Soybean lecithin (90%) was obtained from Alfa Aesar (Ward Hill, MA, USA). DGS-NTA(Ni) was purchased from Avanti Polar Lipids (Birmingham, AL, USA). PLGA (lactide/glycolide (50:50), molecular weight=30,000–60,000) was purchased from Sigma Aldrich (St. Louis, MO, USA). ICG was acquired from Tokyo Chemical Industry (Tokyo, Japan).

First, TRHs were prepared via self-assembly of PLGA, lecithin, and DGS-NTA(Ni) through a single-step nanoprecipitation method. Briefly, PLGA was dissolved in chloroform; then lecithin/DGS-NTA(Ni) (8.5:1.5 molar ratio), containing a 15% wt ratio of PLGA, was dissolved in 4% aqueous ethanol solution; and ICG was dissolved in a mixture of chloroform/methanol (4:1 v/v%). The lecithin/DGS-NTA(Ni) solution was heated to 70°C to ensure clarity. The PLGA and ICG (20 w/w% of total lipids) mixture was added dropwise to the preheated aqueous solution under gentle stirring. The solution was vortexed for 3 min and then stirred gently for 4 hours at room temperature to produce TRH (10 mg/mL). HNPs were formulated without ICG. To prepare F-TRH, the FimH proteins (0.5 mg) were dispersed in the TRH solution with slight stirring for 2 hours to allow NTA(Ni) to conjugate with the His-tagged FimH proteins. Finally, the remaining free lipids, polymers, and FimH proteins were removed using an ultracentrifuge (10,000×g, 60 min). The encapsulation efficiency and loading capacity of ICG in the F-TRHs were determined to be 74.07%±0.51% and 4.3%±0.18%, respectively.

Transmission electron microscopy (TEM) images were obtained using an H-7600 transmission electron microscope (Hitachi, Tokyo, Japan). Fourier-transform infrared spectroscopy (FT-IR, Spectrum 100; Perkin Elmer, Waltham, MA, USA), conducted at the Core Research Support Center for Natural Products and Medical Materials in Yeungnam University, was used to observe the IR spectra of the lyophilized TRHs and F-TRHs. The loaded concentration of ICG in the TRHs was determined using a UV–vis spectrophotometer (Cary 100 Bio; Varian, Palo Alto, CA, USA). A fiber-coupled continuous-wave diode laser (808 nm, 10 W) was obtained from Changchun New Industries Optoelectronics Technology Co., Ltd (Jilin, China). Thermographic images were obtained, and temperature changes were measured using an FLIR ONE imaging system (FLIR Systems, Wilsonville, OR, USA).

The murine colon carcinoma cell line CT-26 (ATCC, CRL-2638; Korean Cell Line Bank, Seoul, Korea) and CT26.WT-iRFP-Neo cells (Imanis Life Sciences, CL091, Rochester, USA) were cultured in RPMI-1640 medium, supplemented with 1× penicillin/streptomycin and 10% fetal bovine serum (FBS) in a 5% CO₂ cell incubator at 37°C. Murine mammary carcinoma 4T1-Fluc-Neo/iRFP-Puro cells (Imanis Life Sciences, CL078) were cultured in RPMI-1640 medium, supplemented with 1× penicillin/streptomycin, 10% FBS, 0.1% G418, and 2 µg/mL puromycin, in a 5% CO₂ cell incubator at 37°C.

CT-26 cells (1×10⁵ cells) were cultured in 24-well plates. Six hours later, the cells were treated with phosphate-buffered saline (PBS), FimH, HNPs, TRHs, or F-TRHs and irradiated with an NIR laser (2 W/cm² for 5 min). The cells were stained with annexin V-FITC and 4′,6-diamidino-2-phenylindole (DAPI, Sigma-Aldrich) for 15 min at room temperature, 24 hours after treatment. The apoptotic cells were analyzed using flow cytometry (NovoCyte; ACEA Biosciences, San Diego, CA, USA).

BALB/c mice (female, 5–8 weeks old) were obtained from Hyochang Science (Daegu, Korea). The mice were kept under pathogen-free conditions at 20°C–24°C and 40%–60% humidity. In all experiments, efforts were made to minimize the suffering of the mice, and the mice were euthanized via CO₂ inhalation.

