated in this fashion can be obtained after a single overnight culture, far more rapidly than existing culture methods which can take up to 7 days; 2) Apoptotic whole cells are used as a source of tumor antigen, ensuring that the spectrum of CTCL antigenicity is incorporated into the DC instead of singular known antigens which may not be optimal or may be lost as the tumor de-differentiates; 3) Passage of monocytes through the column matrix induces the synchronized differentiation of a large population of monocytes into DC, allowing for the generation of a more robust source of APC displaying a spectrum of antigens upon re-injection; 4) The rapidity and ease of manipulating the cellular subpopulations of apoptotic CTCL cells and antigen-presenting cells allows the ability to easily optimize the cell populations and added factors in the hope of creating more effective clinical immunotherapy.
In these experiments, we sought to further our understanding of the conditions that favor the induction of Treg cells following exposure to autologous DC co-cultured with large numbers of apoptotic CTCL cells. The use of the magnetic bead column provides a new avenue for rapid generation of tumor-loaded DC because the variety of cellular components inherent in DC differentiation can be titrated to optimize dosing with this method.
In our studies, CTCL cells were incubated with anti-CD3 antibody prior to passage through the magnetic bead column, rendering the antigen-experienced malignant T cells apoptotic, as had been previously demonstrated. In addition, the use of anti-CD4 antibody allowed for separation of CD4+ CTCL cells to be used as a control population while at the same time not altering the CTCL cells. The membrane perturbation inherent in monocyte passage through the column matrix physically stimulates their transition into DC . DC differentiation was measured by both the reduction of expression of the monocyte marker CD14 and the increased expression of class II and CD86. In all instances, significant changes in the expression of these cell markers were found, indicating that the column passaged monocytes were differentiating into DC and that co-culture of the maturing DC with apoptotic malignant T cells further stimulated their entry into the DC pathway. Differentiation of monocytes utilizing membrane stimulation, as in passage through the column matrix, is advantageous in particular in the setting of cancer immunotherapy since it has been demonstrated that physical perturbation is one of the more effective means of cross-priming phagocytosed peptides into the MHC class I pathway, which is crucial in generation of a CD8 anti-tumor response .
Following the generation of potentially immunogenic DC using the magnetic bead column, we sought to determine if Treg conversion could be induced. By loading dendritic cells with large numbers of apoptotic cells, in a ratio of > 10 apoptotic cells to 1 DC, we were able to show that CTCL cells could be induced to assume some of the phenotypic and functional features of Treg cells. These findings correlate with the work of Berger et al, who has shown that co-culture of freshly purified CTCL cells with autologous DC loaded with large numbers of apoptotic CTCL cells can induce a Treg profile in the responding CTCL cell population, one that has phenotypic and functional characteristics associated with Treg cells . Treg are a subset of T cells that comprise 5–10% of the peripheral T cell population, and are involved in the maintenance of peripheral tolerance and prevention of autoimmune disease, regulating the response to infectious agents, transplanted tissues, and self antigens [19, 20, 21]. Treg are characterized by their ability to suppress immune responses primarily by inhibiting normal T cell antigen driven proliferation [22, 23]. Treg have been found in increased numbers in solid tumors and are currently seen as one of the major hurdles to beneficial clinical responses in existing immunotherapies . Removal of the suppressive Treg population has been shown to improve the outcome of cancer immunotherapy in both mouse models and clinical trials in a number of malignancies .
As the dose of apoptotic CTCL cells fed to a constant number of DC increased, expression of the Treg-associated markers CTLA-4, CD25, and FoxP3 also increased concomitantly. In addition, there appeared to be a threshold ratio of apoptotic cell:DC that corresponded with induction of the Treg phenotype. As the apoptotic cell:DC ratio increased from 5:1 to 10:1, expression of the hallmark of Treg cells, FoxP3, increased nearly four fold, indicating that perhaps there may be some internal DC checkpoint related to the number of processed apoptotic cells that governs the generation of an immune response versus a suppressive one. While specific mechanisms of this process remain unclear, we can speculate that perhaps a CD8 T cell response is favored when low numbers of peptides are presented while induction of an immunosuppressive response requires a higher antigen burden and presentation of elevated levels of peptides in MHC class II. In our experiments, the induced phenotypic Tregs also displayed some functional capabilities, as they were able to suppress CD8-mediated perforin expression in co-culture; an observation, which if borne out in vivo, may explain some of the failures in generating anti-tumor immunity with existing immunotherapies. These findings are in concert with our previous studies which demonstrated that cultured CTCL cells can be stimulated to assume a Treg phenotype and function suppressing normal T cell driven immune responses when exposed to DC that have ingested apoptotic cells . In our present studies, we demonstrate that, in addition, freshly isolated CTCL cells can be driven to adopt a Treg phenotype by overnight exposure to DC loaded with large numbers of apoptotic cells.
