RGD Peptide and PAD4 Inhibitor-Loaded Gold Nanorods for Chemo-Photothermal Combined Therapy to Inhibit Tumor Growth, Prevent Lung Metastasis and Improve Biosafety

Introduction

Peptidyl arginine deiminase 4 (PAD4) is an important enzyme in the post-translational modification of histones.^1–3 It catalyzes the conversion of arginine from histone to citrulline (citrullination).^4,5 Since PAD4 is associated with many human diseases, such as cardiovascular diseases,⁶ ischemia-reperfusion (I/R),^7,8 neutrophil extracellular traps (NETs),⁹ rheumatoid arthritis (RA)¹⁰ and various tumors,^9,11 it has attracted the attention of many scientists. A number of reversible and irreversible inhibitors have been synthesized to inhibit the activity of PAD4 and thus prevent the occurrence of various diseases.^12–14 YW3-56 (356) was one of the inhibitors that has been shown to be able to suppress the citrullination of histone 3 (H3cit) and thus inhibit sarcoma growth.¹⁴ However, its rapid metabolism and the inherent biological toxicity of chemotherapeutic agents restrict the clinical development of 356.

In recent years, photothermal therapy (PTT) is attractive to more and more scientists.^15–18 It could target tumor sites accurately to reduce biological toxicity of systemic tissues, and laser irradiation from outside is a minimally invasive treatment.^19,20 Gold nanoparticles (NPs) have become an excellent photothermal transduction agent (PTA) because of their high photothermal conversion efficiency. Many types of NPs and their surface functionalized preparations have been used as PTA for the treatment of tumors,^21–27 including gold nanospheres (AuNPs)²⁸ and gold nanorods (AuNRs).^29–31 The ability of light-to-heat conversion could trigger tumor cell death through apoptosis or necrosis.^32,33 However, NPs may damage liver, kidney or other tissues because of the different accumulation level in organs and tumors. Many researchers have explored strategies to increase the tumor accumulation of PTAs, such as introducing targeting ligands to NPs³⁴ or changing the NPs size and shape.³⁵ RGD containing peptides are the ligand of integrin which is highly expressed on the surface of tumor cells.^36,37 Many drug-loaded systems or small molecules use RGD containing peptides as the target part of drug-loaded system to improve tumor targeting.^28,38–43 The peptide c(RGDfK) has been shown to interact with integrin receptor αVβ3 which is overexpressed on the surface of tumor cells.⁴⁴

Our aim is to construct a targeted drug delivery system that combines PAD4 inhibitors with PTT of NPs. Synergistic effects of chemotherapy and photothermal therapy could improve the biosafety of drug-loaded systems. Surface-functionalized gold nanospheres (AuNPs-c(RGDfK) and 356-AuNPs-c(RGDfK)) and gold nanorods (AuNRs-c(RGDfK) and 356-AuNRs-c(RGDfK)) were prepared. Their structures were confirmed by UV absorption spectra and transmission electron microscopy (TEM). The anti-tumor and anti-metastatic efficacies of 356-loaded NPs and 356 alone were compared by MTT and transwell assay with or without Matrigel in vitro. Confocal laser scanning microscopy (CLSM) and flow cytometry were used to determine the tumor permeability and uptake pathway of the drug-loaded system. In vivo, S180 sarcoma model was used to investigate the anti-tumor activity, and tumor sections were used to confirm tissue permeability and H3cit inhibition of 356-loaded NPs. 356-loaded gold nanorods were proved to have a better profile of tumor inhibitory activity and anti-metastatic activity compared with 356-loaded gold nanospheres. This may be related to the stronger cell and tissue permeability of 356-loaded gold nanorods. Thus, 356-loaded gold nanorods were selected to be applied for chemo-photothermal combined therapy on A549 tumor-bearing mice in vivo. The gold content in different organs was determined by ICP-MS. In addition, blood routine index and blood biochemical index were measured by a blood cell analyzer and blood biochemistry analyzer, respectively. They were all used to assess the biological toxicity of mice after administration. Besides, apoptosis in tumor tissue was observed by TUNEL assay and TEM pictures.

Patients and Methods

Materials

Please refer to the supporting information for the materials and reagents used. Methods are recorded in the supporting information as follow: synthesis of AuNPs and AuNRs, synthesis of YW3-56, UV-Vis absorption spectra, transmission electron microscopy (TEM), Zeta potential, in vitro growth inhibitory effect of NPs, transwell assay with or without Matrigel, cellular uptake of NPs, uptake pathway determination, immunofluorescence and immunohistochemistry, in vivo biosafety of chemo-photothermal combined therapy, cell apoptosis of tumor tissues in vivo.

Statistical Analysis

All data were expressed as mean ± SD and analyzed by one-way ANOVA using Prism 7.0 software. All values were obtained from at least three independent experiments. p < 0.05 was considered to be statistically significant and p < 0.01 was considered to be extreme significance.

Synthesis

For the synthesis and identifications of YW3-56 and gold nanoparticles, please refer to supporting information 2.2 section.

Synthesis of AuNPs-c(RGDfK) and AuNRs-c(RGDfK)

Citrate-stabilized gold nanoparticles (AuNPs) are commonly synthesized by reducing tetrachloroauric acid with trisodium citrate.³⁹ 0.5 mg HS-PEG-COOH (3.4 kDa) was added into 5 mL aqueous AuNPs (0.1 mg Au/mL) overnight. The AuNPs were centrifuged (11,000 rpm, 10 min) and washed with deionized water to remove the excess free PEG and the AuNP pellet was redispersed in 5 mL deionized water. No aggregation was observed. c(RGDfK) peptide was conjugated onto the carboxyl terminal of AuNPs-PEG-COOH via EDC/NHS coupling reactions. First, EDC⋅HCl (4.98×10⁻² mmol) and NHS (4.98×10⁻² mmol) were individually dissolved in 500 μL PBS buffer (pH=7.4), added to 1 mL of a AuNPs-PEG-COOH solution (0.5 mg Au/mL), and allowed to react 2 h in the dark at room temperature. The reaction solution was centrifuged (11,000 rpm, 10 min) and washed with PBS buffer to remove the excess free EDC/NHS and the AuNP pellet was redispersed in 1 mL PBS buffer. Subsequently, 1 mL c(RGDfK) peptide (4.99×10⁻³ mmol) was added to the reaction mixture, followed by stirring for 6 h in the dark at room temperature. The final product was purified by centrifugation (11,000 rpm for 10 min at 4 °C), redispersed in 1 mL deionized water, and stored at 4 °C in the dark. AuNRs-c(RGDfK) was synthesized following a similar procedure as AuNPs-c(RGDfK).

Synthesis of 356-AuNPs-c(RGDfK) and 356-AuNRs-c(RGDfK)

According to the literature,^38,39,45 compound YW3-56 was synthesized and purified. YW3-56 was loaded on the surface of gold nanoparticles by nanoprecipitation method.⁴⁶ Briefly, 5 mL AuNPs-c(RGDfK) (0.05 mg/mL) or AuNRs-c(RGDfK) (0.05 mg/mL) and 5 mL YW3-56 with different concentrations (0.001 mg/mL, 0.005 mg/mL, 0.01 mg/mL, 0.05 mg/mL and 0.1 mg/mL) were mixed under stirring. The mixture was stirred for another 2 h. After dialyzing in distilled water (to remove unloaded YW3-56), 356-AuNPs-c(RGDfK) or 356-AuNRs-c(RGDfK) was obtained and stored at 4 °C. The solution was concentrated to required drug concentration before use.

In vivo Activity

Female BLAB/C Nude mice and male ICR mice were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. The animal study in this work was approved by the Institutional Animal Care and Use Committee of Capital Medical University, and the ethics number is AEEI-2018-174. Humane care of animals was given in the animal studies followed the protocol and the Regulations on Laboratory Animal Welfare issued by Chinese Ministry of Science and Technology.

Sarcoma 180 (S180) cells of 4×10⁶ prepared from one tumor bearing mice were inoculated subcutaneously at right flank of male ICR mice. Treatments were started 5 days after cell implantation, when tumor in the mice reached average volume of 85~100 mm³. S180 bearing mice were randomly assigned to seven treatment groups, including normal saline (NS), DOX, YW3-56 (356), AuNPs-c(RGDfK), AuNRs-c(RGDfK), 356-AuNPs-c(RGDfK) and 356-AuNRs-c(RGDfK). The mice of the experimental animals were administered by intravenous (i.v.) injection once a day for one week. The treatment efficacy and safety evaluation were assessed by measuring the tumor volume and body weight, respectively. The tumor dimensions and body weights of mice were measured daily. Tumor volume was calculated by the following formula:

Tumor volume (V) = a^2*b/2 (where a and b are the shortest and the longest diameter, respectively, of a given tumor).

Female BALB/C Nude mice were used to evaluate the chemo-photothermal combined therapy in vivo. A549 cells were subcutaneously injected to form solid tumors. The tumors were implanted by injecting 0.1 mL of normal saline (NS) containing 5×10⁶ tumor cells under the skin into the right flank of mice. When the tumor size reached approximately 150 mm³ in volume (13 days after implantation), the mice were randomly divided into the following treatment groups: NS (i.v., 6 mice), 356 (i.v. dose: 10 μmol/kg/day, 6 mice), AuNRs-c(RGDfK) (i.v. dose: 1 mg/kg/day, 6 mice) and 356-AuNRs-c(RGDfK) (i.v. dose: 1 mg/kg/day NPs and 1 μmol/kg/day 356, 6 mice). The drugs were administered by intravenous injection every day. The mice were irradiated with 808 nm laser (1 w/cm²) the other day in 7 days after the beginning of the administration. At 30 min after administration, tumor was irradiated for 4 minutes. The weight and volume of mice were detected the other day. 24 h after the last injection, the mice were weighed and killed with ether anesthesia, and the organs and tumor were immediately obtained.

In vivo Biosafety of Chemo-Photothermal Combined Therapy

The tissues of A549 tumor-bearing mice less than 300 mg were digested in H₂O₂ and HNO₃ (1/1) using MARS5 Microwave digestion system. Then fixed volume of the digested liquid to 10 mL with water to determine the content of Au using Agilent Technologies 8800 ICP-MS Triple Quad.

Blood was taken from the eye. The blood used for blood routine examination was added to tubes containing EDTA. Then, blood routine examination was determined by a blood cell analyzer (URIT-5160Vet). The blood used for blood biochemistry was added to tubes without anticoagulant and centrifuged (3000 rpm for 10 min at 4 °C) to obtain the serum. Serum Crea, Urea, ALT and AST levels were measured with a chemistry analyzer and the appropriate reagents (BS-600, Mindray)

Results and Discussion

Characterization of AuNPs-c(RGDfK) and AuNRs-c(RGDfK)

To identify the synthesized gold nanoparticles and their modifier, UV-vis spectrophotometry, TEM and DLS were used. UV-vis absorbance spectra of AuNPs and its modifiers (Figure 1A) showed that the maximum absorption peak shifted from 527 nm to 531 nm after HS-PEG-COOH modification. The maximum absorption peak of AuNPs-c(RGDfK) also presented a slight red shift from 531 nm to 534 nm. And there was a characteristic absorption peak of c(RGDfK) around 258 nm, which was absent in spectrophotometry of AuNPs-PEG-COOH. UV-vis absorbance spectra of AuNRs and its modifiers (Figure 1B) showed that the maximum absorption peak of AuNRs was 758 nm. Same as the change of AuNPs, the horizontal plasmon resonance absorption (TSPR) of AuNRs does not change significantly, and the longitudinal plasmon resonance absorption (LSPR) has a slight red shift from 758 nm to 764 nm after HS-PEG-COOH modification. After cRGD conjugation, the maximum UV-vis absorption peak shifted to 769 nm. Newly emerging peaks 258 nm was also found in the spectrophotometry of AuNRs-c(RGDfK). These changes indicated successful conjugation of PEG and c(RGDfK) peptide.

Figure 1 Characterization of Gold nanospheres (AuNPs), nanorods (AuNRs) and their modifiers. (A) UV–vis spectra of AuNPs and its modifiers. (B) UV–vis spectra of AuNRs and its modifiers. (C) TEM images of AuNPs and its modifiers. (D) TEM images of AuNRs and its modifiers. (E) Zeta potential of AuNPs and its modifiers. (F) Zeta potential of AuNRs and its modifiers determined by DLS. (G and H) Changes in the UV–vis spectra of AuNPs and AuNRs interaction with different concentrations of YW3-56 (0.001 mg/mL, 0.005 mg/mL, 0.01 mg/mL, 0.05 mg/mL and 0.1 mg/mL). (I and J) TEM images of AuNPs and AuNRs with or without 0.1 mg/mL YW3-56. (K) Synthesis scheme of 356-AuNRs-c(RGDfK) and 356-AuNPs-c(RGDfK).

TEM images (Figure 1C–D) showed that AuNPs/AuNRs and their modifiers were well dispersed with a narrow size distribution. The AuNPs were globular in shape and 30 nm in diameter. The AuNRs exhibited a rod shape with an average aspect ratio of 4.2 (50 nm in length and 12 nm in width). The modification of AuNPs/AuNRs with PEG and conjugation of c(RGDfK) peptide did not change the morphology of the nanoparticles.

Figure 1E showed that the zeta potential of AuNPs was negative charge (−25.2 ± 0.6 mV) due to the use of sodium citrate during the colloidal gold synthesis.¹⁴ When the surface of AuNPs was modified with a long chain structure HS-PEG-COOH molecule and c(RGDfK), the surface maintained a negative value despite the decrease in zeta potential values. Figure 1F showed that the zeta potential of AuNRs was positive charge (20.2 ± 4.0 mV) due to the effect of the cationic nature of cetyltrimethylammonium bromide (CTAB).⁴⁷ When CTAB was replaced with HS-PEG-COOH, the zeta potential of AuNRs-PEG shifted in negative direction to −19.2 ± 0.1 mV due to the carboxyl end group of HS-PEG-COOH molecule dissociated in aqueous solution and showed negative potential. On conjugation with c(RGDfK), zeta potential of AuNRs- c(RGDfK) shifted a little in positive direction to −14.1 ± 1.5 mV due to slight positive charge of c(RGDfK). The shift in the zeta potential also indicated successful conjugation of PEG and c(RGDfK) peptide.

Characterization of 356-AuNPs-c(RGDfK) and 356-AuNRs-c(RGDfK)

To identify the loading of 356, UV-vis spectrophotometry and TEM were applied. Varying concentrations of YW3-56 (0.001 mg/mL, 0.005 mg/mL, 0.01 mg/mL, 0.05 mg/mL and 0.1 mg/mL) were loaded in AuNPs-c(RGDfK) and AuNRs-c(RGDfK). The characteristic peaks of AuNPs-c(RGDfK) and AuNRs-c(RGDfK) are around 520 nm and 810 nm, respectively. And the peaks around 330 nm was the characteristic peak of YW3-56. With the concentration of 356 increasing, the characteristic peaks of gold NPs were slightly weakened and occurred a red shift while the characteristic peak of YW3-56 was enhanced (Figure 1G and H). This suggested that 356-loaded level was not reached saturation. The loaded amount of 356 absorbed on the surfaces of AuNPs-c(RGDfK) and AuNRs-c(RGDfK) was by subtracting the equilibrium concentration of the drug outside the dialysis bag from the total drug concentration.⁴⁸ The concentration of 356 outside the dialysis bags was quantified by a HPLC method (Figure S1). Concentrations of free YW3-56 were below the minimum concentration of 7.8215 μg/mL, which was negligible. In addition, as shown in Table S1, Zeta potential of 356-loaded gold NPs kept stable when 0.025 mg/mL YW3-56 adding to 0.05 mg/mL gold NPs. Thus, we selected 356-loaded gold NPs (w/w = 2:1) to continue to be intensively studied. TEM images in Figures 1I and J and S2 showed that the loading of 0.1 mg/mL 356 did not change the morphology and size of the nanoparticles. As shown in Figure 1K, the driving force for the loading was electrostatic adsorption between 356 and the surfaces of gold nanoparticles.^49–51 In details, YW3-56 was positively charged in aqueous solution because of the nitrogen(+) of the N,N-dimethyl group, while gold nanoparticles were negatively charged as showed by their zeta potential. This led to the formation of YW3-56 and the gold nanoparticles via electrostatic bonding.⁵²

In vitro Growth Inhibitory Effect

The in vitro growth inhibitory effect of nanoparticles (NPs) on the S180, 95D, A549 and MCF-7 cancer cells was evaluated using the MTT assay. In order to ensure the comparability of data, the concentrations of 356 and gold nanoparticles in different groups were consistent. Specifically, 20 μM (10 μg/mL) 356 corresponded to 20 μg/mL gold nanoparticles. The cancer cells treated with of 20 μg/mL AuNPs-c(RGDfK) and AuNRs-c(RGDfK) showed 85% to 90% viability (Figure 2A–D). This suggested that the blank NPs were nontoxic at each of the tested concentrations.⁵³ Figure 2A–D indicated that the growth inhibitory effect of 356-AuNRs-c(RGDfK) and 356-AuNPs-c(RGDfK) on the S180, 95D, A549 and MCF-7 cancer cell was slightly higher than that of free 356. Figure 2E showed the IC₅₀ value of 356-AuNRs-c(RGDfK) on lung cancer cells showed 3 folds decrease in IC₅₀ value comparable to the pure drug while that of 356-AuNPs-c(RGDfK) showed 2 folds decrease. These results suggested that 356-loaded NPs could improve the growth inhibitory effect. However, there was no significant difference in terms of anti-tumor activity on S180 and MCF-7 cells for the two 356-loaded NPs.

Figure 2 In vitro activity. (A–D) Cell viability of S180, 95D, A549 and MCF-7 cells treated with different concentration nanodrugs. The concentration marked on the figure was that of 356 and 20 μM 356-loaded gold NPs were equal to 20 μg/mL gold NPs. (E) IC50 of 356, 356-AuNPs-c(RGDfK) and 356-AuNRs-c(RGDfK) on S180, 95D, A549 and MCF-7 cells. (F) Cell viability of control, AuNRs-c(RGDfK) and 356-AuNRs-c(RGDfK) on A549 cells with or without 808 nm Laser irritation for 1 min (2 w/cm2). (G–J) Representative images and analysis of anti-metastatic activity and anti-invasive activity. a: PBS, b: c(RGDfK), c: AuNPs-c(RGDfK), d: 356-AuNPs-c(RGDfk), e: AuNRs-c(RGDfK), f: 356-AuNRs-c(RGDfK). * represents p < 0.05, ** represents p < 0.01.

Thus, A549 cells was selected to evaluate the cytotoxicity of conjugated AuNRs with 808 nm laser irradiation. The viability of A549 cells exposure to 808 nm laser irradiation and incubation with 1.0 μg/mL AuNRs-c(RGDfK) have a significant decrease, while no decrease in cell viability was observed with laser irradiation or incubation with AuNRs-c(RGDfK) alone (Figure 2F). Although 356-AuNRs-c(RGDfK) could delay tumor growth with or without laser irradiation, there are still significant differences before and after laser irradiation. As shown in Figure S3, the ratio of dead cells increased significantly after laser irradiation compared with that without laser irradiation. This suggested that laser irradiation added to 356-AuNRs-c(RGDfK) could improve the anti-tumor activity of nanodrug in vitro. The stability of 356-AuNRs-c(RGDfK) was detected by UV-vis spectrum as shown in Figure S4. 2 w/cm² laser was irradiated for different time and the UV-vis spectra had no change during or after irradiation. And the UV-Vis spectra kept stable while the UV absorption peaks appeared to blue shift with pH decreasing. This indicated that 356-AuNRs-c(RGDfK) could keep stable with irradiation in physiological environment and could release 356 in the tumor environment.

Anti-Invasive and Anti-Metastatic Effect on A549 Cells

To determine the antimetastatic and anti-invasive ability of NPs in vitro, the transwell assay and Matrigel invasion assay were conducted. Figure 2G and H showed that the number of A549 cells passing through the transwell membrane pores decreased significantly after NPs incubation. And Figure 2I and J illustrated that 1.0 μg/mL NPs also inhibited the invasiveness of these cells. This could be due to the introduction of c(RGDfK).^43,54 The cells treated with 356-AuNRs-c(RGDfK) have higher antimetastatic ability compared with those treated with 356-AuNPs-c(RGDfK).

Cellular Uptake and Colocalization

As shown in Figure 3A and B, compared to control group, faint blue fluorescence (403/462) was detected in the cell area of 356 treated group, and indicated that 365 (4 μM) was not easy to penetrate through cell membranes to be taken up. However, the brighter blue fluorescence was observed in 356-AuNRs-c(RGDfK) and 356-AuNPs-c(RGDfK) group. This indicated that 356-AuNRs-c(RGDfK) and 356-AuNPs-c(RGDfK) entered A549 cells effectively.

Figure 3 The internalization and pathway of NPs into A549 cells. (A) Representative images showing cellular uptake of 356 and the H3cit protein colocalization with 356. (B and C) Analysis of histone H3 citrullination and uptake of 356 and 356-loaded NPs in A549 cells. (D) Uptake pathway determination of 356 and 356-loaded NPs. ** represents p < 0.01.

The amount of histone H3 citrullination was analyzed using anti-histone H3cit antibody (Alexa Fluor 647). The staining revealed strong nuclear labeling in the A549 cells treated with PBS, which was less diffuse in the groups of free 356 and NPs. Notably, 356-AuNRs-c(RGDfK) and 356-AuNPs-c(RGDfK) were able to significantly inhibit the histone H3 citrullination by analyzing the intensity of red fluorescence (Figure 3C). The results showed that 356 modification with NPs could significantly increase the inhibition of PAD4 activity. This might be attributed to the fact that drug-loaded NPs could deliver 356 more efficiently into the cells.

Uptake Pathway Determination

To further investigate potential endocytic mechanisms for 356-AuNRs-c(RGDfK) and 356-AuNPs-c(RGDfK), some selective inhibitors chlorpromazine (CPZ), amiloride hydrochloride (AM) and methyl-β-cyclodextrin (Mβ-CD) were used to check the effects on the uptake efficiency of NPs. First, none of these inhibitors affected the cell viability.⁵⁵ After pre-incubation with the individual inhibitor, cells were co-cultured with NPs for another 2 h. As shown in Figures 3D and S5, the uptake of 356 and 356-loaded NPs decreased slightly after the pre-treatment with chlorpromazine and amiloride hydrochloride, which could specifically inhibit the clathrin-mediated endocytosis and macrocytosis-mediated endocytosis, respectively. In the presence of a specific caveolin-mediated endocytosis inhibitor (Mβ-CD), the internalization efficiencies of 356 and 356-loaded NPs were markedly reduced. Specifically, the internalization of 356 and 356-loaded NPs exhibited an inhibition rate of approximately 80% in A549 cells. The results indicated that 356 entered the cell through specific caveolin-mediated endocytosis. The nanoparticles did not affect the way of uptake.

In vivo Anti-Tumor Activity on S180-Bearing Mice

The anti-tumor activity of NPs on the same dose (356: 10 μmol/kg, 356-loaded NPs containing 1 μmol/kg 356 and 1 mg/kg NPs, NPs: 1 mg/kg) was examined on S180-bearing mice model. Figure 4A showed that 356 and 356-loaded NPs could inhibit the growth of S180 sarcoma. And there was no significant inhibition effect in the AuNPs-c(RGDfK) and AuNRs-c(RGDfK) groups. This suggested that 356-loaded NPs showed enhanced anti-tumor activity compared to the NPs alone and the anti-tumor effect of 356-loaded NPs was consistent with 10-fold free drug 356.

Figure 4 In vivo anti-tumor activity on S180 tumor-bearing mice of 356 and 356-loaded NPs (YW3-56: 10 μmol/kg, 356-loaded NPs containing 1 μmol/kg 356 and 1 mg/kg NPs, NPs: 1 mg/kg). (A) The weight of S180 tumor. (B and C) Analysis of histone H3 citrullination and distribution of 356 and 356-loaded NPs in tumor tissues. (D) Representative pictures of H3 citrullination immunofluorescence and distribution of 356 and 356-loaded NPs. * represents p < 0.05, ** represents p < 0.01.

Distribution of 356 and 356-loaded NPs in Tumor Tissues and Its Effect on H3cit

The pictures in Figure 4D showed that 356 and 356-loaded NPs with blue fluorescence aggregated in the tumor tissue. The analysis in Figure 4C illustrated that the mean fluorescence intensity of 356-AuNRs-c(RGDfK) increased significantly in comparison with that of free drug 356. In addition, the expression of H3cit in tumor tissue sections was observed by analyzed the mean fluorescence intensity of red fluorescence area. Figure 4B showed that red fluorescence intensity in 356 and 356-loaded NPs groups all decreased and there was no significant difference between 10 μmol/kg 356 group and 356-loaded NPs groups (including 1 μmol/kg 356). This indicated that 356 and 356-loaded NPs could significantly inhibit the histone H3 citrullination in tumor tissues. This was consistent with the results in vitro.

In vivo Chemo-Photothermal Combined Therapy on A549 Tumor-Bearing Mice

In comparison with chemotherapy alone, the combined chemo-photothermal therapy exhibited a synergistic effect, resulting in even higher therapeutic efficacy.^56,57 Thus, NPs treatment combined 808 nm laser irradiation was conducted on BALB/C Nude mice implanted with A549 tumor. The tumor weight (Figure 5A) showed that 10 μmol/kg 356, AuNRs-c(RGDfK) and 356-AuNRs-c(RGDfK) were able to inhibit tumor growth significantly after 22 days combined treatment. Although 356-AuNRs-c(RGDfK) only contained 1 μmol/kg 356, it has higher therapeutic efficacy than 10 μmol/kg 356. During the treatment, the tumor volumes were monitored every other day (Figure 5B). Compared to the NS group, all groups could decrease the volumes. The mean tumor volume of 356-AuNRs-c(RGDfK) group on 2nd day after irradiation was significantly smaller than that of the saline group. And the difference in tumor volume between the other groups and control group was seen on the fifth day after irradiation. After 22 days treatment, the tumor volume of the chemotherapy group (356 group) was similar to that of the photothermal treatment group (AuNRs-c(RGDfK) group). However, they were both significantly larger than that of the combined treatment group (356-AuNRs-c(RGDfK) group). In addition, the mice weight exhibited normal increasing trends and there was no difference among all groups (Figure 5C).

Figure 5 Chemo-photothermal combined therapy of 356 and 356-loaded NPs (356: 10 μmol/kg, 356-AuNRs-c(RGDfK) containing 1 μmol/kg 356 and 1 mg/kg AuNRs-c(RGDfK), AuNRs-c(RGDfK): 1 mg/kg) (A) The weight of A549 tumor. (B) Tumor volume changes of the mice during the administration period. (C) Changes in body weight during the administration period. (D) The representative pictures showing temperature changes at the tumor sites during laser irradiation and average temperature changes of the tumor sites. Pictures were taken at the ninth day of drug administration. * represents p < 0.05, ** represents p < 0.01.

Temperature changes on tumor sites of mice were recorded and analyzed at various time points after laser irradiation (Figure 5D). After laser irradiation for 4 min, the tumor site temperature of NS group mice only increased by 9.1 °C, while that of AuNRs-c(RGDfK) and 356-AuNRs-c(RGDfK) groups mice increased by 15.7 °C and 18.9 °C, respectively. And no difference was observed between 356 group and NS group. In the 2nd minute of laser irradiation, the tumor site temperature of the mice treated with AuNRs-c(RGDfK) and 356-AuNRs-c(RGDfK) increased to more than 50 °C. These results indicated that the 356 loading of AuNRs-c(RGDfK) did not affect its photo-thermal conversion.

Lung Metastasis

Lung metastasis was observed by lung photographs and lung H&E staining from A549 tumor-bearing mice (Figure 6A and B). Lung photographs showed that there were macroscopically visible metastatic nodules on the lungs’ surface of the mice injected with normal saline. At the meantime, microscopically visible nodules could be observed by H&E staining. Furthermore, visible metastatic nodules did not be observed in 356 treatment group. However, a marked infiltration of inflammatory cells surrounding blood vessels was observed in lung H&E sections. Compared with the NS and 356 groups, lung tissue sections of mice in AuNRs-c(RGDfK) and 356-AuNRs-c(RGDfK) groups displayed normal alveolar architecture and no inflammatory cells.

Figure 6 (A) Representative photographs of lungs. (B) HE Staining for the lungs of tumor-bearing mice. (C) Gold content in heart, liver, spleen, lung, kidney, brain, tumor and blood tissues measured by ICP-MS. (D) The count of WBC and LYM determined by a blood cell analyzer. (E and F) Representative images and statistical analysis of apoptosis in tumor tissue determined by TUNEL assay. (G) Representative TEM pictures of tumor tissue. ** represents p < 0.01.

In vivo Biosafety of Chemo-Photothermal Combined Therapy

To evaluate the biosafety of these nanodrugs, Au accumulation in various tissues of A549 tumor-bearing mice was determined by ICP-MS (Figure 6C). After 356 loading, the Au content in heart, kidney and brain disappeared compared with the AuNRs-c(RGDfK) and that in liver decreased approximately to the one-third of Au content in AuNRs-c(RGDfK) group. However, the level of Au content in tumor tissues and lung tissues did not change a lot. This suggested that 356 loading reduced liver and kidney injuries caused by gold accumulation in other organs. This may be related to the rapid metabolism of 356. In addition, no obvious histopathological changes were observed in heart, kidney, liver and spleen section (Figure S6A).

Blood biochemistry and blood routine examination were also conducted to evaluate the biosafety of these nanodrugs in vivo. Figure 6D showed that 356 significantly decreased white blood cells (WBC) and lymphocyte (LYM) count. However, 356-loaded AuNRs-c(RGDfK) increased the level of these two parameters which may be due the decrease in the dose of chemotherapy. In addition, Figure S6B–D showed no significant difference between NS group and treatment groups in blood biochemistry parameters, the count of red blood cells (RBC) and platelet (PLT). This illustrated that the nanodrugs did not cause liver/kidney injury, hemolysis, and thrombus. These results were consistent with that previously reported.³⁹

Tumor-Suppressive Mechanisms of Chemo-Photothermal Combined Therapy in vivo

When the tumor site of mice received photothermal agents and NIR irradiation, apoptosis and necrosis were induced in tumor tissues.^58,59 TUNEL staining was used to detected apoptosis (Figure 6E and F). There were few apoptotic cells (TUNEL-positive cells) in NS and 356 groups. However, the number of apoptotic tumor cells in AuNRs-c(RGDfK) and 356-AuNRs-c(RGDfK) treatment groups were significantly increased compared with that in NS and 356 groups. This suggested that laser-induced apoptosis may be one of the reasons that photothermal therapy inhibited the growth of A549 tumors.

The apoptosis and necrosis of tumor tissues also were determined by TEM (Figure 6G). Ultrastructure of the normal tumor cells was complete, nuclear membrane was clear and mitochondrial structure was intact. Many autophagosomes (red arrows) were accumulated in the cytoplasm of tumor cells treated with 356, which was consistent with the literature reported previously.⁴⁵ The tumor tissues in AuNRs-c(RGDfK) treatment group demonstrated a marked appearance of necrosis while those in 356-AuNRs-c(RGDfK) treatment group occurred apoptosis. Cytoplasmic vacuolization and increasing mitochondria were observed in tumor tissues of AuNRs-c(RGDfK) and nuclear construction disappeared. Apoptotic bodies (green arrows) formed around the cell membrane and the chromatin accumulated along the nuclear membrane in tumor tissues of 356-AuNRs-c(RGDfK) treatment group. In addition, the gold nanoparticles were found to be distributed in the nucleus (blue arrows).

Through the above experiments, we found that 356-AuNRs-c(RGDfK) triggered apoptosis higher than necrosis while AuNRs-c(RGDfK) preferred to induce necrosis. This suggested that the introduction of 356 to AuNRs-c(RGDfK) may manage the scar formed around the treated tumor.

Conclusion

In this work, we constructed RGD targeting and PAD4 inhibitor loading gold nanospheres and nanorods. And the 356-loaded NPs had enhanced the cell killing ability in vitro compared with 356 alone or NPs alone. They also exhibited the ability to resist metastasis and invasion in vitro due to RGD functional modification of the NPs surface. Compared with the 356 alone group, the better penetration for tumor cells and tissues of 356-loaded NPs was improved. And the inhibition of histone citrullination in cells was significantly strengthened while the uptake pathway in cells was not affected.

We found that the activity of 356-AuNRs-c(RGDfK) were better than that of 356-AuNPs-c(RGDfK) in vitro and in vivo, including in vitro cell killing ability with or without laser irradiation, anti-metastatic activity, and tissue intake. Thus, 356-AuNRs-c(RGDfK) was selected to evaluate activity of chemo-photothermal combined therapy. Firstly, we used infrared cameras to find that 356 loading did not affect the photothermal conversion ability of gold nanorods in vivo. By monitoring the changes of tumor volume and tumor weight, we found that the anti-tumor effect of 356-AuNRs-c(RGDfK) was significantly enhanced and the dosage of 356 was reduced. Secondly, photographs of the lungs and analysis of lung H&E sections showed that 356-AuNRs-c(RGDfK) also inhibited tumor metastasis and inflammatory infiltration in the lungs. Thirdly, after 356 loading, the accumulation of gold in heart, liver, lung and kidney decreased or even disappeared, but did not affect the accumulation of gold in tumor. This may be due to the fact that rapid metabolism of 356 affects that of gold nanorods. In addition, blood routine test showed that 356 caused a decrease in white blood cells and lymphocytes, while 356-AuNRs-c(RGDfK) did not. It may possibly be related to a decrease in 356 doses. Finally, we found that both AuNRs-c(RGDfK) and 356-AuNRs-c(RGDfK) could cause apoptosis in tumor tissue. This was consistent with the research previously. However, how 356 affected the metabolism of gold nanorods requires further study. Overall, the combination therapy of 356-loaded NPs with PTT is a good strategy for addressing miss effects and biotoxicity of PAD4 inhibitors.