Abstract
Aim: Effects of radiofrequency ablation (RFA) combining immunoadjuvant glycated chitosan (GC) on tumor control and potent cytokine responses were investigated in a syngeneic breast tumor model. Materials and Methods: Murine 4T1 breast carcinoma cells harboring the luciferase reporter gene were used to evaluate the tumor growth rate and metastasis in vivo using bioluminescent imaging. Plasma of RFA/GC-treated tumor-bearing mice was collected for ex vivo cytotoxicity analysis and mouse chemokine array assays. Results: Tumor growth and systemic metastasis were suppressed by combined RFA and GC when tumor size reached 300 mm3, not detected, however, when tumor size reached 800 mm3. The survival rate of mice bearing small tumors was also higher than that of large ones after RFA-GC treatment. Plasma extracted from RFA-GC-treated small tumor-bearing mice exhibited cytotoxicity on cultured 4T1 cells. Moreover, reduced tumor growth-related cytokines and increased antitumor-related cytokines were detected in the plasma collected. Conclusion: RFA combining GC could control tumor progression with induced potent antitumor cytokine responses.
- Glycated chitosan
- radiofrequency ablation
- syngeneic tumor model
- breast cancer
- antitumor
- cytokine responses
Combination of different therapeutic approaches is commonly applied in cancer treatment. Hyperthermia is one of the methods using local heat for tumor control via various physical approaches and modalities, such as radiofrequency (RF) and focused ultrasound (FUS). Hyperthermia is usually combined with radiotherapy and chemotherapy because it may change the tumor's microenvironment and increase blood flow for oxygenation to provide radio-chemosensitization (1, 2). Although the therapeutic efficacy is diverse, combined hyperthermia and other treatments have been applied in the treatment of melanoma, cervical cancer, head and neck cancer, breast cancer, brain cancer, lung cancer, liver cancer and bladder cancer (3-6). Although hyperthermia can also induce cell killing effects at high temperature range, its tumor suppressive effects remain to be addressed using mild hyperthermia.
Radiofrequency ablation (RFA) depends on an electrical generator that hooks a needle electrode to deliver high frequency alternative currents by penetrating through skin to the tumor mass. RFA is regarded the most commonly used modality for tumorous thermal ablation and applied in the treatments of liver cancer, lung cancer, renal carcinomas, breast cancer and prostate cancer (7, 8). RFA-mediated elevation of effective temperature for tumor control may depend on tumor types. For instance, it aims to reach 50°C to 100°C to kill liver tumors (9). Alternatively, there was a report that employed 46.8°C to 79°C for breast cancer treatment (10). The temperature can also be adjusted between 42°C and 44°C to avoid damage of normal tissues surrounding targeted tumors (11). In addition to heat effect, RFA has been reported to activate tumor specific T-cell responses for cancer therapy (12). RFA also induces antitumor immune responses by modulating the immune system of metastatic liver patients (13).
Adjuvants are immunological agents that modify or augment an immune response, usually to a vaccine, without having any specific antigenic effect on their own (14). The most common adjuvants for human application today are aluminum hydroxide and monophosphoryl lipid A (MPL) (14). However, their application has not been widely used in cancer therapy. Glycated chitosan (GC) is derived from chitosan, a linear polysaccharide composed of D-glucosamine and N-acetyl-D-glucosamine through β-[1-4] linkage (15). This modification improves the water solubility of chitosan for various biomedical applications (16). Several lines of evidence have shown that GC is an immunoadjuvant agent that can target primary and metastatic tumors formed in animal and human patients (16-18). By combining laser photothermal and photodynamic therapy, GC is believed to induce immunological responses to attack tumors, so called laser immunotherapy (LIT) (16, 19-23). Additionally, GC has been combined with nanomaterials, such as single-walled carbon nanotubes (SWNTs), and used for improving the immuno-modified effects of GC on tumor suppression through photothermal therapy. Although the optical-mediated induction of immune responses has been broadly used in these researches, photon-based therapy remains uncommonly used in clinic. RFA has been widely used as a modality of hyperthermia in hospitals. It is of interest to investigate whether GC can combine with RFA for better tumor control.