By establishing that Treg cells could be rapidly generated using this method, we then sought to determine if that delicate balance between the DC and surrounding numbers of apoptotic cells could be shifted in the other direction, towards stimulating a more vigorous immune response. By lowering the dose of apoptotic cells processed by a constant number of DC, we were able to induce significant expansion of the CD8 T cell compartment. This expanded CD8 T cell compartment not only contained a greater percentage of perforin expressing cells but also mediated increased T cell death in the added CTCL cell population. In terms of existing immunotherapies, these findings hold major clues into possible mechanisms for tumor survival following treatment. If too aggressive an attempt is made at killing tumor cells in hopes of loading DC with large number of tumor antigens, it may lead to Treg generation and suppression of the desired immune response. A considerable body of literature exists supporting the idea that the difference between the DC's ability to induce tolerance or mediate immunity may be determined by its persistent exposure to antigen. [25, 26] In our experiments, the DC that were exposed to larger numbers of apoptotic cells and hence had more persistent antigen exposure induced suppression of the CD8 T cell population. These findings support earlier work and help to reinforce the role of apoptosis in the induction of immunosuppression. [27, 28]
A better understanding of the role of Treg cells in immunotherapy is crucial because while they hamper the effectiveness of cancer immunotherapy , Treg may be of great value for the therapy of autoimmune disease and transplantation tolerance. [21, 29] The ability to use rapid methods such as the magnetic bead column allows for optimization of existing methodologies so that we can improve our understanding of apoptotic T cell and DC interactions, with the goal of creating more effective cancer immunotherapy.
The authors wish the acknowledge research support from the Doris Duke Clinical Research Foundation, JDS, CLB and RLE.
- 15.Banchereau J, Palucka AK, Dhodapkar M, Burkeholder S, Taquet N, Rolland A, Taquet S, Coquery S, Wittkowski KM, Bhardwaj N, Pineiro L, Steinman R, Fay J: Immune and clinical responses in patients with metastatic melanoma to CD34(+) progenitor-derived dendritic cell vaccine. Cancer Res. 2001, 61 (17): 6451-6458.PubMedGoogle Scholar
- 18.Delamarre L, Holcombe H, Mellman I: Presentation of exogenous antigens on major histocompatibility complex (MHC) class I and MHC class II molecules is differentially regulated during dendritic cell maturation. J Exp Med. 2003, 198 (1): 111-122. 10.1084/jem.20021542.PubMedCentralCrossRefPubMedGoogle Scholar
- 19.Hori S, Carvalho TL, Demengeot J: CD25+CD4+ regulatory T cells suppress CD4+ T cell-mediated pulmonary hyperinflammation driven by Pneumocystis carinii in immunodeficient mice. Eur J Immunol. 2002, 32 (5): 1282-1291. 10.1002/1521-4141(200205)32:5<1282::AID-IMMU1282>3.0.CO;2-#.CrossRefPubMedGoogle Scholar
- 22.Takahashi T, Kuniyasu Y, Toda M, Sakaguchi N, Itoh M, Iwata M, Shimizu J, Sakaguchi S: Immunologic self-tolerance maintained by CD25+CD4+ naturally anergic and suppressive T cells: induction of autoimmune disease by breaking their anergic/suppressive state. Int Immunol. 1998, 10 (12): 1969-1980. 10.1093/intimm/10.12.1969.CrossRefPubMedGoogle Scholar
- 29.Sutmuller RP, van Duivenvoorde LM, van Elsas A, Schumacher TN, Wildenberg ME, Allison JP, Toes RE, Offringa R, Melief CJ: Synergism of cytotoxic T lymphocyte-associated antigen 4 blockade and depletion of CD25(+) regulatory T cells in antitumor therapy reveals alternative pathways for suppression of autoreactive cytotoxic T lymphocyte responses. J Exp Med. 2001, 194 (6): 823-832. 10.1084/jem.194.6.823.PubMedCentralCrossRefPubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.