Shikonin enhances efficacy of a gene-based cancer vaccine via induction of RANTES
© Chen et al; licensee BioMed Central Ltd. 2012
Received: 9 January 2012
Accepted: 12 April 2012
Published: 12 April 2012
Skip to main content
© Chen et al; licensee BioMed Central Ltd. 2012
Received: 9 January 2012
Accepted: 12 April 2012
Published: 12 April 2012
Shikonin, a phytochemical purified from Lithospermum erythrorhizon, has been shown to confer diverse pharmacological activities, including accelerating granuloma formation, wound healing, anti-inflammation and others, and is explored for immune-modifier activities for vaccination in this study. Transdermal gene-based vaccine is an attractive approach for delivery of DNA transgenes encoding specific tumor antigens to host skin tissues. Skin dendritic cells (DCs), a potent antigen-presenting cell type, is known to play a critical role in transmitting and orchestrating tumor antigen-specific immunities against cancers. The present study hence employs these various components for experimentation.
The mRNA and protein expression of RANTES were detected by RT-PCR and ELISA, respectively. The regional expression of RANTES and tissue damage in test skin were evaluated via immunohistochemistry assay. Fluorescein isothiocyanate sensitization assay was performed to trace the trafficking of DCs from the skin vaccination site to draining lymph nodes. Adjuvantic effect of shikonin on gene gun-delivered human gp100 (hgp100) DNA cancer vaccine was studied in a human gp100-transfected B16 (B16/hgp100) tumor model.
Among various phytochemicals tested, shikonin induced the highest level of expression of RANTES in normal skin tissues. In comparison, mouse RANTES cDNA gene transfection induced a higher level of mRANTES expression for a longer period, but caused more extensive skin damage. Topical application of shikonin onto the immunization site before gene gun-mediated vaccination augmented the population of skin DCs migrating into the draining lymph nodes. A hgp100 cDNA gene vaccination regimen with shikonin pretreatment as an adjuvant in a B16/hgp100 tumor model increased cytotoxic T lymphocyte activities in splenocytes and lymph node cells on target tumor cells.
Together, our findings suggest that shikonin can effectively enhance anti-tumor potency of a gene-based cancer vaccine via the induction of RANTES expression at the skin immunization site.
Shikonin, a phytochemical from the traditional herbal medicine Lithospermum erythrorhizon, has been shown to possess diverse pharmacological properties, including anti-oxidant , wound healing [2–4], anti-inflammatory  and anti-tumor properties [6, 7]. Anti-inflammatory activities of shikonin can be mediated by suppression of TNF-α promoter activity  or by regulation of TNF-α pre-mRNA splicing . Wound-healing activity of shikonin resulted in proliferation of fibroblasts and collagen fiber levels in granuloma tissues, and an increase in the CD11b+ cell population in granulation tissues [3, 4]. These findings suggest that shikonin may confer cellular activities that can induce specific chemokines and subsequent chemotactic activities in specific immune-responsive cell types.
Transdermal or dermal delivery as a strategy of administration for specific vaccines, therapeutic agents and experimental medicines, is an attractive approach due to the presence and unique features of key immunocompetent cells in skin tissues, such as keratinocytes and dendritic cells (DCs), which include both dermal and epidermal DCs. DCs are potent professional antigen presenting cells that can act as sentinels that constantly monitor the surrounding milieu. They can be activated by a variety of cellular or microenvironmental signals produced by invading pathogens or neighboring cells [10, 11]. Activated skin DCs can migrate from the epidermis or dermis to the draining lymph nodes where antigen presentation takes place . On the basis of the specific characteristics of DCs, advanced vaccine strategies have been designed to target antigens onto DCs, promoting cross-priming and improving peptide-binding to both MHC class I and class II molecules . Evidence has accumulated showing that DCs can play a critical role in eliciting immunity from DNA-based vaccination [14, 15].
Chemotactic-tumor associated antigen vaccines have been reported to target specific antigens to special immune cell types, and thereby elicit improved immune responses . Previously it was suggested that the majority of cells transfected by gene gun-mediated vaccination (in skin) were non-antigen presenting cells ; however, a relatively small but significant number of transgene-expressing DCs were detected in the draining lymph nodes . Timares and his colleagues  also suggested that directly-transfected DCs can incite immune response in a more efficient way than indirectly-transfected DCs. As compared to conventional vaccines, the success of skin vaccination via gene gun delivery has been challenged by its capacity to elicit only limited immunity . In a recent study, we reported that co-vaccination with a chemokine, mouse RANTES (mRANTES) and a tumor-specific antigen gene, human gp100 (hgp100) within a specific time frame can induce a strong cell-mediated immunity against targeted B16/hgp100-tumor cells . Gene gun-mediated bombardment with RANTES cDNA transgene significantly increased CD11c+, NK1.1+, CD4+ and CD8+ immune cells in test mouse skin. RANTES, which is regulated upon activation of normal T-cell expression and secretion, is a versatile chemokine which can not only recruit leukocytes into the inflamed tissue sites, but can also act as a powerful leukocyte activator . High level induction of RANTES in keratinocytes has been reported for cutaneous wound-repairing .
Here, we evaluated the effects of topical application of shikonin to skin to aid trafficking of skin DCs and thus augment the effect of a DNA-based cancer vaccine. We investigated the effects of shikonin on the induction of RANTES in treated skin tissues that subsequently served as the immunization site for hgp100 gene-based cancer vaccines. Our findings suggest that shikonin can effectively enhance the efficacy of the tumor associated antigen-hgp100 DNA vaccine, and its effect is likely mediated via the induction of mRANTES and an increase in the population of skin DCs at the vaccinated site draining into lymph nodes. Possible implication and application of phytochemical shikonin for generic and topical use as an adjuvant in skin-based vaccinations are discussed.
Male C57BL/6Jnarl mice (6-8-weeks old) were purchased from the National Laboratory Animal Breeding and Research Center, Taipei, Taiwan. All mice were maintained in a laminar airflow cabinet in a room kept at 24 ± 2°C and 40-70% humidity with a 12 h light/dark cycle under specific pathogen-free conditions. All facilities were approved by the Academia Sinica Institutional Animal Care and Utilization Committee, and all animal experiments were conducted under the institutional guidelines established for the Animal Core Facility at Academia Sinica, Taipei.
Shikonin was purchased from Tokyo Chemical Industry. Gallic acid, ferulic acid, curcumin, caffeic acid and epigallocatechin-3-gallate, phorbol ester 12-O-tetradecanoylphorbol-13-acetate were purchased from Sigma.
The B16F10 melanoma cell line, syngeneic in C57BL/6 mice, was purchased from American Type Culture Collection (ATCC). Human gp100-transfected B16 cell line were obtained by stably transfecting B16 cells using the PowderJect gene delivery system (PowderJect Vaccines) as described previously [18, 23]. B16/hgp100 cells were maintained in DMEM supplemented with 10% fetal bovine serum, 2 mM L-glutamine, and 100 μg/ml streptomycin and penicillin.
Human gp100 (hgp100) expression plasmid pWRG1644 was a kind gift from Dr Nicholas Restifo, National Cancer Institute, (Bethesda, MD USA). pWRG1644 plasmid without containing the coding sequence of hgp100 was used as a control . The pORF9-mRANTES mammalian expression vector was purchased from Invivogen (San Diego, CA, USA).
Total RNA was extracted as per the manufacturer's instructions of the TRIzol reagent (Invitrogen). RT-PCR reactions were conducted with the AccessQuick RT-PCR system (Promega, Madison, WI), according to the manufacturer's instructions, at cycle numbers that yielded products within the linear range as determined for each specific pair of primers. Specific primer pairs for mouse RANTES were 5'-ATATGGCTCGGACACCACTC and 5'-CCTTCGAGTGACAAACACGA.
The assay method was modified from a previous study . FITC was dissolved in acetone/dibutylphthalate (4:1) before application. Mice were painted the abdomen skin with 50 μl of 1% FITC solution. After 30 mins, shikonin (0, 50 or 100 μg/3.8 cm2/site) or transgenic bombardments were applied onto the test site. After FITC painting for 24 hours, axillary, brachial and inguinal lymph nodes were obtained, single-cell suspensions were prepared, purified by anti-CD11c microbeads, and then stained with PE-anti-MHC class II antibody. Flow cytometry analysis by test cells was performed on a FACScan.
Protocols for creating skin inflammation in mice were approved by the Institute Ethical Committee and IACUC. For skin inflammation test, female C57BL/6JNarl mice received a topical application on ~3.8 cm2 area/site of the shaved abdomen with 10 nmol TPA. In selected groups, 50 μl of the vehicle control or solutions of test shikonin concentration was applied on test skin 30 min before challenge. Total RNA was extracted from test skin tissues after 24 hours; chemokine expression was analyzed in subsequent time points by RT-PCR. In each experimental group, some of the mice were sacrificed and the test skin samples were excised for snap frozen for immunohistochemistry or histological analysis.
Skin tissue samples were collected and fixed in 10% PBS buffered formalin and embedded in paraffin wax. Specimens were cut into 5 μm-thick sections and mounted on slides for the immunohistochemistry by a standard indirect streptavidin-biotin immunoperoxidase staining method. Specimens were treated with primary rat anti-mouse mRANTES, and subsequent secondary biotin-conjugated anti-rat IgG and tertiary peroxidase-conjugated streptavidin (Serotec) [6, 27]. Primary antibodies were purchased from PharMingen (San Diego, CA, USA) or Serotec. Apoptotic cells were detected with an in situ Cell Death Kit (Roche Diagnostics). All test skin tissue sections were counterstained with hematoxylin.
B16/hgp100 tumor cells were injected subcutaneously (1 × 105 cells/100 μl/mouse) into the right flank of mice. On day 7 post-tumor cell inoculation, test mice were vaccinated with hgp100 cDNA vector (3 μg each/mice). mRANTES cDNA vector (3 μg each/mice) or shikonin (100 μg/site/mice) was transfected or topically applied to equal area of abdominal skin, 24 hours before the hgp100 cDNA vaccination. One primer and two boosters were administered on day 0, 7 and 14. One week after the last booster, splenocytes and lymph node cells were harvested from immunized mice and assayed for CTL activity assayed according to the manufacturer's protocol (DELFIA, Perkin Elmer, MA, USA). The percentage of specific lysis was calculated as (experimental release (counts)-spontaneous release (counts))/(maximum release (counts) - spontaneous release (counts)) × 100.
ANOVA or the T-test statistical method was employed to analyze the data from different groups or treatments of test herbs or phytochemicals.
Topical and transdermal delivery of pharmacological agents including vaccines is receiving increased research attention due to recent expansion of the concept of skin to a fascinating immune system. Our previous study showed that a combination of transgenic expression of RANTES and hgp100 as a chemotactic cancer vaccine can elicit a much more efficient immune response than hgp100 cDNA alone . In this study, we first screened several candidate phytochemicals for their possible induction of mRANTES expression in test skin tissue and found that shikonin was able to induce the highest level of mRANTES expression in mouse skin (Figure 1). Shikonin has been shown to exhibit multiple and sometimes seemingly conflicting bio-activities, e.g., anti-tumor vs. wound-healing effects [7, 30] and anti-inflammation vs. induction of pro-inflammatory chemokines . A number of studies have also reported that shikonin can exhibit anti-tumor effects via inhibition of cell proliferation  and proteasome activity , and by circumventing cancer drug resistance via induction of necroptosis . A herbal medicine formulation made from Lithospermum erythrorhizon containing shikonin has been shown to enhance granuloma formation , and help reconstitution of skin tissue  through induction of neovascularization in granulomatous tissue during wound-healing . It is well-known that shikonin confers a potent anti-inflammatory effect, and we have previously shown that this effect on skin may result from inhibition of the promoter activation  and mRNA splicing  activities for TNFα expression. As a follow up, we showed that shikonin can interfere with or inhibit the maturation of DCs by suppressing certain proinflammatory cytokines. Shikonin treatment substantially affected transgenic and/or endogenous expression of TNF-α, GM-CSF and other cytokine genes in mouse skin tissues in vivo and in DCs or monocytes ex-vivo. In the present study we have shown that under normal physiological conditions both shikonin and curcumin, which have both been claimed to confer various anti-inflammatory activities, were able to induce expression of mRANTES which is generally considered to be a pro-inflammatory chemokine. Secretion of RANTES by keratinocytes is known to play a role in promoting cutaneous wound-healing mediated via interaction with the TWEAK/Fn14 signaling network [22, 35]. This in situ RANTES expression pattern at the histological level (Figure 4a) was distinguishable from the results obtained from ELISA assays at the total skin tissues extract level (Figure 3c). Increasing evidences are showing that the immune systems are highly sophisticated and can no longer been dichotomized simply into "pro-inflammation" versus "anti-inflammation" activities. There are overlapping and parallel factors involved in the regulation of both immune responses, as is known for the "multiple facets" of IL-6 . As for the different effects of shikonin under different physiological/pathological microenvironment, together with our own and previous findings from others, we can speculate some possibilities for elucidating these observations. For example, after TPA treatments, several responsive genes can be highly activated, including NFκB, RANTES, MIP-1β and TNFα. Shikonin has been shown and considered as an anti-inflammatory phytochemical, as we have reported previously [8, 9]. Therefore, the inhibitory effect of shikonin on TPA-induced RANTES and MIP-1β gene expression may result into such anti-inflammatory activity. On the other hand, in addition to inflammatory signals, the expression of RANTES may be regulated by other signals or factors, such as apoptotic signaling, FasL  and calcineurin . The present study indicates that a high dose of shikonin may act as a strong stimulus to the epidermis, which is made up largely of keratinocytes, and induce a high level of RANTES expression, which may partially result from transiently induced apoptotic activities in the epidermis. In comparison with changes seen after gene gun-bombardment of skin, only minor changes in the levels of cell apoptosis in the epidermis and skin damage were seen after shikonin treatment. Although a number of skin vaccination approaches have been explored to boost systemic immunity, new and optimized tissue manipulation techniques for topical vaccination are still needed to address issues such as reducing adjuvant toxicity and disruption of epidermal structure. Within this paradigm, the localized microenvironmental stress induced by shikonin can be considered to be minor. In addition, shikonin treatment did not induce a fast turnover of skin tissues. Overall, the evidence presented here suggests that shikonin may play a "multi-faceted" role in regulating certain activities that induce a diversified repertoire of chemokines expressed in skin, and some of these activities may correlate with the recruitment of DCs.
Several modalities of topical immunotherapy have been developed . A variety of adjuvants with potent immuno-enhancing activities, such as toll-like receptor ligands , have been employed for use in skin vaccination [40, 41]. RANTES/CCL5 is characterized as a proinflammatory chemokine and can attract CCR5-expressing immune cells, including immature DC and T cells. In our previous study, although we showed that the CD11c+ DCs were significantly increased in mRANTES-transfected skin, we were unable to exclude the possibility that the CD11c+ DCs which were increased in population as a response to treatment with RANTES were immature. The results of this study (Figure 4) and a previous report  suggest that a gene gun-mediated DNA vaccine encoding specific tumor-associated transgenes may provide a strong stimulus, due to physical stress and/or a DNA backbone signal which activates skin DCs, that enables DCs fully mature and thus effectively migrate to the draining lymph nodes where T-cell priming can take place. We further suggest that topical application of shikonin can effectively increase immature DC trafficking and accumulation at the immunization site, thus facilitating the effectiveness of subsequent vaccinations.
In conclusion, our present study investigated the mechanistic and multi-targeted molecular effect of shikonin on the trafficking of DCs, and we evaluated its potential application as a DNA vaccine adjuvant. We showed that shikonin, a phytochemical purified from Lithospermum erythrorhizon, can induce the highest level of expression of RANTES in normal skin tissues among a number of anti-inflammatory phytochemicals tested. In comparison with mouse RANTES cDNA gene transfection, shikonin treatment induced a transient high-level expression of mRANTES (up to 24 hours) and this caused less extensive skin tissue damage. A human gp100 cDNA gene vaccination regimen in combination with pretreatment with shikonin as an adjuvant in a human gp100-transfected B16 mouse tumor model significantly increased the cytotoxic T lymphocyte activities on target tumor cells in splenocytes and lymph node cells of the test mice. We believe this shikonin treatment approach may be beneficial/promising for future clinical application to gene-based cancer vaccines. Future optimization of a DNA vaccination schema and regimen may include the efforts to reduce the undesirable effects of shikonin and amplify its beneficial effect. Shikonin, as a natural phytochemical product, warrants further evaluation as an adjuvant for gene-based vaccines, especially for topical application alongside vaccination or immunotherapy. The present study may serve as a pilot approach or system for evaluating the use of an immuno-stimulatory agent (e.g., shikonin) as an adjuvant for topical application on skin alongside vaccines and gene delivery vectors.
regulated upon activation of normal T-cell expression and secretion
cytotoxic T lymphocytes.
This present work was sponsored by the Grant (NSC-097-2320-B-001-012) from the National Science Council, Taiwan and the Grant (99-Academia Sinica Investigator Award-12) from Academia Sinica, Taiwan. We thank the professional editing of this manuscript by Miranda Jane Loney, Heiko Kuhn and Ruth Giodan.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.