|Home | About | Journals | Submit | Contact Us | Français|
There is an urgent need for innovative therapies against ovarian cancer, one of the leading causes of death from gynecological cancers in the United States. Immunotherapy employing Toll-like receptor (TLR) ligands, such as CpG oligodeoxynucleotides (CpG-ODN), may serve as a potentially promising approach in the control of ovarian tumors. The CpG-ODN requires intracellular delivery into the endosomal compartment, where it can bind to TLR9 in order to activate the immune system. In the current study, we aim to investigate whether the antimicrobial polypeptide from the cathelicidin family, LL-37, could enhance the immunostimulatory effects of CpG-ODN by increasing the uptake of CpG-ODN into the immune cells, thus enhancing the antitumor effects against ovarian cancer. We found that treatment with the combination of CpG-ODN and LL-37 generated significantly better therapeutic antitumor effects and enhanced survival in murine ovarian tumor-bearing mice compared with treatment with CpG-ODN or LL-37 alone. We also observed that treatment with the combination of CpG-ODN and LL-37 enhanced proliferation and activation of natural killer (NK) cells, but not CD4+ or CD8+ T cells, in the peritoneal cavity. Furthermore, in vivo antibody depletion experiments indicated that peritoneal NK cells played a critical role in the observed antitumor effects. Thus, our data suggest that the combination of CpG-ODN with LL-37 peptide may lead to the control of ovarian tumors through the activation of innate immunity.
Ovarian cancer is distinguished as the most lethal gynecologic cancer in the United States, with approximately 22,000 new cases and 16,000 deaths occurring annually (Jemal et al., 2007). Despite maximal cytoreductive surgery followed by first-line paclitaxel-based chemotherapy, most women with advanced ovarian cancer will eventually relapse (Markman, 2007). In the recurrent setting, response rates to second-line chemotherapy are substantially diminished (Fung-Kee-Fung et al., 2007), highlighting the crucial need to develop novel therapeutic strategies for the control of ovarian cancer.
Cancer immunotherapy has emerged as an attractive approach to cancer treatment, with high specificity for cancer cells and less toxicity to the host (Rosenberg et al., 2004). For advanced ovarian cancer, direct intraperitoneal seeding is the most common pathway of tumor metastases, although lymphatic and hematogenous spread can also occur (Gerber et al., 2006). Therefore, for immunotherapy to be successful, harnessing the peritoneal immune system for antitumor response is a critical step in the treatment of advanced ovarian cancer. However, systemic investigations of the peritoneal immune system are relative scarce in the literature as compared with other immune systems (Funda et al., 1993; Kubicka et al., 1996; Broche and Tellado, 2001). Thus, it is essential to develop immunotherapeutic strategies targeting the peritoneal immune system for the treatment of ovarian cancer.
One approach for the treatment of ovarian cancer is the employment of “alert” signals such as Toll-like receptor (TLR) ligands that stimulate dendritic cells to mature and differentiate into potent activators of antigen-specific T cells (for a review, see Guermonprez et al., 2002). It is clear that TLRs play a crucial role in enhancing innate and adaptive immune responses (for reviews, see Aderem and Ulevitch, 2000; Akira et al., 2001; Akira and Takeda, 2004; Iwasaki and Medzhitov, 2004). TLR ligands have emerged as a promising new class of vaccine adjuvants, particularly the oligodeoxynucleotides containing one or more unmethylated CpG dinucleotides (CpG-ODN), which target TLR9 (Krieg, 2006). It has been shown that CpG-ODN can stimulate innate immunity and confer protection against a variety of bacterial and viral infections (Deng et al., 2004; Rees et al., 2005). In the presence of antigen, CpG-ODN can trigger the development of predominantly helper T cell type 1 (Th1)-promoting adaptive immune responses (Ioannou et al., 2002). However, although CpG-ODN has the potential to eradicate tumor in the monotherapy setting, it generally needs to be combined with other therapeutic modalities (e.g., surgery, chemotherapy, or radiation) to successfully control large tumors (Weigel et al., 2003; Speiser et al., 2005). This may be because, since TLR9 is located in the endosomal compartment, it is difficult for the CpG-ODN to efficiently penetrate the cell membrane and bind to its receptor (Krieg, 2007). Thus, it is important to develop strategies to improve the uptake of CpG-ODN into the effector cells in order to enhance the antitumor function of CpG-ODN.
One strategy that could potentially improve the uptake of CpG-ODN by immune cells is the coadministration of CpG-ODN with the LL-37 peptide. LL-37 is a member of the cathelicidin family of antimicrobial polypeptides, which are characterized by a highly conserved region (cathelin domain) and a highly variable cathelicidin peptide domain. Cathelicidins serve a critical role in mammalian innate immune defense against invasive bacterial infection (Nizet et al., 2001). LL-37 is a 37-residue helical peptide found throughout the body and has been shown to exhibit a broad spectrum of antimicrobial activity (Durr et al., 2006). LL-37, when complexed with self-DNA or CpG-ODN, has been shown to promote DNA translocation and can significantly increase interferon (IFN)-α production in plasmacytoid dendritic cells (DCs) (Lande et al., 2007).
In the current study, we aimed to investigate whether LL-37 can enhance the immunostimulatory effects of CpG-ODN and thus increase the antitumor effects against ovarian cancer, using a murine ovarian cancer model. We found that treatment with the combination of CpG-ODN and LL-37 generated significant therapeutic antitumor effects and enhanced survival in mouse ovarian surface epithelial cell (MOSEC) tumor-bearing mice. We also observed that treatment with the combination of CpG-ODN and LL-37 enhanced proliferation and activation of natural killer (NK) cells in the peritoneal cavity. Peritoneal NK cells were also shown to play a critical role in the antitumor effects on MOSEC/luc tumors generated by treatment with the combination of CpG-ODN and LL-37 in antibody depletion experiments. The clinical implications of the current study are discussed.
Female C57BL/6 mice (H-2Kb and I-Ab), 5 to 6 weeks of age, were purchased from the National Cancer Institute (Frederick, MD) and kept in the oncology animal facility of the Johns Hopkins Hospital (Baltimore, MD). Animals were used in compliance with institutional animal health care regulations, and all animal experimental procedures were approved by the Johns Hopkins Institutional Animal Care and Use Committee. The mouse ovarian MOSEC cell line (clone ID8), generated as described previously (Roby et al., 2000), was originally derived from mouse ovarian surface epithelial cells. The luciferase-expression MOSEC cell line (MOSEC/luc) was generated as previously described (Hung et al., 2007). MOSEC/luc cells were maintained in RPMI 1640, supplemented with 10% (v/v) fetal bovine serum, penicillin–streptomycin (50U/ml), 2mM l-glutamine, 1mM sodium pyruvate, 2mM nonessential amino acids, and G418 (0.4mg/ml) at 37°C with 5% CO2.
CpG-ODN 1826 (CpG-B no. 1826, TCCATGACGTTCCTGACGTT), with a fully phosphorothioated backbone, was synthesized by Invitrogen (Carlsbad, CA). Fluorescein isothiocyanate (FITC)-conjugated CpG-ODN 1826 was purchased from InvivoGen (San Diego, CA). CpG-ODN 1826 is denoted CpG-ODN hereafter. The antimicrobial peptide, LL-37, was purchased from Sigma-Aldrich (St. Louis, MO).
For the in vitro cell-labeling assay, single-cell suspensions (1×106/ml) were made from peritoneal cells harvested from naive mice and seeded into a 24-well plate. Phosphate-buffered saline (PBS), FITC-conjugated CpG-ODN (10μg/ml), or FITC-conjugated CpG-ODN combined with LL-37 (50μg/ml) was added to each of triplicate wells and incubated at 37°C for 1hr. Cells were then recovered, washed with PBS, and analyzed by flow cytometry for the number of FITC+ cells.
C57BL/6 mice (five per group) were injected intraperitoneally with 2×105MOSEC/luc cells per mouse to induce advanced ovarian cancer (Chang et al., 2007). Four days after tumor inoculation, mice were injected intraperitoneally with CpG-ODN (30μg/mouse) and LL-37 (100μg/mouse) either alone or in combination, starting 4 days after MOSEC/luc inoculation. The injection was repeated three times at 4-day intervals (i.e., on days 4, 8, and 12 after MOSEC/luc inoculation). Mice were imaged with the IVIS-200 system (Xenogen, Alameda, CA) at baseline and every 7 days to monitor for therapeutic effects. The bioluminescence signals were analyzed with Living Image software (Xenogen). To evaluate intraperitoneal tumor seeding, additional mice (five per group) were treated with the same reagents and were killed 50 days after tumor inoculation. Mice were killed when clinical signs of massive tumor burden (e.g., massive ascites, bowel obstruction, and cachexia) occurred.
Groups of naive C57BL/6 mice (three per group) were treated with CpG-ODN (30μg/mouse) and LL-37 (100μg/mouse) either alone or in combination. Two days after the last treatment, mice were killed by CO2 asphyxiation, their abdomens were wiped with 70% alcohol, 10ml of cold sterile PBS was injected into the peritoneal cavity, and peritoneal exudate cells (PECs) were harvested by syringe. Contaminating red blood cells were lysed with ACK buffer (Quality Biological, Gaithersburg, MD). The viability of cells was checked by trypan blue exclusion test. MOSEC/luc cells (when 60–80% confluent) were seeded into a round-bottom 96-well microplate at 5×103/well in complete medium. After 2hr, counted PECs were added to each well in triplicate at titrated effector-to-target ratios. The plates were imaged for bioluminescence activity after 16hr of culture in an incubator (at 37°C with 5% CO2). For parallel experiments, some mice also received in vivo depletion of peritoneal macrophages or NK cells as indicated. For in vivo depletion of peritoneal macrophages, mice were injected intraperitoneally, 1 day before and 3 days after the first time of CpG-ODN and LL-37 treatment, with 0.2ml of clodronate liposomes. For in vivo NK depletion, mice were injected intraperitoneally, 1 day before and 3 days after first time of combined CpG-ODN with LL-37 treatment, with anti-mouse NK1.1 monoclonal antibody (PK136, 200μg/mouse). Depletion efficiency was checked by flow cytometry, with >90% depletion of target cells (data not shown).
Groups of C57BL/6 mice (three per group) were treated with CpG-ODN (30μg/mouse) and LL-37 (100μg/mouse) either alone or in combination as described earlier. PECs were harvested 2 days after the last treatment. PECs were then washed once in FACScan buffer and stained with surface markers for innate and adaptive effectors including phycoerythrin (PE)-conjugated anti-CD4 (L3T4), PE-conjugated anti-CD8 (53-6.7), FITC-conjugated anti-GR-1 (RB6-8C5), PE-conjugated anti-CD19 (1D3), PE-conjugated anti-NK1.1 (PK136), and PE–Cy5-conjugated anti-F4/80 (BM8). All antibodies were purchased from eBioscience (San Diego, CA).
For the analysis of interferon-γ and CD69 expression with Cp-ODN and LL-37 administration either alone or in combination, splenocytes from naive mice were harvested and made into single-cell suspension in complete medium (1×106/ml). Splenocytes were then seeded into a 24-well microplate and cultured with CpG-ODN 1826 (10μg/ml) and/or LL-37 (50μg/ml) for 16hr. GolgiStop (BD Biosciences, San Jose, CA) was added 6hr before harvesting the cells from the culture. Cells were then washed once in FACScan buffer and stained with PE-conjugated anti-mouse NK1.1 (PK136) and FITC-conjugated anti-mouse CD69 (H1.2F3). For intracellular cytokine staining, cells were subjected to intracellular cytokine staining, using a Cytofix/Cytoperm kit (BD Biosciences) according to the manufacturer's instructions. FITC-conjugated anti-IFN-γ antibodies and the immunoglobulin isotype control antibody (Rat IgG1) were purchased from BD Biosciences. Analysis was performed with a FACScan equipped with CellQuest software (BD Biosciences). Interferon-γ and CD69 histograms are presented on gated NK1.1+ cells.
C57BL/6 mice (five per group) were inoculated with MOSEC/luc cells (2×105 per mouse) at baseline, and were administered PBS or combined treatment (LL-37 plus CpG-ODN). For mice given combined treatment, two separate groups of mice also underwent either peritoneal NK cell or peritoneal macrophage depletion. For in vivo depletion of peritoneal macrophages, mice were injected intraperitoneally, 1 day before and 3 days after the first time of CpG-ODN and LL-37 treatment, with 0.2ml of clodronate liposomes and thereafter once per week with 0.1ml of clodronate liposomes until the end of follow-up. For in vivo depletion of peritoneal NK cells, mice were injected intraperitoneally, 1 day before and 3 days after the first time of combined CpG-ODN with LL-37 treatment, with anti-mouse NK1.1 monoclonal antibody (PK136, 200μg/mouse) and thereafter once per week until the end of follow-up. Depletion was performed according to standard protocols as described earlier (van Rooijen et al., 1997; Chen et al., 2000). Depletion efficiency was tested by flow cytometry, with more than 90% depletion of both peritoneal macrophages and NK cells.
Statistical analysis was performed with Prism 3.0 software (GraphPad, San Diego, CA). Cellular proliferation and both CD69 and IFN-γ expression among PECs in the various groups were analyzed by Pearson's χ2 test. The difference in bioluminescence signal between groups was analyzed by Mann–Whitney test. Survival curves were plotted by the Kaplan–Meier method and compared by log-rank test. For all analyses p<0.05 was considered statistically significant. Data are presented as means±standard deviation (SD) unless otherwise specified.
Evidence suggests that antimicrobial peptide LL-37 can significantly enhance the intracellular transport of self-DNA into early endocytic compartments containing TLR9 in dendritic cells (Lande et al., 2007). To determine whether LL-37 forms a complex with CpG-ODN, we performed a gel retardation assay. LL-37 was mixed with CpG at a 1:5 (peptide–DNA) mass ratio and incubated for 30min. Complex formation was assayed by running samples on gels. CpG alone was loaded as a control. As shown in Fig. 1, we observed that the addition of LL-37 to CpG-ODN leads to retardation of the migration of CpG-ODN. This suggests that the addition of the LL-37 peptide to CpG-ODN leads to the formation of a complex.
To determine whether LL-37 could enhance the delivery of CpG-ODN (non-self-DNA) into immune cells derived from the peritoneal cavity, peritoneal cells harvested from naive mice were incubated with PBS, FITC-conjugated CpG-ODN, or FITC-conjugated CpG-ODN combined with LL-37 peptide. Peritoneal cells were washed and analyzed by flow cytometry on the gated lymphocyte population. As shown in Fig. 2, we observed that incubation with CpG in combination with the LL-37 peptide demonstrated a significantly higher percentage of FITC+ cells compared with cells incubated with CpG alone (p<0.05). Taken together, our results indicate that the addition of LL-37 to CpG-ODN leads to the formation of a complex, which may facilitate enhanced delivery of CpG-ODN into peritoneal cells compared with CpG-ODN alone.
We next investigated whether LL-37 could increase the therapeutic antitumor effects of CpG-ODN. Groups of C57BL/6 mice (five per group) were inoculated with murine ovarian cancer cells that express luciferase (MOSEC/luc). Four, 8, and 12 days after tumor inoculation, mice were intraperitoneally administered PBS, LL-37 alone, CpG-ODN alone, or CpG-ODN in combination with LL-37 as depicted in Fig. 3A. Mice were imaged with the IVIS imaging system series 200. As shown in Fig. 3B, MOSEC/luc tumor-bearing mice treated with the combination of CpG-ODN and LL-37 demonstrated a significant reduction in luminescence intensity over time compared with tumor-bearing mice treated with CpG alone or LL-37 alone (*p<0.05). A graphical representation of the luminescence intensity in the treated mice is demonstrated in Fig. 3C. Furthermore, tumor-bearing mice treated with the combination of CpG-ODN and LL-37 showed improved survival compared with tumor-bearing mice treated with CpG alone or LL-37 alone (Fig. 3D). In addition, mice treated with PBS or LL-37 alone showed signs of bloody ascites as compared with mice treated with CpG-ODN alone or with CpG plus LL-37 on day 50 after tumor inoculation (see Supplementary Fig. 1A at www.liebertonline.com/hum). Furthermore, the size of the tumor in the peritoneal cavity was significantly smaller in mice treated with CpG plus LL-37 or with CpG-ODN alone compared with the bulky tumor observed in the peritoneal cavity of mice treated with PBS or LL-37 alone (see Supplementary Fig. 1B). Thus, our data suggest that treatment with the combination of CpG-ODN and LL-37 produces the best therapeutic antitumor effects and long-term survival in MOSEC/luc tumor-bearing mice.
We then characterized the immune cell subsets that were present in the peritoneal cavity after stimulation with CpG-ODN and LL-37 either alone or in combination. Groups of naive C57BL/6 mice (five per group) were treated either with PBS, LL-37 alone, CpG-ODN alone, or CpG-ODN plus LL-37 as described in Fig. 3. Two days after the last treatment, cells from the peritoneal cavity were harvested and stained with surface markers for CD4+ cells, CD8+ cells, B cells, neutrophils, NK cells, and macrophages. The number of the various peritoneal cell types was determined by flow cytometric analysis. As shown in Fig. 4, there was no significant difference in the numbers of CD4+ and CD8+ T cells, B lymphocytes, or neutrophils after stimulation with the various reagents. However, there was a significant increase in the number of NK (NK1.1+) cells and macrophages (F4/80+) after stimulation with CpG-ODN plus LL-37 compared with stimulation with CpG or LL-37 alone (p<0.05). Thus, our data indicate that treatment with the combination of CpG-ODN and LL-37 can enhance proliferation of NK cells and macrophages in the peritoneal cavity.
To determine whether peritoneal cells derived from mice treated with the combination of CpG-ODN with LL-37 are capable of killing MOSEC/luc tumor cells, we performed an in vitro cytotoxicity assay using luminescence imaging. Groups of mice (three per group) were treated with PBS, LL-37 alone, CpG-ODN alone, or CpG-ODN plus LL-37. Two days after the last treatment, cells from the peritoneal cavity were harvested and added to MOSEC/luc cells at various effector-to-target ratios and the plates were imaged for bioluminescence activity. As shown in Fig. 5A, the lowest luciferase activity was observed in the wells incubated with cells derived from mice treated with CpG-ODN plus LL-37 as compared with the wells incubated with cells from mice treated with LL-37 alone or CpG-ODN alone.
Because treatment with the combination of CpG-ODN with LL-37 led to increased numbers of NK cells and macrophages in the peritoneal cavity, we sought to determine which cell subset (or subsets) play a crucial role in the killing of MOSEC/luc tumor cells. Mice were treated with CpG-ODN plus LL-37 and depleted of either peritoneal macrophages, using clodronate liposomes, or NK cells, using NK1.1 antibody. Undepleted mice were used as negative controls. Two days after the last treatment, cells from the peritoneal cavity were harvested and added to MOSEC/luc cells followed by luminescence imaging as described previously. As shown in Fig. 5B, the wells incubated with cells derived from mice depleted of NK cells demonstrated the highest luciferase activity compared with wells incubated with cells derived from mice depleted of macrophages, or from undepleted mice. A graphical representation of the relative fluorescence intensity is depicted in Fig. 5C. Taken together, our data indicate that peritoneal NK cells derived from mice treated with CpG-ODN in combination with LL-37 play a critical role in the killing of MOSEC/luc tumor cells.
Cytokines such as IFN-γ have been shown to be important for antitumor or antiviral effects generated by NK cells (Long et al., 2008; Shey and Ballas, 2008). Furthermore, CpG-ODN has been shown to be able to stimulate CD69 and IFN-γ expression in NK cells (Cowdery et al., 1996; Hartmann et al., 2000; Marshall et al., 2006). Therefore, we sought to determine whether treatment with the combination of CpG-ODN with LL-37 could enhance the expression of CD69 and IFN-γ in NK cells derived from the peritoneal cavity. Cells from the peritoneal cavity harvested from naive mice were incubated with CpG-ODN and/or LL-37. Gated NK cells were characterized for the expression of CD69 and IFN-γ, using flow cytometric analysis. As shown in Fig. 6, the expression of both CD69 and IFN-γ was significantly higher in NK cells treated with CpG-ODN combined with LL-37 compared with cells treated with CpG-ODN alone (*p<0.05; **p<0.01). Thus, our data indicate that treatment with LL-37 can further enhance the expression of CD69 and IFN-γ in peritoneal NK cells stimulated with CpG-ODN.
To confirm the role of peritoneal NK cells in the treatment of MOSEC/luc tumors with the combination of CpG-ODN with LL-37, we performed an in vivo antibody depletion experiment. Groups of mice (five per group) were inoculated with MOSEC/luc cells and treated with CpG-ODN plus LL-37. Mice treated with PBS were used as controls. Mice were depleted of either macrophages or NK cells as described in Materials and Methods. Undepleted mice were used as a control. Tumor growth was monitored with the IVIS imaging system series 200 every 7 days. As shown in Fig. 7, a significantly higher luminescence signal was observed in treated mice depleted of NK cells compared with treated mice depleted of macrophages or without depletion (p<0.05). Thus, our data indicate that peritoneal NK cells play a critical role in the antitumor effects on MOSEC/luc tumors in vivo generated by treatment with the combination of CpG-ODN with LL-37.
Our study demonstrated that peritoneal NK cells play a critical role in the antitumor effects against MOSEC/luc tumors generated by the combinational treatment of CpG-ODN with LL-37. We also showed that the expression of CD69 and IFN-γ on peritoneal NK cells stimulated with CpG-ODN can be enhanced by treatment with the LL-37 peptide, thus leading to further activation of NK cells in the peritoneal cavity. Previous studies have shown that CpG-ODN is highly potent in activating NK cells (for a review see Ballas, 2007). Because NK cells do not express TLR9, it is assumed that CpG-ODN activate dendritic cells (DCs), which in turn activate NK cells. In fact, CpG-ODN have been employed in clinical trials on melanoma patients and have been shown to activate plasmacytoid and myeloid dendritic cells, which in turn activate NK cells (Molenkamp et al., 2007). The importance of the role played by NK cells in antitumor immune responses highlights opportunities for controlling tumors by manipulating the NK cell arm of innate immunity (Orange and Ballas, 2006; Terme et al., 2008).
Our study showed that treatment with the combination of CpG-ODN and LL-37 could enhance proliferation of NK cells and macrophages in the peritoneal cavity (Fig. 4). Although some studies suggest that CpG-ODN may indirectly augment NK cell activity by inducing the secretion of interleukin (IL)-12 in macrophages (Ballas et al., 1996; Cowdery et al., 1996), peritoneal macrophages were not found to play an essential role in the antitumor effects generated by CpG-ODN in combination with LL-37 as demonstrated by depletion experiments. Thus, the role of peritoneal macrophages in the observed antitumor effects caused by CpG-ODN remains to be further characterized.
The encouraging results from the current study serve as an important foundation for future clinical translation, particularly in patients with ovarian cancer, which tends to lead to ascites and cachexia in patients. It has been shown that cachexia syndrome may be caused by cytokines either produced by the tumor or released by the immune cells in response to the tumor and tumor products (for a review see Seruga et al., 2008). Thus, the reduction of tumor load by immunotherapeutic strategies may potentially lead to the improvement of cachexia syndrome, resulting in better survival and quality of life for patients
For clinical translation, it is important to address concerns regarding the potential toxicity associated with administration of the specific drugs or reagents into the body. We have previously developed a Hsp70-secreting ovarian tumor cell-based vaccine for the control of lethal ovarian cancer (Chang et al., 2007). In this respect, the current study is highly translatable because the LL-37 peptide is a natural compound and is expressed in humans (Durr et al., 2006). In addition, CpG-ODN is commercially available and has been used in several clinical trials with no harmful side effects (Link et al., 2006; McHutchison et al., 2007). Thus, the current strategy employing CpG-ODN and LL-37 reagents has significant potential for future clinical translation.
In summary, our study has demonstrated that the LL-37 peptide can potentiate delivery of CpG-ODN into peritoneal immune effectors, leading to potent tumor cytotoxic effects in treated mice. In fact, CpG-ODN is currently being employed in several clinical trials in several cancer models with promising results. Moreover, TLR3, TLR7, TLR8, and TLR9 are expressed in the same intracellular compartments, mainly in the endosomes and the endoplasmic reticulum, and therefore our strategy may be applied to enhance ligand function to these Toll-like receptors. Our results potentially serve as an important foundation for future clinical translation. Furthermore, similar strategies may potentially be applied to other cancer systems.
The authors thank Dr. T.-C. Wu for helpful discussions and critical review of the manuscript. This work was supported by ovarian cancer grants from the Alliance for Cancer Gene Therapy (ACGT), the NCDGG (1U19 CA113341-01), and the American Cancer Society (ACS).