SEARCH

SEARCH BY CITATION

Keywords:

  • bladder;
  • bladder carcinoma;
  • mice;
  • interleukin-23;
  • gene therapy

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

OBJECTIVES

  • • 
    To evaluate the antitumour effects of IL-23 gene transfer into mouse bladder carcinoma (MBT2) cells.
  • • 
    To investigate the mechanisms underlying the subsequent constitutive secrection of IL-23 by the MBT2 cells

MATERIALS AND METHODS

  • • 
    An expression vector containing IL-23 gene was introduced into MBT2 cells by liposome-mediated gene transfer, and secretion of IL-23 was confirmed by ELISA.
  • • 
    The in vivo antitumour effect of IL-23-secreting MBT2 cells (MBT2/IL-23) was examined by injecting the cells into syngeneic C3H mice.
  • • 
    A tumour vaccination study using mitomycin C (MMC)-treated IL-23-secreting MBT2 cells was carried out, and the usefulness of in vivo CD25 depletion for an additional vaccine effect was also investigated.
  • • 
    The mechanisms underlying the antitumour effects were investigated by antibody depletion of CD8 or CD4 T cells, or natural killer cells, and cells infiltrating the tumour sites in vivo were assessed using immunohistochemistry.

RESULTS

  • • 
    Stable transformants transduced with MBT2/IL-23 secreted IL-23 into the culture supernatant.
  • • 
    Genetically engineered IL-23-secreting MBT2 cells were rejected in syngeneic mice.
  • • 
    MBT2/IL-23-vaccinated mice inhibited the tumour growth of parental MBT2 cells injected at a distant site and this vaccine effect was enhanced by combination with in vivo CD25 depletion by an antibody.
  • • 
    The main effector cells for the direct antitumour effect of MBT2/IL-23 were CD8 T cells, which was shown by in vivo depletion and immunohistochemical study.

CONCLUSIONS

  • • 
    IL-23-secreting MBT2 cells were rejected in syngeneic mice by the activation of CD8 T cells.
  • • 
    MMC-treated MBT2/IL-23 can have a tumour vaccine effect for parental MBT2 cells, and this effect was enhanced by combination with in vivo CD25 depletion.

Abbreviations
MBT2

mouse bladder carcinoma

MMC

mitomycin C

MBT2/IL-23

IL-23 gene secreting MBT2 cell

BT

bladder tumour

NK cells

natural killer cells

IFN-γ

interferon-γ

Treg

regulatory T cell

ELISA

enzyme-linked immunosorbent assay

MBT2/β-GAL

MBT2 cells transfected with β-galactosidase

RT

room temperature

INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

In the 1980s, subgroups of patients with T1G3 bladder tumours (BTs) and/or carcinoma in situ had a significantly higher progression rate to carcinoma invading the bladder muscle and a poorer long-term survival, which led to the concept of ‘high-risk’ non-muscle-invasive BT. [1]

In the past, patients with high-risk BTs were candidates for undergoing total cystectomy. [2] However, recently, the accepted standard treatment for high-risk BTs consists of transurethral resection, removing all visible lesions, followed by intravesical therapy, in particular, intravesical instillation of BCG, now recognized as the best treatment for high-risk BT [3,4]. Although BCG was employed to treat many kinds of solid tumours in the 1970s, its usefulness has proven to be limited in most tumours [5,6]. BCG instillation therapy was first introduced for BT by Morales in 1976 [5], and its usefulness was confirmed in numerous reports [3,4]. Thus, BT is one of the few tumours for which immunotherapy is effective.

Two novel IL-12-related cytokines, IL-23 and IL-27, were recently identified [7,8]. IL-23 is composed of p19 and the p40 subunit of IL-12. It is mainly secreted by activated dendritic cells and monocytes/macrophages. IL-23 induces the proliferation of memory Th1 CD4+ T cells and the production of interferon-γ (IFN-γ) from activated T cells. IL-23 and IL-12 play an essential role in linking innate and adaptive immunity. Because IL-12 is known to have strong antitumour effects in vivo, IL-23 may also possess anti-tumour activity [9–12].

In urology, systemic immunotherapy using IL-2 and IFN-γ is applied for RCC. In particular, high-dose IL-2 therapy can result in long-term complete remission in a minority of metastatic RCC patients. However, severe side effects such as fatal vascular leak syndrome limit the application of this therapy. To overcome this problem, immunogene therapy is a possible strategy. Introducing a cytokine gene into tumour cells can maintain a high concentration of cytokine at the tumour site, while keeping systemic levels low, thus diminishing the systemic side effects. We previously reported antitumour activities using different cytokines, such as IL-2 [13,14], IL-12, IL-18 [15–17] and IL-21 [18,19].

In the present study, we evaluated the antitumour effects of IL-23 gene transfer into a mouse bladder carcinoma (MBT2) cell line. We also investigated the mechanisms mediating this antitumour effect and the possibility of developing a cancer vaccine for future clinical use.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

TUMOUR CELL LINES

Mouse bladder carcinoma, of C3H origin (American Type Culture Collection, Manassas, VA, USA), was cultured in Eagle’s minimum essential medium supplemented with 10% heat inactivated fetal bovine serum.

EXPRESSION VECTOR DESIGN

The cDNAs encoding single-chain IL-23, i.e. the p40 chain, a (Gly4Ser)3 linker, and the p19 chain, were amplified from a single-chain IL-23-immunogloblin fusion protein expression plasmid using PCR [20,21]. The DNA constructs for IL-23 were finally cloned into the p3xFLAG-CMV-9 (Sigma Chemical Co., St. Louis, MO, USA) expression vector (p3xFLAG-CMV-mscIL-23). This vector has a preprotrypsin signal peptide and a 3xFLAG-epitope-tag sequence at the NH2 terminals and, hence, expresses a secreted NH2-terminal 3xFLAG fusion protein in mammalian cells.

TRANSFECTION OF EXPRESSION VECTOR INTO A TUMOUR CELL LINE

The expression vector was transfected into MBT2 cells by liposome-mediated gene transfer. Briefly, MBT2 cells (1.8 × 105) were incubated in a 6-cm dish for 24 h before transfection. Purified IL-23 (4 µg), cloned into plasmid expression vector p3xFLAG-CMV-mscIL-23, was added to MBT2 cells after pre-incubation for 20 min with Lipofectamine 2000 TM (Invitrogen, Carlsbad, CA, USA) and serum-free Opti MEM®. Drug selection in 0.5 mg/mL Geneticin® was started 1 day after transfection. Colonies were harvested by cloning ring and expanded to cell lines. Similarly, a control cell producing β-galactosidase was produced.

ELISA FOR MOUSE IL-23

To estimate the cytokine production level, a total of parental and 1 × 106 IL-23 transfected MBT2 cells were cultured in a 25-cm2 flask with 2.0 mL medium for 24 h. Culture supernatants were then collected. In brief, monoclonal anti-mouse IL-12/IL-23 p40 antibody (R&D Systems, Minneapolis, MN, USA) was coated overnight onto a 96-well plate at RT. After incubation with blocking solution (1% BSA) for 1 h at RT, the sample supernatants were added to the coated plate and incubated for 2 h at RT. The samples were then washed (0.05% Tween 20 in PBS), incubated with anti-mouse IL-23 p19 antibody (R&D Systems), which was biotinylated using a Biotin Labelling Kit-NH2 (Dojindo Molecular Technologies, Inc., Rockville, MD, USA) for 1 h at RT, further washed, and incubated with streptavidin-alkaline phosphatase (Vector Laboratories, Burlingame, CA, USA) for 1 h at RT. Finally, SIGMAFASTTM OPD (Sigma Chemical Co.) was added to each well and activity was measured using an MTP-300 microplate reader (CORONA ELECTRIC, Ibaraki, Japan).

IN VIVO TUMORIGENICITY OF EACH SUBLINE

The right flanks of 6- to 8-week-old female syngeneic C3H mice (Nippon Clea, Tokyo, Japan) were injected s.c. with 1 × 106 cells from each subline. Tumour growth was measured at least 3 times weekly using calipers. The longest surface length (a) and the width perpendicular to this dimension (b) were measured and tumour size was described in mm2 as (a) × (b). In all experiments for tumour growth measurement, each group contained five mice and we repeated the experiment at least twice for confirmation. Figures show the representative data as repeated experiments showed the same tendency. All animal experiments were performed according to the Guidelines for Animal Experimentation at Wakayama Medical University. The mice were killed when the tumour diameter reached 15 mm.

RECHALLENGE STUDY

Parental MBT2 cells (1 × 106) were injected s.c. into the left flank of syngeneic C3H mice that had rejected the MBT2/IL-23 after 1 × 106 per mouse had been similarly administered into the right flank. The tumour growth of parental MBT2 cells was measured, as described.

TUMOUR VACCINE STUDY

MBT2/IL-23 was preincubated with mitomycin C (MMC; 50 µg/mL; Sigma Chemical Co.) at 37 °C for 30 min and then washed with PBS twice. On day −7, 5 × 106 cells were injected s.c. into the left flanks as a tumour vaccine treatment. Mice were then challenged with 1 × 106 cells of parental MBT2 cells injected s.c. on day 0 into the right flanks. The prophylactic efficacy of this transfectant was also examined in combination with anti-CD25 mAb (PC61) treatment. In brief, 0.1 mL of ascites fluid was injected i.p. on day −8, followed by a single vaccination with 5 × 106 cells on −7. As a control, the same concentration of rat IgG as anti-CD25 mAb was injected i.p. Tumour growth was measured, as described.

We checked the distribution of CD4+ CD25+ T cells in regional lymph nodes in each group. Mice were killed 14 days after s.c. injection of parental MBT2 cells and we prepared lymphocyte suspension from popliteal lymph nodes. Cells were washed and incubated with the following mAbs for 30 min at 4 °C in PBS: anti-CD4-PE and anti-CD25-FITC mAbs (BD Pharmingen, San Jose, CA, USA). After incubation, the cells were washed, suspended in PBS and analysed using a FACS Calibur (Becton Dickinson, Franklin Lakes, NJ, USA).

We also checked the distribution of CD4+ Foxp3+ T cells in regional lymph nodes in each group. Mice were killed 14 days after s.c. injection of parental MBT2 cells and we prepared lymphocyte suspension from popliteal lymph nodes.

Cells were labelled for surface marker for 30 min at 4 °C with anti-CD4-FITC (BD Pharmingen) before intracellular staining for Foxp3. Staining for Foxp3 was done with the phycoerythrin anti-mouse Foxp3 antibody (e-BioScience, San Diego, CA, USA) staining kit and according to the manufacturer’s protocol. After incubation, the cells were washed, suspended in PBS and analysed using a FACS Calibur. In CD4+ CD25 and CD4+ Foxp3+ staining experiments, each group had three mice and the experiments were repeated once for confirmation.

IN VIVO DEPLETION OF IMMUNE CELLS

Anti-asialo GM1 antiserum (Wako Fine Chemicals, Osaka, Japan; 50 µL per mouse), diluted with 150 µL PBS, was injected i.p. 24 h before tumour inoculation and once every 5 days thereafter to deplete natural killer (NK) cells. CD4+ or CD8+ T cells were depleted in vivo by i.p. administration of rat mAbs GK1.5 (anti-CD4) and 53.6 (anti-CD8). Anti-CD4 or CD8 antibody (500 µg per mouse) was injected i.p. 24 h before tumour inoculation and once every 7 days thereafter. CD4 T or CD8 T cell depletion was confirmed by cytofluorometric analysis, as previously described [15,18].

IMMUNOHISTOCHEMISTRY

Because MBT2/IL-23 was completely rejected, a mixture of MBT2/IL-23 (1 × 106) and parental MBT2 (1 × 106) cells was injected into the right flanks of mice. As a control, parental MBT2 (1 × 106) cells alone were injected. At 10 days the mice were killed and subcutaneous tumours were resected, embedded in Tissue-Tek® without fixation, frozen in liquid nitrogen and stored at −80 °C before sectioning with a cryostat. Sections were incubated with mAbs GK1.5 (anti-CD4) and 53.6 (anti-CD8) for 60 min, followed by peroxidase-conjugated goat anti-rat IgG secondary antibody (Nichirei, Tokyo, Japan) for 30 min. Peroxidase activity was detected using diaminobenzidine peroxidase (Nichirei). Sections were counterstained with Mayer’s Hematoxylin (MUTO PURE CHEMICAL CO. Ltd., Tokyo, Japan) and dehydrated. Each group had three mice and the experiment was repeated once for confirmation.

STATISTICAL ANALYSIS

The significance of differences in means among groups was determined using Dunnett’s test, Student’s t-test or the Tukey–Kramer post hoc test. A P value of <0.05 was considered to indicate statistical significance. Data were presented as mean (sem). All data were tabulated and analysed using JMP® 7.0.1 (SAS Institute Inc., Cary, NC, USA).

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

IL-23 PRODUCTION BY MBT2 TRANSFECTED WITH IL-23 GENE

The MBT2 cell line was transfected with IL-23 p3xFLAG-CMV-mscIL-23. After drug selection, several Geneticin®-resistant stable transfectants were isolated and their IL-23 secretion into the supernatant was measured by sandwich ELISA. Clones that did secrete IL-23 (MBT2/IL-23 ♯4 and MBT2/IL-23 ♯8) produced approximately 20 and 30 ng/mL of IL-23, respectively. Parental MBT2 and MBT2/β-Gal (transfected with the same expression vector encoding β-galactosidase) did not secrete IL-23 into the supernatant. Figure 1 shows the representative data of three independent experiments.

image

Figure 1. IL-23 secretion by transduced tumour cells was assayed by sandwich ELISA in triplicate. The overnight supernatant from 1 × 106 MBT2 (MBT2/P), MBT2/β-Gal, IL-23cDNA-transfected MBT2 sublines (MBT2/IL-23). Note representative results in MBT2/IL-23 sublines, MBT2/IL-23 ♯4, MBT2/IL-23 ♯8. The representative data of three independent experiments are shown. Bars represent sem.

Download figure to PowerPoint

IN VIVO TUMORIGENICITY OF EACH SUBLINE

To determine the effect of IL-23 secretion on tumour growth in vivo, 1 × 106 MBT2/IL-23 ♯4, MBT2/IL-23 ♯8, MBT2/β-Gal or MBT2 cells were injected into the right flanks of syngeneic C3H mice. Tumours arose in mice injected with MBT2 or MBT2/β-Gal 15–20 days after inoculation and rapidly grew. However, syngeneic C3H mice completely rejected MBT2 transfected with IL-23 gene, both MBT2/IL-23 ♯4 and MBT2/IL-23 ♯8 (Fig. 2[P < 0.01, MBT2/IL23 ♯4 or MBT2/IL23 ♯8 vs. MBT2, Dunnett’s test]). In the following experiments, we show representative data using either MBT2/IL-23 ♯4 or MBT2/IL-23 ♯8 cell lines, because both yielded similar results.

image

Figure 2. Tumour growth in syngeneic C3H mice inoculated with parental MBT2 cells (MBT2/P), cells transfected with control vector (MBT2/β-Gal) or IL-23 cDNA (MBT2/IL23 ♯4, MBT2/IL23 ♯8). Mice were injected s.c. with 1 × 106 cells in the right flank on day 0. Tumour growth was measured as the product of maximal and perpendicular diameters. Tumours arose in mice injected with MBT2 or MBT2/β-Gal 15–20 days after inoculation and rapidly grew. However, syngeneic C3H mice completely rejected MBT2 transfected with the IL-23 gene, both MBT2/IL23 ♯4 and MBT2/IL23 ♯8. **, P < 0.01 (MBT2/IL23 ♯4 or MBT2/IL23 ♯8 vs. MBT2 Dunnett’s test).

Download figure to PowerPoint

RECHALLENGE STUDY WITH PARENTAL MBT2 IN MICE THAT REJECTED MBT2/IL-23

Parental MBT2 cells (1 × 106) were injected s.c. into the left flanks of the syngeneic C3H mice that had rejected MBT2/IL-23 cells in the right flank (1 × 106 per mouse). Tumour growth of rechallenged MBT2/P in mice that had rejected MBT2/IL-23 was significantly retarded compared with that in naïve mice. (Fig. 3*, [P < 0.05 **, P < 0.01 unpaired Student’s t-test]).

image

Figure 3. Rechallenge with parental MBT2 (MBT2/P) in five mice that had rejected MBT2/IL-23. Parental MBT2 cells (1 × 106) were injected s.c. into the left flank of syngeneic C3H mice that had rejected MBT2/IL-23 cells after injection of 1 × 106 per mouse similarly administered into the right flank. Tumour growth of parental MBT2 was measured as described. *, P < 0.05; **, P < 0.01 (unpaired Student’s t-test).

Download figure to PowerPoint

TUMOUR VACCINE STUDY

Pre-injection of MBT2/IL-23 treated with MMC into the left flanks significantly inhibited the subsequent challenge of parental MBT2 cells in the right flanks. Furthermore, this tumour vaccine effect was markedly enhanced in mice that were treated with anti-CD25 mAb, leading to complete rejection (Fig. 4, [P < 0.05 Tukey-Kramer test]). We also confirmed that equal concentrations of rat IgG and anti-CD25 mAb ascites did not affect tumour growth of MBT2 (data not shown).

image

Figure 4. Tumour vaccine effect in subcutaneous models. MMC-treated MBT2/IL-23 cells (5 × 106 per mouse) were injected s.c. into the left flank of five mice as the tumour vaccine on day −7. The right flank was injected subcutaneously with 1 × 106 parental MBT2 cells on day 0. The prophylactic efficacy of this transfectant was also examined in combination with anti-CD25 mAb (PC61) treatment. Tumour growth was measured, as described MMC-treated MBT2/IL-23 exerted a tumour vaccine effect for parental MBT2. The prophylactic vaccine effects combined with anti-CD25 mAb were markedly enhanced. *, P < 0.05 (Tukey-Kramer test).

Download figure to PowerPoint

We also checked the distribution of CD4+ CD25+ T cells in regional lymph nodes in each group. Mice were killed 14 days after injection of s.c. parental MBT2 cells and we prepared lymphocyte suspension from popliteal lymph nodes. The percentage of CD4+ CD25+ T cells did not differ significantly among non-treated mice, MBT2-injected mice and MBT2-injected mice that had been vaccinated with MBT2/IL23. Only the mice in the group that had been treated with anti-CD25 mAb showed a notable depletion of CD4+ CD25+ T cells (Fig. 5A,B[P < 0.01 , experimental vs. non-treated mice, Dunnett’s test]).

image

Figure 5. The distribution of CD4 + CD25+ T cells and CD4+Foxp3+ T cells in regional lymph node was analysed in each tumour vaccine study group. **, P < 0.01 (experimental vs. non treated mice, Dunnett’s test). A: Representative flow cytometric data. CD4 + CD25+ T cells delineated by the rectangles. B: Mean percentage of CD4 + CD25+ T cells. Each experimental group consisted of three mice. Bars represent mean percentage of CD4 + CD25+ in three mice. C: Representative flow cytometric data. CD4+ Foxp3+ T cells delineated by the quadrants. D: Mean percentage of CD4+ Foxp3+ T cells. Each experimental group consisted of three mice. Bars represent mean percentage of CD4 + Foxp3+ in three mice.

Download figure to PowerPoint

We also checked the distribution of CD4+ Foxp3+ T cells in regional lymph nodes in each group. Mice were killed 14 days after s.c. injection of parental MBT2 cells and we then prepared a lymphocyte suspension from popliteal lymph nodes. The percentage of CD4+ Foxp3+ T cells did not differ significantly among non-treated mice, MBT2-injected mice and MBT2-injected mice that had been vaccinated with MBT2/IL23. Only the mice in the group that had been treated with anti-CD25 mAb showed a notable depletion of CD4+ Foxp3+ T cells [Fig. 5C,D (P < 0.01, experimental vs. non treated mice, Dunnett’s test)].

DEPLETION OF CD4+, CD8+ OR NK CELLS IN MBT2/IL-23-INJECTED MICE

To determine the nature of the immune cells involved in the rejection of MBT2/IL-23, 1 × 106 of these cells were injected s.c. into the right flanks of syngeneic C3H mice that had been previously administered antibodies against CD4 (GK1.5), CD8 (53.6) or NK cells (anti-asialo GM1). CD4 T or CD8 T cell depletion was confirmed by cytofluorometric analysis on a FACS Calibur (Becton Dickinson) in splenocytes using the mAbs GK1.5 (anti-CD4) or 53.6 (anti-CD8), respectively. NK cell depletion was confirmed by 4-h 51Cr assays, as previously described [17,18]. The antitumour effect of MBT2/IL-23 was slightly reduced in mice depleted of CD4 T cells and NK cells in the initial phase, but was almost completely abrogated in mice depleted of CD8 T cells (Fig. 6).

image

Figure 6. Tumour growth of MBT2/IL23 in CD8-, CD4-, and NK-depleted syngeneic C3H mice. Mice were injected s.c. with 1 × 106 MBT2/IL-23 cells into the right flank on day 0. Antibodies against CD4 (GK1.5), CD8 (53.6) were intraperitoneally injected on day −1 and once every 7 days thereafter. Antibodies against NK cells (anti-asialo GM1) were injected on day −1 and every 5 days thereafter. Tumour growth of parental MBT2 was measured, as described. Each group had three mice and the experiment was repeated once for confirmation. *; P < 0.05 , **; P < 0.01 (experimental vs. MBT2/IL-23, Dunnett’s test).

Download figure to PowerPoint

IMMUNOHISTOCHEMICAL FEATURES

The immunohistochemical features (CD4+, CD8+ T cells) of sites of mixed injection of parental MBT2 and MBT2/IL-23 tumours were examined 10 days after s.c. injection in syngeneic mice. CD4+ T cells appeared to infiltrate MBT2 and the mixture of MBT2 and MBT2/IL-23 tumours to the same degree. In contrast, the number of CD8+ T cells was significantly increased in the mixture of MBT2 and MBT2/IL-23 tumour compared with MBT2 alone (Fig. 7A,B).

image

Figure 7. A: Immunohistochemical features of CD4 + and CD8+ T cells. B: Average number of stained cells per 0.025 mm2 of parental MBT2 and MBT2-IL-23 tumours 10 days after s.c. cell injection in syngeneic C3H mouse, as caluculated in three mice. Bars represent mean percentage of positive cells in three mice). Similar numbers of CD4 + T cells appeared against parental MBT2 and mixed MBT2/IL-23 tumours (P= 0.33). In contrast, the number of CD8 T cells was increased against mixed MBT2/IL-23 tumours (P < 0.001). Each group had three mice and the experiment was repeated once for confirmation. Reduced from ×200. **, P < 0.01 (unpaired Student’s t- test).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

IL-23 is a member of the IL-12 family of heterodimeric cytokines, composed of the p40 subunit shared with IL-12 and the IL-23-specific p19 subunit. IL-23 is normally secreted by activated macrophages and dendritic cells and stimulates proliferation of and IFN-γ production by Th1 CD4+ T cells. IL-23 plays important roles in Th1 immunity similar to IL-12. We have previously established that antitumour activities occur by using IL-12 gene transduction into RCC cell lines[15]. Cells transduced with the IL-12 gene were completely rejected in syngeneic mice but not in athymic nude mice. Cells secreting IL-12 can also act as tumour vaccines. CD8+ T cells and NK cells were essential for this tumour vaccine effect. In the present study, we introduced IL-23 gene, which has a heterodimeric structure similar to that of IL-12, into a murine BT cell line and evaluated the antitumour activities of IL-23 in tumour immunity.

In previous experiments using IL-12, we transfected two expression vectors encoding p40 and p35 simultaneously into RCC cell lines. In the present study using IL-23, we also first applied the same cotransfection methods using expression vectors encoding p40 and p19. However, we were unable to create transfectants expressing the heterodimeric form of IL-23 using this approach. Therefore, we used an expression vector encoding single chain IL-23 composed of the p40 chain, a (Gly4Ser)3 linker, and the p19 chain [20,21]. By using this system, we succeeded in constructing transfectants which secreted the heterodimeric form of IL-23, as detected by sandwich ELISA.

The proliferation rate of clones of MBT2/IL-23 in vitro was not different from that of the parental cells (data not shown). However, MBT2/IL-23 tumours were completely rejected in vivo. In a previous study, we introduced IL-23 and IL-27 genes into a poorly immunogenic melanoma (B16F10) using the same expression vector and evaluated their antitumour activities [22]. Interestingly, IL-23-transfected B16F10 (B16/IL-23) tumours exhibited almost the same growth curve as B16F10 parental tumour until about 20 days after tumour injection, but then showed growth inhibition or even regression. This phenomenon had been observed previously in studies by other investigators [10,12]. In contrast, the MBT2/IL-23 tumour exhibited significant retardation of growth from the early stage. In fact, some authors have reported that expression of IL-23 gene in other tumour cells could cause tumour regression from the early stage in animal tumour models [9,11].

Additionally, the syngeneic C3H mice that had rejected MBT2/IL-23 rejected a subsequent challenge with parental MBT2 cells, suggesting that systemic immunity was conferred by inoculation of MBT2/IL-23. However, with regard to clinical applications, live tumour cell vaccines will not be acceptable. Therefore, we investigated the feasibility of tumour vaccine therapy using MBT2/IL-23 treated with MMC. Treatment with MMC completely inhibits proliferation of tumour cells, while preserving cytokine secretion of MBT2/IL-23 (data not shown). We showed that MMC-treated MBT2/IL-23 cell injection could inhibit the growth of parental MBT2 distant sites, suggesting that these cells could be used as a tumour vaccine. In a previous study[22], we reported that B16/IL-23-vaccinated mice showed significant protective immunity against B16F10 parental tumour cells. Because the tumour vaccine effect shown in this study was also observed in previous investigations [22], we suggest that this may be a useful approach to vaccine therapy in clinical applications.

We investigated the prophylactic vaccine effect combined with anti-CD25 mAb for depletion of regulatory T cells (Tregs). Interestingly, in this setting, the protective immunity against MBT2 parental tumour endowed by MBT2/IL-23 vaccination was markedly enhanced. The immune system has established an elaborate network of central and peripheral tolerance mechanisms to discriminate between self and non-self. An important component of this network are the Tregs, which mediate self-tolerance and immune homeostasis by acting in a dominant cell-extrinsic manner to regulate immune functions[23–25]. Two sets of observations also implicate Tregs in suppression of tumour immunity. First, Tregs accumulation at tumour sites in patients with cancer correlated with disease progression [26]. Second, elimination of Tregs in mice by treatment with a CD25 antibody enhances the immune-mediated rejection of tumours [27,28] and synergizes with vaccination protocols [29,30]. Therefore we analysed the distribution of the CD4+ CD25+ population and the CD4+ Foxp3+ population of regional lymph nodes in a tumour vaccine model. As expected, the CD4+CD25+ and CD4+ Foxp3 population is notably smaller in the mice that had been treated with anti CD25+ mAb. We also expected that the CD4+ CD25+ and CD4+ Foxp3 population would be larger in the mice bearing MBT2, but smaller in the mice vaccinated with MBT2/IL23. However, the populations were similar to those in non-treated mice. In a recent clinical trial, depletion of Tregs in patients with RCC using an IL-2/diphtheria toxin fusion product (ONTAK) led to enhanced vaccine-induced antitumour immune responses [9]. Thus, depletion of Tregs could represent an important addition to cancer immunotherapy.

For clinical applications, these results suggest that IL-23 has great potency in a cytokine-based tumour vaccine and that anti-CD25 treatment could act as an efficient adjuvant in an IL-23-based tumour vaccination.

Mechanisms underlying the antitumour effects observed were investigated in syngeneic mice by depleting CD8 or CD4 T cells, or NK cells using appropriate antibodies. The antitumour effect of MBT2/IL-23 was partially inhibited in mice depleted of CD4 T cells and NK cells at the initial phase, but was almost completely abrogated in mice depleted of CD8 T cells. Using immunohistochemistory, we confirmed that CD8 T cells infiltrated the tumour.

A similar antitumour response has been reported in studies using other tumour types. Wang et al. [10]. reported that the expression of IL-23 in CT26 murine colon carcinoma cells resulted in antitumour effects that were mediated through CD8+T cells secreting IFN-γ. Lo et al. [11] also reported that antitumour activity was mediated through CD8+ T cells but not CD4+ T cells or NK cells. In fact, it seems to be hard to define the mechanism of tumour shrinkage engineered to secrete cytokines. In terms of this complication, Musiani et al. [12] suggested that continual crosstalk between leukocytes and lymphocytes plays a much important role than the single effector population which has been assumed to be stimulated by secreted cytokine. They reconstructed the cell events that take place at the site of the challenge with tumour cells engineered with many kinds of cytokine genes as determined by sequential histological, immunocytochemical and ultrastructural observations. They found that granulocytes induced by T cells at the tumour site play a crucial role in the rejection mechanism. This unexpected and major role for granulocytes seems to be key to understanding the mechanism.

In conclusion, we observed that CD8+ T cells are required for the IL-23-mediated antitumour effects described here. Moreover, MMC-treated IL-23-secreting MBT2 cells acted as a tumour vaccine for parental MBT2 rejection. We found that this vaccine effect was enhanced by combining it with CD25 antibody treatment. We will continue our vaccination studies in preparation for future clinical application.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES