A dopamine-precursor-based nanoprodrug for in-situ drug release and treatment of acute liver failure by inhibiting NLRP3 inflammasome and facilitating liver regeneration
Chenyue Zhan , Guifang Lin , Yong Huang , Ziqian Wang , Fang Zeng *, Shuizhu Wu **
Biomedical Division, State Key Laboratory of Luminescent Materials and Devices, Guangdong Provincial Key Laboratory of Luminescence from Molecular Aggregates, College of Materials Science and Engineering, South China University of Technology, Wushan Road 381, Guangzhou, 510640, China


Acute liver failure Nanodrug Inflammasome Liver regeneration Imaging


Acute liver failure (ALF) is a severe liver disease with high mortality rate. Inflammasome is a newly-found and promising target for effective treatment of immunity-associated diseases including liver disease, and dopamine has recently been proved as an inhibitor for NLRP3 inflammasome. This work demonstrates a diselenide-based nanodrug for ALF treatment through inhibiting NLRP3 inflammasome activation and enhancing liver regener- ation. A diselenide-containing molecule (DSeSeD) has been synthesized via covalently linking two L-Dopa mol- ecules to a diselenide linker, and the resultant molecules form stable nanoparticles in aqueous media and encapsulate SW033291 (an inhibitor of prostaglandin-degrading enzyme that hampers liver regeneration) to produce the nanodrug (SW@DSeSeD). As a nanoscale prodrug, SW@DSeSeD protects its payloads from decomposition in bloodstream upon administration, accumulates in liver of ALF mice, then responds to the overexpressed ROS and thereby releases SW033291 as well as a stable dopamine precursor that can transform into dopamine in hepatic cells, thus achieving significant therapeutic efficacy against ALF through inhibiting NLRP3 inflammasome activation and enhancing hepatic regeneration. Moreover, multiple contrast agents have been loaded onto the nanodrug to achieve fluorescence, optoacoustic and magnetic resonance imaging for nanodrug location and disease evaluation.

1. Introduction

Acute liver failure (ALF) is a life threatening critical hepatic disease which features rapid loss in hepatocyte functions. ALF is associated with multiorgan failure and usually has poor prognosis with high mortality rate [1,2]. The main causes of ALF include hepatitis virus infection which is responsible for majority cases of ALF in developing countries, and the drug-induced liver injury which accounts for about one-half of ALF incidents in Western countries [3,4]. With advances in liver trans- plantation and intensive clinical management, the previously very high mortality rate (~80%) caused by ALF has now declined, but the overall death rate remains at a high level of around 40% [5]. The limited availability of liver transplantation has led to the development of other therapies, including the medication (drug) treatment [6,7]. Currently,

variety of drugs have been used to treat ALF, including the traditional anti-oXidants and corticosteroids, as well as the newly developed ther- apeutic drugs that can offer broadly protective effects against ALF
[8–14]. Considering that the mortality rate from acute liver failure is
still high, continuous efforts have to be made.
OXidative stress and inflammation contribute to the pathogenesis of most acute and chronic liver diseases [15–18]. During liver injury pro-
cess, local immune cells, such as Kupffer cells, are activated and in- flammatory leukocytes including monocytes and lymphocytes are recruited, thus producing excessive reactive oXygen species (ROS) and incurring damage to hepatocytes [19]. Inflammasomes are cytosolic multiprotein oligomers of the innate immune system responsible for the activation of inflammatory responses, they assemble in response to danger signals and initiate the release of some pro-inflammatory

* Corresponding author. Guangdong Provincial Key Laboratory of Luminescence from Molecular Aggregates, College of Materials Science and Engineering, South China University of Technology, Wushan Road 381, Guangzhou 510640, China.
** Corresponding author. Biomedical Division, State Key Laboratory of Luminescent Materials and Devices, Guangdong Provincial Key Laboratory of Luminescence from Molecular Aggregates, College of Materials Science and Engineering, South China University of Technology, Wushan Road 381, Guangzhou, 510640, China.
E-mail addresses: [email protected] (F. Zeng), [email protected] (S. Wu).
Received 11 June 2020; Received in revised form 18 November 2020; Accepted 22 November 2020
Available online 25 November 2020
0142-9612/© 2020 Elsevier Ltd. All rights reserved.

cytokines (such as IL-1β and IL-18) for the activation of inflammatory with diselenide bond have attracted great attention due to their unique
responses [20,21]. Inflammasome activation has been studied in properties including oXidation-responsive properties [28–30]. In

immunity-related diseases including liver diseases and has been identi- fied as a major contributor to hepatocyte damage, immune cell activa- tion and amplification of liver inflammation [22]. NLRP3 inflammasome is a relatively well-studied inflammasome and identified as a potential mediator of various liver disease, and application of potent inhibitor of NLPR3 activation signifies a rational curative strategy to cure liver failure [23]. A recent study has demonstrated that dopamine (DA), which is a neurotransmitter in body and also functions as an important molecule bridging the nervous and immune systems, can effectively control systemic inflammation through inhibition of NLRP3 inflammasome activation [24]. Thus, DA may serve as potential medi- cine to treat acute liver failure.
For ALF treatment, hepatic tissue regeneration is also of great importance because patients must regain adequate hepatic function to survive [1]. Recently, Markowitz and coworkers created a new approach to achieve tissue regeneration by inhibiting prostaglandin-degrading enzyme 15-PGDH, and they also developed a potent small-molecular inhibitor of 15-PGDH. Dubbed as SW033291, the 15-PGDH inhibitor induced rapid tissue regeneration in multiple mice models, including a liver hepatectomy model [25]. We thus envision that, using the inhibitor SW033291 in combination with dopamine in ALF treatment may further improve the therapeutic outcome by facilitating liver regeneration and inhibiting NLPR3 activation simultaneously.
Despite of the recent promising strategies for ALF treatment brought by the newly developed small molecules, the lack of a proper drug de- livery system may limit their applications, as many of the small mole- cules (e. g. dopamine) are not stable and show short half-life in bloodstream [24]. To address this challenge, loading multiple drugs or drug candidates into a single nanocarrier is of particular interest for achieving synergistic activities [26,27]. In consideration of the fact that a nanosized system tends to accumulate and be sequestered in the liver after administration, a nanosized delivery system which can respond to the microenvironment of the injured liver and release the drugs would be highly ideal for achieving effective ALF therapy. Recently, materials

particular, Se–Se bond is prone to cleave in the presence of various ROS [31], and we suppose this property could be exploited to design nano- sized delivery system for treating acutely injured liver where ROS is
Herein we demonstrate a combination therapy for acute liver failure treatment via inhibiting NLRP3 inflammasome activation and facili- tating liver regeneration, as shown in Scheme 1. For this strategy, two L- Dopa molecules were linked to the two sides of a diselenide linker, and the resultant small-molecular compound (DSeSeD) could form stable nanoparticles (NPs) in aqueous media. In addition, the inhibitor SW033291 was embedded during the nanoparticle formation to create a nanodrug system (SW@DSeSeD) for ALF treatment. Upon administra- tion, the nanodrug could accumulate in ALF lesion, respond to the overexpressed ROS in the injured liver of ALF mice and release SW033291 and a dopamine derivative that would be transformed into dopamine by endogenous enzymes, thus achieving significant thera- peutic efficacy for ALF by inhibiting NLRP3 inflammasome activation and enhancing hepatic regeneration. The diselenides also contribute to the therapy by scavenging the excessive ROS. Moreover, the DSeSeD NPs are capable of carrying contrast agents for fluorescent, optoacoustic and MRI imaging, which can be used to track the biodistribution of the nanodrug and monitor evolution of liver function during therapy course.
2. Materials and methods
2.1. Materials

L-Dopa, selenium power, 3-bromo-1-propanol, di-tert- butyldicarbonate (Boc2O), tert-butyldimethylsilyl chloride (TBDMSCl); 1,8-diazabicyclo[5.4.0]-undec-7-ene (DBU), N,N-diisopropylethylamine
(DIPEA), tetrabutylammonium fluoride (TBAF), trifluoroacetic acid (TFA) and 2-(7-azabenzotriazol-1-yl)-N,N,N′,N′-tetramethyluronium
hexafluorophosphate (HATU) were purchased from Aladdin Bio-Chem Technology Co. Ltd. (Shanghai, China). Live/Dead Viability/

Scheme 1. Schematic representation of combination therapy for acute liver failure in mouse model by using SW@DSeSeD nanodrug via inhibiting further liver damage (through blocking NLRP3 inflammasome activation) as well as facilitating liver regeneration (through inhibiting prostaglandin-degrading enzyme 15- PGDH). A dye-doped nanoparticle (ICG-SW@DSeSeD) can be utilized for multi-mode imaging.

CytotoXicity Kit, RIPA buffer (PBS with 0.5% NP-40, 0.5% sodium deoXycholate, 0.1% sodium dodecyl sulfate, 5.5% glycerophosphate, 1 mM dithiothreitol and complete protease and phosphatase inhibitors),
BCA Protein Assay Kit, Western blocking buffer, dichlorofluorescein diacetate (DCF-DA) and DAB (3,3′-diaminobenzidine tetrahydro- chloride) were purchased from KeyGen Biotech Co. Ltd. IL-1β, TNF-α, PGE2, ALT and AST ELISA Kits were purchased from Shanghai Enzyme-
linked Biotechnology Co., Ltd. Primary antibodies (Casepase 3, p-JNK,

Afterwards, 100 μL freshly prepared ICG aqueous solution (5.0 mg
mL—1) was added dropwise into 10 mL dispersion of as-prepared Fe3+- SW@DSeSeD. The miXture was then stirred for 5 h in dark at room
temperature under nitrogen atmosphere, and the ICG/Fe3+-doped nanoparticles ICG-SW@DSeSeD NPs were obtained after dialysis (with
cut-off molecular weight of 3000 Da) against purified water for 24 h in dark to remove the unbound ICG.
To determine the loading capacity of ICG on ICG-SW@DSeSeD NPs,

NLRP3, 15-PGDH, β-actin and Ki67), horseradish and peroXidase-

the prepared ICG-SW@DSeSeD NPs were lyophilized with a vacuum

conjugated secondary antibodies were obtained from Abclonal Technology.
2.2. Instrumentation

Transmission electron microscopy (TEM) was performed on a JEOL JEM-1400 transmission electron microscope. The hydrodynamic diam- eter distribution and zeta potential were determined on a Malvern Nano ZS90 system running DTS software. Absorption spectra were measured on a Hitachi U-3010 UV-VIS spectrophotometer. Fluorescence spectra were obtained with Hitachi F-4600 fluorescence spectrophotometer. Fluorescence microscopic images were obtained using an Olympus IX 71 with a DP72 color CCD. Flow cytometry were performed on a BD C6 flow cytometer. Optoacoustic tomography imaging was performed on an inVision128 multispectral optoacoustic tomographic (MSOT) imaging system (iThera Medical GmbH) and fluorescence imaging was per- formed on an AMI small animal fluorescence imaging system (Spectral Instruments Imaging Co.). The T1-weighted MRI imaging was per-
formed on a 7.0 T MRI system (Bruker 7T PharmaScan 70/16) at 20 ◦C.

2.3. Preparation of DSeSeD and SW@DSeSeD NPs

The DSeSeD NPs were prepared by nano-precipitation method. Briefly, a certain amount of DSeSeD was dissolved in DMSO and then the solution was added into water dropwise. After 5s of ultrasonic treatment with the operating frequency at 50 kHz, the miXture was subject to mechanical stirring (~500 rpm) at room temperature for 2 h and the
resultant dispersion was filtered with a 0.22 μm filter to obtain the
Similarly, both DSeSeD and SW033291 were dissolved in DMSO with a certain weight ratio of DSeSeD to SW033291 and the solution was added dropwise into water. After 5 s of ultrasonic treatment, the miXture was subject to stirring at room temperature for 4 h, so as to allow the free (unloaded) SW033291 molecules (which are unstable in aqueous media)
to precipitate in PBS solution. The resultant dispersion was filtered with a 0.22 μm filter to remove the un-encapsulated SW033291. The as- prepared SW@DSeSeD NPs were used for some measurements or freeze-dried and then redispersed in pH 7.4 PBS and then kept in 4 ◦C for
subsequent experiments.

2.4. Determination of loading of SW033291 in SW@DSeSeD

The as-prepared SW@DSeSeD NPs were lyophilized with a vacuum lyophilizer, 5.0 mg SW@DSeSeD NPs were dissolved in 50 mL DMSO, followed by the absorbance measurement at 315 nm. The SW033291 loading was determined based on absorption spectrometry with the experimentally-determined calibration curves (Supplementary Fig. S14).
2.5. Preparation of Fe3+-SW@DSeSeD and ICG-SW@DSeSeD NPs
To a SW@DSeSeD NPs dispersion (10 mL, 1.0 mg mL—1), a certain of
volume of 100 mg mL—1 FeCl3 aqueous solution was added dropwise under nitrogen atmosphere, the miXture was further stirred for 5 h. The
final solution was sealed in a dialysis bag (with cut-off molecular weight of 3000 Da) and dialyzed against water for 24 h to remove the unbound
Fe3+ to yield the Fe3+-doped nanoparticles Fe3+-SW@DSeSeD.

lyophilizer, and 5.0 mg ICG-SW@DSeSeD NPs were dissolved in 50 mL DMSO for absorbance measurement at 795 nm. The ICG loading ca- pacity was determined as 1.76 wt% using absorption spectrometry with a pre-determined calibration curve shown in Supplementary Fig. S10 b.
2.6. In vitro SW033291 release of the SW@DSeSeD NPs

To determine the release profile of SW033291 from the nanodrug, 2 mL of SW@DSeSeD NPs dispersion was sealed in a dialysis bag (with cut-
off molecular weight of 3000 Da) which was then immersed into 50 mL PBS solution (pH 7.4, containing 5% DMSO) and 200 μM H2O2 or 500 μM at 37 ◦C. At certain time points, 3 mL release solution was withdrawn
for HPLC measurement. Then same amount of preheated solution was added back to replace the withdrawn volume.
2.7. Cell experiments

2.7.1. Cell culture
NCTC-1469 cells were cultured in Dulbecco’s modified Eagle’s me- dium (DMEM) medium, while RAW 264.7 cells were culture in RPMI1640 medium, and the culture media were supplemented with
10% fetal bovine serum and 1% penicillin-streptomycin at 37 ◦C in an
atmosphere of 5% CO2.

2.7.2. Cell viability assay
To investigate the cell toXicity of the different formulations, RAW
264.7 cells were seeded at 1 104 cell/well in a 96-well plate and incubated for 24 h. Then, different formulations with indicated con- centrations were added and incubated for additional 24 h. Afterwards, the cells were washed 3 times with PBS, and cell viability was deter- mined by using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT).
To evaluate the effect of the different formulations against ROS- induced damage to cells, the RAW 264.7 cells were pretreated with 1
mM H2O2 or 200 μM for 4 h, and then incubated with different formu-
lations for 20 h. After that, the cells were washed 3 times with PBS, and the cell viability was determined by using MTT.
2.7.3. Evaluation of effect of different formulations on H2O2-induced oxidative cell damage and intracellular reactive oxygen
RAW 264.7 cells were seeded in 6-well plates at a density ca. 1 105
cells per well and incubated for 24 h. For ROS scavenging activity measurement, cells were treated with 200 μM H2O2 for 4 h, and then incubated with different formulations for 4 h and washed with PBS 3 times. Afterwards the cells were treated with 20 μM dichlorofluorescein diacetate (DCF-DA) for 30 min and washed three times with PBS again.
The oXidized DCF in cells was observed on a fluorescent microscope.
2.7.4. Live/dead cell staining
RAW 264.7 cells were seeded in 6-well plates at a density of ca. 1 105 cells per well and incubated for 24 h. Afterwards, 1 μg mL—1 of LPS was used to pre-treat the cells, and after 4 h the cells were incubated
with different formulations for 20 h. The supernate (SN) of culture medium of RAW 264.7 cells was then obtained by centrifugation.
NCTC-1469 cells were also seeded in 6-well plates at a density of 1 105 cells per well and incubated for 24 h. The culture medium of NCTC- 1469 cells was replaced with the supernate of pretreated RAW 264.7

cells, the NCTC-1469 cells were incubated for another 4 h before treatment with LIVE/DEADViaility/CytotoXicity Kit (KeyGen Biotech Co. Ltd.) according to the protocol. The cells were then observed on a fluorescent microscope.