BALB/c mice were subcutaneously injected with 1×10⁶ CT-26 or CT-26-iRFP cells. Once the tumor size at the longest dimension reached approximately 5.0 mm on day 7, the mice were randomly separated into 10 groups: PBS, FimH, HNP, TRH, and F-TRH with or without NIR laser irradiation. After intratumorally injecting into the mice, the tumor site was irradiated with an NIR laser at 2 W/cm² power for 5 min. The elevated temperature was imaged using the FLIR One Thermal Imaging System (FLIR Systems). The tumor volume was calculated using the formula V ¼ 1/2 (longest dimension/2 shortest dimension).

iRFP expression in the mice was measured using a fluorescence in vivo imaging system, FOBI (Cellgentek, Cheongju, Republic of Korea), on day 10 after the first CT-26-iRFP challenge or day 12 after the second CT-26-iRFP challenge.

Mouse antibodies (Abs) and isotype control Abs (IgG1, IgG2a, and IgG2b), anti-CD3 (17A2), CD4 (GK1.5), CD8α (53–6.7), tumor necrosis factor (TNF)-α (MP6-XT22), perforin (S16009A), CD11c (N418), CD40 (3/23), CD80 (16-10A1), CD86 (GL-1), anti-interferon (IFN)-γ (B27), MHC class I (H2Kd, 28-8-6), and MHC class II (I-A/I-E, M5/114.15.2) were purchased from BioLegend (San Diego, CA, USA).

Mouse interleukin (IL)−6, IL-12p40, IFN-γ, and TNF-α ELISA kits were obtained from BioLegend. A mouse perforin/pore-forming protein (PF/PFP) ELISA kit was purchased from ABBKINE (Wuhan, China). A Mouse Granzyme B ELISA kit was obtained from Abcam (Cambridge, UK). The concentrations of TNF-α, IFN-γ, and perforin were measured in triplicate using ELISA kits according to the manufacturer’s protocol.

BALB/c mice were injected intravenously with CT-26 carcinoma cells (0.5×10⁶/100 µL of PBS). The survival of the mice was monitored for 25 days after cancer cell injection.

Lung, colon, and liver tissue samples were fixed in 4% formaldehyde and embedded in paraffin. The tissue paraffin blocks were then sectioned at 5 µm thickness and stained with H&E.

IFN-γ production in response to cancer Ag was measured using ELISPOTs according to the manufacturer’s protocol (BioLegend). In brief, capture Abs were coated on the plate and 5×10⁴ splenocytes were seeded in the wells. The cells were then treated with 10 µg/mL of cancer Ag and incubated at 37°C for 24 hours. The wells were stained with detection Abs, and the plates were counted automatically using a CTL ELISPOT reader (CTL Europe GmbH, Bonn, Germany).

To prepare the CT-26 lysate, 1×10⁷ CT-26 cells were washed with PBS and subjected to freezing and thawing to obtain a crude lysate. After centrifugation (10,000×g for 5 min at 4°C), the concentration of the CT-26 lysate was determined in the supernatant (Bradford assay), and the lysate was treated with 10 µg/mL of splenocytes. The CT-26 lysate was incubated with splenocytes for 6 hours. The cells were incubated with Fc block Abs and unlabeled isotype control Abs. After 20 min, the cells were incubated with surface Abs (FITC-anti-CD3, BV605-anti-CD4, and APC/Cy7-anti-CD8α) on ice for 20 min. After washing, the cells were fixed with fixing buffer and permeabilized (both from BioLegend) for 30 min. After washing again, the cells were stained with intracellular Abs (PE/Cy7-anti-IFN-γ, Alexa 647-anti-TNF-α, and PE-anti-Perforin) for 30 min. The cells were then washed and re-suspended in PBS and analyzed using NovoCyte (ACEA Biosciences) and NovoExpress software. Dead cells were excluded from the analysis using the Zombie Violet Fixable Viability Kit (BioLegend).

After labeling with 200 nM CFSE, the splenocytes were coated with 1 µg/mL CT-26 Ag. Other splenocytes were labeled with 10 mM CellTracker Orange CMTMR (Life Technologies) and coated with the control protein. The CFSE-labeled and CMTMR-labeled cells were mixed in a 1:1 ratio and the mixture was transferred into tumor-treated BALB/c mice. The spleen was harvested 6 hours after transfer, and the levels of CFSE and CMTMR in the splenocytes were analyzed using Novocyte (ACEA Biosciences).

Anti-CD4 (GK1.5, 1 mg/mL) or anti-CD8 (YTS169.4, 1 mg/mL) Abs (both from BioXcells, West Lebanon, NH, USA) were intraperitoneally injected every 2 days in the BALB/c mice from 25 days after the first tumor challenge (3 days before the second challenge of cancer cells). Cell depletion was confirmed using a Novocyte flow cytometer.

Data are expressed as the mean±SE of the mean. The p value was analyzed using SPSS (Chicago, Illinois, USA), and p values of <0.01 were considered to represent significant differences.

F-TRH, comprising a spherical core–shell lipid-polymeric hybrid nanoparticle with surface expression of FimH, was formulated by self-assembly and the single-step nanoprecipitation method wherein PLGA and ICG were dissolved in chloroform, and lecithin and DGS-NTA(Ni) were dispersed in an aqueous solution. The lipid mixture was added dropwise to either PLGA alone or a mixture of PLGA and ICG, under gentle constant stirring until it became clear, causing the precipitation of HNP or TRHs. During stirring, the lipids self-assembled around the PLGA core via hydrophobic interactions, resulting in the formation of TRHs. Subsequently, the His-tagged FimH proteins attached to the NTA-Ni²⁺ moiety of the TRHs via the NTA-Ni²⁺/His affinity interaction (figure 1A). Observation of a series of HNPs, TRHs, and F-TRHs, using TEM, revealed well-defined spherical structures, with the average diameters of three NPs being 89.42±0.86, 152.6±4.73, and 179.8±7.49 nm (figure 1B,C). To determine the stability of F-TRHs, they were incubated for 7 days at 4°C to examine whether ICG incorporated into the F-TRHs was released. F-TRH released approximately 26.87% of ICG by day 7 (online supplemental figure S1). The conjugation of the FimH proteins on the surface of the HNPs was analyzed using FT-IR spectroscopy. As shown in figure 1D, the spectrum of FimH conjugated to HNP presented peaks at 1596, 1406, and 1251 cm⁻¹, which corresponded to the amide groups, compared with that of HNP. Moreover, TRH and F-TRH were encapsulated in ICG, as confirmed from the UV–vis absorbance spectra, which showed a strong peak at 800 nm and a red shift, indicating that ICG was successfully trapped in the hydrophobic moiety of the PLGA core (figure 1E). These data indicate that F-TRHs are well-defined spherical nanoparticles incorporated with FimH and ICG.

The thermal responses of F-TRH were evaluated under laser irradiation at 808 nm for 5 min. Irradiation with NIR elicited an elevation in temperature in an F-TRH dose-dependent manner (figure 2A). The temperatures of 1.0, 2.0, and 5.0 mg/mL F-TRH reached 47, 60, and 70°C, respectively, within 5 min, while HNP did not show an increase in temperature against laser irradiation (figure 2A,B). Furthermore, the photothermal conversion efficiency of F-TRH was found to be 18.7%, which was higher than that of free ICG (12.03%), using a method described in a previous study (online supplemental figure S2).35 The photothermal effects of TRH and F-TRH were then examined for the induction of apoptosis in CT-26 carcinoma cells. Treatment with PBS, FimH, or HNP did not induce apoptosis in CT-26 cells with or without laser irradiation (figure 2C,D). TRH and F-TRH induced a remarkable increase in the apoptotic death of CT-26 cells after irradiation with the NIR laser, while the cells did not die without laser irradiation (figure 2C,D). These data indicate that TRH and F-TRH can induce apoptosis in cancer cells via the photothermal effect.

Since F-TRH could promote the apoptosis of CT-26 cells through the photothermal effect in vitro, we next examined whether F-TRH can induce anticancer effects in mice in vivo. Once the CT-26 tumor was established in the BALB/c mice, the mice were intratumorally treated with PBS, FimH, HNP, TRH, F-TRH, or NIR irradiation. As shown in figure 3A, NIR laser irradiation increased the temperature of the tumors of TRH-treated and F-TRH-treated mice, to 62°C and 64°C, respectively. However, the temperature of the tumors in mice treated with PBS, FimH, and HNP did not increase in response to NIR laser irradiation (figure 3A). Moreover, TRH-mediated PTT did not promote liver toxicity (online supplemental figure S3). On day 10 of tumor injection, the tumors in TRH-treated and F-TRH-treated mice showed a burnt tumor and the scar remained at the tumor site (figure 3B,C). On day 21, the tumor disappeared following the TRH and F-TRH treatment with laser irradiation, and the skin and hair were regenerated in the mice (figure 3D). In addition, TRH and F-TRH treatment with laser irradiation inhibited tumor growth, whereas PBS, FimH, or HNP treatment did not show any inhibitory effect on tumor growth (figure 3E). Furthermore, the mice treated with PBS, FimH, or HNP died within 28 days of tumor injection, while TRH-treated and F-TRH-treated mice survived without recurrence of the tumor (figure 3F). These findings suggest that TRH and F-TRH can treat tumors via PTT.

Lipid-based nanoparticles melt above the melting point of lipids.21 The NIR laser irradiation in F-TRH promoted FimH-conjugated lipid release from the F-TRH by increasing the temperature (figure 4A). Since FimH is an immune stimulatory recombinant protein, we next examined whether the released FimH can induce the activation of DCs in the tumor-draining lymph node (tdLN). DCs in the tdLN were defined as lineage^–CD11c⁺ cells, which were further divided into CD8α⁺ and CD8α^– DCs (figure 4B). Eighteen hours after F-TRH treatment and laser irradiation, expression levels of the co-stimulator and MHC class I and II increased significantly in both CD8α⁺ and CD8α^– DCs, which was almost similar to the results obtained with FimH treatment alone (figure 4C). In addition, F-TRH treatment induced the activation of splenic DCs, which controlled systemic immune responses (online supplemental figure S4). Furthermore, the levels of IL-6, IL-12, and TNF-α levels in the serum were substantially upregulated by F-TRH and laser irradiation (figure 4D). These results indicate that F-TRH and laser irradiation induced the activation of DCs in the tdLN and spleen of mice.

Our data, showing that F-TRH and laser irradiation induced the apoptosis of tumors and activation of DCs, prompted us to examine whether F-TRH and laser irradiation can elicit tumor Ag-specific immune responses. Therefore, we next assessed the second cancer challenge in mice that survived the first tumor challenge using F-TRH and NIR laser irradiation. On day 28 of the first tumor challenge, TRH-treated and F-TRH-treated mice were injected intravenously with CT-26 cells. As controls, the mice treated with PBS, FimH, and HNP were also inoculated intravenously with CT-26 cells (online supplemental figure S5). The F-TRH and NIR laser-mediated cured mice from the first tumor were protected from the second cancer challenge (figure 5A). However, the mice that were cured of the first tumor by TRH died within 21 days of the second tumor inoculation (figure 5A). The mice treated with PBS, FimH, and HNP died within 14 days of the second tumor injection (figure 5A). In addition, the second challenge CT-26 cells did not infiltrate the lungs of the mice that were protected from the first tumor challenge by F-TRH (figure 5B–D). The cancer cells infiltrated the lungs of mice treated with PBS, FimH, and HNP on day 12 after the second cancer cell injection (figure 5B–D). Although the infiltration of cancer cells was much less than that in the mice treated with PBS, FimH, and HNP, the lungs of the TRH-mediated first tumor-protected mice were infiltrated with the second challenge CT-26 cells 12 days after the second challenge (figure 5D). The first and second tumor-challenged mice, which were treated with F-TRH and NIR laser irradiation, did not show any inflammation in the kidney, colon, and liver on day 49 after the first tumor challenge (online supplemental figure S6).

Since F-TRH had a protective effect against the second challenge of CT-26 tumors, we further examined whether F-TRH can protect mice from different types of cancers, such as the 4T1 breast cancer. Consistent with CT-26 treatment, TRH and F-TRH treatment with NIR irradiation abolished the first challenge of 4T1 tumor growth in mice (online supplemental figure S7A,B). In addition, the second challenge of 4T1 cell infiltration in the lungs was also completely prevented in the mice that were cured by F-TRH treatment, whereas TRH-mediated cured mice showed 4T1 cell infiltration in the lungs (online supplemental figure S7C). These data suggest that F-TRH protects from the first tumor and also prevents the second challenge from inducing lung metastatic cancer growth in mice.

Our finding that F-TRH induces the activation of DCs and protects against second challenged cancer prompted us to examine whether F-TRH can elicit cancer Ag-specific immunity to protect mice from second challenged cancer. The splenocytes of F-TRH-treated mice showed a significant increase in the production of IFN-γ in response to CT-26 cell lysate, as shown using ELISPOT analysis, while other control splenocytes did not produce IFN-γ (figure 6A). In addition, the concentrations of TNF-α, IFN-γ, and perforin in the culture medium were significantly elevated in the splenocytes from F-TRH mice in response to CT-26 cell lysate, whereas other control splenocytes did not produce cytokines (figure 6B). Moreover, stimulation with CT-26 antigen induced intracellular production of IFN-γ, TNF-α, and perforin in CD4 and CD8 T cells in F-TRH-treated splenocytes (online supplemental figure S8). Next, we examined Ag-specific cytotoxicity and found that, in mice cured of the first tumor by F-TRH treatment, CT-26 Ag-coated splenocytes were selectively killed (figure 6C). Other controls, as well as the TRH-mediated first tumor-cured mice, failed to kill CT-26 Ag-coated splenocytes (figure 6C). Furthermore, the protective effect of F-TRH on second cancer challenge in mice diminished with the depletion of CD4 or CD8 cells (figure 6D). These findings suggest that the F-TRH-induced protective effect against the second challenge of cancer is mediated by T-cell immunity.