Supplementary Figure 1. Fullerence has poor solubility in water while the C60S and LC60S nanoparticles can be stably dispersed in water.

Transcription

1 Supplementary Figure 1. Fullerence has poor solubility in water while the C60S and LC60S nanoparticles can be stably dispersed in water. (a) A typical photograph of fullerence (C60), C60S nanoparticles, and LC60S nanoparticles in deionized water (2 mg ml -1 ). The C60S and LC60S nanoparticles can be well dispersed in deionized water while fullerene forms aggregates and sinks down at the bottom of the cuvette. (b) A typical photograph of the aqueous solutions of C60, C60S nanoparticles, and LC60S nanoparticles after shining a red laser beam (arrow) through them in dark. As a result of the Tyndall effect (i.e., scattering of laser beam by nanoparticles in solution), a light track (indicated by asterisks) is visible in dark in the solutions of C60S and LC60S nanoparticles. This indicates both the C60S and LC60S nanoparticles can be stably dispersed in deionized water while the aqueous solubility of C60 is negligible. 1

2 Supplementary Figure 2. Fourier transform infrared spectroscopy data showing successful synthesis of the C60S and LC60S nanoparticles. The Fourier transform infrared spectroscopy (FTIR) spectra of C60, C60S nanoparticles, and LC60S nanoparticles showing (1) the presence of C60 (indicated by the C-C bond at 1429 cm -1 ) and silica (indicated by the Si-O-Si bond at 788 cm -1 ) in the C60S nanoparticles and (2) the presence of C60, silica, and DPPC phospholipid (indicated by the C-H bond at 2900 cm -1 ) in the LC60S nanoparticles. It is worth noting that although the peak of fullerene at 1429 cm -1 is evident in the spectra of C60S and LC60S nanoparticles, the peak at 526 and 575 cm -1 of fullerence disappears and shows up as a tiny peak, respectively, in the spectra of the two nanoparticles. Similar phenomena were reported in the literature on the FTIR spectrum of fullerene modified silica surface (Masayoshi et al., J. Mater. Chem., 2003, 13, and Lee et al., Environ. Sci. Technol. 2010, 44, ). Therefore, there might be some interactions between fullerene and silica that cause the change to the two peaks at 526 and 575 cm -1 when combining fullerene and silica together. 2

3 Supplementary Figure 3. Characterization of the morphology and size of the C60S-A nanoparticles with electron microscopy. (a) Transmission electron microscopy (TEM) image of C60S-A nanoparticles together with an inset showing the core-shell structure of a fullerene core (dark appearance) embedded in mesoporous silica in the nanoparticles. (b) Scanning electron microscopy (SEM) of C60S-A nanoparticles showing a round morphology and homogeneous size distribution. 3

4 Supplementary Figure 4. Ultra-high encapsulation efficiency and loading content of DOX with the LC60S nanoparticles. (a) encapsulation efficiency (EE) and (b) loading content (LC) with the LC60S nanoparticles at six different feeding ratios (DOX:nanoparticle = 1:20, 1:10, 1:5, 1:1, 1:0.5, and 1:0.25). The encapsulation was achieved by simply mixing the nanoparticles with DOX at room temperature for 30 min. Error bars represent S.D. (n = 3). 4

5 Supplementary Figure 5. Spontaneous release of DOX from LC60S-D nanoparticles is minimal in phosphate buffered saline at ph 7.4 and 37 C for 96 h. The minimal spontaneous release allows for the opportunity to control the drug release specifically in tumor or other diseased tissue. The ratio of DOX to LC60S nanoparticle for making the LC60S-D nanoparticles was 1:20. Error bars represent S.D. (n = 3). 5

6 Supplementary Figure 6. FTIR spectra showing successful encapsulation of DOX and ICG in nanoparticles and the laser irradiation controlled release of DOX. The peak at 1581 cm -1 is due to the stretching vibration of the two carbonyl groups of the anthracene ring in DOX. The peak at 1429 cm -1 is referred to C=C ring of ICG. The peak at 1581 cm -1 of DOX is visible in the LC60S-D and LC60S-DI nanoparticles, indicating successful encapsulation of DOX in them. The peak at 1429 cm -1 of ICG is visible in the LC60S-DI nanoparticles, but not the LC60S or LC60S-D nanoparticles. In addition, a peak at 925 cm -1 in the LC60S-DI nanoparticles becomes less evident after NIR laser (L) irradiation (1 min at 1.5 W cm -2 ) to partially release DOX, suggesting this unique peak in the LC60S-DI nanoparticles is due to the interaction between DOX and ICG in the nanoparticles. The ratios of DOX to LC60S nanoparticle and DOX to ICG for making the DOX and/or ICG-laden nanoparticles were 1:20 and 1:1, respectively. 6

7 Supplementary Figure 7. Ultra-high encapsulation efficiency and loading content of DOX and ICG with the C60S nanoparticles. (a) encapsulation efficiency (EE) and (b) loading content (LC) with the C60S nanoparticles at two different feeding ratios (DOX:nanoparticle = 1:20 and 1:0.5). (c) EE of ICG and (d) loading content of both DOX and ICG at the two feeding ratios of DOX:nanoparticle with the feeding ratio of DOX to ICG being 1:1. The encapsulation was achieved by simply mixing the nanoparticles first with DOX and then ICG at room temperature for 30 min each. Error bars represent S.D. (n = 3). 7

8 Supplementary Figure 8. Ultra-high loading content of theranostic agents in the LC60S-DI nanoparticles revealed by electron microscopy. (a) TEM image of LC60S-DI nanoparticles made from a low drug feeding ratio (1:20) of DOX to LC60S nanoparticles with a low LC of the DOX and ICG. These LC60S-DI nanoparticles retain the core-shell appearance of the LC60S nanoparticles without any agents under TEM. (b) TEM images of LC60S-DI nanoparticles made from a high drug feeding ratio (1:0.5) of DOX to LC60S nanoparticles with a high LC of DOX and ICG. The core-shell configuration of these LC60S-DI nanoparticles is not evident under TEM and they have a much darker appearance than the LC60S-DI with a low LC of the agents, due to the high content of DOX and ICG that could be negatively stained and visualized under TEM. In order to compare the difference of the LC60S-DI nanoparticles with different loading content of DOX and ICG, we didn t use uranyl acetate solution to stain the nanoparticles to avoid the strong artificial background because the two agents could also be stained by uranyl acetate. Therefore, the phospholipid membrane is not clearly visible in the TEM image of the LC60S-DI nanoparticles with a low drug loading content. However, it is more evident in the LC60S-DI nanoparticles with a high drug loading content, probably due to the high contrast induced by the darkened silica matrix as a result of the high drug loading content. (c) SEM image of LC60S-DI made from the high feeding ratio of 1:0.5 (DOX:nanoparticle) showing a round morphology and uniform size distribution. The ratio of DOX to ICG for making the LC60S-DI nanoparticles was 1:1. 8

9 Supplementary Figure 9. Surface zeta potential of LC60S-DI nanoparticles in deionized water, medium, and blood showing the LC60S-DI nanoparticles stay negatively charged after drug encapsulation. The ratios of DOX to LC60S nanoparticles for making the LC60S-DI nanoparticles are 1:20 (a) and 1:0.5 (b). The ratio of DOX to ICG was 1:1. 9

10 Supplementary Figure 10. Free ICG is not stable in deionized water after 5 days and either mixing ICG with DOX or encapsulating ICG and DOX in the LC60S-DI nanoparticles improves the aqueous stability of ICG. Data on UV-Vis absorbance of free DOX, free ICG, a simple mixture of DOX and ICG (DOX&ICG), and LC60S-DI nanoparticles of different concentrations on five different days are shown. DOX is stable in deionized water for all the formations during the five days. The ratios of DOX to LC60S nanoparticle and DOX to ICG for making the LC60S-DI were 1:20 and 2:1, respectively. 10

11 Supplementary Figure 11. Spontaneous release of DOX from LC60S-DI nanoparticles is minimal in phosphate buffered saline (PBS) at ph 7.4 and 37 C for 96 h. The nanoparticles were made by encapsulating DOX and ICG (1:1) into the C60S nanoparticles first and then coated with the phospholipid membrane. Error bars (small and barely visible) represent S.D. (n = 3). 11

12 Supplementary Figure 12. Release of DOX from C60S-DI nanoparticles is ph dependent and can be controlled with NIR laser irradiation. The NIR laser irradiation indicated by arrows at three different times was at 1.5 W cm -2 for 1 min. The ratios of DOX to C60S nanoparticle and DOX to ICG for making the C60S-DI were 1:20 and 1:1, respectively. Error bars (small and barely visible) represent S.D. (n = 3). 12

13 Supplementary Figure 13. The photothermal effect of LC60S-DI nanoparticles is dependent on the concentration of ICG and the power and duration of NIR laser irradiation. (a) The irradiation time and ICG concentration dependent increase of temperature in bulk PBS and medium and in the bulk solutions of LC60S, LC60S-D, and LC60S-DI nanoparticles in PBS with 0-4 µg ml -1 ICG when laser irradiation power is 0.7 versus 1.5 W cm -2. It is worth noting that in this work, the laser irradiation power of 1.5 W cm -2 was used for in vitro cell studies while a lower power of 0.7 W cm -2 was used for all in vivo studies. This is because for in vitro studies, the baseline temperature was 22 C (room temperature) while for in vivo studies, the starting temperature was 37 C. (b) Higher temperature increase could be achieved by using higher ICG concentration (in PBS), which could be used for selective photothermal therapy to treat cancer. The ratios of DOX to LC60S nanoparticle and DOX to ICG for making the DOX and/or ICG-laden nanoparticles were 1:20 and 1:1, respectively. 13

14 Supplementary Figure 14. Data on DOX release from the LC60S-D and LC60S-DI nanoparticles showing a threshold temperature of ~70 C to trigger the drug release. (a) No triggered drug release from LC60S-D nanoparticles with NIR laser irradiation (indicated by arrow), due to the absence of ICG in the nanoparticles and lack of significant photothermal effect. (b) No triggered drug release from the LC60S-DI nanoparticles when heated at 37 C and 45 C in hot water bath for a total of 5 h. Triggered drug release from the LC60S-DI nanoparticles is noticeable when heated at 50 C (c), 60 C (d), 70 C (e), and 75 C (f) in hot water bath for 30 min. The ratios of DOX to LC60S nanoparticle and DOX to ICG for making the DOX and/or ICG-laden nanoparticles were 1:20 and 1:1, respectively. Error bars (small and barely visible) represent S.D. (n = 3). The NIR laser irradiation was for 3 min at 1.5 W cm

15 Supplementary Figure 15. A schematic illustration of the mechanisms for loading DOX and ICG with the LC60S nanoparticles and controlling the drug release from the nanoparticles using NIR laser irradiation and high temperature. (a) DOX can be loaded into the LC60S nanoparticles in 30 min with ultra-high encapsulation efficiency (EE) and loading content (LC) as a result of electrostatic interaction between DOX and silica and π-π stacking interaction between DOX and fullerene. Subsequently, ICG can be loaded into the DOX-laden nanoparticles in 30 min with ultra-high EE and LC as a result of the electrostatic interaction between ICG and DOX. (b) NIR laser irradiation-controlled release of DOX from the LC60S-DI nanoparticles, but not the LC60S-D nanoparticles due to the lack of ICG-mediated photothermal effect. However, release of DOX from the LC60S-D nanoparticles can be triggered by heating to a temperature higher than ~70 C. 15

16 Supplementary Figure 16. Cytotoxicity and cellular uptake of C60S nanoparticles for delivering theranostic agents. Viability of (a) PC-3 and (b) MDA-MB-231 cancer cells after treated with C60S, C60S-D, and C60S-DI nanoparticles and free DOX at different concentrations with and without laser irradiation. Error bars represent S.D. (n = 3). (c) DIC and confocal fluorescence images showing the cell morphology and intracellular distribution of free DOX, free ICG, C60S-D nanoparticles, C60S-DI nanoparticles, and C60S-DI nanoparticles with NIR laser irradiation (C60S-DI+L). DIC: differential interference contrast. The ratios of DOX to C60S nanoparticle and DOX to ICG for making the DOX and/or ICG-laden nanoparticles were 1:20 and 1:1, respectively. The NIR laser irradiation was at 1.5 W cm -2 for 3 min. 16

17 Supplementary Figure 17. The LC60S-DI nanoparticles without NIR laser irradiation are significantly less cytotoxic compared to C6S-DI nanoparticles and free DOX. Viability of (a) PC-3 and (b) MDA-MB-231 cancer cells after treated with LC60S-DI nanoparticles, C60S-DI nanoparticles, and free DOX. Error bars represent S.D. (n = 3). *: p < (Mann-Whitney U test). The ratios of DOX to empty nanoparticle and DOX to ICG for making the DOX and ICG-laden nanoparticles were 1:20 and 1:1, respectively. 17

18 Supplementary Figure 18. Cellular uptake of LC60S-DI nanoparticles and their distribution inside PC-3 and MDA-MB-231 cancer cells observed by confocal microscopy. The cells were cultured on 35 mm dishes with a glass surface for 24 h, followed by incubating first with LC60S-DI for 3 h at 37 C and then LysoTracker Green to stain late endosomes/lysosomes for 30 min at 37 C before examination using confocal microscopy. The ratios of DOX to LC60S nanoparticle and DOX to ICG for making the DOX and/or ICG-laden nanoparticles were 1:20 and 1:1, respectively. 18

19 Supplementary Figure 19. A higher DOX loading content (LC) renders more DOX release from the LC60S-DI nanoparticles into cells under NIR laser irradiation of the same dose. Typical images of PC-3 cells after incubating them with LC60S-DI nanoparticles made from different feeding ratios of DOX to nanoparticles (1:20, 1:5, and 1:0.5) for different times (3, 6, and 12 h) at 37 C. The cells were treated with NIR laser at 1.5 W cm -2 for 1 min immediately before imaging. The ratio of DOX to ICG for making the LC60S-DI nanoparticles was 1:1. The concentration of DOX was 10 g ml

20 Supplementary Figure 20. The photodynamic effect of LC60S and LC60S-DI nanoparticles examined by the production of singlet oxygen in PBS under NIR laser irradiation. (a) The production of singlet oxygen by the LC60S nanoparticles with and without NIR irradiation as compared to deionized water. (b) The production of singlet oxygen by the LC60S-DI versus LC60S-D nanoparticles with and without NIR irradiation. The ratios of DOX to LC60S nanoparticle and DOX to ICG for making the DOX and/or ICG-laden nanoparticles were 1:20 and 1:1, respectively. The NIR laser irradiation was at 1.5 W cm -2 for 1 min. 20

21 Supplementary Figure 21. The photodynamic effect of LC60S, LC60S-D, and LC60S-DI nanoparticles examined by the production of reactive oxygen species (ROS) in cancer cells under NIR laser irradiation. The ROS in MDA-MB-231 cancer cells (a) quantified using a PerkinElmer VICTORX3 Multilabel microplate reader and (b) detected using fluorescence microscopy. Error bars represent S.D. (n = 3). *: p < (Kruskal-Wallis H test). The ratios of DOX to LC60S nanoparticle and DOX to ICG for making the DOX and/or ICG-laden nanoparticles were 1:20 and 1:1, respectively. The NIR laser (L) irradiation was at 1.5 W cm -2 for 3 min. 21

22 Supplementary Figure 22. LC60S-DI nanoparticles either with a low or high drug loading content (LC) enhance drug delivery to tumor in vivo. In vivo whole animal imaging of ICG at 12 h after intravenous injection via the tail vein in the form of LC60S-DI nanoparticles made from both low (1:20, DOX:nanoparticle) and high (1:0.5) drug feeding ratios together with ex vivo imaging of ICG in the tumor and five different organs collected after sacrificing the mice at 12 h. The ratio of DOX to ICG for making the LC60S-DI nanoparticles was 2:1. Both LC60S-DI nanoparticles can accumulate in tumor with only slight difference in their biodistribution. We attribute the slight difference in their biodistribution mainly to the difference in 1) the total number of nanoparticles and 2) loading content of ICG in each nanoparticle given the same total dose of ICG administered intravenously. 22

23 Supplementary Figure 23. The C60S-DI nanoparticles without lipid coating show significant systemic toxicity probably due to their poor stability in blood, which can be overcome by using the LC60S-DI nanoparticles. (a) Animal survival showing that the treatment with C60S-DI nanoparticles killed 50% mice in two days after the drug injection through the tail vein. No mice died after treating with saline, LC60S nanoparticles, LC6S-DI nanoparticles, LC60S-D nanoparticles with NIR laser irradiation (LC60S-D+L), and LC60S-DI nanoparticles with NIR laser irradiation (LC60S-DI+L). (b) Photograph of C60S-DI and LC60S-DI nanoparticles in mouse blood plasma (10% in PBS) after incubating at room temperature for 7 days. The C60S-DI nanoparticles easily form aggregates and sink down at the bottom of the cuvette in 5 h while the LC60S-DI nanoparticles are stable in blood for at least 7 days. The ratios of DOX to empty nanoparticle and DOX to ICG for making the DOX and/or ICG-laden nanoparticles were 1:20 and 2:1, respectively. The NIR laser (L) irradiation was at 0.7 W cm -2 for 3 min. 23

24 Supplementary Figure 24. No evident toxicity to normal organs is observable for the LC60S-DI nanoparticles in vivo. Representative hematoxylin&eosin (H&E) stained tissue of various organs in mice after treated with saline, the simple mixture of free DOX and ICG (DOX&ICG), and LC60S-DI nanoparticles. Systemic toxicity of the free drug is not evident either because only one injection at a low dose (2.5 mg per kg body weight) was applied to the animals. The ratios of DOX to LC60S nanoparticle and DOX to ICG for making the LC60S-DI nanoparticles were 1:20 and 2:1, respectively. 24

25 Supplementary Figure 25. A schematic illustration of the mechanisms of the eukaryotic cell-like nanoplatform (EukaCell) for enhancing drug delivery to improve the safety and efficacy of cancer therapy. The combination of silica and fullerence enables efficient encapsulation of DOX and ICG with the EukaCell. The biomimetic (i.e., eukaryotic cell-like) configuration with a phospholipid membrane renders the EukaCell prolonged half-life and robust stability in blood circulation. Due to its nanoscale size (~60 nm), the EukaCell can preferentially accumulate in tumor rather than normal tissue as a result of the enhanced permeability and retention (EPR) effect of tumor but not normal vasculature. NIR laser irradiation can be used to precisely control the drug release specifically in tumor, leading to effective tumor destruction as a result of the combined chemotherapeutic, photothermal, and photodynamic effects. Moreover, the ultra-high drug loading content further enhances the safety and efficacy of cancer therapy by enabling more drug release from the nanoparticles with a shorter laser irradiation and allowing for much reduced amount of empty nanoparticles being used to deliver the drug at a given dose in vivo. 25

26 Supplementary Note 1 List of Abbreviations: EukaCell: Eukaryotic cell-like hybrid nanoparticle platform LC: Loading content EE: Encapsulation efficiency C60: Fullerene C60 DPPC: 1,2-dihexadecanoyl-sn-glycero-3-phosphocholine NIR: Near infrared PTT: Photothermal therapy PDT: Photodynamic therapy DOX: Doxorubicin hydrochloride ICG: Indocyanine green Silica: SiO 2 TEOS: Tetraethyl orthosilicate APTMS: (3-Aminopropyl) trimethoxysilane C60S: Fullerene-embedded silica nanoparticles C60S-D: DOX-laden C60S C60S-DI: DOX&ICG-laden C60S C60S-A: APTMS-coated C60S LC60S: Lipid-coated C60S LC60S-D: DOX-laden LC60S LC60S-DI: DOX&ICG-laden LC60S FBS: Fetal bovine serum FTIR: Fourier transform infrared spectroscopy TEM: Transmission electron microscopy SEM: Scanning electron microscopy PBS: Phosphate buffered saline ROS: Reactive oxygen species DCFH-DA: Dichloro-dihydro-fluorescein diacetate IACUC: Institutional Animal Care and Use Committee H&E: Haematoxylin&eosin SD: Standard deviation 26