| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Transfection into a Rodent Tumor Cell Line Increases Intratumoral Dendritic Cell Infiltration but Enhances (Facilitates) Tumor Growth and Decreases Immunogenicity1
* Molecular Medicine Program and Department of Immunology, Mayo Clinic, Rochester, MN 55905; and
Institut National de la Santé et de la Recherché Médicale Unité 517, Faculty of Medicine, Dijon, France
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
which is strongly chemotactic for DCs. Surprisingly, MIP-3-transfected tumor cells grew faster than the mock-transfected tumor cells. Histological analysis and tumor dissociation confirmed that the MIP-3-transfected tumors contain three to four times more DCs than mock-transfected tumors. FACS analysis of the intratumor DCs showed that they were predominantly immature. Functional analysis showed that the alloreactivity mediated by these infiltrating MIP-3-transfected tumor DCs is strongly reduced. In conclusion, MIP-3 is an efficient chemokine for attracting DCs in vivo, but the high density of DCs in the tumor site injection is not a sufficient condition to induce an immune response. Furthermore, this attraction of immature DCs may always have an adverse effect by inducing a tolerance to the tumor cells. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The PRO tumor cell line, which has been established from a chemically induced tumor in rats, gives rise to progressive, metastatic, and lethal tumors (7). The development of tumors indicates an impairment of the immune response to these cells. Using FITC-labeled tumor cells, we have already demonstrated (8) that this impairment is partly attributable to the absence of tumor Ag engulfment by the DCs and their migration into the draining lymph nodes (1). We reasoned, therefore, that enhanced tumor infiltration by DCs may facilitate the induction of the immune response. Since MIP-3 is a powerful chemoattractive protein for DCs (9, 10, 11) and since it has been reported that adenovirus-mediated gene transfer of MIP-3 to tumors has inhibited tumor growth (12), we transfected a plasmid containing mouse MIP-3 cDNA into the PRO cells.
Surprisingly, we showed that, after s.c. injection, transfected PRO tumor cells that secreted high amounts of MIP-3 gave rise to tumors that grew faster than the mock-transfected PRO tumor cells. Using histological analysis and tumor dissociation methods, we confirmed that this chemokine had indeed recruited immature DCs into tumors. Cytofluorometric analysis showed that these DCs were immature since they express weakly MHC class II and did not express activating costimulatory molecules as B7. These immature DCs were not able to activate the immune response and functional analysis showed that they could not induce proliferation of allogenic T lymphocytes, whereas the DCs infiltrating mock-transfected tumor were able to induce a weak proliferation. These results have important implications for the development of cancer vaccine approaches using gene transfer.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A mouse MIP-3 ELISA kit was purchased from R&D Systems (Oxon, U.K.). The mouse MIP-3 cDNA was inserted into the PCR3.1 plasmid (Invitrogen Life Technologies, Groningen, The Netherlands).
Animals, cell lines, and tumorigenicity assays
Animals used in these experiments were BD-IX strain rats and nude rats purchased from Charles River Breeding Laboratories (Wilmington, MA). The progressive variant DHD-K12/TRb (PROb) was established from the tumor DHD, a colon adenocarcinoma induced by 1,2-dimethylhydrazine in a BD-IX strain female rat (7). These tumor cells were cultured in Ham’s F10 medium complemented with 10% FBS, as previously described (13).
For the tumorigenicity assays, 106 tumor cells in 100 µl of serum-free Ham’s F10 medium were injected s.c. into the anterolateral thoracic wall of syngeneic BD-IX rats. Tumor volume was evaluated weekly using a caliper to measure two perpendicular diameters.
To test the efficacy of MIP-3 transient secretion in vaccination assays, BD-IX rats were vaccinated twice with 1 x 106 irradiated PRO-PCR (eight rats) or irradiated PRO-MIP-3 cells (eight rats). One week later, all of the rats received a s.c. injection of 1 x 106 PRO-PCR cells. Tumor volumes were evaluated weekly. Six weeks after the tumor challenge, tumor-free rats were counted in each group.
Histological study of the tumor cell injection site
Animals were sacrificed 35 days after tumor cell injection. The tumor was resected, embedded in Tissue-Tek, and snap-frozen in methylbutane cooled in liquid nitrogen. An immunohistochemical study of tumor-infiltrating inflammatory cells was performed on acetone-fixed 5-µm cryostat sections (14, 15, 16). Mouse mAb to rat immature DCs and monocytes (ED1), MHC class II (OX-17), CD80 (B7-1), and IgG isotype-matched control were obtained from Serotec (Oxford, U.K.). After incubation with specific mAbs, sections were incubated with biotinylated sheep Ab to mouse IgG (Amersham, Little Chalfont, U.K.), then with streptavidin-peroxidase and stained with aminoethylcarbazole.
Isolation of DCs from the tumor and spleen
DCs were isolated from spleen, mock, or MIP-3-transfected tumors according to two different procedures which permitted us to obtain routinely from 80 to 90% of DCs: metrizamide gradient and Ab coupled with magnetic microbeads. Briefly, tumor and spleen fragments were digested with collagenase D (Sigma-Aldrich, Saint-Quentin Fallavier, France). According to the metrizamide gradient procedure, cell suspension was layered onto 4 ml of 14.5% (w/v) metrizamide gradient and was centrifuged for 13 min at 1800 x g at 4°C. Low-density cells were recovered, washed twice, resuspended, and cultured in complete medium overnight. Nonadherent cells were then further enriched for DCs by centrifugation over a 14.5% metrizamide column and subjected to two rounds of plastic adherence for 1 h at 37°C. We also used CD11c Ab to isolate DCs. We did not use ED1 Ab because ED1 is a intracellular marker of immature DCs. Cell isolation was conducted according to the manufacturer’s instructions (Miltenyi Biotec, Paris, France). Briefly, after 15 min of incubation in MACS buffer at 4°C with CD11c Ab, cells were washed twice and incubated for 25 min at 4°C with rat anti-mouse Ig associated with magnetic beads. Positively selected cells were isolated using MACS LS separation columns (Miltenyi Biotec).
Bone marrow-derived DCs
Rat bone marrow cells were isolated from tibias and femurs and cultured with rat GM-CSF and rat IL-4 as previously described (17).
Cytofluorimetry analysis
Cells were washed in PBS supplemented with 0.5% BSA and 0.01% sodium azide, adjusted to 1 x 105 cells/100 µl, and analyzed with a FACScan (BD Biosciences, Mountain View, CA). The CellQuest software was used to determine the percentage of fluorescent cells.
Functional study of tumor-infiltrating DCs
Allogeneic splenic T cells were obtained from Fisher rats by mechanical dissociation and depleted of adherent cells by overnight incubation at 37°C. Tumor-infiltrating DCs were treated with mitomycin C (25 mg/ml for 30 min) and cocultured with CSFE (Neuroprobe, Gaithersburg, MD)-labeled T cells (1 x 105 cells/well) at a 1:10 E:T ratio for 3 days. After washing, cytofluorometric analysis was performed. BrdU incorporation assay (Roche, Penzberg, Germany) was performed following the manufacturer’s instructions. BrdU was added after 3 days of coculture of DCs infiltrating PRO or PRO-MIP tumors and allogeneic T cells at different ratios with or without LPS. BrdU incorporation was measured by ELISA 8 h later.
Statistics
Statistical significance was determined by the two-tailed Student’s t test, covariance analysis applied to means, or Mantel and Haenszel test. Differences were considered to be significant when the p values were <0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although it has been reported that most tumor cell lines of epithelial origin (renal, colon breast adenocarcinomas) secrete MIP-3 in vivo (9), we confirmed that the PRO wild-type cells do not produce MIP-3 using an ELISA (Fig. 1). Therefore, PRO cells were transfected with either control plasmid or a murine MIP-3 cDNA-containing PCR3.1 plasmid. Several clones of geneticin-resistant transfected cells were obtained. Some of which secreted high levels of MIP-3 (Fig. 1). The higher level producer clone PRO-MIP-3-2B5 was used in the subsequent experiments (PRO-MIP-3); the control cell line was the PRO cells transfected by the plasmid PCR3.1 alone (PRO-PCR). Expression of MIP-3 was stable after multiple passages in vitro. MIP-3 secretion by tumor cells isolated from tumors harvested from the rats 7 wk after the s.c. injection was approximately the same as before injection, confirming that PRO-MIP-3 cells, injected in vivo did not loose their capacity to produce this chemokine (Fig. 1).
|
increases the tumorigenicity of PRO cells
We injected s.c. one million PRO-MIP-3 tumor cells into syngeneic animals. Compared with the tumors generated after s.c. injection of PRO-PCR cells, PRO-MIP-3 cell injection gave rise to tumors which grew significantly faster in three independent experiments (Fig. 2A). PRO-MIP-3, PRO-PCR, and PRO wild-type cells grew at the same rate in vitro (data not shown). Because MIP-3 can have nonimmune effects that can affect tumor growth (18), we demonstrated that its effects in our system are immune mediated by comparing tumor growth in nude rats. In these T cell-depleted animals, the tumor growth of PRO-MIP-3 and PRO-PCR did not differ significantly (Fig. 2B). Moreover, when both PRO and PRO-MIP-3 tumor cells were injected at different sites, PRO tumors grew slightly faster in rats bearing MIP-3-secreting tumors but without significant statistical differences (Fig. 2C).
|
attracts rat DCs in vivo
To check whether MIP-3 secretion by the transfected tumor cells attracted immature DCs into the tumor site, immunohistological analysis was performed on 35-day-old PRO-PCR and PRO-MIP-3 tumors (Fig. 3). Tumors from the control animals injected with PRO-PCR or from animals injected with PRO-MIP-3 demonstrated some features that were identical to those of the PRO wild-type tumors (19). The tumor was surrounded by a corona of inflammatory cells including CD4+ and CD8+ T cells and a majority of mononuclear cells. Most of these mononuclear cells were immature DCs because they are ED1+ (a marker of immature DCs and monocytes (20, 21)) and MHC class II+ and demonstrated a dendritic-like morphology (data not shown). The main difference between these two tumors was the high density of the ED1+ and MHC class II+ dendritic-like cells infiltrating into PRO-MIP-3 tumors (Fig. 3, A and C) compared with the PRO-PCR tumors (Fig. 3, B and D). Similar to our previous observation in PRO wild-type tumors, neither PRO-MIP-3 (Fig. 3E) nor PRO-PCR (Fig. 3F) tumor-infiltrating DCs expressed the B7 costimulatory molecule analyzed in each group (magnification, x25). We also counted the ED-1+ cells in random fields in PRO-MIP-3 or PRO-PCR tumors and observed a 3- to 4-fold increase in the number of ED-1+ cells in PRO-MIP-3 tumors compared with PRO-PCR tumors.
|
tumors
Thirty-five-day-old tumors were mechanically dissociated before enzymatic digestion. The total numbers of cells isolated from PRO-MIP-3 or from PRO-PCR tumors were approximately the same, at 
25 x 106 cells/g. Nevertheless, there was a 3- to 4-fold increase in the number of tumor-infiltrating DCs isolated by metrizamide gradient from the PRO-MIP-3 tumors compared with PRO tumors (Fig. 4).
|
attracts mostly immature DCs
To confirm that the PRO-MIP-3 tumors were infiltrated by higher numbers of DCs than the PRO-PCR tumors, we dissociated the tumors using collagenase and isolated the DCs by gradient centrifugation. We confirmed that a 4-fold increase in the number of immature DCs could be recovered from PRO-MIP-3 compared with PRO-PCR tumors for the same volume (data not shown). Cytofluorometric analysis showed that these tumor-infiltrating DCs have an immature phenotype, namely, low levels of MHC class II and nearly no B7 molecules (Fig. 5). The striking difference in MHC class II expression (as seen by strong labeling in immunohistology in tumors and weak expression in cytofluorometric analysis) is explained by the fact that these DCs were immature and the MHC class II expression is predominantly intracytoplasmic.
|
secretion does not inhibit DC migration to the draining lymph nodes
We studied whether MIP-3 could play a role in maturation and migration of the DCs. First, we compared the phenotype of bone marrow-derived immature DCs cultured with PRO tumor cells either secreting MIP-3 or not added at different times (days 5 and 8). We did not find any significant difference between the groups (data not shown), suggesting that MIP-3 does not inhibit DC maturation. We also studied whether MIP-3 could inhibit the migration of DCs to the lymph nodes. Although the numbers of DCs in the lymph nodes draining the PRO-MIP-3 tumors were lower in three different experiments, we did not find significant differences in the number of DCs in the lymph nodes draining the PRO-PCR or the PRO-MIP-3 tumors (Fig. 6).
|
tumors demonstrate reduced potency on allogeneic MLR
We compared the ability of tumor-infiltrating DCs from the two different tumors, PRO-MIP-3 or PRO-PCR, to induce proliferation of allogeneic T cells. We used splenic DCs from a naive syngeneic BD-IX rat as a positive control. PRO-MIP-3 tumor-infiltrating DCs were not able to induce proliferation of allogeneic T cells even after stimulation with LPS. The data about PRO-PCR tumors are consistent with our previous findings that PRO-PCR tumor-infiltrating DCs were able to induce a weak proliferation of allogeneic T cells only compared with the splenic DCs of a naive animal. (Fig. 7).
|
secretion is not able to induce an immune response
Therefore, to investigate whether transient secretion of MIP-3 could induce an effective immune response, we compared the protection conferred by two injections of irradiated PRO-MIP-3 cells vs irradiated PRO-PCR cells. Irradiated PRO-MIP-3 cells continued to secrete MIP-3 for at least 3 days before dying (data not shown).
Our data suggest that continuous secretion of MIP-3 may establish chronic attraction of immature DCs to the tumor site and may inhibit their maturation. Therefore, to investigate whether transient secretion of MIP-3 could induce an effective immune response, we compared the protection conferred by two injections of irradiated PRO-MIP-3 cells vs irradiated PRO-PCR cells. Irradiated PRO-MIP-3 cells continued to secrete MIP-3 for at least 3 days before dying (data not shown), and this MIP-3 secretion attracts 2-fold more DCs into 7-day-old tumors compared with the PRO wild-type tumors. Transient secretion by irradiated PRO-MIP-3 cells actually decreased the protection against PRO tumor challenge compared with irradiated PRO-PCR cells. In two different experiments, irradiated PRO-MIP-3 cell injections conferred a protection against PRO tumor challenge only in 13 (1 of 8) to 25% (2 of 8) of the animals, whereas irradiated PRO-PCR cell injections protected 75 (6 of 8) to 88% (7 of 8) animals. We have also injected PRO-MIP-3 irradiated cells mixed with PRO-PCR nonirradiated cells. The transient MIP-3 secretion is not able to induce an efficient immune response, allowing rejection of the PRO-PCR tumor (Fig. 8).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
appears to be one of the most chemotactic chemokine for immature DCs or Langerhans cell precursors (10, 11). Murine MIP-3 shares 71% amino acid sequence homology with rat MIP-3. It has also been demonstrated that rat and mouse MIP-3 attract mouse cells transfected by MIP-3 human ligand hCCR6 cDNA and that human MIP-3 attracts mouse DCs (12). Therefore, before performing in vivo experiments, we confirmed that murine MIP-3 could attract immature rat DCs using a chemotaxis and calcium mobilization (data not shown). Using transfected tumor cells secreting MIP-3, we showed that this chemokine is very efficient in vivo in attracting immature DCs into the tumors. Nevertheless, this high concentration of immature DCs into the tumor did not induce an antitumor immune response. Even though the immature DCs may engulf a high amount of tumor Ag, they were retained in an immature phenotype and could not induce T cell activation even of allogeneic T cells. The inhibition of DC maturation could explain the impairment of the immune response both in our experimental model (14) and also in human cancer (22, 23, 24). Even, when we stimulated DC generation by Flt-3 ligand, a potent DC growth factor, we could not lift the inhibition of DC maturation in the experimental tumor model PRO/REG (19). Therefore, inhibition of DC maturation is likely to be a potent mechanism by which tumor cells block the immune response to themselves. It has been demonstrated that the specific T cells are inhibited after injection of immature DCs (25) and that there is an increased production of immature myeloid DCs in cancer patients (26, 27). Mechanisms of inhibition of DC maturation included production of vascular endothelial growth factor or IL-6 by tumor cells (28, 29, 30). We did not find significant amounts of secretion of IL-6 from PRO cells (data not shown). Surprisingly, we did find significant production of vascular endothelial growth factor by the PRO tumor cells but the regressive variant REG isolated from the same initial tumor secretes higher amounts, suggesting that it has not a very important role in our model (our unpublished data). We did not find any role of MIP-3 at the concentration secreted by PRO- MIP-3, nor in DC maturation in vitro, nor in DC migration to the lymph nodes.
Contrary to Fushimi et al. (12) who demonstrated that intratumor injection of adenovirus-mediated gene transfer of MIP-3 could suppress tumor growth, we have found that the secretion of MIP-3 by the PRO tumor cells accelerates tumor growth. A striking difference between our results and those reported by Fushimi et al. (12) is that in our experimental model, as in human breast carcinomas which secrete high levels of MIP-3 (24), attracted DCs in the tumor site did not mature. In the Fushimi experimental model, "danger signals" inducing the maturation of the MIP-3-attracted DCs could be induced by the local inflammation caused by the intratumoral injection or by the viral infection. Indeed, viral infection induces IFN- production which has been demonstrated to induce the DC maturation (31). Furthermore, MIP-3 secretion induced by adenovirus transfection is transient and not continuous as in our transfected clones. High levels of MIP-3 may explain the high number of immature DCs infiltrating human breast carcinomas. To check whether transient MIP-3 secretion could induce an antitumor response in our model, we have compared the vaccination efficiency of irradiated PRO-MIP-3 vs PRO-PCR cells. Transient MIP-3 secretion decreased significantly vaccination efficiency with irradiated tumor cells. We have also studied the tumorigenicity of PRO cells mixed with irradiated PRO-MIP-3 cells. In this condition, no antitumor immune response was induced.
Another possible mechanism by which immature DCs might induce tolerance to tumor cells may involve regulatory T cells. The immune suppressive function of regulatory T cells has been recently confirmed in humans (22). These cells are also able to down-regulate the antitumor response in rodents (32). Although the mechanism of suppressor T cell generation is not completely known (33), one major cause has been already identified, repetitive contact with immature DCs (23). The high number of immature DCs attracted into the tumor by MIP-3 associated with a decreased immune response in the tumor-bearing MIP-3 animals has prompted us to check a possible role of regulatory T cells in the inhibition of the antitumor immune response. We did not find that the T cells from MIP-3 or mock-transfected tumor-bearing rats exerted any suppressive activity able to down-regulate an allogeneic response.
In conclusion, MIP-3 is an efficient chemokine for attracting immature DCs, but in this model, the high density of DCs in the tumor site injection is not a sufficient condition to induce an immune response. Furthermore, this attraction of immature DCs could have an adverse effect in inducing a more important tolerization to the tumor cells. These results demonstrate that unexpected results may be produced by cytokine gene transfer into tumor cells depending on the tumor type and experimental system that is used.
|
Footnotes |
|---|
1 This work was supported by National Institutes of Health Grant RO1 CA 094180 and by the Mayo Foundation, the French National League against Cancer (National, Burgundy, and Saône-et-Loire Committees), the Association for Research on Cancer, and the Regional Council of Burgundy. 
2 Address correspondence and reprint requests to Dr. Richard Vile, Molecular Medicine Program, Guggenheim 18, Mayo Clinic, 200 First Street SW, Rochester, MN 55905. E-mail address: vile.richard{at}mayo.edu
3 Abbreviation used in this paper: DC, dendritic cell.
Received for publication July 21, 2004. Accepted for publication August 18, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
is expressed at inflamed epithelial surfaces and is the most potent chemokine known in attracting Langerhans cell precursors. J. Exp. Med. 192:705. is involved in the constitutive trafficking of epidermal Langerhans cells. J. Exp. Med. 190:1755. transgene attracts dendritic cells to established murine tumors and suppresses tumor growth. J. Clin. Invest. 105:1383.
in systemic lupus erythematosus. Science 294:1540.This article has been cited by other articles:
|
|
 |
|
  Y.-s. Wang, D. Li, H.-s. Shi, Y.-j. Wen, L. Yang, N. Xu, X.-c. Chen, X. Chen, P. Chen, J. Li, et al. Intratumoral Expression of Mature Human Neutrophil Peptide-1 Mediates Antitumor Immunity in Mice Clin. Cancer Res., November 15, 2009; 15(22): 6901 - 6911. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Ramakrishna, N. Woller, B. Mundt, S. Knocke, E. Gurlevik, M. Saborowski, N. Malek, M. P. Manns, T. Wirth, F. Kuhnel, et al. Antitumoral Immune Response by Recruitment and Expansion of Dendritic Cells in Tumors Infected with Telomerase-Dependent Oncolytic Viruses Cancer Res., February 15, 2009; 69(4): 1448 - 1458. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Nicolas, D. Cathelin, N. Larmonier, J. Fraszczak, P.-E. Puig, A. Bouchot, A. Bateman, E. Solary, and B. Bonnotte Dendritic Cells Trigger Tumor Cell Death by a Nitric Oxide-Dependent Mechanism J. Immunol., July 15, 2007; 179(2): 812 - 818. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Bergeron, F. El Hage, M. Kambouchner, D. Lecossier, and A. Tazi Characterisation of dendritic cell subsets in lung cancer micro-environments Eur. Respir. J., December 1, 2006; 28(6): 1170 - 1177. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