2.8. ELISA assay

RAW 264.7 macrophage cells were seeded in 96-well plates at a density of 5 103 cells per well and incubated for 24 h. Afterwards, 1 μg mL-1 LPS was used to pre-treat the cells, and after 4 h the cells were then
respectively incubated with different formulations for 20 h. The me-
diums were collected and centrifuged at 8000 rpm to wipe off cell debris and residuary nanoparticles. The levels of IL-1β, TNF-α and PGE2 were assayed using corresponding ELISA kit (Shanghai Enzyme-linked
Biotechnology Co., Ltd.).

2.9. Animal experiments

All animal experiments were performed in SPF environment in Laboratory Animal Center of South China Agricultural University in accordance with the Regulation on Administration of Laboratory Animal of Guangdong Province and Protocol of Animal Ethics Committee of South China Agricultural University, and all experiments have been pre- approved by the Animal Ethics and Welfare Committee of South China Agricultural University. During the animal procedures, every effort was made to minimize suffering.
8-10 week-old C57BL/6 male mice were randomly allocated to
different groups, after 20 h of starvation, APAP was i.p. injected with a sublethal dose of 300 mg kg—1 for most liver injury studies and with a lethal dose of 500 mg kg—1 for survival evaluation. The treatment was
initiated at 6 h post APAP injection. The mice (5 mice per group) received i.v. injections 4 times of PBS, dopamine (4.3 mg kg—1 in PBS), SW033291 (0.98 mg kg—1, dispersed in PBS with 5 wt% Tween-80), DSeSeD NPs (9.02 mg kg—1 in PBS) or SW@DSeSeD NPs (10.0 mg kg—1 in PBS) at 6 h, 12 h, 24 h and 48 h post APAP injection. To assess
the injurious effects of APAP on liver histology and function, serum was collected from mice via orbit puncture. Serum samples were taken at indicated post-APAP time points at 12 h, 24 h and 72 h. ALT and AST levels were determined in blood specimens using an ALT and AST assay kit (Shanghai Enzyme-linked Biotechnology Co., Ltd.).

2.10. Multispectral optoacoustic tomography (MSOT) imaging

In vitro and in vivo MSOT imaging was performed on a multispectral optoacoustic tomography system (iThera Medical). For phantom imag- ing, the control (PBS solution) or the test solution was fully filled in a commercial Wilmad NMR tube, and then fiXed on the holder of the in-
strument for imaging.
For cellular internalization imaging, 1 105 RAW 264.7 cells were
dissociated from 25 mm culture dishes and grouped by co-incubation time (0 h, 1 h, 2 h, 4 h, 6 h) with ICG-SW@DSeSeD (50 μg mL—1) NPs.
Cells were then digested and washed 3 times with PBS and dispersed in 2 mL PBS per dish. The cells suspensions were fully filled in the Wilmad
NMR tubes and then fiXed on the holder of the instrument for imaging. For in vivo imaging of ICG-SW@DSeSeD NPs, 10.0 mg kg—1 ICG-
SW@DSeSeD was injected into mice via tail vein. Optoacoustic signals were recorded under multiple excitation wavelengths (680, 710, 740,
770, 800, 830, 860 and 900 nm). After data acquisition, MSOT images were reconstructed using a standard back projection algorithm. The linear spectral unmiXing was utilized to separate signals from the ICG and those from the absorbers in tissue (e.g., hemoglobin). Upon gener- ation of cross sectional MSOT images, the z-stack images were rendered as orthogonal MIP images with 3D information. Three mice were tested for each group.

2.11. T1-weighting MRI imaging

The T1-weighted MRI images of ICG-SW@DSeSeD NPs dispersion were obtained on a 7.0 T MRI system (Bruker 7T PharmaScan 70/16 US) at 20 ◦C. ICG-SW@DSeSeD NPs were dispersed in pH 7.4 PBS to produce
dispersions with varied concentrations respectively for magnetic reso- nance imaging. Pure PBS solution was also imaged as the control.
For in vivo MRI imaging, C57BL/6 mice of different groups were
injected with ICG-SW@DSeSeD NPs dispersion intraperitoneally (with
ICG-SW@DSeSeD dose of 25 mg kg—1). During imaging the mice were continuously anesthetized with 2.5% isofluorane delivered using a nose
count mounted on the mouse holder. Liver slices images were acquired using a multislice spin-echo (SEMS) sequence with TR/TE 520/15 ms. FOV = 80 × 80 mm2, 256 × 256 matrices.
3. Results and discussion
3.1. Fabrication and characterization of the nanodrug, as well as its response to reactive oxidative species (ROS)
The main architecture (DSeSeD) of the nanodrug was synthesized by covalently linking two L-Dopa (a direct precursor of dopamine) mole- cules on each side of a diselenide linker, and the detailed synthetic route is shown in Scheme S1. The structures of the DSeSeD and main in-
termediates were confirmed by high-resolution mass spectrometry (HR- MS) and 1H nuclear magnetic resonance (1H NMR) spectrometry, as shown in Supplementary Fig. S1-S8 in the mation. The oil-water balance
(partition) coefficient (Log P) for DSeSeD was determined as Log P 2.0
0.2, suggesting the hydrophobic nature of the molecule; while the critical aggregation concentration (CAC) of DSeSeD in water was
determined as 1.5 10—6 mol L—1 by dynamic light scattering (Sup- plementary Fig. S9). Moreover, we found DSeSeD could form stable
nanoparticles as its DMSO solution was slowly added into aqueous medium. The driving forces for the formation of stable nanoparticles could be hydrophobic nature of DSeSeD molecule, the diselenide bond
insertion and intermolecular π – π stacking, as well as other interactions
between the components in DSeSeD molecules [29,31]. Some properties of DSeSeD-based nanoparticles are presented in Fig. 1. The formation process of the relevant nanoparticles is shown in Fig. 1A. The inhibitor SW033291 could be embedded into the nanoparticles during the nano-precipitation process to afford the nanodrug SW@DSeSeD.
To achieve multi-mode imaging, after formation of SW@DSeSeD NPs, Fe3+ (in FeCl3 aqueous solution) was added dropwise into the NPs
dispersion under nitrogen atmosphere to chelate with the catechol groups in L-Dopa moieties on DSeSeD, thus forming Fe3+-doped nano- particles Fe3+-SW@DSeSeD NPs. After purification of the nanoparticles by dialysis, ICG solution was added into the dispersion of Fe3+-
SW@DSeSeD NPs, and the negatively-charged ICG molecules (with a sulphonic group in its structure) could be absorbed on the Fe3+- SW@DSeSeD NPs with positively-charged Fe3+ on their surface via electrostatic interaction. The π–π interaction between ICG and L-Dopa
moiety may also facilitate the absorption [32]. The resultant Fe3+ and
ICG-doped nanoparticles are designated as ICG-SW@DSeSeD NPs and are used for multi-mode imaging only. Some properties of ICG-SW@DSeSeD NPs dispersion are shown in Supplementary Fig. S10. The loading capacity of ICG on ICG@DSeSeD NPs was determined as
1.76 wt% by absorption spectrometry using a pre-determined calibra- tion curve shown in Supplementary Fig. S10 b. Also, we recorded the particle diameter, the absorption and emission spectra of the dispersion of the ICG-SW@DSeSeD NPs (in PBS) after different time periods of incubation, and found that the changes in these properties are insig- nificant, indicating the nanoparticles are quite stable in PBS solutions, as shown in Fig. S10a, S10c and S10d.
The diameter distribution and morphology for DSeSeD NPs, SW@DSeSeD NPs and ICG-SW@DSeSeD NPs were determined by dy- namic light scattering (DLS) and transmission electronic microscopy

Fig. 1. (A) Schematic illustration for formation of DSeSeD NPs, SW@DSeSeD NPs (for therapy only) and ICG-SW@DSeSeD NPs (for imaging only). (B) Hy- drodynamic diameter distribution in water as deter- mined by DLS for DSeSeD-based nanoparticles. The
initial DSeSeD molecule concentration was 1.0 mg mL—1. (C) TEM images showing morphologies of
DSeSeD-based NPs. Scale bar: 100 nm. (D) Absorption spectra for DSeSeD (in DMSO, 50 μg mL—1), free SW033291 (in DMSO, 7.5 μg mL—1) and SW@DSeSeD
NPs (in pH 7.4 PBS, 50 μg mL—1). (E) Absorption (red)
and emission (blue, λex = 730 nm) spectra for ICG- SW@DSeSeD NPs (100 μg mL—1 in pH 7.4 PBS). (F)
Change in particle diameter as determined by DLS for
SW@DSeSeD NPs at varied time upon treatment with H2O2 of different concentrations at 37 ◦C in pH 7.4 PBS. (G) Release profile of SW033291 from
SW@DSeSeD NPs upon treatment with 200 μM or 500 μM H2O2 for 4 h at 37 ◦C in pH 7.4 PBS con- taining 5% DMSO. (H) High-resolution mass spec-
trum (HR-MS) for molecular DSeSeD before (I) and after (II) treatment with 200 μM H2O2 in methanol. (For interpretation of the references to color in this
figure legend, the reader is referred to the Web version of this article.)

(TEM), and the results are presented in Fig. 1B, 1C and Supplementary
Fig. S11. It was found that the diameter of DSeSeD NPs could be manipulated from ca. 50 nm–200 nm by varying the initial concentra-
tion (before being added into water) of DSeSeD in DMSO, and higher initial concentration led to higher particle diameter, as shown in Sup-
plementary Fig. S11. The average hydrodynamic diameter for the par- ticles prepared with the initial concentration of 1.0 mg mL—1 DseSeD in
DMSO is about 60 nm, which was found stable in aqueous media and suitable for accumulation in liver, were used in subsequent experiments. TEM images reveal that the nanoparticles all exhibited spherical shape. The Zeta potentials for different nanoparticles were determined and shown in Supplementary Fig. S12, which provides evidence for the incorporation of ICG and iron ions onto SW@DSeSeD NPs. It can be seen from Supplementary Fig. S13, the SW@DSeSeD NPs exhibit good dis- persibility and stability in pH 7.4 PBS.
Absorption spectra for DSeSeD, SW033291 and SW@DSeSeD NPs were shown in Fig. 1D, which suggest the successful incorporation of SW033291 in DSeSeD NPs. The absorption and emission spectra of ICG-

SW@DSeSeD NPs were shown in Fig. 1E, which indicates that the nanoparticle can be used for bio-imaging. The drug loading of SW033291 into DSeSeD NPs were determined by absorption spectrom- etry according to calibration curves shown in Supplementary Fig. S14. For the samples used in the subsequent experiments, the loading ca- pacity of SW033291 in SW@DSeSeD NPs is 9.8 wt%. The response of SW@DSeSeD NPs to H2O2 (a reactive oXidative species) was observed on TEM, and TEM images (Supplementary Fig. S15) show that after incu- bation with H2O2, deformation and aggregation of SW@DSeSeD parti-
cles occurred. Fig. 1F also shows that, the hydrodynamic diameters for SW@DSeSeD increased upon incubation with 200 μM or 500 μM H2O2,
indicating the gradual aggregation of SW@DSeSeD nanoparticles upon treatment by H2O2; in contrast, under normal physiological condition, the hydrodynamic diameter for SW@DSeSeD NPs remained unchanged for a long period of time (7 days). The release profile of SW033291 from SW@DSeSeD was measured by absorption spectrometry and the result is given in Fig. 1G. After 4 h of incubation with H2O2, about 85% of SW033291 had been released. High-resolution mass spectrometry (HR-

MS) results shown in Fig. 1H demonstrate that the DSeSeD turned into selenic acid upon incubation with H2O2. Notably, as shown in Supple- mentary Fig. S16, L-Dopa, which is the direct precursor of dopamine, was detected by HR-MS after DSeSeD was treated with H2O2 and then with the esterase from rabbit liver. These results clearly demonstrate that the reactive oXidative species such as H2O2 can disrupt the nano- drug SW@DSeSeD and result in the release of SW033291 and L-Dopa- containing selenic acid. The latter can be transformed into L-Dopa by liver esterase and then into dopamine by dopa decarboXylase, which are highly present in liver [33].

3.2. Cell imaging using multispectral optoacoustic tomography (MSOT)

Multispectral optoacoustic tomography (MSOT) imaging was per- formed to evaluate cellular uptake of the ICG-doped nanodrug (ICG-

SW@DSeSeD NPs). The multispectral optoacoustic tomography (MSOT), which is a spectral optoacoustic technique achieved by irra- diating an object with multiple wavelengths and allowing the system to detect ultrasound waves from different photo-absorbing substances in the tissue, has shown its great potential in a wide range of biological
imaging [34–39]. The optoacoustic intensity at different wavelengths
for ICG-SW@DSeSeD NPs was performed in vitro, and the result is given in Supplementary Fig. S17 and Fig. 2A, which confirms the nano- particles could be used for optoacoustic imaging. The RAW264.7 cells were then used for cellular imaging after they were incubated with ICG-SW@DSeSeD NPs for 0, 1, 2, 4, 6 h, washed three times with PBS and suspended in culture medium. As shown in Fig. 2B, the imaging of cell suspension by MOST indicates that the nanodrug could be readily internalized by RAW264.7 cells.

Fig. 2. (A) Optoacoustic (OA) images and corresponding mean OA signal intensities for ICG-SW@DSeSeD NPs of different concentrations in pH 7.4 PBS. EXcitation wavelength: 805 nm. (B) MSOT images and mean MSOT intensities for RAW 264.7 cells incubated with ICG-SW@DSeSeD NPs (50 μg mL—1) for different time. (C) Cell viability of RAW 264.7 cells after treatment with SW@DSeSeD NPs (200.0 μg mL—1), DSeSeD NPs (180.4 μg mL—1), dopamine (86.6 μg mL—1) or free SW033291 (19.6 μg mL—1) for 24 h, respectively. The untreated cells were used as control. (D) Cell viability of RAW 264.7 cells pretreated with H2O2 1 mM for 4 h and then
incubated with PBS, SW@DSeSeD NPs (50.0 μg mL—1), DSeSeD NPs (45.0 μg mL—1), dopamine (21.6 μg mL—1) or free SW033291 (4.9 μg mL—1) for 20 h, respectively.
(E) Cell viability of RAW 264.7 cells pretreated with 200 μM H2O2 for 4 h and then incubated with various formulations of different concentrations for 20 h, respectively. (F) and (G) Inhibitory effects of different formulations on IL-1β and TNF-α production in LPS-stimulated (1 μg mL—1) RAW 264.7 cells, respectively. (H)
PGE2 levels in LPS-stimulated RAW-264.7 cells upon incubation with different formulations for 20 h. For (F), (G) and (H), the concentrations for dopamine, SW033291, DSeSeD and SW@DSeSeD are all 21.6 μg mL—1, 4.9 μg mL—1, 45.0 μg mL—1 and 50.0 μg mL—1 respectively.

3.3. Evaluation of the actions of the nanodrug against ROS-induced damage to cells
First, the cell viabilities for RAW-264.7 cells upon treatment with various formulations are shown in Fig. 2C and Supplementary Fig. S18, which demonstrates that the cells treated with different formulations all exhibit relatively high viability and the formulations are of low cyto-
toXicity. Next, H2O2 (1 mM or 200 μM) was added in the culture media
to simulate the oXidative damage microenvironment before cells were incubated with different drug formulations. As shown in Fig. 2D and 2E, treatment with H2O2 significantly reduced cell viability, and subsequent incubation with various formulations could increase cell viabilities to varying degrees. Among these formulations, the nanodrug SW@DSeSeD led to highest increase in cell viability and thus exerted highest cyto- protective effect, probably due to the combined action by the released drugs and ROS scavenging effect by diselenides.
Also, we assessed the inhibitory effect of different formulations on
some pro-inflammatory cytokines using test kits. Tumor necrosis factor- alpha (TNF-α) and interleukin-1β (IL-1β) are two cytokines that induce inflammatory responses in macrophages and are highly produced in
LPS-stimulated macrophages such as RAW-264.7. As shown in Fig. 2F and 2G, both DSeSeD and SW@DSeSeD NPs significantly inhibited IL-1β

production and moderately inhibited the production of TNF-α in RAW
264.7 cells pretreated with LPS. Since IL-1β is NLRP3-inflammasome- dependent cytokine, while TNF-α is NLRP3-inflammasome- independent one [40], the different inhibition capabilities of DSeSeD
and SW@DSeSeD towards the two cytokines suggest that the two drugs may inhibit the formation of NLRP3 inflammasome by releasing the dopamine precursor.
Prostaglandin E2 (PGE2) is a bioactive lipid that elicits a wide range of biological effects associated with inflammation, and their biosyn- thesis is significantly increased in inflamed tissue [41]. On the other hand, it also serves as a cell proliferation promoter that supports expansion of several types of tissue stem cells and has recently become a candidate therapeutic target for promoting tissue regeneration [42,43]. Thus, the higher cellular PGE2 level is beneficial to cell and tissue regeneration. In this study, the levels of PGE2 in RAW 264.7 cells pre- treated with LPS and then treated with various formulations were determined, as shown in Fig. 2H. As expected, pretreatment with LPS enhanced PGE2 level in RAW 264.7 cells and subsequent incubation with SW033291 or the nanodrug SW@DSeSeD NPs further increased PGE2 level, as SW033291 can inhibit 15-hydroXyprostaglandin dehy- drogenase (15-PGDH, a major degrading enzyme of PGE2) and therefore enhance the expression of PGE2. In contrast, the DSeSeD, which

Fig. 3. (A) Fluorescence images of RAW 264.7 cells pre-treated with 200 μM H2O2 and then incubated with SW@DSeSeD NPs (two concentrations), DSeSeD NPs (two concentrations) and free drugs (21.6 μg mL—1 dopamine, 4.9 μg mL—1 SW033291) for 4 h, respectively. The cells were stained with ROS probe DCFH-DA for intercellular ROS evaluation. Scale bar: 20 μm. (B) Representative flow cytometry profiles (DCFH-DA staining) for RAW 264.7 cells pre-treated with 200 μM H2O2 and then treated with various formulations (50 μg mL—1 SW@DSeSeD NPs, 45.0 μg mL—1 DSeSeD NPs, 21.6 μg mL—1 dopamine, 4.9 μg mL—1 SW033291). (C) Live/
dead staining with Calcein-AM/PI for NCTC-1469 cells incubated with supernate (SN) of culture media of RAW 264.7 cells pre-treated with LPS and then incubated with different formulates. “Control-”: negative control without any treatment; “Control+“: positive control with LPS pretreatment only. Scale bar: 20 μm.

contained no SW033291, exhibited little effect on PGE2 level. Moreover,
the nanodrug SW@DSeSeD caused even higher PGE2 level than the molecular SW033291, and the reason is that, the molecular Sw033291’s water solubility is very low (Figure S13), it tends to precipitate in
aqueous media (including the cell culture medium). This reduces its bioavailability and compromises its capability in promoting PGE2 level in cells. The SW@DSeSeD, on the other hand, can release SW033291 slowly in cells, thereby showing higher effect than molecular SW033291.
3.4. Capability of SW@DSeSeD NPs in ROS eliminating and cytokines secretion inhibition
Fluorescence imaging was first applied to investigate ROS elimi- nating capability of SW@DSeSeD NPs in RAW 264.7 cells using fluo- rescence imaging (Fig. 3A). A ROS indicator DCFH-DA (DCF), which exhibits green fluorescence in the presence of ROSs such as H2O2 in live cells, was applied for monitoring the ROS levels during the incubation. The nanodrug SW@DSeSeD and some other formulations (dopamine,
SW033291 and DSeSeD NPs) were incubated for 4 h with RAW264.7 cells pretreated with 200 μM H2O2. Compared to other formulations, the
DSeSeD NPs and SW@DSeSeD NPs could reduce the intracellular H2O2 levels more greatly. We suppose the diselenide moiety in DSeSeD NPs and SW@DSeSeD NPs may have served as a strong ROS scavenger and thereby eliminated the overdosed H2O2 in cells. The drug SW033291 promotes cell and tissue regeneration by facilitating PGE2 expression through inhibiting 15-PGDH, however, no previous research results have proved or implied that it can clean ROS in cells and tissues. That is the possible reason why SW033291 relieved the ROS-induced damage to macrophages (Fig. 2D), but could not scavenge the ROS. The flow cytometry analysis also supports this observation, as shown in Fig. 3B. Further, we determined the generation of L-Dopa in RAW264.7 cells by measuring the L-Dopa concentrations in cell lysates using HPLC, as shown in supplementary Fig. S19. This result verified the intracellular release of L-Dopa.
Liver is a main metabolic organ, and the prevailing assumption suggests that the majority of the nanoparticles are taken by stationary Kupffer cells (a kind of hepatic macrophage) or internalized by the patrolling macrophages and further transfer to other hepatocytes [44, 45]. Previous researches demonstrated that, delivery of nanosized drugs to hepatic macrophages can ameliorate liver injury by blocking mac- rophages from producing and secreting pro-inflammation cytokines [46]. In this study, to explore how the cytokines secretion of macro- phages affects liver cells, we incubated various formulations with the LPS-pretreated RAW264.7 cells (a common hepatic macrophage simu- lant [47]) for 20 h, and extracted the resultant culture supernatants by centrifugation, which were then used to incubate with NCTC-1469 cells (an in vitro model of hepatocytes [48]) for 4 h. The live/dead cell staining was performed, with living cells stained by calcein (green) and dying cells stained by propidium iodide (PI, red). As shown in Fig. 3C, compared to other formulations and the positive control, the SW@DSeSeD group showed much less dying cells, suggesting that after treatment with SW@DSeSeD NPs, the macrophages secreted less pro-inflammation cytokines which could lead to apoptosis of other he- patic cells in the microenvironment [49].
3.5. Establishment of ALF mouse model and multi-mode imaging of ALF in vivo
We established an ALF mouse model by injecting N-acetyl-p-ami- nophenol (APAP) at a sublethal dose of 300 mg kg—1 into male C57BL/6
mice. We performed immunohistochemical (IHC) assays to confirm the accumulation of macrophages in ALF lesion which is suitable for retention of the nanodrug in liver. We chose F4/80-antibody as a mouse mature macrophage marker to locate the mature macrophages in mice liver [50]. As shown in Fig. 4A, the macrophages in WT (healthy) mice

liver were normal in quantity and evenly distributed all over the liver section, whereas the macrophages vastly concentrated at the specific areas in ALF mice liver. This result confirms the enrichment of macro- phages in ALF lesion. We then used three imaging modalities (near infrared fluorescence, multispectral optoacoustic tomography and magnetic resonance imaging) to track the accumulation of the nanodrug in liver region, as shown in Fig. 4B-G. A simplified diagram of mouse liver anatomy was given in Fig. 4B to display the positions of 4 lobes of mouse liver. Fig. 4C shows the near infrared (NIR) fluorescence imaging of wild-type mice (the healthy mice as the control) and ALF mice at varied time upon i.v. injection of ICG-SW@DSeSeD NPs. Clearly, the ICG-doped nanodrug could accumulate in both healthy mice and ALF mice shortly after nanodrug administration, verifying the accumulation of nanodrug in liver region upon administration. We then employed multispectral optoacoustic tomography (MSOT) to obtain images with more detailed information for the nanodrug in liver (Fig. 4D). After injection of ICG-doped nanodrug, the cross-sectional MSOT signal in liver region of the healthy mice and ALF mice emerged and gradually increased; and in contrast to that in healthy ones, more intense MSOT signal in ALF mice can be identified at the medial and left lobe of live by referring a cryostat section (Fig. 4E) at the same cross-section of the male mouse, in which the lobes of the mouse liver were marked based on literature report [51]. By selecting a region of interest (ROI) in the medial lobe (ML) and left lobe (LL) of the mice of the two groups and obtaining the relative mean MSOT intensities of ROI, we could demonstrate the MSOT signal intensity difference between the WT and ALF mice more clearly. As shown in Fig. S20, more intense MSOT signals could be observed in the medial lobe and left lobe of liver in ALF mice than those in WT mice. Moreover, in contrast to the observation that fluorescence signals began to decrease from 60 min, the MSOT signals almost remained unchanged at 120 min post-injection. This discrepancy was observed previously by other group [52], which may be due to the different detection mechanism between fluorescence and MSOT.
Moreover, ICG-SW@DSeSeD NPs, which contain iron ions, allowed us to use magnetic resonance imaging (MRI) to further confirm the location of the nanodrug. Fig. 4F reveals that, the nanodrug displays concentration-dependent MRI signal intensities in PBS, suggesting that the nanodrug is suitable for MRI imaging. Fig. 4G reveals that, the positive signal enhancement of T1-weighted MRI could be observed in both healthy and ALF mice after intraperitoneal injection of ICG- SW@DSeSeD NPs. After 120 min, the nanodrug retained in almost the whole liver area for the healthy ones, while it retained mainly in middle and left lobe of liver in ALF mice. This MRI result is largely in consistent with that of the MSOT, except that the MRI images could provide more detailed information in depth due to the highest penetration depth of MRI among three imaging modalities.
ICG has been widely used in liver function test. After i.v. injection it binds to plasma proteins, and is then extracted by healthy livers almost exclusively, thus the plasma disappearance rate of ICG (ICG-PDR) is the most commonly-used parameter for clinical and experimental assess- ment of liver function [53,54]. Based on the similar principle, we used MSOT to monitor the ICG-doped SW@DSeSeD in the blood passing through the jugular vein to access the liver function of the healthy and ALF mice. The cross-sectional MSOT images at mice neck shown in
Fig. 4H reveal that, after i.v. injection, the ICG-doped nanodrug’s signal
increased quickly in the jugular vein of the healthy and ALF mice. Af- terwards, the signal intensities gradually decreased, but the decreasing rate for ALF mice was obviously lower than that for healthy (Fig. 4I). This indicates that the elimination (extraction) of nanodrug from the bloodstream by the ALF mice is much less efficient, which is caused by the poorer liver function of the ALF mice. The above observations indicate that, the nanodrug can accumulate in liver region and may serve as a potential indicator for ALF evaluation.

Fig. 4. (A) Representative images of IHC analysis of liver section from WT and ALF mice at different time points post APAP injection. The F4/80+ macrophage cells were stained with brown color. Scale bar: 200 μm. (B) Simplified diagram of mouse liver anatomy (in upright position). ML: medial lobe, LL: left lobe, RL: right lobe, CL: caudate lobe. (C) Representative NIR fluorescence images of WT and ALF mice at varied time after intravenous injection of ICG-SW@DSeSeD. λex = 808 nm. (D) Cross-sectional MSOT images at liver site for WT and ALF mice at varied time upon injection of ICG-SW@DSeSeD. The images were merged ones with both signals
from ICG (in color) and the background (gray-scale). Label 1 represents the location of spinal cord. (E) A cryosection image of a male mouse with the cross section’s location comparable to those shown in (D). The lobes of the liver are marked according to a literature report. (F) T1 relaxation rate for ICG-SW@DSeSeD and the T1-
weighted MR images for pure PBS and ICG-SW@DSeSeD of varied concentrations. (G) In vivo T1-weighted MR images for WT and ALF mice upon injection with ICG- SW@DSeSeD NPs for varied time. (H) Cross-sectional MSOT images at neck site for WT and ALF mice before and after ICG-SW@DSeSeD injection. (I) Mean cross- sectional MOST intensities at jugular vein site. Data are represented as the mean ± SD (n = 3). (For interpretation of the references to color in this figure legend, the
reader is referred to the Web version of this article.)

3.6. Therapeutic efficacy evaluation

We then evaluated the therapeutic efficacy of different formulations with the experimental outline shown in Fig. 5A. Male C57BL/6 mice were randomly distributed into several groups with 5 mice in each
group. One group was used as the control (WT), and the mice in treat- ment groups were injected with APAP (200, 300 or 500 mg kg—1 for
different experiments) and then respectively subject to treatment with free dopamine (4.3 mg kg—1), free SW033291 (0.98 mg kg—1, dispersed in 5% Tween-80), DSeSeD NPs (9.02 mg kg—1) or SW@DSeSeD NPs (10.0 mg kg—1) at 6 h, 12 h, 24 h and 48 h post APAP injection. To
evaluate the therapeutic efficacy, serum alanine aminotransferase (ALT) and aspartate transaminase (AST) levels were measured at different time points (Fig. 5B and 5C). The level of serum ALT and AST were dramat- ically elevated by APAP intoXication, which indicated the acute liver injury. After treatment with SW@DSeSeD, serum ALT and AST were significantly reduced, whereas the treatment with free SW033291, free dopamine (which is very unstable in biological system) or DSeSeD resulted in a certain degree of reduction in liver function markers as

compared to the positive control group. Furthermore, after intraperi-
toneal injection of lethal dosage of APAP (500 mg kg—1), all 5 mice without treatment were dead within 24 h, whereas mice treated with
SW@DSeSeD showed highest survival rate, as shown in Fig. 5D. In addition, we also performed the survival experiment by intraperitoneal
injection of sub-lethal dosage of APAP (300 mg kg—1), and no death was found for all groups, as shown in Fig. S21.
The therapeutic efficacies of different formulations were also eval- uated by MSOT imaging as we demonstrated the diagnostic ability of our ICG-doped nanodrug in Fig. 4. The ALF mice were respectively treated with different formulations (SW033291, dopamine, DSeSeD or SW@DSeSeD) for 72 h, and then injected the ICG-doped nanodrug (ICG- SW@DSeSeD) into the bloodstream of the mice via tail vein to evaluate the therapeutic efficacy of the different formulations by tracking the MSOT signal intensities of ICG-SW@DSeSeD in bloodstream at jugular vein. As shown in Fig. 5E, the MSOT signal for the mice treated with nanosized formulations (especially the SW@DSeSeD) decreased more quickly than those treated with molecular ones (dopamine and SW033291), since the cured mice with improved liver function could

Fig. 5. Evaluation of therapeutic efficacy of different formulations. (A) EXperimental outline for treatment
course of ALF mice. The dosage of APAP injection was 300 mg kg—1 for most experiments, and that for sur- vival curve determination was 500 mg kg—1. (B) and
(C) Serum ALT and AST levels determined by ELISA assay for WT group (control) and treatment groups at three time points. (D) Survival curve for WT and five treatment groups (n 5 per group) after injection of a
lethal dose of APAP (500 mg kg—1). (E) Upper panel:
representative cross-sectional MSOT images at neck site for ALF mice at 10 min post injection of ICG- SW@DSeSeD after the ALF mice were treated with different formulations for 72 h. Lower panel: the corresponding mean MSOT intensities in ROI at different time post injection of ICG-SW@DSeSeD for the mice. (F) Representative cross-sectional MSOT images (upper) at liver site for ALF mice at 120 min post-injection of ICG-SW@DSeSeD after the mice were treated with different formulations for 72 h and the corresponding mean cross-sectional MOST in- tensities (lower). Label 1 represents the location of spinal cord. (G) Western blotting analyses illustrating the level of p-JNK, caspase 3, NLRP3 and 15-PGDH in liver tissues dissected from the control (WT) and different treatment groups after euthanasia: APAP
injection (200 mg kg—1 or 300 mg kg—1) only, and
treatment with dopamine, SW, DSeSeD NPs or SW@DSeSeD NPs after injection of 300 mg kg—1 APAP. All data are represented as the mean ± SD. *P
< 0.05, **P < 0.01, and ***P < 0.001. (H) Repre- sentative photographs of dissected liver from WT, untreated ALF or SW@DSeSeD-treated ALF mouse after euthanasia. (I) representative H&E staining and IHC analyses (caspase-3 and Ki67) for different groups. Scale bar: 200 μm for H&E and 100 μm for IHC. metabolize the dye-doped nanodrug more quickly. Compared to the MSOT result for healthy (WT) and ALF (without treatment) mice (Fig. 4H and I), the mice treated with SW@DSeSeD showed a much quicker rate of nanoparticle clearance than ALF mice and a slightly lower clearance rate than WT mice. In contrast, the mice treated with free drugs exhibited an obviously slower clearance rate than WT mice and a quicker rate than ALF mice. These imaging results suggest that, the nanosized formulations generally exhibited higher therapeutic efficacies than the molecular ones. It is worth noting that, despite of lower elim- ination capability of the damaged liver, the nanosized drugs would eventually accumulate in liver for ALF therapy [55]. In addition to the images at jugular vein, we also obtained MSOT images of the ICG-doped nanodrug at liver site of different groups of mice, as shown in Fig. 5F. We found the differences in MSOT intensities at liver site among different groups are less significant than those at jugular vein. Following the treatment assessments, we evaluated the level of key proteins involving in the liver injury (p-JNK, caspase-3 and NLRP3) and liver regeneration (15-PGDH) in liver tissue. Fig. 5G shows that, the treatment with SW@DSeSeD significantly attenuated the level of p-JNK (the phosphorylated c-Jun N-terminal kinase in the JNK signaling pathway that plays an active role in inflammatory responses) and cleaved caspase-3 (an executioner caspase that induces apoptosis of cells including hepatic cells), as well as NLRP3 and 15-PGDH. Notably, since ALF has recently been found a NLRP3-related disease [23], suppression of NLRP3 (NOD-, LRR- and pyrin domain-containing protein 3) could inhibit the activation of NLRP3 inflammasome and attenuate the release of pro-inflammatory cytokines (such as IL-1β and IL-18) and the subse- quent recruitment of inflammatory cells, thus alleviating the excessive inflammation and preventing further hepatic cell death [24]. In addi- tion, suppression of 15-PGDH demonstrated the successful delivery of SW033921 from SW@DSeSeD NPs as expected, which can promote tissue regeneration by increasing PGE2 level. The other formulations, however, could not significantly attenuate the levels of all four proteins. Typical photographs (Fig. 5H) for the liver dissected from the WT, ALF and SW@DSeSeD-treated mice show that, the liver injury can be observed as liver congestion and increased roughness and these symp- toms disappeared after the treatment of SW@DSeSeD, indicating the high efficacy of SW@DSeSeD in ALF treatment. 4. Conclusions In summary, we have constructed a diselenide based nanosystem for treatment of ALF, which is the first effort of utilizing a nano-prodrug formed by molecular diselenide in liver disease treatment. With the inherent anti-inflammation property of diselenides and through controlled release of SW033921 and large dose of dopamine precursor in liver injury lesion, this system can not only prevent further liver injury by inhibiting formation of NLRP3 inflammasome but also stimulate liver’s regeneration via downregulation of 15-PGDH, thereby achieving superior therapeutic efficacy for ALF. In addition, upon incorporation of a NIR dye or ferric ion on its surface, this nanosystem serves as a multifunctional platform which can be employed to track its bio- distribution and monitor liver function evolution through NIR fluores- cence, optoacoustic and T1-weighting MRI imaging. This strategy may provide a new approach for designing theranostic platforms for treat- ment of other inflammasome-related diseases. Data availability statement The data supporting the findings of this study are available within the article and the supplementary data. CRediT authorship contribution statement Chenyue Zhan: Conceptualization, Methodology, Investigation, Data curation, Writing - original draft. Guifang Lin: Methodology, Investigation, Data curation, Writing - original draft. Yong Huang: Methodology, Investigation, Data curation. Ziqian Wang: Methodol- ogy, Investigation, Data curation. Fang Zeng: Conceptualization, Methodology, Supervision, Writing - review & editing. Shuizhu Wu: Conceptualization, Methodology, Supervision, Writing - review & editing. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence Furthermore, as revealed in Fig. 5I, hematoXylin-eosin (H&E) staining of liver tissue sections from different groups indicates that, APAP caused severe liver injury evidenced by distinct leukocyte infil- tration and disruption of hepatocytes; while the treatment with SW@DSeSeD suppressed the liver injury significantly, and other for- mulations were not so effective. Furthermore, we observed that treat- ment with SW@DSeSeD also led to reduced level of cleaved caspase-3 and increased level of Ki67 as determined by IHC analysis (Fig. 5I and Fig. S22). These results further confirm that application of SW@DSeSeD in ALF mice results in reduction of apoptosis and increase of liver regeneration (via enhancing Ki67 expression). The aforementioned results indicate that, the nanodrug SW@DSeSeD can achieve combined therapy against acute liver failure. SW@DSeSeD has abundant diselenide bonds in its structure, which scavenge excessive reactive oXidative species and reduce the level of some pro- inflammatory cytokines such as IL-1beta and TNF-alpha. It releases a molecular drug SW033291 to achieve liver regeneration by enhancing the PGE2 level, and also releases dopamine precursor that can be transformed into dopamine by endogenous enzymes to inhibit the for- mation of NLRP3. In contrast, free SW033291 or dompamine does not exhibit multiple functions, the former enhances PGE2 level and the latter inhibits NLRP3 formation. Moreover, free SW033291 exhibits low solubility in aqueous media and dopamine is of low stability in biolog- ical systems, while the nanodrug is water dispersible and responsive to inflammatory environment. The multi-function property makes SW@DSeSeD more effective in ALF treatment than other formulations.

the work reported in this paper.


This work was supported by NSFC (21875069 and 51673066), the Natural Science Foundation of Guangdong Province (2016A030312002) and the Fund of Guangdong Provincial Key Laboratory of Luminescence from Molecular Aggregates (2019B030301003).

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.biomaterials.2020.120573.

[1] W. Bernal, J. Wendon, Acute liver failure, N. Engl. J. Med. 369 (2013) 2525–2534, https://doi.org/10.1056/NEJMra1208937.
[2] M.D. Torres, R.D. Stevens, A. Gurakar, Acute liver failure: a management challenge for the practicing gastroenterologist, Gastroenterol. Hepatol. 6 (2010) 444–450.
[3] E.S. Bjo¨rnsson, Drug-induced liver injury: an overview over the most critical
compounds, Arch. ToXicol. 89 (2015) 327–334, https://doi.org/10.1007/s00204-
[4] N. Magee, R. Convery, Medical mystery—the answer, N. Engl. J. Med. 349 (2003) 1974, https://doi.org/10.1056/nejm200311133492020.
[5] S.A. Brown, E. AXenfeld, E.G. Stonesifer, W. Hutson, S. Hanish, J.P. Raufman, N.
H. Urrunaga, Current and prospective therapies for acute liver failure, Dis. Mon. 64 (2018) 493–522, https://doi.org/10.1016/j.disamonth.2018.10.001.
[6] W. Bernal, G. Auzinger, A. Dhawan, J. Wendon, Acute liver failure, Lancet 376 (2010) 190–201, https://doi.org/10.1016/s0140-6736(10)60274-7.

[7] L.J. Chun, M.J. Tong, R.W. Busuttil, J.R. Hiatt, Acetaminophen hepatotoXicity and acute liver failure, J. Clin. Gastroenterol. 43 (2009) 342–349, https://doi.org/ 10.1097/mcg.0b013e31818a3854.
[8] M. Doulberis, G. Kotronis, D. Gialamprinou, O. O¨ zgüler, A.K. EXadaktylos,
V. Oikonomou, P. Katsinelos, I. Romiopoulos, S.A. Polyzos, D. Tzivras, G. Deretzi,
E. Dardiotis, J. Kountouras, Acute liver failure, J. Clin. Gastroenterol. 53 (2019) 89–101, https://doi.org/10.1097/mcg.0000000000001162.
[9] M. Wang, X. Yang, P. Zhang, L. Cai, X. Yang, Y. Chen, Y. Jing, J. Kong, X. Yang, F.
L. Sun, Sustained delivery growth factors with polyethyleneimine-modified nanoparticles promote embryonic stem cells differentiation and liver regeneration, Adv. Sci. 3 (2016) 1500393, https://doi.org/10.1002/advs.201500393.
[10] Y. Tang, Z. Zeng, X. He, T. Wang, X. Ning, X. Feng, SiRNA crosslinked nanoparticles for the treatment of inflammation-induced liver injury, Adv. Sci. 4 (2017) 1600228, https://doi.org/10.1002/advs.201600228.
[11] H. Liang, K. Huang, T. Su, Z. Li, S. Hu, P. Dinh, E.A. Wrona, C. Shao, L. Qiao, A.
C. Vandergriff, M.T. Hensley, J. Cores, T. Allen, H. Zhang, Q. Zeng, J. Xing, D.
O. Freytes, D. Shen, Z. Yu, K. Cheng, Mesenchymal stem cell/red blood cell- inspired nanoparticle therapy in mice with carbon tetrachloride-induced acute liver failure, ACS Nano 12 (2018) 6536–6544, https://doi.org/10.1021/
[12] E. Ko, D. Jeong, J. Kim, S. Park, G. Khang, D. Lee, AntioXidant polymeric prodrug microparticles as a therapeutic system for acute liver failure, Biomaterials 35
(2014) 3895–3902, https://doi.org/10.1016/j.biomaterials.2014.01.048.
[13] Y. Go, H. Lee, L. Jeong, S. Sun, E. Hong, E. Jung, C. Ko, J. Noh, S. Park, M. Lee,
C. Song, D. Lee, Acid-triggered echogenic nanoparticles for contrast-enhanced
ultrasound imaging and therapy of acute liver failure, Biomaterials 186 (2018) 22–30, https://doi.org/10.1016/j.biomaterials.2018.09.034.
[14] Z. Liu, Y. Li, W. Li, C. Xiao, D. Liu, C. Dong, M. Zhang, E. Ma¨kil¨a, M. Kemell,
J. Salonen, J.T. Hirvonen, H. Zhang, D. Zhou, X. Deng, H.A. Santos, Multifunctional nanohybrid based on porous silicon nanoparticles, gold nanoparticles, and acetalated dextran for liver regeneration and acute liver failure theranostics, Adv. Mater. 30 (2018) 1703393, https://doi.org/10.1002/adma.201703393.
[15] A.H.A. Moustafa, E.M. Ali, T.M. Mohamed, H.I. Abdou, OXidative stress and thyroid hormones in patients with liver diseases, Eur. J. Intern. Med. 20 (2009)
703–708, https://doi.org/10.1016/j.ejim.2009.08.003.
[16] H. Cichoz˙-Lach, A. Michalak, OXidative stress as a crucial factor in liver diseases,
World J. Gastroenterol. 20 (2014) 8082–8091, https://doi.org/10.3748/wjg.v20. i25.8082.
[17] R.F. Schwabe, D.A. Brenner, Mechanisms of liver injury. I. TNF-α-induced liver injury: role of IKK, JNK, and ROS pathways, Am. J. Physiol. Gastrointest. Liver Physiol. 290 (2006) G583–G589, https://doi.org/10.1152/ajpgi.00422.2005.
[18] H. Jaeschke, A. Ramachandran, Reactive oXygen species in the normal and acutely injured liver, J. Hepatol. 55 (2011) 227–228, https://doi.org/10.1016/j. jhep.2011.01.006.
[19] S. Sutti, F. Tacke, Liver inflammation and regeneration in drug-induced liver injury: sex matters!, Clin. Sci. 132 (2018) 609–613, https://doi.org/10.1042/ cs20171313.
[20] S. Mariathasan, K. Newton, D.M. Monack, D. Vucic, D.M. French, W.P. Lee,
M. Roose-Girma, S. Erickson, V.M. DiXit, Differential activation of the inflammasome by caspase-1 adaptors ASC and Ipaf, Nature 430 (2004) 213–218, https://doi.org/10.1038/nature02664.
[21] T. Strowig, J. Henao-Mejia, E. Elinav, R. Flavell, Inflammasomes in health and disease, Nature 481 (2012) 278–286, https://doi.org/10.1038/nature10759.
[22] G. Szabo, J. Petrasek, Inflammasome activation and function in liver disease, Nat.
Rev. Gastroenterol. Hepatol. 12 (2015) 387–400, https://doi.org/10.1038/ nrgastro.2015.94.
[23] B.L. Woolbright, H. Jaeschke, Role of the inflammasome in acetaminophen- induced liver injury and acute liver failure, J. Hepatol. 66 (2017) 836–848, https:// doi.org/10.1016/j.jhep.2016.11.017.
[24] Y. Yan, W. Jiang, L. Liu, X. Wang, C. Ding, Z. Tian, R. Zhou, Dopamine controls systemic inflammation through inhibition of NLRP3 inflammasome, Cell 160
(2015) 62–73, https://doi.org/10.1016/j.cell.2014.11.047.
[25] Y. Zhang, A. Desai, S.Y. Yang, K.B. Bae, M.I. Antczak, S.P. Fink, S. Tiwari, J.
E. Willis, N.S. Williams, D.M. Dawson, D. Wald, W. Chen, Z. Wang, L. Kasturi, G.
A. Larusch, L. He, F. Cominelli, L.D. Martino, Z. Djurice, G.L. Milne, M. Chance,
J. Sanabria, C. Dealwis, D. Mikkola, J. Naidoo, S. Wei, H. Tai, S.L. Gerson, J.
M. Ready, B. Posner, J.K.V. Willson, S.D. Markowitz, Inhibition of the prostaglandin-degrading enzyme 15-PGDH potentiates tissue regeneration, Science 348 (2015) aaa2340, https://doi.org/10.1126/science.aaa2340.
[26] M. Li, X. Sun, N. Zhang, W. Wang, Y. Yang, H. Jia, W. Liu, NIR-activated polydopamine-coated carrier-free “nanobomb” for in situ on-demand drug release, Adv. Sci. 5 (2018) 1800155, https://doi.org/10.1002/advs.201800155.
[27] H. Wang, J. Wu, K. Xie, T. Fang, C. Chen, H. Xie, L. Zhou, S. Zheng, Precise engineering of prodrug cocktails into single polymeric nanoparticles for combination cancer therapy: extended and sequentially controllable drug release,
ACS Appl. Mater. Interfaces 9 (2017) 10567–10576, https://doi.org/10.1021/
[28] D. Shao, M. Li, Z. Wang, X. Zheng, Y.H. Lao, Z. Chang, F. Zhang, M. Lu, J. Yue,
H. Hu, H. Yan, L. Chen, W. Dong, K.W. Leong, Bioinspired diselenide-bridged mesoporous silica nanoparticles for dual-responsive protein delivery, Adv. Mater. 30 (2018) 1801198, https://doi.org/10.1002/adma.201801198.
[29] B. Sun, C. Luo, X. Zhang, M. Guo, M. Sun, H. Yu, Q. Chen, W. Yang, M. Wang,
S. Zuo, P. Chen, Q. Kan, H. Zhang, Y. Wang, Z. He, J. Sun, Probing the impact of sulfur/selenium/carbon linkages on prodrug nanoassemblies for cancer therapy, Nat. Commun. 10 (2019) 1–10, https://doi.org/10.1038/s41467-019-11193-X.

[30] Y. Li, Y. Li, W. Ji, Z. Lu, L. Liu, Y. Shi, G. Ma, X. Zhang, Positively charged polyprodrug amphiphiles with enhanced drug loading and reactive oXygen species-
responsive release ability for traceable synergistic therapy, J. Am. Chem. Soc. 140 (2018) 4164–4171, https://doi.org/10.1021/jacs.8b01641.
[31] H. Xu, W. Cao, X. Zhang, Selenium-containing polymers: promising biomaterials
for controlled release and enzyme mimics, Acc. Chem. Res. 46 (2013) 1647–1658, https://doi.org/10.1021/ar4000339.
[32] D. Hu, C. Liu, L. Song, H. Cui, G. Gao, P. Liu, Z. Sheng, L. Cai, Indocyanine green- loaded polydopamine-iron ions coordination nanoparticles for photoacoustic/ magnetic resonance dual-modal imaging-guided cancer photothermal therapy,
Nanoscale 8 (2016) 17150–17158, https://doi.org/10.1039/c6nr05502h.
[33] M. Ando-Yamamoto, H. Hayashi, T. Sugiyama, H. Fukui, T. