In Vivo Hydrodynamic Delivery of cDNA Encoding IL-2: Rapid, Sustained Redistribution, Activation of Mouse NK Cells, and Therapeutic Potentia
http://www.100md.com
免疫学杂志 2005年第14期
Abstract
In the present study, we have tested the ability of hydrodynamically delivered IL-2 cDNA to modulate the number and function of murine leukocyte subsets in different organs and in mice of different genetic backgrounds, and we have evaluated effects of this mode of gene delivery on established murine tumor metastases. Hydrodynamic administration of the IL-2 gene resulted in the rapid and transient production of up to 160 ng/ml IL-2 in the serum. The appearance of IL-2 was followed by transient production of IFN- and a dramatic and sustained increase in NK cell numbers and NK-mediated cytolytic activity in liver and spleen leukocytes. In addition, significant increases in other lymphocyte subpopulations (e.g., NKT, T, and B cells) that are known to be responsive to IL-2 were observed following IL-2 cDNA plasmid delivery. Finally, hydrodynamic delivery of only 4 μg of the IL-2 plasmid to mice bearing established lung and liver metastases was as effective in inhibiting progression of metastases as was the administration of large amounts (100,000 IU/twice daily) of IL-2 protein. Studies performed in mice bearing metastatic renal cell tumors demonstrated that the IL-2 cDNA plasmid was an effective treatment against liver metastasis and moderately effective against lung metastasis. Collectively, these results demonstrate that hydrodynamic delivery of relatively small amounts of IL-2 cDNA provides a simple and inexpensive method to increase the numbers of NK and NKT cells, to induce the biological effects of IL-2 in vivo for use in combination with other biological agents, and for studies of its antitumor activity.
Introduction
An explosion of understanding of the intricacies of the immune system over the past decade has sparked many new approaches to modulate tumor-host interactions and create novel strategies for the biological therapy of cancer (1). A major focus of these effects has been on the use of potent cytokines that increase the numbers and/or functions of the NK and T lymphocyte subsets. In particular, IL-2, which is a potent regulator of both NK and T cell function, has shown considerable antitumor activity in some patients suffering from metastatic melanoma (2, 3) or renal cell carcinoma (4). In addition, even more pronounced antitumor effects have been achieved in experimental mouse models when IL-2 has been combined with other potent NK or T cell-stimulating cytokines such as IL-12 (5, 6) or IL-18 (7) or in cancer patients when IL-2 has been used to support adaptively transferred tumor reactive T cells (8, 9) or other vaccine approaches (3).
Despite these promising results, much remains unknown about factors that are vital for success in some settings or failure in others. In this regard, the complex interplay between key innate (NK and NKT cells) and adaptive T cell components of the immune system remains under active study (10, 11, 12). Specifically, NK cells (13, 14) effector T cells (15), and NKT cells (16, 17) can have potent antitumor effects (3, 8, 9), whereas NKT cells (18) and T regulatory cells (19) can profoundly inhibit the development of immune responses. IL-2 has profound effects on all of these cell types, and therefore, new approaches to perform proof-of-principle studies of its common or unique effects on these leukocyte subsets in vivo may provide insight into how to predictably modulate the function of these cells to achieve more consistent antitumor effects. NK cells can recognize tumors that might evade T cell killing by detecting aberrant MHC expression and thus may provide a first line of recognition during the development of neoplasia (13, 20). Specifically, infiltration and recognition of tumors by NK cells can result in direct NK-mediated lysis of tumors by NK cells (13, 20) and/or the production of potent cytokines such as IFN- that can engage other important antitumor mechanisms mediated via dendritic cells or T lymphocytes (11, 21).
Thus, a better understanding of the interaction between IL-2 and NK cells may yield important insight into how to tip the balance of immunity in favor of host recognition and response to tumors. These types of studies are often limited by the extremely rapid clearance of exogenously delivered IL-2 protein (22), which necessitates a need for repeated injection of large amounts of IL-2 that is often accompanied by substantial toxicity (4). Hydrodynamic delivery of naked DNA (23, 24) encoding gene sequences for secreted mouse or human proteins offers the ability to rapidly perform in vivo proof-of-principle studies in the absence of a need for large amounts of purified proteins. In addition, conditions where exposure of leukocytes to sustained levels of the desired gene product can be effectively achieved using this approach.
In this study, we have shown that hydrodynamic activity of the IL-2 gene results in high levels of biologically active IL-2 and IFN- proteins in mouse serum and subsequent recruitment of NK cells to major parenchymal organs. Finally, the hydrodynamic approach for IL-2 delivery was also effective in reducing the number of pre-existing mouse renal cell carcinoma metastasis in liver, and this antitumor activity could be achieved in the absence of NKT cells. These results provide an experimental platform that can be used to dissect IL-2-dependent effects of NK cells from NKT cells in host tumor immunity.
Results
Exogenous administration of IL-2 protein dramatically perturbs the NK and T cell compartments and can induce antitumor activity under some conditions as a single agent (2, 4, 30) or also when administered in combination with some other cytokines (5, 6, 7). However, these activities usually require large amounts of IL-2 and can be accompanied by significant toxicity. In this study, we show that hydrodynamic delivery of small amounts of IL-2 cDNA recapitulates the major biologic effects of IL-2 without a need for large amounts of purified protein and without the often substantial toxicity that accompanies active protein-based therapeutic regimens. The data shown in Fig. 1A show that the administration of IL-2 cDNA resulted in a rapid increase (up to >150 ng/ml) of mouse IL-2 in the serum by 6 h that was sustained through 48 h. Doses of DNA >4 μg/mouse resulted in significant toxicity and/or lethality to mice, whereas doses of 2–4 μg/mouse resulted in >100 ng of serum IL-2 that was sustained for 24 h, whereas 0.5–1 μg/mouse resulted in substantial but lower levels of IL-2 (15–20 ng/ml). Subsequent to the peak of IL-2 production, serum levels of IFN- were also detected with peak amounts seen at 24 or 48 h, depending on the dose of IL-2 cDNA (Fig. 1B). In some cases, IL-2 cDNA could sustain detectable IFN- beyond 6 days. Doses of cDNA <1 μg/mouse did not result in consistently detectable levels of IFN- in the serum. Neither IL-2 or IFN- were observed when cDNA encoding the control protein GFP was used (data not shown).
The results presented to this point demonstrate that hydrodynamically delivered IL-2 provides a rapid, highly efficient approach for modulating key elements of the immune system, provides an easy, cost-effective tool for obtaining large numbers of leukocyte subsets for biological studies from the liver and spleen, and illustrates the possible efficacy of this approach in the context of tumor models.
Therefore, the therapeutic impact of hydrodynamic delivery of IL-2 cDNA was studied in an experimental metastasis model using the Renca renal cell carcinoma in BALB/c mice. Previous studies from our laboratory (14, 20) and others (13, 31, 32) have shown that NK cells can have potent antitumor effects. Previous reports (20, 30) have confirmed a role for NK cells in the regulation of metastases in the liver in that the pretreatment with the NK cell-depleting reagent anti-asialo GM1 resulted in a dramatic increase in experimentally established metastases in the absence of NKT cell modification (data not shown). Therefore, we used experimental models of liver (intrasplenic injection) and lung (i.v. injection) to compare and contrast the effects of IL-2 protein and cDNA on progression of metastases in different organ microenvironments. The data shown in Fig. 6 show results for IL-2 cDNA against established liver (Fig. 6A) and lung (Fig. 6B) metastases. The results showed that hydrodynamic delivery of IL-2 cDNA to mice bearing established metastases inhibited progression to about the same extent (60%) as did an established high-dose regimen of IL-2 protein (100,000 U delivered twice daily) for 3 days. However, when the same treatment approaches were compared for effectiveness against established lung metastases, neither strategy was effective in reducing the number of metastases. These results demonstrate that hydrodynamic delivery of IL-2 cDNA is active against established metastases in the liver. However, the failure of both IL-2 protein and IL-2 cDNA against lung metastases derived from the same tumor highlights the importance of unique organ microenvironments in the success or failure of IL-2 therapy and suggests that the liver and the lung contain distinctly different constitutive or inducible effector cell subsets.
The disparity in therapeutic efficacy in liver vs lung suggests important differences in key elements of the immunological microenvironments of those two organs. One specific difference in immune cell composition of these organs is the incidence of NKT cells that are a prominent subset in the liver but not the lung. At present there are no specific Abs available with which to selectively deplete NKT cells. However, we have shown recently that -galactosylceramide (GalCer)3 (C12;GalCer) can selectively deplete NKT cells in vivo without activating NK or other leukocytes (27). Therefore, we used C12;GalCer to selectively impair NKT cell function in the absence of NK cell activation to determine the possible role of NKT cells in the antimetastatic effects of IL-2 cDNA in the liver (Fig. 7). The impact of C12;GalCer was compared in these studies to the effect of GalCer, which stimulates and depletes NKT cells, and results in downstream activation of NK cells. However, when GalCer, which selectively binds to and depletes NKT cells was given, a small reduction (NS; Fig. 7 statistics shown in Table III) in the number of metastases was observed, whereas IL-2 cDNA achieved the expected significant (p < 0.01) decrease in metastases. However, interestingly, when both GalCer (to remove NKT cells) and IL-2 cDNA (to activate NK cells) were coadministered (p < 0.05), a reduction in liver metastasis that was quantitatively similar to that achieved with GalCer was observed. To further investigate the possible contribution of NKT cells to IL-2-induced metastatic effects in the liver, experiments were performed in CD1d knockout mice (Fig. 7). Interestingly, the number of metastases in untreated mice is lower but not significantly different from those seen in mice where NKT cells are depleted with GalCer or GalCer (Table III). In addition, IL-2 cDNA remained as effective in reducing the number of liver metastases in these CD1d–/– mice as it was in mice selectively depleted of NKT cells by GalCer. Overall, these results show that the effectiveness of IL-2 cDNA against liver metastasis is not dependent on NKT cells, and the increased effectiveness of IL-2 cDNA in liver vs lung is most likely due to its profound effect on NK cells and NK cell progenitors. However, the significantly reduced number of metastases seen in wild-type mice treated with GalCer (selectively removes NKT cells; Ref. 27) and CD1d–/– mice, which have no NKT cells, suggest that NKT cells may actually actively inhibit the ability of other elements of the immune system to control of newly established metastases in the liver, an observation that requires further study.
Overall, these results show that hydrodynamically administered IL-2 cDNA has profound immunoregulatory and antitumor effects in the liver microenvironment, provides a unique cost-effective approach for obtaining large numbers of NK cells for biological study, and provides a platform for proof-of-principle studies where IL-2 can be used in combination with other immune modifiers.
Discussion
The results of this study have demonstrated that hydrodynamic-based gene delivery can provide a simple method to express biologically active IL-2 protein. The results show that IL-2 cDNA administration is accompanied by a rapid production of IL-2 protein that appeared in serum by 6 h and remained detectable for >48 h. Although the duration of proteins expressed from various genes may vary based on the nature of the cDNA construct and the specific gene, the levels of IL-2 protein achieved (>100 ng/ml serum) were active biologically. This was evident by the rapid subsequent production of IFN- that followed the peak and nadir of IL-2 in the serum. In addition, the in vivo translation of IL-2 protein achieved qualitatively and quantitatively similar perturbations in critical innate and adaptive immune leukocyte subsets to those induced by large amounts of highly purified rIL-2 protein. Specifically, the degree and timing of the increase in NK cell number and frequency that were observed with hydrodynamic gene delivery was similar to that induced by protein IL-2. The effects of hydrodynamic injection of IL-2 cDNA were dependent on the IL-2 gene because vectors encoding irrelevant protein did not have similar immunomodulatory effects. The hydrodynamic delivery of IL-2 cDNA resulted in a sustained activation of NK cells that was confirmed by increased cell numbers, increased cell size based on flow cytometric forward scatter, and increased expression of CD69 (an early activation Ag on lymphocytes), and an increase in NK cell-mediated lytic potential in both NK- and Ab-dependent cellular cytotoxicity assays was measurable after several hours (data not shown). The increase in NK cells is accompanied by an increase in lytic activity per cell (LU (specific activity)). The treatment with IL-2 cDNA also increases both total cell number and total lytic activity. Thus, the IL-2 cDNA treatment results in substantially increased lytic potential and an increase in the number of NK cells.
Our studies with numerous strains of wild-type and mutant mice have shown that the effects of IL-2 cDNA gene delivery are predictable and reproducible in most mice with immune defects. These results suggest that the use of hydrodynamic gene delivery in conjunction with selected knockout mice may be a powerful tool with which to rapidly study unique mechanisms of immunological response against infectious agents or cancer without a need for large amounts or repeated administration of purified proteins. In this setting, selected cytokines or other unique proteins of interest could be administered by hydrodynamic gene delivery at different times during an immune response in knockout mice specific for the same protein to determine the importance of specific cytokines or proteins in the initiation, duration, and termination of various unique immune responses.
In addition, we demonstrated that hydrodynamic delivery of IL-2 cDNA potently modified the cellular immune response profile of peripheral organs via dramatic effects in the bone marrow. Specifically, using CFSE-labeled precursors to trace the origin of cells, we detected 10-fold increases in progeny of adoptively transferred bone marrow progenitors. This finding supports the contention that most of the increases in peripheral leukocyte subsets induced by IL-2 result from increased bone marrow differentiation and redistribution to the periphery.
Finally, we were able to use a unique approach of hydrodynamic IL-2 cDNA injection combined with several different NKT cell binding ligands to analyze the role of NKT cells in tumor rejection. NKT cells have been implicated previously as a contributing element in tumor rejection by virtue of the ability of GalCer (a potent NKT cell-stimulating compound) to induce antitumor effects. However, these results are complicated by the fact that GalCer not only alters NKT function but also strongly activates other effector leukocytes due to the cytokine storm created in vivo (27). Using a renal cell carcinoma model that can selectively establish metastases in either the liver or lung, we combined hydrodynamic administration of IL-2 cDNA with GalCer, an agent that we have shown recently to remove but not activate NKT or NK cells to elucidate the role of these effector cells. This approach demonstrated that an additive antitumor effect was achieved against metastases in the liver. In addition, treatment with GalCer alone, which only removed NKT cells, also showed a significant ability to reduce metastases, suggesting that these cells may actually inhibit other antitumor mechanisms in the liver. The antitumor effects of NKT cell depletion by GalCer were further amplified by IL-2 cDNA that potently engages the antimetastatic effects of NK cells.
In summary, we have demonstrated that hydrodynamic delivery of naked DNA encoding secreted mouse IL-2 can dramatically increase the number and function of various leukocyte subsets and preferentially increase the number and biological functions of NK cells. These results show that hydrodynamic delivery of IL-2 cDNA provides a simple, efficient, and inexpensive way of delivering new genes in vivo, as well as an inexpensive way to dissect antitumor and immunoregulating functions of IL-2 in vivo.
Acknowledgments
We thank Anna Mason for her technical assistance, Tim Back, John Wine, and Erin Lincoln for their technical support in animal care and experimentation, as well as Susan Charbonneau for preparation of this manuscript.
Disclosures
The authors have no financial conflict of interest.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This project has been funded in whole or in part with federal funds from the National Cancer Institute, National Institutes of Health, under Contract No. N01-CO-12400.
2 Address correspondence and reprint requests to Dr. John R. Ortaldo, National Cancer Institute, Center for Cancer Research, Laboratory of Experimental Immunology, Building 560, Room 31-93, Frederick, MD 21702-1201. E-mail address: ortaldo@ncifcrf.gov
3 Abbreviation used in this paper: GalCer, galactosylceramide.
Received for publication November 2, 2004. Accepted for publication April 29, 2005.
References
Blattman, J. N., P. D. Greenberg. 2004. Cancer immunotherapy: a treatment for the masses. Science 305: 200-205.
Phan, G. Q., P. Attia, S. M. Steinberg, D. E. White, S. A. Rosenberg. 2001. Factors associated with response to high-dose interleukin-2 in patients with metastatic melanoma. J. Clin. Oncol. 19: 3477-3482.
Rosenberg, S. A.. 2001. Progress in the development of immunotherapy for the treatment of patients with cancer. J. Intern. Med. 250: 462-475.
Yang, J. C., R. M. Sherry, S. M. Steinberg, S. L. Topalian, D. J. Schwartzentruber, P. Hwu, C. A. Seipp, L. Rogers-Freezer, K. E. Morton, D. E. White, et al 2003. Randomized study of high-dose and low-dose interleukin-2 in patients with metastatic renal cancer. J. Clin. Oncol. 21: 3127-3132.
Wigginton, J. M., E. Gruys, L. Geiselhart, J. Subleski, K. L. Komschlies, J. W. Park, T. A. Wiltrout, K. Nagashima, T. C. Back, R. H. Wiltrout. 2001. IFN- and Fas/FasL are required for the antitumor and antiangiogenic effects of IL-12/pulse IL-2 therapy. J. Clin. Invest. 108: 51-62.
Wigginton, J. M., R. H. Wiltrout. 2002. IL-12/IL-2 combination cytokine therapy for solid tumours: translation from bench to bedside. Expert. Opin. Biol. Ther. 2: 513-524.(John R. Ortaldo2,*, Robin)
In the present study, we have tested the ability of hydrodynamically delivered IL-2 cDNA to modulate the number and function of murine leukocyte subsets in different organs and in mice of different genetic backgrounds, and we have evaluated effects of this mode of gene delivery on established murine tumor metastases. Hydrodynamic administration of the IL-2 gene resulted in the rapid and transient production of up to 160 ng/ml IL-2 in the serum. The appearance of IL-2 was followed by transient production of IFN- and a dramatic and sustained increase in NK cell numbers and NK-mediated cytolytic activity in liver and spleen leukocytes. In addition, significant increases in other lymphocyte subpopulations (e.g., NKT, T, and B cells) that are known to be responsive to IL-2 were observed following IL-2 cDNA plasmid delivery. Finally, hydrodynamic delivery of only 4 μg of the IL-2 plasmid to mice bearing established lung and liver metastases was as effective in inhibiting progression of metastases as was the administration of large amounts (100,000 IU/twice daily) of IL-2 protein. Studies performed in mice bearing metastatic renal cell tumors demonstrated that the IL-2 cDNA plasmid was an effective treatment against liver metastasis and moderately effective against lung metastasis. Collectively, these results demonstrate that hydrodynamic delivery of relatively small amounts of IL-2 cDNA provides a simple and inexpensive method to increase the numbers of NK and NKT cells, to induce the biological effects of IL-2 in vivo for use in combination with other biological agents, and for studies of its antitumor activity.
Introduction
An explosion of understanding of the intricacies of the immune system over the past decade has sparked many new approaches to modulate tumor-host interactions and create novel strategies for the biological therapy of cancer (1). A major focus of these effects has been on the use of potent cytokines that increase the numbers and/or functions of the NK and T lymphocyte subsets. In particular, IL-2, which is a potent regulator of both NK and T cell function, has shown considerable antitumor activity in some patients suffering from metastatic melanoma (2, 3) or renal cell carcinoma (4). In addition, even more pronounced antitumor effects have been achieved in experimental mouse models when IL-2 has been combined with other potent NK or T cell-stimulating cytokines such as IL-12 (5, 6) or IL-18 (7) or in cancer patients when IL-2 has been used to support adaptively transferred tumor reactive T cells (8, 9) or other vaccine approaches (3).
Despite these promising results, much remains unknown about factors that are vital for success in some settings or failure in others. In this regard, the complex interplay between key innate (NK and NKT cells) and adaptive T cell components of the immune system remains under active study (10, 11, 12). Specifically, NK cells (13, 14) effector T cells (15), and NKT cells (16, 17) can have potent antitumor effects (3, 8, 9), whereas NKT cells (18) and T regulatory cells (19) can profoundly inhibit the development of immune responses. IL-2 has profound effects on all of these cell types, and therefore, new approaches to perform proof-of-principle studies of its common or unique effects on these leukocyte subsets in vivo may provide insight into how to predictably modulate the function of these cells to achieve more consistent antitumor effects. NK cells can recognize tumors that might evade T cell killing by detecting aberrant MHC expression and thus may provide a first line of recognition during the development of neoplasia (13, 20). Specifically, infiltration and recognition of tumors by NK cells can result in direct NK-mediated lysis of tumors by NK cells (13, 20) and/or the production of potent cytokines such as IFN- that can engage other important antitumor mechanisms mediated via dendritic cells or T lymphocytes (11, 21).
Thus, a better understanding of the interaction between IL-2 and NK cells may yield important insight into how to tip the balance of immunity in favor of host recognition and response to tumors. These types of studies are often limited by the extremely rapid clearance of exogenously delivered IL-2 protein (22), which necessitates a need for repeated injection of large amounts of IL-2 that is often accompanied by substantial toxicity (4). Hydrodynamic delivery of naked DNA (23, 24) encoding gene sequences for secreted mouse or human proteins offers the ability to rapidly perform in vivo proof-of-principle studies in the absence of a need for large amounts of purified proteins. In addition, conditions where exposure of leukocytes to sustained levels of the desired gene product can be effectively achieved using this approach.
In this study, we have shown that hydrodynamic activity of the IL-2 gene results in high levels of biologically active IL-2 and IFN- proteins in mouse serum and subsequent recruitment of NK cells to major parenchymal organs. Finally, the hydrodynamic approach for IL-2 delivery was also effective in reducing the number of pre-existing mouse renal cell carcinoma metastasis in liver, and this antitumor activity could be achieved in the absence of NKT cells. These results provide an experimental platform that can be used to dissect IL-2-dependent effects of NK cells from NKT cells in host tumor immunity.
Results
Exogenous administration of IL-2 protein dramatically perturbs the NK and T cell compartments and can induce antitumor activity under some conditions as a single agent (2, 4, 30) or also when administered in combination with some other cytokines (5, 6, 7). However, these activities usually require large amounts of IL-2 and can be accompanied by significant toxicity. In this study, we show that hydrodynamic delivery of small amounts of IL-2 cDNA recapitulates the major biologic effects of IL-2 without a need for large amounts of purified protein and without the often substantial toxicity that accompanies active protein-based therapeutic regimens. The data shown in Fig. 1A show that the administration of IL-2 cDNA resulted in a rapid increase (up to >150 ng/ml) of mouse IL-2 in the serum by 6 h that was sustained through 48 h. Doses of DNA >4 μg/mouse resulted in significant toxicity and/or lethality to mice, whereas doses of 2–4 μg/mouse resulted in >100 ng of serum IL-2 that was sustained for 24 h, whereas 0.5–1 μg/mouse resulted in substantial but lower levels of IL-2 (15–20 ng/ml). Subsequent to the peak of IL-2 production, serum levels of IFN- were also detected with peak amounts seen at 24 or 48 h, depending on the dose of IL-2 cDNA (Fig. 1B). In some cases, IL-2 cDNA could sustain detectable IFN- beyond 6 days. Doses of cDNA <1 μg/mouse did not result in consistently detectable levels of IFN- in the serum. Neither IL-2 or IFN- were observed when cDNA encoding the control protein GFP was used (data not shown).
The results presented to this point demonstrate that hydrodynamically delivered IL-2 provides a rapid, highly efficient approach for modulating key elements of the immune system, provides an easy, cost-effective tool for obtaining large numbers of leukocyte subsets for biological studies from the liver and spleen, and illustrates the possible efficacy of this approach in the context of tumor models.
Therefore, the therapeutic impact of hydrodynamic delivery of IL-2 cDNA was studied in an experimental metastasis model using the Renca renal cell carcinoma in BALB/c mice. Previous studies from our laboratory (14, 20) and others (13, 31, 32) have shown that NK cells can have potent antitumor effects. Previous reports (20, 30) have confirmed a role for NK cells in the regulation of metastases in the liver in that the pretreatment with the NK cell-depleting reagent anti-asialo GM1 resulted in a dramatic increase in experimentally established metastases in the absence of NKT cell modification (data not shown). Therefore, we used experimental models of liver (intrasplenic injection) and lung (i.v. injection) to compare and contrast the effects of IL-2 protein and cDNA on progression of metastases in different organ microenvironments. The data shown in Fig. 6 show results for IL-2 cDNA against established liver (Fig. 6A) and lung (Fig. 6B) metastases. The results showed that hydrodynamic delivery of IL-2 cDNA to mice bearing established metastases inhibited progression to about the same extent (60%) as did an established high-dose regimen of IL-2 protein (100,000 U delivered twice daily) for 3 days. However, when the same treatment approaches were compared for effectiveness against established lung metastases, neither strategy was effective in reducing the number of metastases. These results demonstrate that hydrodynamic delivery of IL-2 cDNA is active against established metastases in the liver. However, the failure of both IL-2 protein and IL-2 cDNA against lung metastases derived from the same tumor highlights the importance of unique organ microenvironments in the success or failure of IL-2 therapy and suggests that the liver and the lung contain distinctly different constitutive or inducible effector cell subsets.
The disparity in therapeutic efficacy in liver vs lung suggests important differences in key elements of the immunological microenvironments of those two organs. One specific difference in immune cell composition of these organs is the incidence of NKT cells that are a prominent subset in the liver but not the lung. At present there are no specific Abs available with which to selectively deplete NKT cells. However, we have shown recently that -galactosylceramide (GalCer)3 (C12;GalCer) can selectively deplete NKT cells in vivo without activating NK or other leukocytes (27). Therefore, we used C12;GalCer to selectively impair NKT cell function in the absence of NK cell activation to determine the possible role of NKT cells in the antimetastatic effects of IL-2 cDNA in the liver (Fig. 7). The impact of C12;GalCer was compared in these studies to the effect of GalCer, which stimulates and depletes NKT cells, and results in downstream activation of NK cells. However, when GalCer, which selectively binds to and depletes NKT cells was given, a small reduction (NS; Fig. 7 statistics shown in Table III) in the number of metastases was observed, whereas IL-2 cDNA achieved the expected significant (p < 0.01) decrease in metastases. However, interestingly, when both GalCer (to remove NKT cells) and IL-2 cDNA (to activate NK cells) were coadministered (p < 0.05), a reduction in liver metastasis that was quantitatively similar to that achieved with GalCer was observed. To further investigate the possible contribution of NKT cells to IL-2-induced metastatic effects in the liver, experiments were performed in CD1d knockout mice (Fig. 7). Interestingly, the number of metastases in untreated mice is lower but not significantly different from those seen in mice where NKT cells are depleted with GalCer or GalCer (Table III). In addition, IL-2 cDNA remained as effective in reducing the number of liver metastases in these CD1d–/– mice as it was in mice selectively depleted of NKT cells by GalCer. Overall, these results show that the effectiveness of IL-2 cDNA against liver metastasis is not dependent on NKT cells, and the increased effectiveness of IL-2 cDNA in liver vs lung is most likely due to its profound effect on NK cells and NK cell progenitors. However, the significantly reduced number of metastases seen in wild-type mice treated with GalCer (selectively removes NKT cells; Ref. 27) and CD1d–/– mice, which have no NKT cells, suggest that NKT cells may actually actively inhibit the ability of other elements of the immune system to control of newly established metastases in the liver, an observation that requires further study.
Overall, these results show that hydrodynamically administered IL-2 cDNA has profound immunoregulatory and antitumor effects in the liver microenvironment, provides a unique cost-effective approach for obtaining large numbers of NK cells for biological study, and provides a platform for proof-of-principle studies where IL-2 can be used in combination with other immune modifiers.
Discussion
The results of this study have demonstrated that hydrodynamic-based gene delivery can provide a simple method to express biologically active IL-2 protein. The results show that IL-2 cDNA administration is accompanied by a rapid production of IL-2 protein that appeared in serum by 6 h and remained detectable for >48 h. Although the duration of proteins expressed from various genes may vary based on the nature of the cDNA construct and the specific gene, the levels of IL-2 protein achieved (>100 ng/ml serum) were active biologically. This was evident by the rapid subsequent production of IFN- that followed the peak and nadir of IL-2 in the serum. In addition, the in vivo translation of IL-2 protein achieved qualitatively and quantitatively similar perturbations in critical innate and adaptive immune leukocyte subsets to those induced by large amounts of highly purified rIL-2 protein. Specifically, the degree and timing of the increase in NK cell number and frequency that were observed with hydrodynamic gene delivery was similar to that induced by protein IL-2. The effects of hydrodynamic injection of IL-2 cDNA were dependent on the IL-2 gene because vectors encoding irrelevant protein did not have similar immunomodulatory effects. The hydrodynamic delivery of IL-2 cDNA resulted in a sustained activation of NK cells that was confirmed by increased cell numbers, increased cell size based on flow cytometric forward scatter, and increased expression of CD69 (an early activation Ag on lymphocytes), and an increase in NK cell-mediated lytic potential in both NK- and Ab-dependent cellular cytotoxicity assays was measurable after several hours (data not shown). The increase in NK cells is accompanied by an increase in lytic activity per cell (LU (specific activity)). The treatment with IL-2 cDNA also increases both total cell number and total lytic activity. Thus, the IL-2 cDNA treatment results in substantially increased lytic potential and an increase in the number of NK cells.
Our studies with numerous strains of wild-type and mutant mice have shown that the effects of IL-2 cDNA gene delivery are predictable and reproducible in most mice with immune defects. These results suggest that the use of hydrodynamic gene delivery in conjunction with selected knockout mice may be a powerful tool with which to rapidly study unique mechanisms of immunological response against infectious agents or cancer without a need for large amounts or repeated administration of purified proteins. In this setting, selected cytokines or other unique proteins of interest could be administered by hydrodynamic gene delivery at different times during an immune response in knockout mice specific for the same protein to determine the importance of specific cytokines or proteins in the initiation, duration, and termination of various unique immune responses.
In addition, we demonstrated that hydrodynamic delivery of IL-2 cDNA potently modified the cellular immune response profile of peripheral organs via dramatic effects in the bone marrow. Specifically, using CFSE-labeled precursors to trace the origin of cells, we detected 10-fold increases in progeny of adoptively transferred bone marrow progenitors. This finding supports the contention that most of the increases in peripheral leukocyte subsets induced by IL-2 result from increased bone marrow differentiation and redistribution to the periphery.
Finally, we were able to use a unique approach of hydrodynamic IL-2 cDNA injection combined with several different NKT cell binding ligands to analyze the role of NKT cells in tumor rejection. NKT cells have been implicated previously as a contributing element in tumor rejection by virtue of the ability of GalCer (a potent NKT cell-stimulating compound) to induce antitumor effects. However, these results are complicated by the fact that GalCer not only alters NKT function but also strongly activates other effector leukocytes due to the cytokine storm created in vivo (27). Using a renal cell carcinoma model that can selectively establish metastases in either the liver or lung, we combined hydrodynamic administration of IL-2 cDNA with GalCer, an agent that we have shown recently to remove but not activate NKT or NK cells to elucidate the role of these effector cells. This approach demonstrated that an additive antitumor effect was achieved against metastases in the liver. In addition, treatment with GalCer alone, which only removed NKT cells, also showed a significant ability to reduce metastases, suggesting that these cells may actually inhibit other antitumor mechanisms in the liver. The antitumor effects of NKT cell depletion by GalCer were further amplified by IL-2 cDNA that potently engages the antimetastatic effects of NK cells.
In summary, we have demonstrated that hydrodynamic delivery of naked DNA encoding secreted mouse IL-2 can dramatically increase the number and function of various leukocyte subsets and preferentially increase the number and biological functions of NK cells. These results show that hydrodynamic delivery of IL-2 cDNA provides a simple, efficient, and inexpensive way of delivering new genes in vivo, as well as an inexpensive way to dissect antitumor and immunoregulating functions of IL-2 in vivo.
Acknowledgments
We thank Anna Mason for her technical assistance, Tim Back, John Wine, and Erin Lincoln for their technical support in animal care and experimentation, as well as Susan Charbonneau for preparation of this manuscript.
Disclosures
The authors have no financial conflict of interest.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This project has been funded in whole or in part with federal funds from the National Cancer Institute, National Institutes of Health, under Contract No. N01-CO-12400.
2 Address correspondence and reprint requests to Dr. John R. Ortaldo, National Cancer Institute, Center for Cancer Research, Laboratory of Experimental Immunology, Building 560, Room 31-93, Frederick, MD 21702-1201. E-mail address: ortaldo@ncifcrf.gov
3 Abbreviation used in this paper: GalCer, galactosylceramide.
Received for publication November 2, 2004. Accepted for publication April 29, 2005.
References
Blattman, J. N., P. D. Greenberg. 2004. Cancer immunotherapy: a treatment for the masses. Science 305: 200-205.
Phan, G. Q., P. Attia, S. M. Steinberg, D. E. White, S. A. Rosenberg. 2001. Factors associated with response to high-dose interleukin-2 in patients with metastatic melanoma. J. Clin. Oncol. 19: 3477-3482.
Rosenberg, S. A.. 2001. Progress in the development of immunotherapy for the treatment of patients with cancer. J. Intern. Med. 250: 462-475.
Yang, J. C., R. M. Sherry, S. M. Steinberg, S. L. Topalian, D. J. Schwartzentruber, P. Hwu, C. A. Seipp, L. Rogers-Freezer, K. E. Morton, D. E. White, et al 2003. Randomized study of high-dose and low-dose interleukin-2 in patients with metastatic renal cancer. J. Clin. Oncol. 21: 3127-3132.
Wigginton, J. M., E. Gruys, L. Geiselhart, J. Subleski, K. L. Komschlies, J. W. Park, T. A. Wiltrout, K. Nagashima, T. C. Back, R. H. Wiltrout. 2001. IFN- and Fas/FasL are required for the antitumor and antiangiogenic effects of IL-12/pulse IL-2 therapy. J. Clin. Invest. 108: 51-62.
Wigginton, J. M., R. H. Wiltrout. 2002. IL-12/IL-2 combination cytokine therapy for solid tumours: translation from bench to bedside. Expert. Opin. Biol. Ther. 2: 513-524.(John R. Ortaldo2,*, Robin)