Te-I Liu ,Yun-Chung Tsi ,Te-Ming Wng ,Siou-Hn Chng ,Ying-Chieh Yng ,Hsin-Hung Chen ,Hsin-Cheng Chiu,*

a Department of Biomedical Engineering and Environmental Sciences,National Tsing Hua University,Hsinchu City,Taiwan

b Department of Radiology,National Taiwan University Hospital Hsin-Chu Branch,Hsinchu City,Taiwan

Keywords:Radiotherapy Drug delivery Immunomodulation Combination therapy Breast cancer

ABSTRACT Immunotherapy is an ascendant approach in cancer treatment.It shows more pronounced effects on killing cancer cells in a specific manner in particular against metastasis more than traditional techniques,such as chemotherapy or surgery.However,tumor immunosuppression limits the response of the immune system to cancer development.In this study,we developed a lipid-based nanocarrier doubly loaded with imiquimod(R837),a toll-like receptor 7 agonist,and caffeine,an adenosine receptor antagonist.This R837/caffeine loaded nanocarrier served as a nano-immunomodulator (RC-nIM) for combination treatment with radiotherapy (RT)against orthotopic breast cancer.RT-induced immunogenic cell death facilitated the production of tumor antigen and elicited the immune response in corporation with R837-medaited activation of antigen-presenting cells(APCs)while RC-nIMs being adopted.Additionally,caffeine,an adenosine analog,can successfully compete with adenosine in the tumor.The tumor-bearing mice that received RT together with RC-nIMs experienced the best antitumor effects and exhibited higher levels of T cells and APCs within the tumor;the growth of secondary tumors was also limited.This work serves as a proof-of-concept study for the development of a new immunotherapy strategy against cancer.

1.Introduction

Immunotherapy quickly increased in popularity in recent years after several immune therapeutics obtained positive outcomes as anticancer treatments.For example,pembrolizumab and nivolumab are immune checkpoint inhibitors preventing the inhibition of cytotoxic T-cell activity [1,2].Additionally,the CAR-T technique is used to help T cells recognize cancer cells.Furthermore,immune adjuvants,such as the TLR agonist that activates antigen presenting cells (APCs),are used to enhance the tumor-specific immune response [3,4].These therapeutic approaches provide an overview of the immunotherapeutic strategy,which broadly addresses issues of tumor antigen generation,APC maturation,and T cell activation [5,6].Radiotherapy (RT),one of the general approaches for local cancer treatment,can directly break the DNA double helix and generate reactive oxygen species,which induces cell damage and thus falcitates immunogenic cell death (ICD) [7].ICD can in fact be induced by several general anticancer treatment strategies,including hyperthermia,chemotherapy,and radiotherapy,which result in the generation of tumor antigens and expression of danger signals,such as the calreticulin(CRT)and HMGB1 proteins[8,9].Thus,antigen-specific adaptive immune responses can be elicited through the ICD pathway,driving the activation and polarization of APCs loaded with tumor antigen-specific cargo.After the APCs migrate to draining lymph nodes,T cells(CD4+and CD8+T lymphocytes)are then trained to enable the effective anticancer adaptive immune response [10,11].CD8+T lymphocytes,also known as cytotoxic T lymphocytes (CTLs),play a critical role in tumor elimination through the secretion of the proinflammatory cytokines,such as IFN-γ,TNF-α,and IL-2,as well as perforin and granzymes [12,13].

However,even if a high dose of X-ray irradiation can be effective in tumor elimination and tumor antigen generation,its therapeutic effi-cacy is often limited by the high probability of side effects [14,15].Conversely,a range of immunosuppression pathways in the tumor microenvironment can hinder the activity of CTLs [16].For example,regulatory T(Treg)cells can suppress the antitumor ability of CTLs with anti-inflammatory cytokines and largely induce adenosine accumulation by activating CD73 and CD39 during apoptosis [17].This high adenosine concentration causes the activation of the adenosine A2A receptor (A2AR) on the CTL membrane and thus inhibits the secretion of proinflammatory cytokines,resulting in immunosuppression[18,19].Zou et al.have found that apoptotic Treg cells induced by oxidative stress can abruptly enhance the suppressor activity,leading to the development of additional resistance in checkpoint blockage therapy.Further study showed that the A2AR inhibitor(A2ARi),ZM241385,can abolish the suppression of T cell effector cytokine mediated by apoptotic Treg cells [20].The suppressed TNF-α and IL-2 secretions were rescued following the A2ARi treatment.Interestingly,caffeine,a prevalent and accessible natural compound with an adenosine-like structure,has been reported to act as an AR antagonist [21].However,the relevant application of caffeine in combination with other immunotherapeutic approaches or anticancer approaches remains poorly understood.

Here,we report the development of a novel biodegradable lipidbased nanocarrier acting as a R837/caffeine-loaded nano-immunomodulator (RC-nIM) for the enhancement of therapeutic efficacy of radiotherapy.RC-nIM was loaded with the TLR7 agonist imiquimod,also known as R837,and the AR antagonist,caffeine.The use of nanocarrier systems for the delivery of therapeutics has been demonstrated as an effective strategy for enhancing drug accumulation at tumor sites and improving drug stability in circulation [22].By incorporating RT and RC-nIM,the therapeutic approach developed herein aimed to strengthen cancer inhibition of the ICD-induced immune response by modulating the activation of APCs and CTLs.The design of the therapeutic strategy is presented in Scheme 1.The physical characteristics of RC-nIMs were examined by dynamic light scattering(DLS)and UV/Vis and fluorescence spectrometry.The ICD of X-ray irradiation was confirmed by measuring in vitro CRT and HMGB1 expression.The enhancement of APC activation was elucidated through the expression of iNOS and MHC-I.The in vivo therapeutic validation of the combination treatment of RT and immune modulation through the RC-nIM system accounted for the inhibition of both primary tumor and secondary metastasis and was conducted on orthotopic breast tumorbearing mice.

Scheme 1.Schematic illustration of the RC-nIM system in modulating the X-ray-induced immune response for cancer inhibition.

2.Materials and methods

2.1.Materials

Trilaurin,oleic acid,imiquimod (R837),caffeine,Pluronic F68,3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium bromide (MTT)and hematoxylin and eosin (H&E) staining were obtained from Sigma-Aldrich.All organic solvents were purchased from Alfa Aesar,and deionized water (DIW) was produced by the Milli-Q Synthesis System(18 MΩ,Millipore).All chemicals were of reagent grade and used as received.For the cell culture studies,Dulbecco's modified Eagle's medium (DMEM),fetal bovine serum,and 0.25% trypsin–EDTA and penicillin–streptomycin solutions were purchased from Thermo Fisher Scientific (Waltham,MA,USA).Murine 4T1 breast cancer cell line and murine macrophage-like Raw264.7 cell line were obtained from the Food Industry Research and Development Institute (Hsinchu City,Taiwan).Hoechst 33342 and various antibodies used for immunohistochemical (IHC) staining and flow cytometry were acquired from Abcam.

2.2.Preparation and characterization of R837/caffeine-loaded nanoimmunomodulator (RC-nIM)

The nIMs used in this study were prepared by the high-pressure homogenization technique as reported previously with slight modifications [23].The caffeine solution (16 mg caffeine in 200 μL DIW)was added dropwise into 0.48 mL of EtOH/DIW solution (v/v 5:1)containing trilaurin (16.0 mg) under magnetic stirring (2000 rpm) at 50°C.After 5 min stirring,50 μL EtOH containing 3 mg R837 and 22.5 mg oleic acid was added dropwise into the reaction solution,followed by 3 min stirring.The resulting solution was then added dropwise into the Pluronic F68 aqueous solution (7 mg in 200 μL DIW)under magnetic stirring (2000 rpm) at 50°C for 3 min.Then,4 mL of 4°C DIW was rapidly added into the previous solution,followed by sonication in ice bath for 7 min.The crude sample was purified through ultrafiltration (MWCO 10 K) and stored at 4°C until further use.

The drug loading efficiency (L.E.) and loading content (L.C.) were determined by measuring the absorbance of caffeine at 273 nm(Hitachi U-2900 Double Beam Spectrophotometer) and the fluorescence intensity of R837 at 365 nm with an excitation wavelength of 325 nm(Hitachi F-7000 Fluorescence Spectrophotometer) in DMSO at 25°C.The L.E.and L.C.were calculated as follows:

Drug L.E.%=(weight of loaded drug)/(weight of drug in feed) × 100%

Drug L.C.%=(weight of loaded drug)/(weight of NPs) × 100%.

The hydrodynamic size,polydispersity,and zeta potential were determined by DLS (Malvern Zetasizer Nano ZS90).The colloidal stability of RC-nIM was determined by monitoring the size variation of the 50-fold PBS-diluted sample at 37 °C for 1 week.The cumulative drug release of drug-loaded nIMs was measured via dialysis.The nIM suspension (1.0 mL,0.1 mg/mL) was loaded into a dialysis bag (MWCO 12 K–14 K) against 20 mL pH 7.4 PBS at 37 °C.At prescribed time intervals,1 mL outer solution was withdrawn to quantify the amount of caffeine or R837 via UV/Vis or fluorescence spectrometry.After each sampling,1 mL fresh buffer was added to the outer aqueous phase.The cumulative drug release was determined as the ratio of the amount of caffeine(or R837)released into the outer buffer to the original amount caffeine(or R837)in the nIMs.The influence of enzymatic degradation was investigated through the change of the particle size distribution of the nIMs and the cumulative drug release.During the experiment,lipase (1000 units/mL) was co-incubated with the nIM suspension.

2.3.In vitro cellular uptake and cytotoxicity of RC-nIMs

Murine breast cancer 4T1 cells or murine macrophage-like RAW 264.7 cells were seeded into a 6-well plate (3 × 105cells/well) and incubated for 24 h Cy5.5-labeled R-nIMs or RC-nIMs were co-incubated with cells for preset time intervals,followed by PBS washing.To determine the uptake level of R837 by both cell lines,the nIM-treated cells were destroyed with DMSO,followed by a fluorescence intensity measurement (Ex.325 nm and Em.365 nm).For confocal laser scanning microscopic(CLSM)imaging,a 22×22 mm coverslip was placed in each well before cells were seeded.After co-incubating with Cy5.5-labeled nIMS and washing with PBS,the cell nuclei were stained with DAPI.The cell images were obtained by CLSM equipped with a Hoechst set(Ex.401 nm and Em.410–490 nm)and a Cy5.5 set(Ex.650 nm and Em.675–720 nm).The integrated optical density (IOD) of Cy5.5 was analyzed using ImagePro Plus software based on 10 random imaging fields.

The cytotoxicity of free drugs,R-nIMs,and RC-nIMs on 4T1 and RAW264.7 cells was evaluated using the MTT assay.Briefly,4T1 cells or RAW264.7 cells were seeded in 96-well plates(8×103cells/well for 4T1;6×103cells/well for RAW264.7)for 24 h.The spent medium was then replaced with fresh medium containing free drugs,R-nIMs,and RC-nIMs at varied drug concentrations,and the cells were incubated for additional 4 h.After the treatment,the cells were then washed three times with PBS,followed by an additional 24 h incubation.The MTT assay was then conducted to determine cell viability using the SpectraMax M5 microplate reader.

2.4.In vitro evaluation of radiation-induced immunogenic cell death

4T1 cells were seeded in 6-well plates(3×105cells/well)for 24 h and then detached with trypsin.The detached cells were collected by centrifugation and then stored at 4°C for X-ray irradiation experiments.X-ray irradiation (6 and 15 Gy) was produced by a linear accelerator(Elekta SLI-5951,Sweden) at the National Taiwan University Hospital Hsin-Chu Branch.After X-ray irradiation,the cells were again seeded in 6-well plates (3 × 105cells/well) for 24 h.Immunofluorescence staining was employed for CLSM imaging and flow cytometry.The primary antibodies anti-CRT (ab2907,Abcam) and anti-HMGB1(ab77302,Abcam) and the secondary antibody conjugated with Alexa Fluor 488 (ab150117,Abcam) were used herein.Immunofluorescence staining for CRT and HMGB-1 identification was conducted as described in the operating guidelines provided by Abcam.Additionally,cell nuclei were stained with DAPI for CLSM imaging.The stained cells were analyzed by flow cytometry with an FITC set(Ex.488 nm and Em.525 nm) and by CLSM with a Hoechst set (Ex.401 nm and Em.410–490 nm) and an FITC set (Ex.488 nm and Em.515–530 nm).The relative fluorescence intensity of Alexa Fluor 488 was analyzed using the ImagePro Plus software based on 10 random imaging fields.

2.5.In vitro evaluation of macrophage polarization and activation

RAW264.7 cells were seeded in 6-well plates (3 × 105cells/well)for 24 h,followed by the addition of IL-4 (25 ng/mL) and then incubation for another 48 h.The spent medium was then replaced with fresh medium containing free drugs,R-nIMs,and RC-nIMs with drug concentrations of 40 μM caffeine and 45 μM R837 before an additional 4 h incubation.The cells were collected by centrifugation and fixed with 5% BSA.The expression of iNOS was determined by flow cytometry.The primary antibody anti-iNOS (ab15323,Abcam) and the secondary antibody conjugated with Alexa Fluor 488 (ab150117,Abcam) were used.Immunofluorescence staining for iNOS identification was conducted as described in the operating guidelines provided by Abcam.Finally,the samples were analyzed by flow cytometry with an FITC set (Ex 488 nm and Em 525 nm).

The transwell method was utilized to investigate macrophage activation via immunogenic cell death.RAW264.7(3×105cells/well)and 4T1 (3 × 105cells/well) cells were seeded and incubated in the transwell bottom chamber and transwell insert for 24 h separately.The spent medium in the chamber was replaced with fresh medium containing free drugs,R-nIMs,or RC-nIMs at drug concentrations of 40 μM caffeine and 45 μM R837 before the addition of 4 h incubation.After washing with PBS,the insert with the 4T1 cells being irradiated with 15 Gy X-rays was placed into the chamber.An additional 24-h co-incubation was then conducted;afterwards,the RAW264.7 cells were washed with PBS and collected for flow cytometry.The primary antibody,anti-MHC-I (ab95572,Abcam),and the secondary antibody conjugated with Alexa Fluor 488 (ab150117,Abcam),were used to detect MHC-I expression.The samples were analyzed by flow cytometry with an FITC set (Ex 488 nm and Em 525 nm).The staining index was calculated based on the following formula [24]:

Staining index=(MFIpos-MFIneg)/[(84% MFIneg-MFIneg)/0.995]

2.6.Animal and tumor model

Five-week-old male BALB/cByJNarl mice were purchased from the National Laboratory Animal Center (Taiwan) (approved number:IACUC 107018)and cared in accordance with the Guidance Suggestions for the Care and Use of Laboratory Animals,approved by the Administrative Committee on Animal Research at the National Tsing Hua University,Taiwan.The orthotropic breast animal model was established by injecting 4T1 cells (1 × 106cells/100 μL) into the mammary fat pad under the number 4 mammary gland of the mice.The tumor volume was monitored daily,and the tumor volume (V) was calculated as follows:V=L × W 2/2,where W is the tumor width at the widest point and L is the tumor length at the longest point.All of the experiments were conducted when the tumor volume reached 60 mm3.

2.7.In vivo imaging and biodistribution

To study the time point of the maximum NP accumulation,tumorbearing mice were intravenously injected with Cy5.5-labeled R-nIMs and RC-nIMs followed by in vivo imaging(IVIS,Caliper IVIS Spectrum)to investigate the accumulation of the Cy5.5-containing nIMs within the tumor.The fluorescence images and signals of Cy5.5 at 4,6,18,24,48,and 72 h post-injection were then collected by IVIS examination (with Ex.660 nm and Em.720 nm).The treated mice were then sacrificed under CO2.The tumor and major organs,including the heart,liver,spleen,lungs,and kidneys,were collected to examine the ex vivo nanoparticle distribution via a similar IVIS approach.The fluorescence intensity was analyzed via the region of interest function in the IVIS software.

2.8.In vivo tumor growth inhibition

Murine 4T1 breast cancer in BALB/c mice (5 weeks old) was employed as an orthotopic tumor model for evaluating the in vivo RC-nIMassisted ablation of breast cancer.The orthotopic breast tumor model was established by implanting 4T1 cells(1.0×106cells/50 μL of PBS)into the mammary gland of the mice as described previously.When the tumor volume in the mice reached 60 mm3,the mice were randomly divided into five groups(five in each group)and intravenously injected with various drug formulations with a caffeine dosage of 1.3 mg/kg body weight and an R837 dosage of 1.4 mg/kg body weight once per day for consecutive 5 days.The tumor-bearing mice were anesthetized with Zoletil-Rompun by intraperitoneal injection.The X-ray irradiation was produced by a linear accelerator (Elekta SLI-5951,Sweden) at the National Taiwan University Hospital Hsin-Chu Branch.The radiation was tuned with a collimator.The 15 Gy X-ray irradiation was applied at day 0 before I.V.injection.The experimental groups were a PBS control,15 Gy X-ray,15 Gy X-ray+R-nIMs,15 Gy X-ray+RC-nIMs,and 15 Gy X-ray+free drugs.The tumor volumes and body weight of various groups were measured daily until 12 days following I.V.injection.The tumors were then isolated upon sacrifice of the euthanized mice and weighed.

To evaluate the immune-mediated antitumor effect on secondary tumor as a rechallenge study,4T1 cell suspension(5×105cells,50 μL)was subcutaneously injected into the left leg of the orthotopic breast tumor-bearing mice in every group(n=5)at day 5 post-treatment.The tumor volumes of each group were then measured daily.

2.9.IHC examination of tumor tissues

For CLSM imaging,the tumors harvested from the treated mice were embedded in OCT and stored at -80 °C until the tissue section was conducted for IHC examination.The organ tissue sections(10 μm)were mounted onto slides,and the tissue slides were then fixed with precooled methanol for 5 min and rinsed twice with PBS.IHC staining to identify CD3 T cells and CD11c APCs was conducted according to the operating guidelines provided by Abcam.The stained tissue slides were examined under CLSM.

2.10.Evaluation of immune cell populations in tumor tissues

The tumors harvested from the tumor-bearing mice were cut into small pieces and then washed with PBS.The tissue was then suspended in 10 mL PBS containing dispase II (100 μL) and bovine pancreas deoxyribonuclease I (10 mg/mL) under stirring for 15 min.After filtration with a 0.1 μm filter,the filtrate was centrifuged for 5 min to collect the pellet.Then,1 mL Red Blood Cell Lysis Buffer(eBioscience™1X RBC Lysis Buffer) was added and the incubation was performed for 5 min,followed by the addition of 3 mL PBS to stop the reaction.The resulting solution was centrifuged for 5 min,and the supernatant was removed.About 150 μL of the pellet was re-suspended with PBS containing 4% goat serum and 0.5% Fc blocker (purified rat anti-mouse CD16/CD32 Mouse BD Fc Block™,553,142,BD) and incubated for 30 min at 4°C.The antibody solution was then added to the suspension and incubated for 30 min,followed by PBS washing and flow cytometry measurements.The antibodies used herein included FITC hamster antimouse CD3e (553,061,BD),PE-Cy™7 rat anti-mouse CD45 (552,848,BD),PE rat anti-mouse CD4 (557,308,BD),APC rat anti-mouse CD8a(553,035,BD),FITC anti-mouse CD11c antibody(117,305,BioLegend),PE anti-mouse CD86 antibody (105,007,BioLegend),and APC antimouse CD80 antibody (104,713,BioLegend).

2.11.Evaluation of cyclic AMP (cAMP) in tumor tissues

Level of cAMP released from T cells within tumor regions after nIMassisted radiotherapy were examined by enzyme-linked immunosorbent assay (ELISA) according to the manufacturer's instruction.In brief,tumor tissue (80 mg) was cut into small pieces and homogenized in 1 mL PBS containing 0.1 N HCl (w/v=1/5) with an ultrasonic processor (VCX750,30 kHz,750 W) for 5 min.The cell suspension was then centrifuged (10,000 g),and the cell pellet was collected.The suspension was then neutralized with 1 N NaOH,and the amount of cAMP was determined using a cAMP ELISA Kit (ab133,038,Abcam).

2.12.Statistical analysis

All data are expressed as mean ± SD.A one-way ANOVA and Bonferroni correction for multiple comparisons were used to evaluate the statistical significance;n.s.＞0.05,＊P ＜0.05,＊＊P ＜0.01,and＊＊＊P ＜0.001.

3.Results and discussion

3.1.Characterization of drug-loaded nIMs

The lipid-based nIMs developed in this study were fabricated via high-pressure homogenization with some modifications [23].As presented in Fig.1a and Table 1,the drug-loaded nIMs exhibited a mean hydrodynamic diameter of ca 150 nm with mono-modal distribution(PdI ＜0.15).Fig.1b shows acceptable colloidal stability of nIMs most presumably as a result of the PEG stabilization and surface negative charge [25].Additionally,the storage stability of the nIMs was also examined at 4°C (Fig.S1).Both R-nIMs and RC-nIMs remained invariant in size at low temperature over 2 months.The amounts of caffeine and R837 in the loaded RC-nIMs were about 5.9 and 6.3 wt%,respectively.Interestingly,caffeine could only be loaded into the nIMs while R837 was being loaded,possibly due to the stabilization via π–π stacking between R837 and caffeine,which are both aromatic compounds [26,27].Drug release from the R-nIMs and RC-nIMs was then examined using the dialysis technique.As presented in Fig.1c and d,the release curves of both encapsulated caffeine and R837 in RC-nIMs reach a plateau at 12 h with ca 24% drug release,suggesting that both caffeine and R837 can be embedded into the lipid core of the nIMs and avoid premature drug leakage under approximately physiological conditions.In contrast,when the nIMs were co-incubated with lipase,significantly increased drug release was detected in both the caffeine and R837 release profiles– nearly 100% release of R837 was achieved at 24 h.Furthermore,as presented in Fig.S2,a significant change in the size distribution of lipase-treated RC-nIMs compared with intact RCnIMs was detected.These results indicate that lipid-based RC-nIMs can be enzymatically degraded to fully liberate the payload.

Table 1 DLS data and drug loading characteristics of nIMs.

3.2.In vitro cellular uptake and cytotoxicity

Fig.1.DLS measurement of nIMs in terms of(a)size distribution and(b)colloidal stability in pH 7.4 PB S(I=0.01 M)(n=5).(c)Caffeine release profiles of RCnIMs in the absence and presence of lipase.(d) R837 release profiles of RC-nIMs and R-nIMs in the absence or presence of lipase.Lipase (20 units/mL)was used to evaluate the effect of nIM degradation on drug release (n=5).

To confirm the cellular uptake of drug-containing nIMs,RC-nIMs and R-nIMs labeled with Cy5.5 for fluorescence imaging were incubated with murine 4T1 breast cancer cells and murine macrophage-like RAW264.7 cells,respectively,for 1–6 h.Afterwards,the amount of cellular uptake of nIMs was determined by imaging and analyzing the intracellular Cy5.5 signal (Fig.S3).The results indicate that RC-nIMs were successfully internalized by 4T1 and RAW264.7 cells and that the time point of maximum drug/carrier uptake was 4 h in both cell types.RC-nIMs must be non-toxic when being used as an auxiliary tool for enhancing the therapeutic outcome of ICD-induced antitumor treatment.The cytotoxicity of RC-nIMs was thus evaluated in the 4T1 and RAW264.7 cell lines(Fig.S4).No apparent cytotoxicity was observed in the free R837 and free caffeine groups,but slight cytotoxicity of nIMs was observed at rather high concentrations (30 μM R837 and 33.3 μM caffeine) due to the use of surfactant in nIM preparation.The concentrations of R837 and caffeine were set at 20 μM and 22.2 μM,respectively,for the remaining experiments.

3.3.In vitro immunogenic cell death

Immunogenic cell death was investigated in 4T1 cells irradiated by X-rays at dosages of 0,6,and 15 Gy.Crucially,X-rays induce apoptosis,leading to tumor antigen generation.The expression of CRT,as determined by flow cytometry and CLSM,gradually increased with X-ray dosage (Fig.2).An insignificant difference between 0 and 6 Gy was observed in both CRT and HMGB1 analyses,indicating that 6 Gy may not be strong enough to induce tumor antigen production.Contrarily,a 15 Gy dose of X-ray irradiation significantly increased CRT and HMGB1 expression compared with the 6 Gy dose.Hence,15 Gy was the selected dosage for the following in vitro and in vivo experiments.

3.4.The influence of RC-nIMs on macrophage polarization and activation

Macrophage polarization is strongly associated with tumorigenesis.The pro-inflammatory form of the macrophage,known as the M1 macrophage,is considered to have tumoricidal activity by upregulating various pro-inflammatory genes,cytokines,and chemokines [28–30].The high expression of iNOS can upregulate NO generation that triggers tumor cell apoptosis[31].To confirm the presence of RC-nIM-mediated M1 macrophage polarization,RAW264.7 cells were first incubated with IL-4 to obtain M2 macrophages as a negative control,followed by coincubation with free drugs,R-nIMs,or RC-nIMs.As presented in Fig.3a,the expression of iNOS decreased from 24.5% to 19.4% while treating the cells with IL-4,which indicates that the phenotype of macrophages has been mostly normalized to M2.After the RC-nIM treatment,the expression of iNOS significantly increased from 19.4% to 96.1% as compared with the IL-4-treated control.Interestingly,the expression of iNOS mainly decreased when caffeine was absent from the nIM formulation (50.8% in the R-nIM group),suggesting that not only TLR7 agonists but also AR antagonists can induce M1 polarization.However,the treatment with free caffeine alone did not successfully induce M1 polarization.It has been reported that the activation of adenosine receptors can suppress the production of pro-inflammatory cytokines and induce the secretion of anti‐inflammatory cytokines,leading to macrophage polarization toward M2.However,this does not affect the expression of M1-associated pro-inflammatory genes [32,33].Therefore,only the combined treatment incorporated with R837 and caffeine can efficiently induce the M1 macrophage phenotype.

Fig.2.Immunogenic cell death induced by radiotherapy.(a–d)Flow cytometric profiles of 4T1 cells irradiated with(a)0 Gy,(b)6 Gy,and(c)15 Gy for the analysis of the expression levels of CRT.The CRT positive ratio is summarized in(d)(n=5).(e)CLSM images(left)and the corresponding quantification analysis(right)of 4T1 cells after X-ray irradiation.CRT and HMGB1 were identified by the corresponding antibody(in green and red,respectively).Nuclei were stained by DAPI(blue).Scale bars are 25 μm.

Fig.3.Effects of various treatments on macrophage polarization and activation.(a)iNOS expression of RAW264.7 cells treated with either free drugs or drug-loaded nIMs.IL-4 was used for M2 normalization in every group(n=5).(b)Schematic illustration of the experimental timeline and the staining index of MHC-I determined by flow cytometry (n=5).A dosage of 15 Gy was chosen for X-ray irradiation.The concentrations of R837 and caffeine were 20 and 22.2 μM,respectively.

The expression of MHC-I on the surface of macrophages was determined by flow cytometry with staining index analysis to observe the influence of ICD and RC-nIMs on macrophage activation.RAW264.7 cells were incubated with various treatment formulations for 4 h,followed by co-incubation with X-ray-pretreated 4T1 cells via the transwell apparatus for 24 h.As presented in Fig.3b and S5,X-ray irradiation combined with RC-nIMs exhibited the highest MHC-I expression among all groups,indicating that the combination of the X-rayinduced ICD of 4T1 cells and the immune modulation induced by RCnIMs successfully activated macrophages,which can potentially activate T cells [34].When treatment with R-nIMs was compared with treatment with free R837,an apparent increase in MHC-I expression was observed,implying that the nanocarrier-based delivery system can assist effective transportation to macrophages and enhance the effect of the drugs [35].

3.5.In vivo biodistribution of RC-nIMs

To realize the maximum RC-nIM accumulation within the tumor,4T1 orthotropic tumor-bearing mice were treated with Cy5.5-labeled(0.5 wt%) R-nIMs and RC-nIMs.Cy5.5 was used to visualize the location of the nIMs.After intravenous injection of the nIMs,the fluorescence intensity gradually increased and reached a maximum at 24 h,followed by a decrease at 48 and 72 h(Fig.4a).In the ex vivo imaging,the nIMs largely accumulated in liver and spleen due to the clearance of local macrophages [36].Fortunately,the RC-nIMs do not exhibit apparent cytotoxicity (Fig.S4),and the lipid-based carrier can be degraded by the biological system (Fig.S2).

3.6.In vivo antitumor effect of RC-nIM-assisted radiotherapy

The antitumor effect of nIMs was investigated by measuring tumor growth inhibition.The results are shown in Fig.5a.X-ray irradiation with the dosage of 15 Gy was applied at day 0,followed by daily intravenous injection of various drug formulations (1.4 mg/kg of R837 and 1.3 mg/kg of caffeine)for five doses.Both X-ray therapy alone and X-ray supplemented with free drugs elicited similar levels of tumor inhibition;a reduction of the tumor growth was observed compared with the PBS control (Fig.5b).Importantly,X-ray irradiation seems to dominate the therapeutic effect,implying that the insufficient therapeutic outcome of both drugs was caused by the lack of assistance provided by a drug delivery system.Interestingly,RC-nIMs had the best tumor inhibition ability among all treatment groups and performed significantly better than R-nIMs,suggesting that the caffeine can improve the therapeutic outcome of R837 by modulating the tumor microenvironment,possibly in association with T cells [16,18].The AUC of tumor growth inhibition curves and tumor weight (Fig.5c and d)indicate that the tumors treated with X-ray+RC-nIMs exhibited the lowest AUC and lightest tumor weight among all groups.This confirms that RC-nIMs can improve the therapeutic efficacy of radiotherapy.The effect of immunomodulation from the nIMs on RT was also quantitatively evaluated based upon the AUC values of the tumor growth profiles shown in Fig.5c.The enhanced RT index proposed herein is defined as the ratio of(AUCPBS–AUCtreatment)to(AUCPBS–AUCX-ray).The values of the enhanced RT index for various treatments are summarized in Table S1.The result illustrated that the combination of RC-nIMs with X-ray significantly increase the RT effect from 1.0 (X-ray only) to 2.4,suggesting that the immune modulation strategy developed in this study can effectively improve the outcome of radiotherapy.Furthermore,the H&E staining of lung tissue sections(Fig.S6)also showed the enhanced effect of the X-ray+RC-nIM treatment on inhibiting lung metastasis compared to the only X-ray treatment.Insignificant changes in body weight of mice receiving various treatments further indicate the absence of acute side effects (Fig.5e).Interestingly,an approximately threefold decrease in the amount of cyclic AMP between the PBS control group and the X-ray+RC-nIM group demonstrates the success of AR inhibition,which may improve the activation of CTLs[20](Fig.S7).To further identify the distributions of T cells and APCs,CD3 and CD11c cells were stained for CLSM imaging,as presented in Fig.5f.The obvious fluorescence signal from CD3+T cells was found in the Xray+R-nIM and X-ray+RC-nIM groups,and the quantified CD3 fluorescence intensity of the X-ray+RC-nIM group was significantly higher than the X-ray+R-nIM group.The APC fluorescence images also showed that the X-ray+RC-nIM group exhibited the highest fluorescence intensity.Additionally,as presented in Fig.S8,by analyzing the cell population within the tumor,the X-ray+RC-nIM group exhibited the highest level of CD3+CD8+cells,as well as CD11c+CD80+CD86+cells,indicating that the combination of radiotherapy and RC-nIM indeed enhanced the immune response to anticancer treatment.

Fig.4.Biodistribution of Cy5.5-labeled R-nIMs and RC-nIMs in 4T1 tumor-bearing mice with IVIS imaging.(a) In vivo and(b)ex vivo fluorescence imaging and the corresponding fluorescent quantification analyses (n=3).

Fig.5.The in vivo antitumor effect of nIM-assisted radiotherapy.(a) Schematic illustration of the treatment timeline.(b) Tumor growth profiles and (c) the area under the curves (AUC) of the tumor growth profiles of 4T1 tumor-bearing mice receiving various treatments.(d) Tumor mass at day 12 post-treatment.(e) Body weights of the tumor-bearing mice post-treatment.(f)Fluorescence imaging and its quantification in tumor sections harvested from the tumor-bearing mice at day 12 after treatment.Cell nuclei,CD3,and CD11c in tumor sections were stained and identified.Scale bars are 200 μm.

Furthermore,to evaluate the antitumor ability of the RC-nIM-assisted systemic immune response,tumor-bearing mice were inoculated with secondary tumors on day 5 following RT treatment (Fig.6a).The growth of the primary tumor in Fig.6b exhibited a similar trend,as compared with Fig.5b,suggesting that the influence of the secondary tumor inoculation was negligible.Although the primary tumor was not entirely eliminated by RT alone or RC-nIM-assisted RT,the tumor growth of the secondary tumor was apparently suppressed after receiving RC-nIM-assisted RT(Fig.6c).This result confirms that RC-nIMassisted RT can successfully facilitate the systemic immune response,hindering tumor metastasis.

Fig.6.Evaluation of secondary tumor growth inhibition.(a) Schematic illustration of experimental design and tumor growth inhibition profiles of (b) the primary tumor and (c) the secondary tumor on 4T1 tumor-bearing mice receiving various treatments (n=5).

4.Conclusion

This work demonstrates the successful development of a new nanoimmunomodulator,RC-nIM,which improves the therapeutic efficacy of RT.With in vitro and in vivo studies,it is revealed that RC-nIM can effectively induce M1 polarization of macrophages and activate them,as well as facilitate the accumulation of T cells and APCs within the tumor,resulting in the suppression of the growth of primary and secondary tumors.Although the tumor was not entirely eliminated,this study aims to provide a new strategy for improving current therapeutic options.Different therapeutic timelines or other kinds of adjunct therapeutic treatments should be considered for enhancing the antitumor outcome.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work is supported by the Ministry of Science and Technology,Taiwan (MOST 107-2221-E-007-032-MY3) and the National Tsing Hua University (109Q2707E1).

Appendix A.Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.pnsc.2020.09.014.