Coexpression of Flt3 ligand and GM-CSF genes modulates ... - Nature

7 downloads 0 Views 416KB Size Report
Aug 17, 2007 - DNA vaccine and dendritic cells (DCs)-based vaccine have emerged as promising strategies for cancer immunotherapy. Fms-like tyrosine ...
Cancer Gene Therapy (2007) 14, 904–917 r

2007 Nature Publishing Group All rights reserved 0929-1903/07 $30.00

www.nature.com/cgt

ORIGINAL ARTICLE

Coexpression of Flt3 ligand and GM-CSF genes modulates immune responses induced by HER2/neu DNA vaccine Y-T Yo1, K-F Hsu2, G-S Shieh3, C-W Lo4, C-C Chang1, C-L Wu1,5 and A-L Shiau1,4 1

Institute of Basic Medical Sciences, National Cheng Kung University Medical College, Tainan, Taiwan; Department of Obstetrics and Gynecology, National Cheng Kung University Medical College, Tainan, Taiwan; 3Department of Urology, Tainan Hospital, Department of Health, Executive Yuan, Taiwan; 4 Department of Microbiology and Immunology, National Cheng Kung University Medical College, Tainan, Taiwan and 5Department of Biochemistry and Molecular Biology, National Cheng Kung University Medical College, Tainan, Taiwan 2

DNA vaccine and dendritic cells (DCs)-based vaccine have emerged as promising strategies for cancer immunotherapy. Fms-like tyrosine kinase 3-ligand (Flt3L) and granulocyte–macrophage-colony-stimulating factor (GM-CSF) have been exploited for the expansion of DC. It was reported previously that combination of plasmid encoding GM-CSF with HER2/neu DNA vaccine induced predominantly CD4 þ T-cell-mediated antitumor immune response. In this study, we investigated the modulation of immune responses by murine Flt3L and GM-CSF, which acted as genetic adjuvants in the forms of bicistronic (pFLAG) and monocistronic (pFL and pGM) plasmids for HER2/neu DNA vaccine (pN-neu). Coexpression of Flt3L and GM-CSF significantly enhanced maturation and antigen-presentation abilities of splenic DC. Increased numbers of infiltrating DC at the immunization site, higher interferon-g production, and enhanced cytolytic activities by splenocytes were prominent in mice vaccinated with pN-neu in conjunction with pFLAG. Importantly, a potent CD8 þ T-cell-mediated antitumor immunity against bladder tumors naturally overexpressing HER2/neu was induced in the vaccinated mice. Collectively, our results indicate that murine Flt3L and GM-CSF genes coexpressed by a bicistronic plasmid modulate the class of immune responses and may be superior to those codelivered by two separate monocistronic plasmids as the genetic adjuvants for HER2/neu DNA vaccine. Cancer Gene Therapy (2007) 14, 904–917; doi:10.1038/sj.cgt.7701081; published online 17 August 2007 Keywords: Flt3 ligand; GM-CSF; genetic adjuvant; HER2/neu; bladder cancer; DNA vaccine

Introduction

When activated, antigen (Ag)-loaded dendritic cells (DC) migrate to the draining lymph nodes where they prime Ag-specific CD4 þ and CD8 þ T cells.1,2 Administration of soluble recombinant Fms-like tyrosine kinase 3-ligand (Flt3L) and granulocyte–macrophage-colony-stimulating factor (GM-CSF) to mice exerts an additive effect on the expansion of splenic DC numbers.3 Flt3L possesses a growth-stimulatory effect on DC precursors and is capable of generating large numbers of DC in vivo.4–6 In addition, GM-CSF can recruit Ag-presenting cells, primarily DC.7 Introductions of viral vectors8 or separate Correspondence: Professor A-L Shiau, Department of Microbiology and Immunology, National Cheng Kung University Medical College, 1 Dashiue Road, Tainan 70101, Taiwan. E-mail: [email protected] or Professor C-L Wu, Department of Biochemistry and Molecular Biology, National Cheng Kung University Medical College, 1 Dashiue Road, Tainan 70101, Taiwan. E-mail: [email protected] Received 2 March 2007; revised 21 May 2007; accepted 24 June 2007; published online 17 August 2007

naked plasmid DNA encoding Flt3L or GM-CSF, alone or in combination, into cells by different inoculation routes and methods have been investigated aiming at enhancing DC-mediated antitumor immunity.9–13 However, to our knowledge, there have been no reports describing the use of both murine Flt3L and GM-CSF, which were encoded either by two separate monocistronic plasmids or by a bicistronic plasmid, as genetic adjuvants to modulate the immune responses induced by a DNA vaccine against a specific tumor-associated Ag. ErbB-2/HER2/neu oncogene encodes for a 185-kDa transmembrane glycoprotein that contains tyrosine kinase activity and belongs to the epidermal growth factor receptor family.14,15 HER2/neu is one of the most appropriate target antigens for anticancer therapy and also a target for both cellular and humoral immune responses. Tumor rejection mediated by HER2/neu DNA vaccination requires CD4 þ T cells.16,17 Furthermore, CD4 þ T cells are the primary antitumor effectors induced by HER2/neu and GM-CSF DNA immunization.18,19 In this study, we exploited HER2/neu DNA vaccine to study the adjuvant effect of murine Flt3L and GM-CSF in different forms of plasmid construction. We show that

Coexpression of Flt3L and GM-CSF for DNA vaccine Y-T Yo et al

coexpression of Flt3L and GM-CSF as genetic adjuvants induced a CD8 þ T-cell-mediated antitumor immune response. These results were somewhat different from earlier reports showing that the protection mediated by HER2/neu DNA vaccine in conjunction with GM-CSFencoded plasmid is predominantly CD4 þ T-cell dependent.18,19 Surprisingly, the bicistronic expression plasmid appeared to be more potent as the genetic adjuvant than combination of two monocistronic plasmids in activating DC and enhancing cytolytic activities of natural killer (NK) and T cells in the murine mouse bladder tumor-2 (MBT-2) tumor model. Taken together, our results suggest that coexpression of Flt3L and GM-CSF represents a promising strategy to modulate the class of antitumor immune responses and enhance the efficacy of DNA vaccines against tumor-associated antigens.

Materials and methods

Cell lines, mice, monoclonal antibodies and rat HER2/neu peptide MBT-2 and COS-7 (African green monkey kidney) cells were cultured in Dulbecco’s modified Eagle’s medium plus 10% fetal bovine serum with standard supplements. YAC-1 (mouse lymphoma) cells were cultured in RPMI 1640 medium. Mouse splenocytes were cultured as described previously.20 Female C3H/HeN and BALB/cJ mice were obtained from the Laboratory Animal Center of National Cheng Kung University, and used between 6 and 8 weeks of age. The experimental protocol adhered to the rules of the Animal Protection Act of Taiwan and was approved by the Laboratory Animal Care and Use Committee of National Cheng Kung University. All monoclonal antibodies used in this study were purchased from BD Biosciences (San Diego, CA), except where otherwise stated. The rat HER2/neu peptide (PDSLRDLSVF, amino acids 420–429)21 was synthesized by DigitalGene Biosciences Corp. (Nei-Hu, Taipei, Taiwan). Plasmid construction and preparation for vaccination The Flt3L expression vector, designated pFL, was derived from pcDNA3.1() (Invitrogen, Carlsbad, CA) by insertion of a 580-bp fragment encompassing the signal peptide and extracellular domain of murine Flt3L into its XbaI and EcoRI sites.9 The coding region of murine GM-CSF was amplified from pGEM-4/GM-CSF22 by PCR with forward primer 50 -GGGATCCATGTGGCTG CAGAATTTA and reverse primer 50 -GAAGCTTTCAT TTTTGGACTGGGTTT, which introduced BamHI and HindIII sites onto the 50 - and 30 -ends, respectively. The resulting 425 bp PCR product was digested with BamHI and HindIII, and cloned into the BamHI/HindIII sites of pcDNA3.1() and pFL to generate pGM and pFL.GM-CSF, respectively. For constructing pFLAG coexpressing Flt3L and GM-CSF linked by an internal ribosome entry site (IRES) element, a B600-bp fragment containing the IRES sequence of encephalomyocarditis virus with EcoRI site at its 50 -end and BamHI site at its 30 -end obtained by PCR amplification was cloned into the

EcoRI/BamHI sites of pFL, yielding pFL.IRES. The Flt3L-IRES fragment was then excised from pFL.IRES by XbaI and BamHI digestion and cloned into the large vector-containing fragment isolated from pFL.GM-CSF by digestion with XbaI and BamHI, resulting in pFLAG. Meanwhile, the coding region encompassing the N-terminal fragment of rat p185neu protein (N-neu) was excised from pRc/CMV-N0 -neu by HindIII and NotI digestion23 and inserted into pcDNA3.1( þ ) to generate pN-neu. Plasmids were purified using DNA extraction kit (GeneAid, Taoyuan, Taiwan) and Endofree Qiagen Mega kits (Qiagen, Hilden, Germany) for in vitro and in vivo experiments, respectively.

Detection of transgene expression in vitro COS-7 cells transfected with 2 mg of pFLAG or pcDNA3.1 using lipofectamine 2000 (Invitrogen) were fixed 24 h later and examined for Flt3L expression using polyclonal antibody against mouse Flt3L (F-19; Santa Cruz Biotechnology, Santa Cruz, CA) by immunohistochemical staining as described previously.24 The GM-CSF contents in the supernatants of COS-7 cells transfected with pFLAG or pGM were quantified by enzyme-linked immunosorbent assay (ELISA) (R&D, Minneapolis, MN) at 24 and 36 h post-transfection. Meanwhile, COS-7 cells transfected with pN-neu or pcDNA3.1 were examined for HER2/neu expression by intracytoplasmic staining using the Cytofix/Cytoperm kit according to the manufacturer’s instructions (BD Biosciences) and analyzed by flow cytometry 24 h later as described previously.23 Phenotypic and functional analyses of splenocytes from mice injected with cytokine plasmids Groups of three C3H/HeN mice were immunized intramuscularly (i.m.) thrice at weekly intervals with pFLAG, pFL plus pGM or control vector (50 mg for each plasmid) diluted in 50 ml saline. Seven days after the last immunization, the mice were killed and their splenocytes harvested for phenotypic and functional analyses. Splenic DC were purified using anti-mouse CD11c-coated magnetic microbeads and separation columns (Miltenyi Biotech, Bergisch–Gladbach, Germany) and stained with fluorescein isothiocyanate (FITC)-conjugated anti-major histocompatibility complex (MHC) class II (MHC-II) (I-EK) (14-4-4S) or anti-CD86 (GL1) monoclonal antibody for flow-cytometric analysis as described previously.20 Mixed lymphocyte reaction was performed as described previously.11 Briefly, purified DC (107) that had been treated for 1 h with 25 mg ml1 of mitomycin C served as stimulator cells. Allogeneic responder cells were harvested from the spleen of BALB/cJ (H-2d) mice, stained with R-phycoerythrin-conjugated anti-CD3e (145-2C11) and FITC-conjugated anti-CD4 (GK1.5), gated, and sorted using FACSAria (Becton Dickinson, San Jose, CA). Varying numbers of mitomycin C-treated CD11c þ splenic DC from C3H/HeN mice injected with cytokine plasmids were cultured with 105 allogeneic CD4 þ T cells from BALB/cJ mice to achieve desired stimulator/responder (S/R) ratios for 3 days. T-cell

Cancer Gene Therapy

905

Coexpression of Flt3L and GM-CSF for DNA vaccine Y-T Yo et al

906

proliferation was monitored by (3-(4,5-dimethylthiazol-2yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2Htetrazolium, inner salt) (MTS)-based, CellTiter 96 Aqueous One Solution Cell proliferation Assay kit (Promega, Madison, WI).

Histochemical and immunofluorescent analyses Groups of three C3H/HeN mice were immunized i.m. at days 0 and 14 with pN-neu in conjunction with pFLAG, pFL plus pGM or control vector, with 50 mg for each plasmid diluted in 50 ml saline. The site of injection was marked by the yellow dye picric acid. Seven days after the second immunization,25 the mice were killed, and the muscle around the injection site was excised and snapfrozen. Serial cryostat sections with 5 mm thickness were stained with hematoxylin–eosin, and reacted with FITC-conjugated anti-CD11c (HL-3), FITC-conjugated anti-MHC-II (14-4-4S) and R-phycoerythrin-conjugated anti-CD8 (Ly-2) monoclonal antibodies. Nuclei were stained with 4,6-diamidino-2-phenylindole. Vaccination, tumor challenge and depletion of T cells in mice To evaluate vaccine efficacy, we first tested whether vaccination with pFLAG alone had non-Ag-specific antitumor effect. C3H/HeN mice were injected i.m. thrice at weekly intervals with pFLAG or control vector (50 mg of plasmid diluted in 100 ml saline), or saline. Seven days after the last immunization, the mice were challenged subcutaneously (s.c.) with MBT-2 cells (1.5  106). We then evaluated whether vaccination with pN-neu DNA vaccine combined with pFLAG or with pFL plus pGM protected mice against MBT-2 tumor challenge in a prophylactic setting. Mice were immunized twice at a 2-week interval with pN-neu in conjunction with pFLAG, pFL plus pGM or control vector, with 100 mg for each plasmid diluted in 100 ml saline. Two weeks after the last immunization, the vaccinated mice were inoculated s.c. with MBT-2 cells (1.5  106). In a therapeutic setting, mice were first inoculated with MBT-2 cells (1.5  106) on day 0 followed by three vaccinations at weekly intervals with pN-neu (100 mg) in conjunction with pFLAG (100 mg), pFL (100 mg) plus pGM (100 mg), or control vector (100 mg). Saline-treated, MBT-2-bearing mice served as vehicle controls in both prophylactic and therapeutic settings. Palpable tumors were measured every 2 or 3 days in two perpendicular axes with a tissue caliper and the tumor volume was calculated as: (length of tumor)  (width of tumor)2  0.45. The animals were killed when their primary tumor reached 2500 mm3. The recorded day of death or the killing of animals was used to calculate survival time. To determine the major class of immune responses in the vaccinated mice, we performed in vivo T-cell depletion experiments as described previously.23 Mice vaccinated with pN-neu combined with pFLAG in the prophylactic protocol were treated intraperitoneally (i.p.) with antimouse CD4 (GK1.5) and/or anti-mouse CD8 (2.43) monoclonal antibodies at 9 and 2 days before tumor challenge. Depletion of CD4 þ and/or CD8 þ T cells was

Cancer Gene Therapy

confirmed by flow-cytometric analysis of peripheral blood samples with anti-CD4 and anti-CD8 monoclonal antibodies and was shown to be 499%. To maintain efficient depletion of T-cell population, antibody treatment was continued at weekly intervals until 40 days after tumor challenge.23

Detection of cell surface markers and intracytoplasmic cytokine staining The assays were performed as described previously with minor modifications.9 Splenocytes from naive or vaccinated mice were incubated with the rat HER2/neu peptide (PDSLRDLSVF, amino acids 420–429) containing the MHC class I epitope21 for detecting HER2/neu-specific CD8 þ T-cell precursors. The HER2/neu peptide was added at a concentration of 2.5 mg ml1 for 19 h. During the final 4 h, Golgistop (BD Biosciences) was added; and cells were then harvested and ready for staining. A portion of the cells was double-stained with R-phycoerythrin-conjugated anti-CD8 (Ly-2) and FITC-conjugated anti-CD25 (7D4) monoclonal antibodies. The remaining cells stained with anti-CD8 antibody were subjected to intracytoplasmic cytokine staining using the Cytofix/ Cytoperm kit with FITC-conjugated anti-interferon (IFN)-g (XMG1.2) antibody and the immunoglobulin isotype control antibody (rat IgG1) and analyzed by flow cytometry as described previously.9 Mouse splenocytes stimulated with PMA (5 ng ml1; Sigma, St Louis, MO) and ionomycin (500 ng ml1; Sigma) for 4 h were used as the positive control for IFN-g-secreting cells. Assays of IFN-g production as well as cytolytic and NK-cell activities of splenocytes in vaccinated mice Mice were immunized at days 0 and 14 with pN-neu combined with pFLAG, pFL plus pGM or control vector (100 mg for each plasmid). At day 28, they were killed and splenocytes harvested for assessing cytolytic activities using the CytoTox 96 non-radioactive cytotoxicity assay (Promega)26 based on the release of lactate dehydrogenase.9 For assessing NK-cell activity, splenocytes were incubated with YAC-1 target cells for 6 h. For determining cytolytic function, freshly isolated splenocytes were re-stimulated in vitro with 20 U ml1 of recombinant human interleukin-2 (IL-2) (R&D) and mitomycin C-treated MBT-2 cells for 5 days. Nonadherent cells were harvested and incubated with MBT-2 target cells at different effector-to-target ratios for 6 h followed by determination of their cytolytic activity. In some experiments, splenocytes were restimulated in vitro with mitomycin C-treated MBT-2 cells for 5 days and their supernatant collected for quantifying IFN-g content by enzyme-linked immunosorbent assay (R&D).10

Results

Construction and characterization of various expression vectors for vaccination We constructed pFLAG, which coexpressed GM-CSF and Flt3L linked by an IRES element, as well as pFL,

Coexpression of Flt3L and GM-CSF for DNA vaccine Y-T Yo et al

pGM and pN-neu, which expressed Flt3L, GM-CSF and N-terminal extracellular domain of rat p185neu, respectively. COS-7 cells transfected with pFLAG, pFL, pN-neu or pcDNA3.1 plasmid DNA were examined for Flt3L or p185neu expressions. Figure 1a shows that Flt3L expression was detectable in COS-7 cells transfected with pFLAG or pFL by immunohistochemical staining. Furthermore, COS-7 cells transfected with pN-neu, but not with the control vector, expressed p185neu as determined by intracytoplasmic staining using flow cytometry (Figure 1b). When COS-7 cells were transfected with pFLAG, the levels of GM-CSF in the supernatant at 24 and 36 h post-transfection were 2709.28720.26 and 4194.06738.79 pg ml1, respectively, as determined by enzyme-linked immunosorbent assay. Likewise, the concentrations of GM-CSF in the supernatant of pGMtransfected COS-7 cells at 24 and 36 h post-transfection were 3082.13727.61 and 4862.37751.31 pg ml1, respectively.

Injection with pFLAG delayed tumor formation and enhanced Ag-presentation ability of splenic DC To determine whether injection with pFLAG alone protected mice against tumors, mice that had received three doses of pFLAG, control vector, or saline at weekly intervals were challenged with MBT-2 cells and tumor growth was monitored. As shown in Figure 1c, tumor formation was found as early as 6 days following tumor challenge in some of the control mice. However, in the pFLAG-treated mice, tumors were not detectable until 11 days after tumor challenge. Furthermore, at day 33 which was the last recorded day while all of the mice were still alive, the mean tumor volume was 1438.917 269.41 mm3 for pFLAG-treated mice, 2377.097427.08 mm3 for control vector-treated mice, and 2424.387423.96 mm3 for saline-treated mice. Although by 22 days tumor formation was detected in all of the mice receiving pFLAG, the difference in the tumor free ratios determined by the Kaplan–Meier analysis between the pFLAG-treated mice and the control vector-treated mice (P ¼ 0.0249) or unvaccinated mice (P ¼ 0.0031) was statistically significant (Figure 1c). Nevertheless, the survival of the mice receiving pFLAG did not significantly prolong compared with their control counterparts (data not shown). These results indicate that injection of pFLAG alone could delay tumor progression, but had no benefit in enhancing the survival. By flow-cytometric analysis, we found that the population of DC that expressed both CD11c and MHC-II molecules was increased in total splenocytes in mice following injection with pFLAG or pFL/pGM, but not with control plasmid (data not shown). We then purified CD11c þ splenocytes and detected their MHC-II expression. More CD11c þ cells obtained from mice injected with pFLAG (mean7s.d. 50.3273.81%) or pFL/pGM (47.4973.67%) expressed MHC-II compared with those injected with control vector (40.8375.22%) or saline (41.0072.62%), as determined from four independent experiments. The differences between pFLAG and vector (P ¼ 0.02), pFLAG and saline (P ¼ 0.006), and pFL/pGM

and saline (P ¼ 0.02) were statistically significant. Figure 1d shows the representative histogram of splenic CD11c þ cells expressing MHC-II from three mice in each group from one experiment. However, the expression level of CD86 was unchanged in splenic CD11c þ cells in mice following pFLAG or pFL/pGM administration (data not shown). Although there were various immune cell populations in total splenocytes, their proliferation rates did not differ among the four groups (data not shown). Furthermore, we harvested and enriched DC from the vaccinated mice and used them as stimulator cells in the mixed lymphocyte reaction. DC from all the treatment groups stimulated different levels of allogeneic T-cell proliferation in a dose-dependent manner. Notably, DC isolated from mice receiving pFLAG were the most efficient stimulators in mixed lymphocyte reaction (Figure 1e).

Vaccination of pN-neu combined with plasmid DNA encoding Flt3L and GM-CSF increased infiltration of DC at the injection site We examined serial cryostat sections of muscle samples derived from the injection site in mice vaccinated with DNA vaccines by hematoxylin–eosin and immunofluorescence staining. Figure 2a shows that the degree of mononuclear cells infiltrating into the injection site was in the order of pFLAG4pFL/pGM4vector4saline, as detected by hematoxylin–eosin stain. The types of infiltrating cells in the serial cryostat sections were also determined by immunofluorescence staining with specific antibodies. Immunofluorescence reveals that mature DC that expressed CD11c and MHC-II were more abundantly found in mice receiving pN-neu plus pFLAG (Figure 2b) than in mice receiving pN-neu plus pFL/pGM (Figure 2c), whereas DC were hardly seen in their control counterparts (Figures 2d and e). Notably, cells that expressed MHC-II were also found in CD11c cells, suggesting that macrophages may also infiltrate into the injection site in mice vaccinated with pN-neu combined with pFLAG or, to a lesser extent, with pFL/ pGM. Furthermore, CD8 þ cells were also detected from the mice vaccinated with pN-neu plus pFLAG (Figure 2b), and, to a lesser extent, from those receiving pN-neu plus pFL/pGM (Figure 2c). Notably, CD8 expression was observed in some, but not in all DC that coexpressed CD11c and MHC-II (Figures 2b and c). However, T cells were not found in all the muscle sections examined, as detected by immunohistochemical staining with anti-CD3 antibody (data not shown). These results suggest that both CD8 þ lymphoid and CD8 myeloid DC infiltrated into the vaccination site in mice receiving pN-neu combined with pFLAG or pFL/pGM. Prophylactic vaccination with pN-neu combined with pFLAG-induced potent antitumor immunity Since nonspecific adjuvant effect of cytokines encoded by pFLAG may have been too weak to suppress tumor growth, combination with DNA vaccine encoding specific tumor Ag, such as HER2/neu, may be required for inducing potent antitumor immunity in the MBT-2 tumor

Cancer Gene Therapy

907

Coexpression of Flt3L and GM-CSF for DNA vaccine Y-T Yo et al

908 pFLAG

pFL 0.24% M1

128

Vector

R2 pN-neu

Control

Control Events

Events

Vector

37.57% M1

128

R2

Control 0 0 10

101

0 100

102 103 Empty

101

102 103 Empty

104

Rat p185neu expression (ICS)

Tumor-free mice (%)

100

pFLAG (n=8) Vector (n=8)

75

Saline (n=8)

50 25 0 0

4

6 9 12 15 18 21 Days following tumor challenge

1

3.2

pFL/pGM Vector Events

Saline

T cell growth index

pFLAG

2.8 2.4

24

pFLAG pFL/pGM Vector Saline

2.0 1.6 0.0

0 0 10

101

102 Empty

103

104

1:160 1:80 1:40 1:20 1:10 S:R ratio

MHC-II Figure 1 Characterization of pN-neu and cytokine expression vectors in vitro and in vivo. (a) Immunohistochemical detection of Flt3L expression in COS-7 cells transfected with pFLAG, pFL or control vector. The control panel was performed in which anti-Flt3L antibody was omitted in the staining of COS-7 cells transfected with pFLAG. (b) Flow-cytometric analysis of HER2/neu expression in COS-7 cells transfected with pN-neu. Transfected COS-7 cells were incubated with isotype IgG2a for background staining. The percentage of N-neu-expressing cells is shown. ICS, intracytoplasmic staining. (c) Delay in tumor formation in mice injected with pFLAG followed by tumor challenge. C3H/HeN mice were injected thrice at weekly intervals with pFLAG or control vector (50 mg), or saline. Seven days after the last immunization, the mice were challenged with MBT-2 cells (1.5  106). Data are shown as the percentage of tumor free mice following tumor challenge in each group (n ¼ 8) and are representative of two independent experiments. P ¼ 0.0249 for pFLAG vs control vector and P ¼ 0.0031 for pFLAG vs saline by log-rank test. (d) Detection of surface expression of MHC-II on CD11c þ splenocytes from mice injected with pFLAG or pFL/pGM. Mice immunized with pFLAG (50 mg), pFL (50 mg) plus pGM (50 mg), or control vector (50 mg) as above were killed 7 days after the last immunization, and their splenocytes were phenotypically and functionally analyzed. The percentages (mean7s.d.) of splenic CD11c þ cells that expressed MHC-II calculated from four independent experiments for pFLAG, pFL/pGM, vector and saline were 50.3273.81, 47.4973.67, 40.8375.22 and 41.0072.62%, respectively. P ¼ 0.02 for pFLAG vs vector and pFL/pGM vs saline; P ¼ 0.006 for pFLAG vs saline. Representative histogram from three mice of four independent experiments was shown. (e) Enhanced MLR stimulation by DC isolated from mice receiving pFLAG. Allogeneic responder T cells (105) from BALB/cJ mice were stimulated for 3 days with mitomycin C-treated DC from C3H/HeN mice receiving various cytokine plasmids at various S:R ratios. T-cell proliferation was determined by MES-based cell proliferation kit. The T-cell growth index was calculated as the ratio of the OD490 nm value of stimulated responder cells to that of non-stimulated responder cells. Data shown were the mean7s.d. (n ¼ 4 or 5), which were consistent in two independent experiments. Po0.001 for pFLAG vs pFL/pGM and for pFL/pGM or vector vs saline by two-way ANOVA. ANOVA, analysis of variance; DC, dendritic cells; Flt3L, Fms-like tyrosine kinase 3-ligand; MBT-2, mouse bladder tumor-2; MHC, major histocompatibility complex; MLR, mixed lymphocyte reaction.

Cancer Gene Therapy

Coexpression of Flt3L and GM-CSF for DNA vaccine Y-T Yo et al

909 pN-neu+pFLAG

pN-neu+pFLAG CD11c

pN-neu+pFL/pGM

MHC-II

pN-neu+Vector CD8

CD11c

DAPI

Merge

pN-neu+pFL/pGM

pN-neu+Vector CD11c

Saline

MHC-II

Saline CD8

CD11c

DAPI

Merge

Figure 2 Characterization of the infiltrated cells at the immunization site. Groups of three mice were immunized at days 0 and 14 with pN-neu (50 mg) in conjunction with pFLAG (50 mg), pFL (50 mg) plus pGM (50 mg), or control vector (50 mg), and killed at day 21, and the muscle around the injection site was snap-frozen. Serial cryostat sections were examined by H&E staining and immunofluorescence. (a) The degree of mononuclear cells infiltrating into the injection site was in the order of pFLAG4pFL/pGM4vector4saline, as detected by H&E stain (original magnification  400). Meanwhile, expressions of CD11c, MHC-II and CD8 were detected by immunofluorescence in the cells infiltrating into the injection site from mice receiving pN-neu plus pFLAG (b) and, to a lesser extent, from mice receiving pN-neu plus pFL/pGM (c). Note that CD11c þ cells were hardly detected at the injection site from mice receiving pN-neu plus control vector (d) or receiving saline (e). Nuclei were counterstained with DAPI. The merged column represents the superposition of the cells stained with CD11c, MHC-II, or CD8 and DAPI to visualize colocalization. DAPI, 4,6-diamidino-2-phenylindole; H&E, hematoxylin–eosin; MHC, major histocompatibility complex.

model. We then evaluated vaccine efficacy by combination of pN-neu with pFLAG or pFL/pGM. We first assessed if prophylactic vaccination induced significant protective immunity against tumor growth and prolonged survival. Mice were challenged with MBT-2 cells after being given two doses of pN-neu in combination with various genetic adjuvants at a 2-week interval. As shown in Figure 3a, tumor developed in all of the mice except those receiving pN-neu plus pFLAG. Tumor growth in mice vaccinated with pN-neu combined with pFLAG or pFL/pGM was significantly suppressed within 28 days after tumor challenge compared with that in the two control groups. Of note, only two from seven mice vaccinated with pN-neu plus pFLAG developed tumors, in which one animal achieved complete tumor regression and apparent cure of tumor, and the other had very small tumor burden. Remarkably, when pN-neu and pFLAG were administered in combination, 100% of the mice survived the 100-day experimental period (Figure 3b). In marked contrast, all of the unvaccinated mice receiving saline died within day 58. Although the survival time of the mice vaccinated with pN-neu combined with pFL/

pGM was also significantly increased compared with that of unvaccinated mice (P ¼ 0.0002), pN-neu DNA vaccine combined with pFL/pGM only conferred 28.6% (2/7) survival 100 days after tumor challenge (Figure 3b).

Therapeutic vaccination with pN-neu combined with pFLAG reduced tumor burden and prolonged survival in mice bearing established tumors To investigate the therapeutic effect of the vaccines against established tumors, which more closely mimicked real-life conditions, we inoculated MBT-2 cells into the mice; and after 8 days of tumor development, we vaccinated the tumor-bearing mice thrice at weekly intervals with pN-neu combined with various genetic adjuvants. Tumor progression appeared to be slower in the tumor-bearing mice vaccinated with pN-neu plus pFLAG than in the remaining three groups of mice, except for 1 mouse in the pN-neu plus pFL/pGMvaccinated group bearing very small tumor burden and subsequently achieving complete tumor regression (Figure 4a). Notably, mice vaccinated with pN-neu combined with pFLAG had a significantly better survival

Cancer Gene Therapy

Coexpression of Flt3L and GM-CSF for DNA vaccine Y-T Yo et al

910

Tumor volume (mm3)

250

pN-neu+pFLAG

2500

200

2000

150

1500

100

1000

50

500

pN-neu+pFL/pGM

0

0 0 2500

0

5 10 15 20 25 30 35 40 45 pN-neu+Vector

2500

2000

2000

1500

1500

1000

1000

500

500

0 0

5

5 10 15 20 25 30 35 40 45 Saline

0 0 5 10 15 20 25 30 35 40 45 10 15 20 25 30 35 40 45 Days following tumor challenge

100 pN-neu+pFLAG (n=7)

Survival (%)

80

pN-neu+pFL/pGM (n=7) pN-neu+Vector (n=6)

60

Saline (n=6)

40

20

0 0

10

20 30 40 50 60 70 80 Days following tumor challenge

90

100

Figure 3 Induction of antitumor immunity in mice vaccinated with pN-neu combined with pFLAG or pFL/pGM in the prophylactic MBT-2 model. C3H/HeN mice were immunized twice at a 2-week interval with pN-neu (100 mg) in conjunction with pFLAG (100 mg), pFL (100 mg) plus pGM (100 mg), or control vector (100 mg). Unvaccinated mice received saline only. Two weeks after the second immunization, the mice were inoculated with MBT-2 cells (1.5  106). (a) Tumor volumes in each group of mice are shown up to day 42 (n ¼ 6 or 7). Each line represents one individual tumor. Note that mice receiving pN-neu and pFLAG had the smallest tumor volume compared with the remaining three groups. (b) Kaplan–Meier survival curves at day 100 are shown. The recorded day of death or killing, when the primary tumor reached 2500 mm3, was used to calculate survival time. Data are representative of two independent experiments. P ¼ 0.0002 for pFLAG or pFL/pGM vs saline; P ¼ 0.0021 for pFLAG vs vector; P ¼ 0.0065 for pFLAG vs pFL/pGM by log-rank test. MBT-2, mouse bladder tumor-2.

when compared with those vaccinated with pN-neu plus control vector (P ¼ 0.0027) or with unvaccinated mice (P ¼ 0.0027) (Figure 4b).

CD8 þ T cells played a major role, whereas CD4 þ T cells played a partial role in the antitumor immunity induced by pN-Neu in conjunction with pFLAG Since pN-Neu DNA vaccine combined with pFLAG exerted protective antitumor immunity against HER-2/ neu-overexpressing MBT-2 tumors in the prophylactic setting, we applied this vaccination protocol to determine

Cancer Gene Therapy

the relative importance of T-cell subsets during the protective immunity by in vivo antibody depletion experiments. At 30 days following tumor challenge, none of the 13 mice vaccinated with pN-neu plus pFLAG developed tumors with volume exceeding 200 mm3. However, eight of nine vaccinated mice, when treated with anti-CD8 antibody, produced tumors exceeding 200 mm3. By comparison, only 3 of 11 vaccinated mice treated with anti-CD4 antibody had tumors larger than 200 mm3. All of the unvaccinated mice died of tumor growth within 76 days after tumor implantation. By

Coexpression of Flt3L and GM-CSF for DNA vaccine Y-T Yo et al

911

Tumor volume (mm3)

2500

pN-neu+pFLAG

2500

2000

2000

1500

1500

1000

1000

500

500

0

0 0 5 10 15 20 25 30 35 40 45 50

0 5 10 15 20 25 30 35 40 45 50 2500

pN-neu+Vector

2500

2000

2000

1500

1500

1000

1000

500

500

0

pN-neu+pFL/pGM

Saline

0 0 5 10 15 20 25 30 35 40 45 50 0 5 10 15 20 25 30 35 40 45 50 Days after tumor inoculation

100

pN-neu+pFLAG (n=5) pN-neu+pFL/pGM (n=4)

Survival (%)

80

pN-neu+Vector (n=4) Saline (n=4)

60

40

20

0 0

15

25

35 45 55 65 75 Days after tumor inoculation

85

95

Figure 4 Therapeutic effect of pN-neu combined with pFLAG against MBT-2 tumors. C3H/HeN mice were inoculated with MBT-2 cells (1.5  106) on day 0 followed by vaccination thrice at weekly intervals with pN-neu (100 mg) in conjunction with pFLAG (100 mg), pFL (100 mg) plus pGM (100 mg) or control vector (100 mg). Unvaccinated mice received saline only. (a) Tumor volumes in each group of mice are shown up to day 50 (n ¼ 4 or 5). Each line represents one individual tumor. (b) Kaplan–Meier survival curves at day 90 are shown. The recorded day of death or killing, when the primary tumor reached 2500 mm3, was used to calculate survival time. P ¼ 0.0027 for pFLAG vs vector or saline by log-rank test. MBT-2, mouse bladder tumor-2.

marked contrast, more than 90% of the animals receiving pN-neu plus pFLAG were tumor free (Figure 5a) and survived longer than 120 days (Figure 5b) after tumor challenge. Depletion of CD4 þ T cells in the vaccinated mice significantly reduced antitumor efficacy of the vaccines, as determined by tumor free status (Figure 5a) and survival (Figure 5b). Remarkably, depletion of CD8 þ T cells almost completely abrogated the antitumor efficacy of the vaccines. However, depletion of both CD4 þ and CD8 þ T cells did not further abolish the effect. Taken together, the antitumor immunity induced by prophylactic vaccination with pN-neu and pFLAG in the MBT-2 tumor model was dependent on both CD4 þ

and CD8 þ T cells, with CD8 þ T cells playing a dominant role and CD4 þ T cells acting a helper role.

Vaccination with pN-neu in conjunction with pFLAG enhanced CD8 þ T-cell-mediated immune responses The number of T-cell population in total splenocytes did not change significantly in the mice receiving pN-neu combined with pFLAG, pFL/pGM, control vector, or saline (data not shown). According to the T-cell depletion experiment shown in Figures 5a and b, CD8 þ T cells were crucial effectors for inducing antitumor immunity. To further confirm the HER2/neu-specific CD8 þ T-cell precursors generated by pN-neu DNA vaccine combined

Cancer Gene Therapy

Coexpression of Flt3L and GM-CSF for DNA vaccine Y-T Yo et al

912 Tumor-free mice (%)

100 Vaccine (n=13) Vaccine+anti-CD4 (n=11) Vaccine+anti-CD8 (n=9) Vaccine+anti-CD4/CD8 (n=9) Saline (n=12)

80 60 40 20 0 0

20

40

60

80

100

Days following tumor challenge 100

Survival (%)

80 60 40 20 0 0

20 40 60 80 100 Days following tumor challenge

120

Figure 5 CD8 þ T cells played a major role, whereas CD4 þ T cells played a partial role in the antitumor immunity induced by pN-neu combined with pFLAG. Mice were immunized with pN-neu (100 mg) in conjunction with pFLAG (100 mg) at days 28 and 14 and challenged with MBT-2 cell at day 0. Furthermore, the mice were injected with anti-mouse CD4 and/or anti-mouse CD8 monoclonal antibodies at weekly intervals from day 9 to day 40. (a) Data are shown as the percentage of tumor free mice following tumor challenge and are representative of two independent experiments. Po0.0001 for vaccine plus anti-CD4, vaccine plus anti-CD8, vaccine plus anti-CD4/CD8, or saline vs vaccine alone by log-rank test. (b) Kaplan–Meier survival curves at day 120 are shown. Data are representative of two independent experiments. Po0.0001 for vaccine plus anti-CD8 or saline alone vs vaccine alone, Po0.0005 for vaccine plus anti-CD4/CD8 vs vaccine alone, and Po0.01 for vaccine plus anti-CD4 vs vaccine alone by log-rank test. MBT-2, mouse bladder tumor-2.

with pFLAG or pFL/pGM in mice, we detected surface expression of CD25 and IFN-g by splenic CD8 þ T cells. CD25, which is not expressed by resting mature T cells, appears at the surface of activated cells. Splenic T cells were activated with a rat HER2/neu peptide containing the MHC class I epitope21 for 19 h and the expression of CD25 by CD8 þ T cells was determined with flow-cytometric analysis. Mice vaccinated with pN-neu plus pFLAG generated the highest number of HER2/neu-specific CD8 þ CD25 þ T-cell precursors, followed by those immunized with pN-neu plus pFL/ pGM (Figure 6a). By contrast, mice vaccinated with pNneu plus control vector did not generate higher precursors compared with those receiving saline. Moreover, we

Cancer Gene Therapy

conducted double staining for CD8 surface marker and intracytoplasmic IFN-g on splenocytes obtained from immunized mice, followed by flow-cytometric analysis. Mice vaccinated with pN-neu combined with pFLAG or pFL/pGM generated higher numbers of HER2/ neu-specific CD8 þ IFN-g þ T-cell precursors compared with those receiving either pN-neu plus control vector or saline (Figure 6b). Together, these results indicate that pN-neu combined with genetic adjuvants of Flt3L and GM-CSF enhanced CD8 þ T-cell-mediated immune responses. We further investigated the specific cytolytic activity of splenocytes obtained from mice receiving different DNA vaccine combinations. The splenocytes were co-cultured with MBT-2 cells pretreated with mitomycin C for 5 days, and the amount of IFN-g in the supernatant was measured by enzyme-linked immunosorbent assay. The splenocytes from mice vaccinated with pN-neu plus pFLAG produced the highest amount of IFN-g, followed by those vaccinated with pN-neu plus pFL/pGM, whereas those receiving either pN-neu plus control vector or saline secreted a negligible amount of IFN-g (Figure 7a). We also used lactate dehydrogenase release assay to further measure the cytolytic activity of splenocytes against MBT-2 cells. The cytolytic activity was in the order of pFLAG4pFL/pGM4vector4saline (Figure 7b). Previous reports also indicated that Flt3L protein can induce DC differentiation and NK-cell activation.27–29 We therefore isolated splenocytes and assessed their NK-cell activity against YAC-1 target cells.28 Mice vaccinated with pN-neu combined with pFLAG or with pFL/pGM induced higher levels of NK-cell activity than mice vaccinated with either pN-neu plus control vector or saline did (Figure 7c). Collectively, these results demonstrate that the immune response, especially the T helper (Th) 1 response, was induced in mice vaccinated with pN-neu combined with plasmid DNA encoding Flt3L and GM-CSF.

Discussion

Plasmid DNA encoding Flt3L or GM-CSF induces DC expansion and recruitment to the immunization site, which can elicit potent Ag-specific T-cell immune responses.9,10,30 Furthermore, co-administration of Flt3L- and GM-CSF-encoding plasmids by the intradermal route enhances DC recruitment to a cutaneous inoculation site in calves.30 The route of DNA vaccination is a critical parameter when using cytokine genetic adjuvants.31 In this study, we demonstrate that intramuscular administration of plasmid DNA encoding Flt3L and GM-CSF was capable of modulating the class of antitumor immune responses in mice receiving HER2/neu DNA vaccine. Numerous studies establish that both CD4 þ and CD8 þ T cells are required to achieve systemic antitumor immunity. While previous studies have found that protection mediated by HER2/neu DNA vaccine alone16,17 or combined with GM-CSF-encoded plasmid18,19

Coexpression of Flt3L and GM-CSF for DNA vaccine Y-T Yo et al

913 pN-neu+pFLAGp

pN-neu+pFL/pGM

104

104 382

239

3

3

10 Empty

Empty

10

102 101

102 101

100 100

101

102 Empty

103

100 100

104

101

pN-neu+vector

102 Empty

103

Saline

104

104 184

160 103 Empty

Empty

103 102 101

102 101

100 100

CD8 (surface)

104

101

102 Empty

103

100 100

104

101

102 Empty

103

104

CD25 (surface) pN-neu+pFLAGp

pN-neu+pFL/pGM

104

104 1.37 10

102 101

102 101

100 100

101

102

103

104

100 100

101

pN-neu+vector

103

Positive 104

0.79

R2

103 Empty

10

102

0.68

101

102

103

104

10

100 100

102 101

101

102

103

Empty

Empty

1.60

10

2

101

101

R2

3

Empty

R2

3

104

Saline 104

104

Empty

102 Empty

Empty

100 100

0.95

R2

3

Empty

Empty

10

R2

3

104

100 100

101

102

103

104

Empty

IFN- γ (ICS) þ

Figure 6 Quantitation of HER2/neu-specific CD8 T-cell precursors in C3H/HeN mice vaccinated with pN-neu combined with pFLAG or pFL/ pGM. Mice were vaccinated with various DNA vaccines as described in Figure 3. Fourteen days later, splenocytes of three mice from each group were pooled and cultured in vitro with the HER2/neu peptide encompassing amino acids 420–429 for 19 h. During the final 4 h, Golgistop was added. Cells were then harvested and stained for (a) surface CD8 and CD25 and (b) surface CD8 and intracytoplasmic IFN-g. The numbers of CD8 þ CD25 þ or CD8 þ IFN-g þ T-cell precursors in mice immunized with various DNA vaccines were quantified by flow cytometry. Data are expressed as the number of CD8 þ CD25 þ cells/1.5  105 splenocytes and the percentage of CD8 þ IFN-g þ cells in splenocytes. Data are representative of two independent experiments. ICS, intracytoplasmic cytokine staining; IFN-g, interferon-g.

is predominantly CD4 þ T-cell dependent, we show in the present study that the antitumor immunity induced by pNneu and pFLAG was dependent on both CD4 þ and CD8 þ T cells, with CD8 þ T cells playing a dominant role.

Flt3L and GM-CSF proteins have been documented to exert different effects on generation and differentiation of DC.32,33 It was also suggested that distinct DC subsets generated by Flt3L and GM-CSF may differentially

Cancer Gene Therapy

Coexpression of Flt3L and GM-CSF for DNA vaccine Y-T Yo et al

regulate the class of an immune response.33 The predominant form of Flt3L is synthesized as a transmembrane protein from which the soluble form is generated, presumably by proteolytic cleavage. The soluble form of Flt3L (extracellular domain) is functionally similar to Flt3L.34 In this study, we used the soluble form of murine Flt3L as the genetic adjuvant.9 It has been shown that both soluble and membrane-bound forms of Flt3L can

450



∗∗∗

∗∗∗

400

IFN-γ (pg/ml)

350 300 250 200 150 100 50 N.D.

0 + neu

-

pN

70

Specific lysis (%)

60 50

r M AG cto pG pFL +Ve FL/ u p e -n u+ pN -ne pN

ine

Sal

pNeu+pFLAG pNeu+pFL/pGM pNeu+Vector Saline

40 30 20 10 0 6.25

12.5

25

50

E:T ratio 90 75 Specific lysis (%)

914

60

pNeu+pFLAG pNeu+pFL/pGM pNeu+Vector Saline

45 30 15 0 6.25

12.5

25

E:T ratio

Cancer Gene Therapy

50

enhance antitumor immunity.35 The soluble form of Flt3L predominantly generates a lymphoid-type DC subset with pronounce effect on stimulating Th1 immune response,3,4 whereas GM-CSF only induces a myeloidtype DC subset.3,33 Thus, we suggest that combination of Flt3L in the HER2/neu DNA vaccine in our study may contribute greatly to the modulation of the class of immune responses, as Flt3L generated murine DC, which could prime naive CD8 þ T cells and generated memory cells in vivo.6 In the present study, cells infiltrating into the injection site in mice vaccinated with pN-neu combined with pFLAG appeared to consist of both CD8 þ lymphoid and CD8 myeloid DC subsets. Moreover, pFLAG may be more potent than pFL plus pGM in inducing lymphoid-type DC subset. Recently, it was reported that combination treatment of mice bearing established tumors with FltL3 and GM-CSF proteins resulted in the recruitment of large numbers of tumorinfiltrating DC and induction of a CD8 þ T-cell-mediated immune response.36 Mouse splenic DC could be expanded by intramuscular delivery of either Flt3L or GM-CSF cDNA.11 However, a more potent T-cell proliferation was observed in the mixed lymphocyte reaction when stimulator DC originated from mice treated with Flt3L cDNA, compared

Figure 7 Induction of cell-mediated immunity in mice vaccinated with pN-neu combined with pFLAG or pFL/pGM. Mice were vaccinated with various DNA vaccines as described in Figure 3. Fourteen days later, pooled spleens from three mice were aseptically removed and splenocytes prepared. ND, not detectable. (a) Higher IFN-g production in mice vaccinated with pN-neu plus either pFLAG or pFL/pGM. Splenocytes were re-stimulated in vitro with mitomycin C-treated MBT-2 cells for 5 days and the supernatant was assessed for IFN-g content by enzyme-linked immunosorbent assay. Data shown were the mean7s.d. (n ¼ 3), which were consistent in two independent experiments. *Po0.05; ***Po0.001 by Student’s t-test. (b) Enhanced cytolytic activity of splenocytes in mice vaccinated with pN-neu plus either pFLAG or pFL/pGM. Splenocytes were restimulated in vitro with 20 U ml1 of human recombinant IL-2 and mitomycin C-treated MBT-2 cells for 5 days. Splenocytes were then harvested, incubated with MBT-2 target cells for 6 h, and determined by their cytolytic activity. For pFLAG vs vector, Po0.0001 at an E/T ratio of 50, Po0.001 at E/T ratios of 25 and 12.5 and Po0.01 at an E/T ratio of 6.25; for pFLAG vs saline, Po0.0001 at all the E/T ratios tested; for pFL/pGM vs vector, Po0.01 at E/T ratios of 50 and 25; for pFL/pGM vs saline, Po0.0001 at E/T ratios of 50, 25 and 12.5, Po0.001 at an E/T ratio of 6.25 by Student’s t-test. (c) Enhanced NK-cell activity in mice vaccinated with pN-neu plus either pFLAG or pFL/pGM. Freshly isolated splenocytes were incubated with YAC-1 target cells for 6 h followed by assessment of cytolytic activity. For pFLAG vs vector or saline, Po0.05 at an E/T ratio of 50, Po0.0001 at E/T ratios of 25, 12.5 and 6.25; for pFL/pGM vs vector, Po0.0001 at E/T ratios of 25, 12.5 and 6.25; for pFL/pGM vs saline, Po0.0001 at E/T ratios of 25 and 12.5, Po0.01 at an E/T ratio of 6.25 by Student’s t-test. In (b and c), cytolytic activity was determined with the LDH assay, and E/T ratio denotes effector cell to target cell ratio. Data shown were the mean7s.d. (n ¼ 3), which were consistent in two independent experiments. ELISA, enzyme-linked immunosorbent assay; IFN-g, interferon-g; LDH, lactate dehydrogenase; MBT-2, mouse bladder tumor-2; NK, natural killer.

Coexpression of Flt3L and GM-CSF for DNA vaccine Y-T Yo et al

with those derived from mice receiving both Flt3L and GM-CSF plasmids.11 This suggested that lymphoidderived DC expanded favorably by Flt3L were more efficient in stimulating the proliferation of allogeneic T cells.11 In the present study, we observed DC originated from mice receiving pFLAG stimulated higher allogeneic T-cell proliferation than those isolated from mice receiving pFL/pGM did. Giving that combination of two cytokines may have an additive or even synergistic adjuvant effects, previous studies have used plasmids coexpressing two cytokines for vaccine and gene therapy applications. A retroviral vector expressing the GM-CSF and IL-2 fusion protein could induce greater antitumor effect than both cytokine vectors in combination, when used to generate cytokine-modified tumor vaccines.37 A plasmid vector coexpressing GM-CSF and IL-12 driven by different promoters within the same plasmid was reported to function as genetic adjuvants for influenza DNA vaccine.38 Furthermore, a bicistronic retroviral vector coexpressing IFN-a and IFN-g was used to transduce chronic myeloid leukemia cells.39 However, in these studies,38,39 simultaneous coexpression of the two cytokines were not compared with co-delivery of two separate plasmids expressing one cytokine for their immunostimulatory efficacy. Although we cannot exclude the possibility that the expressions of Flt3L and GM-CSF from pFLAG may be higher than those from pFL and pGM in vivo, our results suggest that the bicistronic vector may be superior to the combination of two monocistronic vectors in serving as the genetic adjuvant for HER2/neu DNA vaccine. Regarding the use of genetic adjuvants, superiority in immunogenicity of bicistronic constructs coexpressing cytokine and Ag rather than two coadministered plasmids has been demonstrated.12,40 Hence, it is conceivable that precise temporal and spatial codelivery of FltL3 and GM-CSF may be beneficial for optimal induction of antitumor immunity by HER2/neu DNA vaccine. In the work described here, although pFLAG alone neither prevented tumor formation nor enhanced survival in the prophylactic MBT-2 model, tumor growth was decreased and tumor progression was delayed in mice vaccinated with pFLAG alone compared with their control counterparts. This result suggests that pFLAG exerted some non-Ag-specific antitumor effect against tumors. One of the possible protective mechanisms of pFLAG vaccination may be attributed to the enhancement of NK-cell activity. Furthermore, our results also demonstrate that splenocytes from mice vaccinated with pN-neu plus pFLAG or pFL/ pGM had more potent cytolytic function, which included NK-cell activity. Because Flt3L can expand and activate murine splenic NKDC, a unique subset of cells possessing the characteristics of both NK and DC, and Flt3L-expanded NKDC have potent cytolytic function and increased T-cell stimulatory capacity,41 we cannot exclude the possibility that NKDC or other effector cells may participate in the antitumor immune responses. To this end, DC engineered to express Flt3L were demonstrated to enhance tumor-specific

cytotoxic T- and NK-cell responses and induce antitumor immunity.42 In the present study, we observed that coexpression of Flt3L and GM-CSF encoded by the bicistronic pFLAG plasmid through the intramuscular administration enhanced DC infiltration at the injection site. Whether the distribution of infiltrating DC at the injection site of mice receiving pFLAG affected the protection efficacy of the cytokine genetic adjuvants; however, remains unclear and warrants further investigation. Consistent with previous findings,13,25 we did not observe T-cell infiltration at the injection site in any group of the treated mice. In conclusion, coexpression of Flt3L and GM-CSF by a bicistronic plasmid with HER2/neu DNA vaccine enhanced antitumor immune responses against murine bladder cancer naturally overexpressing HER2/neu. Infiltration of DC at the immunization site and enhancement of T- and NK-cell activities were prominent in mice concomitantly administered with HER2/neu DNA vaccine and pFLAG genetic adjuvant. The strategy of using Flt3L and GM-CSF coexpressed by a bicistronic vector can be further exploited in various immunotherapeutic studies. Acknowledgements

We are indebted to TC Wu (Department of Pathology, Johns Hopkins University), LH Hwang (Hepatitis Research Center, National Taiwan University Hospital, Taiwan), and MD Lai (Department of Biochemistry and Molecular Biology, National Cheng Kung University, Taiwan) for providing pFL, pGEM-4/GM-CSF and pRc/CMV-N0 -neu plasmids, respectively. This work was supported by Grants NSC 89-2318-B-006-014-M51 and 90-2318-B-006-006-M51 to A-L Shiau and NSC 89-2318B-006-013-M51 and 90-2318-B-006-003-M51 to C-L Wu from National Science Council, Taiwan.

References 1 Oelke M, Moehrle U, Chen JL, Behringer D, Cerundolo V, Lindemann A et al. Generation and purification of CD8+ melan-A-specific cytotoxic T lymphocytes for adoptive transfer in tumor immunotherapy. Clin Cancer Res 2000; 6: 1997–2005. 2 Randolph GJ, Angeli V, Swartz MA. Dendritic-cell trafficking to lymph nodes through lymphatic vessels. Nat Rev Immunol 2005; 5: 617–628. 3 Daro E, Butz E, Smith J, Teepe M, Maliszewski CR, McKenna HJ. Comparison of the functional properties of murine dendritic cells generated in vivo with Flt3 ligand, GM-CSF and Flt3 ligand plus GM-SCF. Cytokine 2002; 17: 119–130. 4 Maraskovsky E, Brasel K, Teepe M, Roux ER, Lyman SD, Shortman K et al. Dramatic increase in the numbers of functionally mature dendritic cells in Flt3 ligand-treated mice: multiple dendritic cell subpopulations identified. J Exp Med 1996; 184: 1953–1962. 5 Kataoka K, McGhee JR, Kobayashi R, Fujihashi K, Shizukuishi S, Fujihashi K. Nasal Flt3 ligand cDNA elicits

Cancer Gene Therapy

915

Coexpression of Flt3L and GM-CSF for DNA vaccine Y-T Yo et al

916 6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

CD11c+CD8+ dendritic cells for enhanced mucosal immunity. J Immunol 2004; 172: 3612–3619. Angelov GS, Tomkowiak M, Marcais A, Leverrier Y, Marvel J. Flt3 ligand-generated murine plasmacytoid and conventional dendritic cells differ in their capacity to prime naive CD8T cells and to generate memory cells in vivo. J Immunol 2005; 175: 189–195. Miller G, Pillarisetty VG, Shah AB, Lahrs S, Xing Z, DeMatteo RP. Endogenous granulocyte–macrophage colonystimulating factor overexpression in vivo results in the longterm recruitment of a distinct dendritic cell population with enhanced immunostimulatory function. J Immunol 2002; 169: 2875–2885. Bernt KM, Ni S, Tieu AT, Lieber A. Assessment of a combined, adenovirus-mediated oncolytic and immunostimulatory tumor therapy. Cancer Res 2005; 65: 4343–4352. Hung CF, Hsu KF, Cheng WF, Chai CY, He L, Ling M et al. Enhancement of DNA vaccine potency by linkage of antigen gene to a gene encoding the extracellular domain of Fms-like tyrosine kinase 3-ligand. Cancer Res 2001; 61: 1080–1088. Fong CL, Hui KM. Generation of potent and specific cellular immune responses via in vivo stimulation of dendritic cells by pNGVL3-hFLex plasmid DNA and immunogenic peptides. Gene Therapy 2002; 9: 1127–1138. Peretz Y, Zhou ZF, Halwani F, Prud’homme GJ. In vivo generation of dendritic cells by intramuscular codelivery of FLT3 ligand and GM-CSF plasmids. Mol Ther 2002; 6: 407–414. Chang SY, Lee KC, Ko SY, Ko HJ, Kang CY. Enhanced efficacy of DNA vaccination against Her-2/neu tumor antigen by genetic adjuvants. Int J Cancer 2004; 111: 86–95. Fong CL, Mok CL, Hui KM. Intramuscular immunization with plasmid coexpressing tumour antigen and Flt-3L results in potent tumour regression. Gene Therapy 2006; 13: 245–256. Schechter AL, Stern DF, Vaidyanathan L, Decker SJ, Drebin JA, Greene MI et al. The neu oncogene: an erb-Brelated gene encoding a 185 000-Mr tumour antigen. Nature 1984; 312: 513–516. Bargmann CI, Hung MC, Weinberg RA. The neu oncogene encodes an epidermal growth factor receptor-related protein. Nature 1986; 319: 226–230. Foy TM, Bannink J, Sutherland RA, McNeill PD, Moulton GG, Smith J et al. Vaccination with Her-2/neu DNA or protein subunits protects against growth of a Her-2/neuexpressing murine tumor. Vaccine 2001; 19: 2598–2606. Curcio C, Di Carlo E, Clynes R, Smyth MJ, Boggio K, Quaglino E et al. Nonredundant roles of antibody, cytokines, and perforin in the eradication of established Her-2/neu carcinomas. J Clin Invest 2003; 111: 1161–1170. Pilon SA, Kelly C, Wei WZ. Broadening of epitope recognition during immune rejection of ErbB-2-positive tumor prevents growth of ErbB-2-negative tumor. J Immunol 2003; 170: 1202–1208. Lindencrona JA, Preiss S, Kammertoens T, Schuler T, Piechocki M, Wei WZ et al. CD4+ T cell-mediated HER-2/ neu-specific tumor rejection in the absence of B cells. Int J Cancer 2004; 109: 259–264. Shiau AL, Chen CC, Yo YT, Chu CY, Wang SY, Wu CL. Enhancement of humoral and cellular immune responses by an oral Salmonella choleraesuis vaccine expressing porcine prothymosin alpha. Vaccine 2005; 23: 5563–5571. Ercolini AM, Machiels JP, Chen YC, Slansky JE, Giedlen M, Reilly RT et al. Identification and characterization of the

Cancer Gene Therapy

22

23

24

25

26

27

28

29

30

31

32

33

34

35

immunodominant rat HER-2/neu MHC class I epitope presented by spontaneous mammary tumors from HER-2/ neu-transgenic mice. J Immunol 2003; 170: 4273–4280. Hsieh CL, Pang VF, Chen DS, Hwang LH. Regression of established mouse leukemia by GM-CSF-transduced tumor vaccine: implications for cytotoxic T lymphocyte responses and tumor burdens. Hum Gene Ther 1997; 8: 1843–1854. Lin CC, Chou CW, Shiau AL, Tu CF, Ko TM, Chen YL et al. Therapeutic HER2/Neu DNA vaccine inhibits mouse tumor naturally overexpressing endogenous neu. Mol Ther 2004; 10: 290–301. Lee CH, Wu CL, Shiau AL. Systemic administration of attenuated Salmonella choleraesuis carrying thrombospondin-1 gene leads to tumor-specific transgene expression, delayed tumor growth and prolonged survival in the murine melanoma model. Cancer Gene Ther 2005; 12: 175–184. Sumida SM, McKay PF, Truitt DM, Kishko MG, Arthur JC, Seaman MS et al. Recruitment and expansion of dendritic cells in vivo potentiate the immunogenicity of plasmid DNA vaccines. J Clin Invest 2004; 114: 1334–1342. Corr M, von Damm A, Lee DJ, Tighe H. In vivo priming by DNA injection occurs predominantly by antigen transfer. J Immunol 1999; 163: 4721–4727. Shaw SG, Maung AA, Steptoe RJ, Thomson AW, Vujanovic NL. Expansion of functional NK cells in multiple tissue compartments of mice treated with Flt3-ligand: implications for anti-cancer and anti-viral therapy. J Immunol 1998; 161: 2817–2824. Fernandez NC, Lozier A, Flament C, Ricciardi-Castagnoli P, Bellet D, Suter M et al. Dendritic cells directly trigger NK cell functions: cross-talk relevant in innate anti-tumor immune responses in vivo. Nat Med 1999; 5: 405–411. Tourkova IL, Yamabe K, Chatta G, Shurin GV, Shurin MR. NK cells mediate Flt3 ligand-induced protection of dendritic cell precursors in vivo from the inhibition by prostate carcinoma in the murine bone marrow metastasis model. J Immunother 2003; 26: 468–472. Mwangi W, Brown WC, Lewin HA, Howard CJ, Hope JC, Baszler TV et al. DNA-encoded fetal liver tyrosine kinase 3 ligand and granulocyte macrophage-colony-stimulating factor increase dendritic cell recruitment to the inoculation site and enhance antigen-specific CD4+ T cell responses induced by DNA vaccination of outbred animals. J Immunol 2002; 169: 3837–3846. Kwissa M, Kroger A, Hauser H, Reimann J, Schirmbeck R. Cytokine-facilitated priming of CD8+ T cell responses by DNA vaccination. J Mol Med 2003; 81: 91–101. Weigel BJ, Nath N, Taylor PA, Panoskaltsis-Mortari A, Chen W, Krieg AM et al. Comparative analysis of murine marrow-derived dendritic cells generated by Flt3L or GMCSF/IL-4 and matured with immune stimulatory agents on the in vivo induction of antileukemia responses. Blood 2002; 100: 4169–4176. Pulendran B, Smith JL, Caspary G, Brasel K, Pettit D, Maraskovsky E et al. Distinct dendritic cell subsets differentially regulate the class of immune response in vivo. Proc Natl Acad Sci USA 1999; 96: 1036–1041. Lyman SD, James L, Vanden Bos T, de Vries P, Brasel K, Gliniak B et al. Molecular cloning of a ligand for the flt3/flk-2 tyrosine kinase receptor: a proliferative factor for primitive hematopoietic cells. Cell 1993; 75: 1157–1167. Alsheikhly AR, Zweiri J, Walmesley AJ, Watson AJ, Christmas SE. Both soluble and membrane-bound forms of

Coexpression of Flt3L and GM-CSF for DNA vaccine Y-T Yo et al

Flt3 ligand enhance tumor immunity following ‘suicide’ gene therapy in a murine colon carcinoma model. Cancer Immunol Immunother 2004; 53: 946–954. 36 Berhanu A, Huang J, Alber SM, Watkins SC, Storkus WJ. Combinational FLt3 ligand and granulocyte macrophage colony-stimulating factor treatment promotes enhanced tumor infiltration by dendritic cells and antitumor CD8(+) T-cell cross-priming but is ineffective as a therapy. Cancer Res 2006; 66: 4895–4903. 37 Stagg J, Wu JH, Bouganim N, Galipeau J. Granulocyte– macrophage colony-stimulating factor and interleukin-2 fusion cDNA for cancer gene immunotherapy. Cancer Res 2004; 64: 8795–8799. 38 Iwasaki A, Stiernholm BJ, Chan AK, Berinstein NL, Barber BH. Enhanced CTL responses mediated by plasmid DNA immunogens encoding costimulatory molecules and cytokines. J Immunol 1997; 158: 4591–4601.

39 Salesse S, Lagarde V, Ged C, de Verneuil H, Reiffers J, Mahon FX. Retroviral coexpression of IFN-alpha and IFN-gamma genes and inhibitory effects in chronic myeloid leukemia cells. J Interferon Cytokine Res 2000; 20: 577–587. 40 Lee SW, Cho JH, Sung YC. Optimal induction of hepatitis C virus envelope-specific immunity by bicistronic plasmid DNA inoculation with the granulocyte–macrophage colony-stimulating factor gene. J Virol 1998; 72: 8430–8436. 41 Chaudhry UI, Katz SC, Kingham TP, Pillarisetty VG, Raab JR, Shah AB et al. In vivo overexpression of Flt3 ligand expands and activates murine spleen natural killer dendritic cells. FASEB J 2006; 20: 982–984. 42 Liu Y, Huang H, Chen Z, Zong L, Xiang J. Dendritic cells engineered to express the Flt3 ligand stimulate type I immune response, and induce enhanced cytoxic T and natural killer cell cytotoxicities and antitumor immunity. J Gene Med 2003; 5: 668–680.

Cancer Gene Therapy

917