SUPPORTING INFORMATION

Programmed degradation of a hierarchical nanoparticle with redox and light responsivity for self-activated photo-chemical enhanced chemodynamic therapy

Shenqiang Wang a,b,1, Letao Yang b,1, Hyeon-Yeol Cho b, Sy-Tsong Dean Chueng b, Hepeng

Zhang a, Qiuyu Zhang a* and Ki-Bum Lee b,c*

S. Wang, H. Zhang, Prof. Q. Zhanga MOE Key Laboratory of Material Physics and Chemistry under Extraordinary Conditions, School of Natural and Applied Sciences, Northwestern Polytechnical UniversityXi'an 710129, China.Email: qyzhang@nwpu.edu.cn, Tel: +86-29-8843-1653

Dr. L. Yang, Dr. H.-Y. Cho, Dr. S.T.D. Chueng, Prof. K.-B. Leeb Department of Chemistry and Chemical, Rutgers University, Piscataway, NJ 08854, USAEmail: kblee@rutgers.edu, Tel: +1-848-445-2081

Prof. K.-B. Leec Department of Life and Nanopharmaceutical Science, College of Pharmacy, Kyung Hee University, Seoul 02447, Korea

(1 These authors have contributed equally to this manuscript)

KEYWORDS: Programmed degradation; Tumor microenvironment; Hierarchical Nanoparticle; Self-activated Chemodynamic therapy; Photothermal therapy; Drug Delivery

S1

TABLE OF CONTENTSA. Methods and Experimental details

B. Supplementary (SI) Figures, Tables

Figure S1. Characterization of the hollow structure of Fe3O4-C nanoparticles

Figure S2. Precise size control of Fe3O4-C hollow core

Figure S3. Structural characterization of (redox and light responsive) RLR nanoparticles

Figure S4. Precise shell thickness control of RLR nanoparticles

Figure S5. Biodegradability of RLR nanoparticles

Figure S6. In vitro biodegradation monitoring of RLR nanoparticles

Figure S7. Iron ion induced oxidative stress via Fenton reaction

Figure S8. RLR nanoparticle-induced ferroptosis pathways.

Figure S9. Photothermal performance of RLR and Fe3O4-C nanoparticles

Figure S10. The mechanism of photothermal hyperthermia-mediated cell apoptosis

Figure S11. The generation of ROS (H2O2) induced by DOX

Figure S12. Cytotoxicity analysis of RLR nanoparticles

Figure S13. The therapeutic universality of RLR nanoparticles

Figure S14. Synergistically overcoming drug-resistance of MDA-MB-231 cells

Figure S15. Schematic illustration of iRGD conjugation

Figure S16. In vitro MRI imaging capacity

Figure S17. In vivo real-time MRI monitoring the degradation of RLR nanoparticles

Figure S18. In vivo photothermal setup

Figure S19. RLR nanoparticle-based upregulation of pro-apoptotic genes in vivo

Figure S20. Tumor H&E staining

Figure S21. In vivo compatibility assay.

Figure S22. Histological analysis of major organs

Table S1. Summary of the primers used for quantitative PCR

Table S2. IC50 of DOX on MDA-MB-231 cells

S2

A. Methods and Experimental details

Characterization

The morphologies of the samples were observed using scanning electron microscopy (SEM,

SIGMA, Zeiss, Germany) and transmission electron microscopy (JEOL JEM-2010F high-

resolution TEM). The hydrodynamic diameter and zeta potential of the nanoparticles were

characterized using a Nano-ZS 90 Nanosizer (Malvern Instruments Ltd.) The UV-vis-NIR

adsorption was recorded on a U-3310 spectrophotometer. The specific surface area was

evaluated and calculated through the Brunauer-Emmett-Teller (BET) and Barrett-Joyner-

Halenda (BJH) methods. The fluorescence (FL) emission and excitation spectra were acquired

on a Hitachi F-7000 spectrophotometer under ambient conditions. Element concentration was

determined by Agilent 700 Series Inductively coupled plasma mass spectrometry (ICP-MS).

Raman spectra were obtained to confirm the structural nature of the carbon matrix of Fe3O4-C

nanoparticles using a Renishaw inVia Raman microscope. Hydroxyl radical signal was

collected by electron spin resonance spectroscopy (Bruker X-band A200, EPR). Powder X-

ray diffraction (XRD) pattern was acquired on a Shimadzu XRD-7000s diffraction instrument

with Cu Kα radiation (λ=1.542 Å) over the scan range 10°-80°.

Photothermal conversion efficiency was calculated as previously reported [1].

The cuvette containing Fe3O4-C or RLR nanoparticles (0.2 mL, 100 ppm). The temperature

was recorded by a photothermal camera (Fluke Ti450) every 30 s until reaching maxima. The

photothermal conversion efficiencies were calculated as follows:

）（1QQQ/dtdTcm surDisNCi

ii .

Where m and C are the mass and heat capacity of water (solvent), T is the temperature of the

solution, respectively. QNC is the photothermal energy imputed by the nanoparticles, which

can be determined by:

TmcΔQ)Ths(TQ

)10ηI(1Q

Dis

sursur

ANC

808

Where I=0.5 W is the laser power used in the experiment, and A808 is the absorbance of the

nanoparticles at the used wavelength. The is the photothermal conversion efficiency.

Moreover, Qur expresses the heat input absorbed by the solvent and the container, where h is

the heat transfer coefficient and s is the surface area of the container. T is the maximum

temperature change during the irradiation.

S3

After reaching the maximum temperature, the input and output energies come to a balance.

surDisNC QQQ

The photothermal conversion efficiency could be:

）（2)10I(1Q)ThS(Tη

808A-Dissur

Where two rate constants are defined as :

Shmc

τ

and :

hsmcτ,3

TTTTθ

surmax

sur

）（

dTdθ)T(T surmax

mcQQQdT/dt

dtdθ)T(T surDisNC

surmax

0=Q+Q DisNC

τTT-

mc)Ths(T-

mcQ-

dtdθ)T(T sursursur

surmax

So, we got:

T = - Ln()

To get the hs, we measured the cooling curve and can be determined. From the fit line, of

RLR or Fe3O4-C nanoparticles is 155.01 and 155.92, respectively. At the maximum steady

temperature, the final photothermal conversion efficiency can be determined as:

）（2)10I(1Q)ThS(Tη

808A-Dissur

The absorbance of Fe3O4-C nanoparticles is 0.2959, and the absorbance of RLR nanoparticles

is 0.3026. Also, the m is 0.2 g and the c is 4.2 J g-1. The QD is measured independently to be

0.1344 mW. With all the data, the photothermal conversion efficiency of RLR and Fe3O4-C

nanoparticles can be calculated to be 26.8% and 24.4%.

Calculation of the synergy index among CDT, PTT and chemotherapy

The calculation of synergy index was based on a modified equation from previous literature

[2]. More specifically, the synergy factor equals the result from experimental value of

combined treatment divided by the theoretical multiplication of values from individual

treatment. In our case, we define the experimental value of cell death rate from the combined

S4

treatment from CDT, PTT and chemotherapy as DChemo·PTT·CDT. similarly, the theoretical

multiplication of values from individual treatment as D’Chemo·PTT·CDT. The synergy index:

I Syn=DChemo·PTT·CDT

D 'Chemo·PTT·CDT

Besides, considering the percentage of dead cell and survived cells equal 100%, we can

conclude that:

DChemo·PTT·CDT=1−SChemo·PTT·CDT

D 'Chemo·PTT·CDT=1−S 'Chemo·PTT·CDT

In these equations, SChemo·PTT·CDT is the cell survival rate after the combined therapy from our

experiment; S’Chemo·PTT·CDT is the theoretical value where the three therapeutic modalities are

orthogonal to each other (no synergistic effects).

Considering that CDT alone is toxic to cancer cells, we can assume that PTT combined with

CDT and chemotherapy combined with CDT will result in lower survival than PTT and

chemotherapy alone (SPTT·CDT < SPTT; SChemo·CDT < SChemo).

In theory, when there is no crosstalk (synergistic) effect between combined CDT, PTT and

chemotherapy, the survival rate of cancer cells from the combinatorial therapies can be

expressed as following:

S 'Chemo· PTT ·CDT=SChemo × SPTT × SCDT

Based on the assumption that SPTT·CDT < SPTT and SChemo·CDT < SChemo, we can conclude that when

there is no synergistic effect between combined CDT, PTT and chemotherapy:

S 'Chemo· PTT ·CDT=SChemo × SPTT × SCDT>SChemo·CDT × SPTT·CDT × SCDT

And I Syn=DChemo·PTT·CDT

D 'Chemo·PTT·CDT>

1−SChemo·PTT·CDT

1−SChemo·CDT × SPTT·CDT × SCDT

REFERENCES

[1] Z. Tang, H. Zhang, Y. Liu, D. Ni, H. Zhang, J. Zhang, Z. Yao, M. He, J. Shi, W. Bu, Antiferromagnetic Pyrite as the Tumor Microenvironment-Mediated Nanoplatform for Self-Enhanced Tumor Imaging and Therapy, Adv. Mater. 29(47) (2017) 1701683.[2] M. Cortina-Borja, A. D. Smith, O. Combarros, D. J. Lehmann, The synergy factor: a statistic to measure interactions in complex diseases, BMC. Res. Notes. 2(1) (2009) 105.

S5

B. SI FIGURES AND TABLES

Figure S1. Characterization of the hollow structure of Fe3O4-C nanoparticles a-c,

Representative TEM (a) and HRTEM (b, c) images of Fe3O4-C hollow nanoparticles indicate

that ultra-small Fe3O4 nanoparticles (2 nm)-embedded in the carbon matrix. d-e, TEM image

(d) and EDS elemental mappings (e) of Fe3O4-C nanoparticles indicate that the Fe ions are

homogeneously located in the carbon shell. f, EDS element distribution of Fe3O4-C

nanoparticles show the content of each component. g-h, XRD spectrum (g) and Raman

spectrum (h) further prove the crystallization of Fe3O4 and the carbon shell.

S6

Figure S2. Precise size control of Fe3O4-C hollow core. a-e, Representative SEM images of

Fe3O4-C hollow core with different amount of H2O2 addition: 30 mL (a), 35 mL (b), 40 mL

(c), 45 mL (d), and 50 mL (e). f, DLS size distribution of Fe3O4-C hollow core under different

conditions. Scale bar: 200 nm.

S7

Figure S3. Structural characterization of RLR nanoparticles. a-b, TEM image (a) and

EDS elemental mappings (b) of RLR nanoparticles.

Figure S4. Precise shell thickness control of RLR nanoparticles. a-c, TEM images of RLR

nanoparticles with different thickness of the shell. A redox reaction between the carbon

(reductant) and KMnO4 (oxidant) initiated the in situ growth of MnO2 nanoshell onto the core

nanoparticles. By varying the concentration of KMnO4, the shell thickness can be easily

tuned. In particular, lower concentration of KMnO4 (0.1 mM) resulted in a relatively thin shell

with a thickness of 2~6 nm (a); While increasing the concentration of KMnO4 to 1 mM, we

could get a thicker shell (13 nm) (b); Continuously increasing the concentration of KMnO4

(10 mM), the shell thickness could reach 28 nm (c).

S8

Figure S5. Biodegradability of RLR nanoparticles. Photographic images of RLR

nanoparticles incubated with 10 mM GSH in pH=5.0 (dibasic sodium phosphate-citric acid)

buffer at different time points.

S9

Figure S6. In vitro biodegradation monitoring of RLR nanoparticles. The RLR nanoparticles undergo sequential degradation of MnO2 shell and Fe3O4-C core in response to the high GSH levels and acidic conditions. a-d, Representative TEM images of RLR nanoparticles incubated with 10 mM GSH (pH=5.0) at different time points. a, Day 1, The MnO2 shell on RLR nanoparticles reacted with the extensive GSH to dissolve. b-c, Day 3-5, After the dissolution of the outer shell, the iron oxide nanoparticles in the core would degrade to produce iron ions due to combined effects from GSH and the acidic buffer. d, Day 9, Lastly, the degradation of iron oxide destabilized the carbon matrix, leading to the dissociation of carbon into small graphitic segments.

S10

Figure S7. Iron ion-induced oxidative stress via Fenton reaction. As a proof-of-concept,

methylene blue trihydrate (MB) was used to monitor the generation of ·OH. a, UV-vis

absorption spectra and photo (inset) of MB after incubating with i) H2O2 (10 mM); ii) Fe3+ (0.5

mM); and iii) H2O2 (10 mM) plus Fe3+ (0.5 mM). b, Relative cell viability of MDA-MB-231

cells incubated with different experimental conditions for 24 h. c, The variation of

intracellular ROS was evaluated by detecting the fluorescence of DCF (λex = 488 nm, λem =

525 nm) with a fluorescence microscope after the MDA-MB-231 cells exposed to different

components for 20 min. Scale bar: 100 m.

Figure S8. RLR nanoparticle-induced ferroptosis pathways. The viability of MDA-MB-231 cells treated with varied concentration of RLR nanoparticles for 48 h under pH 6.5.

S11

Figure S9. Photothermal performance of RLR and Fe3O4-C Nanoparticles. a, UV-vis-

NIR absorption spectra of RLR and Fe3O4-C nanoparticles with a concentration of 100 ppm.

b, Photothermal effect of an aqueous dispersion of RLR and Fe3O4-C nanoparticles under

irradiation with NIR laser (808 nm, 1.5 W cm-2), and then the laser was shut off when

reaching the steady maximum temperature. c-d, Linear time data vs. -ln gained from the

cooling period of RLR (c) and Fe3O4-C nanoparticles (d).

S12

Figure S10. The mechanism of photothermal hyperthermia-mediated cell apoptosis.

Quantitative reverse transcription PCR (RT-PCR) showed significant downregulation of anti-

apoptosis gene (BCL-2) and upregulation of apoptosis genes (BAX, Caspase 3). Error bar

represent mean s.d.; n=4, ***p<0.001, **p<0.01.

Figure S11. The generation of ROS (H2O2) induced by DOX. The variation of intracellular

ROS was evaluated by detecting the fluorescence of DCF (λex = 488 nm, λem = 525 nm) with a

fluorescence microscope after the MDA-MB-231 cells exposed to different concentration of

DOX for 2 h. Scale bar: 50 m.

S13

Figure S12. Cytotoxicity analysis of RLR nanoparticles. The viability of HDFs, ADSCs

and C2C12 cells treated with varied concentration of RLR nanoparticles for 48 h based on the

Presto blue assay.

S14

Figure S13. The therapeutic universality of RLR nanoparticles. The therapeutic

universality of RLR nanoparticles was conducted using two different cancer cell lines: Hela

cervical cancer cell line, DU-145 prostate cancer cell. a-b, Relative cell viability of Hela cells

(a) and DU-145 cells (b) incubated with various concentrations of Fe3O4-C and RLR

nanoparticles for 24 h. c-d, Relative cell viability of Hela cells (c) and DU-145 cells (d) under

different treatments: i) DOX alone; ii) DOX-loaded RLR nanoparticles without NIR (RLR-

DOX); iii) DOX-loaded RLR nanoparticles followed by NIR (808 nm, 1.5 W cm -2, 5 min)

treatment (RLR-NIR+DOX). Error bar represent mean s.d.; n=4, ***p<0.001, **p<0.01,

*p<0.05.

S15

Figure S14. Synergistically overcoming drug-resistant of MDA-MB-231 cells.

Dramatically reduced IC50 of DOX was achieved via a combination of CDT, PTT and

chemotherapy. a-c, The viability of MDA-MB-231 cells treated with different doses of DOX

under different conditions: i) DOX alone (a); ii) DOX-loaded RLR nanoparticles (50 g mL-1)

without NIR (RLR-DOX) (b); iii) DOX-loaded RLR nanoparticles (50 g mL-1) followed by

NIR (808 nm, 1.5 W cm-2, 5 min) treatment (RLR-NIR-DOX) (c).

S16

Figure S15. Schematic illustration of iRGD conjugation.

Figure S16. In vitro MR imaging capacity. The unique properties of RLR nanoparticles for

releasing T1 MRI contrast agent Mn2+ were conducted. a, The MRI image of RLR

nanoparticles incubated with 10 mM GSH or PBS for 20 min; b, T1 transverse relaxation rate

as a function of Mn concentration.

S17

Figure S17. In vivo real-time MRI monitoring the degradation of RLR nanoparticles. A

constant increase of MRI intensities was observed throughout the time course between 1 h to

3 h. Thereafter, MRI intensity plateaued between 5 to 24 h followed by a signal decrease until

96 h after injection.

Figure S18. In vivo photothermal setup. The mice were anaesthetized before laser

irradiation. a-b, The laser was fixed and precisely irradiated at the tumor location. c, After 5

min irradiation (808 nm, 1.5 W cm-2), no obvious skin damage was detected.

S18

Figure S19. The mechanism of PTT-mediated reduction of drug efflux in vivo. In vivo

qRT-PCR experiments reveal the significantly suppressed cancer pro-survival genes (BAX-2)

and upregulated pro-apoptotic genes (BCL-2 and Caspase-3) in the experimental condition

(RLR nanoparticle delivered with DOX and treated by NIR) compared to the controls (no

treatment, injection of DOX only and RLR nanoparticles delivered DOX).

Figure S20. Tumor H&E staining. The H&E staining images further reveal the significantly

higher apoptosis in the tumor site of the experimental condition compared to the control

groups. Arrows indicate possible apoptotic cancer cells.

S19

Figure S21. In vivo compatibility assay. The body weights were measured every 3-4 day to

evaluate the in vivo toxicity.

Figure S22. Histological analysis of major organs. H&E stained tissue section of major

organs (heart, kidney, liver, lung, spleen.) from mice with different treatments. In particular,

the DOX-treated group showed obvious damage in heart due to the circulated DOX.

S20

SI TABLES

Gene Forward Primer Reverse Primer

GAPDH CCGCATCTTCTTTTGCGTCG GCCCAATACGACCAAATCCGT

HSF-1 AGATGGAAGATGGGAGAGGGGTA

G

TGGAAAAGTGCTCATCAGTGCG

MDR-1 (ABCB-

1)

GGGATGGTCAGTGTTGATGGA GCTATCGGTGGCAAACAATA

TP53 ACCCAGGTCCAGATGAAG CACTCGGATAAGATGCTGA

BCL-2 ACA ACA TCG CCC TGT GGA TGA C ATA GCT GATTCG ACG TTTTGC

C

BAX GGA ATT CTG ACG GCA ACT TCA

ACT GGG

GGA ATT CTT CCA GAT GGT

GAG CGA GG

Caspase 3 AGAACTGGACTGTGGCATTGA GCTTGTCGGCATACTGTTTCAG

Table S1. Table of the primers used for RT-qPCR.

Cell type Pretreatment IC50 (M)

MDA-MB-231 DOX 30.17

RLR-DOX 1.73

RLR-NIR-DOX 0.38

Table S2. IC50 of DOX on MDA-MB-231 cells. The half-maximal inhibitory concentrations

of DOX for MDA-MB-231 cells that incubated with RLR Nanoparticles (50 g mL-1) with or

without 808 nm laser irradiation (1.5 W cm-2, 5 min).

S21