In this study, we combined GC and RFA to treat a murine breast cancer using a syngeneic tumor model. This combination revealed that small tumors were more sensitive than larger tumors. Plasma extracted from whole blood of mice bearing small tumors, but not large tumors, with RFA-GC treatment also showed higher toxicity on cultured 4T1 cells than that from individual treatment. Moreover, chemokine array analysis demonstrated that several cytokines related to tumor development were reduced in plasma. Therefore, the current results support that GC combined with RFA would be a novel strategy for breast cancer control.
Materials and Methods
Cell culture and chemicals. 4T1 murine breast carcinoma cells with stably transfected multi-cistronic reporter genes have been reported previously, so called 4T1_PB3R (24). Cells were cultured in RPMI1640 medium (Gibco; Invitrogen Inc., Carlsbad, CA, USA) containing 10% fetal bovine serum (FBS; HyClone; Thermo, Waltham, MA, USA), 1% penicillin–streptomycin solution (Caisson Laboratories Inc., North Logan, UT, USA) and 1% l-glutamine (200 mM) (Sigma-Aldrich Co., St. Louis, MO, USA). Cells were cultured in a 37°C, humidified incubator containing 95% air and 5% CO2 and passaged every 2 days. Preparation of glycated chitosan (10 mg/ml, dissolved in deionized distilled water) has been described previously (16). It was then stored in the refrigerator until used.
Syngeneic breast tumor model. The immune competent female Balb/C mice were used for establishing the syngeneic tumor model (National Laboratory Animal Center, Taipei, Taiwan). 4T1_PB3R tumor cells (1×106 in 100 μl phosphate-buffered saline) were subcutaneously injected into the upper backs of 4-week-old mice. Tumors were formed after one week post-injection and their dimensions were measured using caliper twice a week. The tumor volume was determined by the equation: volume=length (mm) × width (mm)2/2. The animal experiments had been reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) of National Yang-Ming University (approval number: 1040722).
Radiofrequency ablation with GC. The tumor-bearing mice were anesthetized and placed on an aluminum plate attached with an electricity-conducting pad. An RFA electrode (a cool-tip RF Ablation Electrodes with cool-tip™, ACT1510; COVIDIEN, Minneapolis, MN, USA) was penetrated into the middle of the tumors. RFA treatments were done using a radiofrequency generator (CTRF117; COVIDIEN). The temperature was set between 42°C and 44°C for 10 seconds of treatment. For combined treatment, 100 μl of GC solution (10 mg/ml) was intratumorally injected into the center of tumor at three random peripheral positions of each tumor using a 27-G needle. The same strategy was also used for GC treatment alone. The tumor responses were then examined daily.
In vivo bioluminescent imaging. Tumor-bearing mice were intraperitoneally injected with 150 mg/kg D-luciferin (Caliper Co., Hopkinton, MA, USA) for 15 min and anesthetized with 2% isofluorane in a small cabinet. Mice were then removed from the cabinet and placed into the chamber of IVISTM50 system (Xenogen Co., Alameda, CA, USA) for detecting the luminescent signals. Regions of interest were drawn around whole body and luminescent signal was quantified by the number of photons detected per second (photons/sec/cm2/sr). Steradian (sr) represents the International System of Units (SI) of solid angle (25).
Isolation of plasma. Tumor-bearing mice were sacrificed using cervical dislocation after different treatments. Blood was immediately collected from cardiac puncture using a 26 G needle. Extracted blood was immediately centrifuged using 1,300 rpm at 4°C for 10 minutes. The supernatant was then collected and centrifuged again. The plasma was then passed through a 0.22 μm Super Membrane Low Protein Binding filter (PALL Inc., Port Washington, NY, USA).
Cell viability assay. Plasma was diluted ten-fold by culture medium and added into a 96-well plate seeded with 4T1_PB3R cells and incubated under 37°C for 2 days. After removal of the supernatant, 1 mg/ml 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenylterazoliumbromide (MTT) solution (Sigma-Aldrich) was mixed with serum-free medium and added to each well, incubated at 37°C for another 4 h. Dimethyl sulfoxide (DMSO) was then added to dissolve crystals. The plate was then quantified using an ELISA reader (Bio-Tek Instruments, Winooski, VT, USA) with a light absorption at 570 nm.
Assay for chemokines in plasma. All monoclonal capture antibodies and cytokines were obtained from RayBiotech Inc. (Norcross, GA, USA) and assay procedures were followed according to the manufacturer's instructions. Plasma was obtained as described above. Ten-fold diluted plasma was incubated with the Mouse Chemokine Array C1 (RayBiotech Inc.) in blocking buffer at 4°C overnight. The secondary antibody, conjugated with horseradish peroxidase, was then replaced with the diluted plasma and the membrane was agitated for 2 h at room temperature. The membrane was rinsed with Western lightning plus-enhanced chemiluminescence (ECL) (Perkin-Elmer Inc., Waltham, MA, USA) and chemoluminescent signals were detected using the LAS-4000 gel imaging system (GE Healthcare Inc., Wauwatosa, WI, USA). The dot densities were quantified using ImageJ software (version 1.46; https://imagej.nih.gov/ij/download.html).
Statistical analysis. Experimental data are presented as the mean of three independent experiments±standard deviation (S.D). Data were analyzed with Student's t-test or two-way ANOVA (for animal experiments) and p<0.05 indicated statistical significance.
Results
The experimental flowchart for treatment of syngeneic breast tumor model by combining GC and RFA. Tumor-bearing mice were grouped into control, GC-treated only, RFA-treated only and combination. Additionally, treatments were initiated when tumor volume reached 300 mm3 or 800 mm3. The tumor progression was then monitored using bioluminescent imaging and caliper (Figure 1A). The RF instrument and the position of electrode on implanted tumor are shown in Figure 1B.
Effects of GC and RFA on the tumor growth rate. Tumor growth was determined by measuring the tumor volume at different time points following treatment. When tumor volume initially reached 300 mm3, GC combined with RFA treatment exhibited dramatic tumor killing effects compared to GC or RFA individual treatment or untreated control (Figure 2A). On the other hand, the same treatments on 800 mm3 tumors showed little tumor suppressive effects (Figure 2B). These results suggested that a combination of GC and RFA could diminish the growth of small tumors but not that of large tumors.
Effects of GC and RFA on tumor metastasis. Because the combined GC and RFA treatment could suppress tumor growth at primary site, we were interested in further investigating whether tumor metastasis could be repressed accordingly. To this end, we used bioluminescent imaging to track the development of 4T1 tumors harboring luciferase reporter genes in vivo (see Materials and Methods). The results showed that, in animals with small tumors, tumor growth was completely suppressed and remote metastasis was not detected after RFA-GC combined treatment compared to control and separate treatments (Figure 3A). However, the growth and metastases of animals with large tumors were not suppressed by any of these treatments (Figure 3B). The photon fluxes were also quantified and compared among these regimes (Figure 3C and 3D). These data suggest that combined GC and RFA treatment could significantly suppress tumor growth and subsequent remote metastases of small syngeneic tumors but not large ones.
Effects of GC and RFA on survival of tumor-bearing mice. We next examined the survival of these mice using different treatments. Our results showed that mice bearing small tumors treated with RFA-GC could survive longer than untreated control or mice treated by RFA or GC alone (Figure 4). For large tumor-bearing mice, survival rates showed no significant differences among the animals under different treatments (Figure 4). Actually, the RFA-GC-treated small tumor-bearing mice remained tumor-free for over 4 months. Therefore, it appeared that small tumors were more sensitive to combined GC and RFA treatment, with the animals surviving much longer.
Blood plasma extracted from combined GC and RFA-treated tumor-bearing mice exhibited cytotoxicity effects. To further investigate whether the tumoricidal effects caused by combination of GC and RFA treatment is associated with immunological responses, blood plasma was extracted from tumor-bearing mice (both small and large tumors) at 3 weeks after various treatments. The plasma was diluted in culture medium and added to cultured 4T1 cells in a 96-well plate for the MTT assay (Figure 5A). Results showed that plasma of mice bearing small tumors exhibited stronger cytotoxicity than that of mice bearing large tumors after RFA-GC treatment (Figure 5B). Therefore, the GC and RFA combined treatment might induce stronger immune responses in mice with small tumors than those with large tumors.
Expression of immunologically-related molecules in plasma extracted from RFA-GC-treated tumor-bearing mice. To understand if any immune responses could be modulated by combined GC and RFA, we used the Mouse Chemokine Array C1 to analyze the expression of immune-related cytokines in plasma after treatment. Twenty-five different plasma cytokines were detected after tumor-bearing mice when treated with GC, RFA or RFA-GC. The patterns of chemokine array staining were acquired and visualized using a gel imaging system (Figure 6A). After densitometric quantification, fifteen of these chemokines exhibited signals sufficient to be quantified. Compared to control, CXCL13 (B lymphocyte chemoattractant (BLC)) and CCL2 (monocyte chemoattractant protein-1 (MCP-1)) were induced by RFA but were suppressed by RFA-GC combined treatment (Figure 6B). On the other hand, CCL20 (macrophage inflammatory protein 3 alpha (MIP-3-α)) and CXCL4 (platelet factor 4 (PF-4)) were boosted by combined RFA-GC treatment to different levels, no matter whether they were affected by RFA alone or not (Figure 6B). Although several cytokines, such as CCL27 (cutaneous T-cell-attracting chemokine (CTACK)), CCL11 (Eotaxin-1), CXCL1 (KC), monocyte chemotactic protein-5 (MCP-5), MIP-2 and CCL-5 (regulated on activation, normal T cell expressed and secreted (RANTES)), were also induced by RFA-GC combined treatment they showed little difference from that in RFA treatment alone. Taken together, the current analysis suggested that RFA-GC treatment might change the expression of cytokines in blood for modulating potent immune response to suppress tumor growth.
Discussion
GC, when combined with photothermal therapy, has shown great promise in the treatment of aggressive DMBA-4 breast tumors in rats with significant antitumor immune response in cured rats that completely resisted repeated tumor challenges. Furthermore, the transfer of splenocytes from cured DMBA-4 breast tumor-bearing rats provided 100% immunity in the naïve recipients against the same tumor cells, demonstrating that LIT induced adaptive immunity. In this study, we used RFA, an FDA approved minimal invasive therapy for soft tissue tumors, followed by GC to investigate the effects of tumor suppression in a syngeneic breast tumor model (8). Surprisingly, this combination not only eradicated tumor growth at the primary site but also the remote metastasis in mice with small tumors as compared to untreated controls or mice under GC or RFA treatment alone. However, the combined treatment did not show better tumor suppressive effect than RFA alone in mice with large tumors at the time of treatment. Several reports using RFA for human breast cancer treatment usually select an average tumor volume from 0.775 cm to 2.08 cm or at early stage (8, 26-31). It is reasonable that heat transduction should be more efficient in small tumors. Additionally, it has been reported that pre-heating at 37°C or 43°C can enhance cell sensitivity to bleomycin (BLM) or cis-diammine-dichloroplatinum(II) (cis-DDP) (32). Although GC is not regarded a chemotherapeutic agent, pre-heating a small syngeneic tumor followed by GC injection also exhibited apparent tumor suppression. To the best of our knowledge, this is the first report showing that pre-heating treatment using RFA combined with a non-chemotherapeutic agent can suppress tumor growth. Based on the immunoadjuvant property of GC, enhancement of immune responses by this combination could be one of the effects to ablate tumor growth.
Moderate hyperthermia can prevent normal tissues from complications; however, the trade-off effect could be inefficient therapy. In fact, hyperthermia is an adjuvant approach for conventional radiotherapy and chemotherapy (33-36). Accumulated evidence supports that hyperthermia combining radio-chemotherapy can lead to better local control and patient survival with different types of high-risk cancers (4, 37). Although GC is not considered as a chemotherapeutic agent, it may alter the immune responses in tumor microenvironment after hyperthermia. Hyperthermia has been believed to induce immunity due to induction of heat shock proteins (HSPs) and local fever (38-40). A recent report has also shown that hyperthermia regulates the interaction between HSPs and mismatch repair (MMR) proteins to influence the cisplatin sensitivity of MMR-deficient/proficient colon cancer cell lines (41). On the other hand, abundant expression of HSPs in tumors correlates to the viability of cancer cells and resists to hyperthermia-induced cytotoxicity (42, 43). Because GC combined with RFA exhibited dramatic tumor suppressive effects compared to RFA alone in small tumors, it is of interest to further investigate whether GC can affect the expression of HSPs in the future.
One of the interesting results was that the plasma obtained from RFA-GC-treated mice with small tumors exhibited significant cytotoxicity in cultured 4T1_PB3R cells. Although little is known about the underlying mechanism, the current findings are relevant to those of a previous report that hyperthermia may augment specific immune reaction via antibody-complement cytotoxicity on tumor cells (44). Here, by using chemokine array analysis, it was found that cytokines, including CXCL13 (BLC) and CCL2 (MCP-1), could be affected by RFA. However, the expression of these cytokines was reversely expressed by RFA-GC treatment. CXCL13 and CCL2 were overexpressed in breast cancer and associated with lung metastasis of 4T1 cells, respectively (45, 46). This finding suggests that GC combining RFA could repress tumor-related cytokines. On the other hand, the most apparent RFA-GC-induced cytokines were CCL20 and CXCL4, that could boost CD8+ T-cell-dependent antitumor response and inhibit angiogenesis to control tumor invasion, respectively (47, 48). Moreover, several detected cytokines were induced by RFA or RFA-GC combined treatment to similar levels. Whether they also play important roles to influence the therapeutic efficacy of RFA-GC combination remains to be studied. Since the cytokines affected by RFA-GC combined treatment are associated with tumor control, it is plausible that GC may modulate the antitumor immunity following hyperthermia. Furthermore, the plasma of tumor-bearing mice on inhibiting the tumor cell viability (Figure 5) indicated the possible generation of specific antibody(ies) by the RFA-GC treatment, which could lead to antibody-dependent complement-mediated cytotoxicity. It is possible the RFA-GC activated both cellular and humoral arms of the host immune responses. This will be an interesting topic in our future studies, such as ELISA-based detection of the difference in IgM and IgG antibody titers.
Additionally, although photothermal application can release a large amount of different antigens for the generation of antitumor immunity, the response of the host immune system is limited (49, 50). When combined with GC, CD80 expression on dendritic cells (DCs) could be significantly enhanced, so could be the antigen acquisition and antigen presentation by antigen presenting cells (APCs), thus activating cytotoxic T-lymphocytes (CTLs) (50). Therefore, immunologic intervention is required to invoke the immune system to achieve an effective and protective immune response against residual tumor cells. Whether GC combining RFA can also induce CD8+ T-cell activation, related mechanisms should be important to clarify in the future, i.e. the levels of innate and humoral immune responses for tumor suppression.
The limitation of this study is that the temperature recorded by RFA was based on the hooked electrode rather than the tumor mass per se. It is not surprising to see that small tumors responded better than large tumors to treatment because the temperature distribution of small tumors should be greater even than that of larger tumors after RFA treatment. Furthermore, our results also provide evidence that RFA-GC, like any modality, loses its potential when tumor burden becomes too big, simply because of the limitation of any treatment method in controlling an overwhelming tumor burden. Without effective control of the primary tumor, the immune system may not be strong enough to control metastasis. Although we used mild hyperthermia and short temperature exposure time, we also tried longer exposure time (60 seconds) under the same temperature range. However, the results were not improved using RFA alone (data not shown). The pain of animals, according to their body responses, seemed also enhanced at longer temperature exposure. Therefore, the current regime of RFA should be ideal for other tumor models using immune competent animals in the future.
Conclusion
The current data suggest that the immunoadjuvant GC can enhance the therapeutic efficacy of hyperthermia in small breast tumors. Use of a syngeneic, immune competent tumor model demonstrated that RFA-induced immune responses could be modulated by GC. The combination of RFA and GC not only eradicated tumor but also suppressed metastasis in vivo. As GC has been proven to be a non-toxic agent and RFA has been used in clinical settings, our results may provide a novel potential strategy to combine these methods for human tumor control in the future.
Acknowledgements
This study was supported, in part, by Minister of Science and Technology of Taiwan (MOST 105-2628-B-010-013-MY3) and Taipei City Hospital, RenAi Branch (No.16). Additionally, the grants of US National Institutes of Health (R21 EB0155091-01 and RS20132225-106) and the Oklahoma Center for Advancement of Science and Technology (HR16-085) also contributed to related research. The Authors also thank the Taiwan Mouse Clinic (MOST 104-2325-B-001-011) for technical support of the in vivo imaging experiment.
Footnotes
↵* These Authors contributed equally to this study.
- Received March 3, 2017.
- Revision received April 17, 2017.
- Accepted April 18, 2017.
- Copyright© 2017, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved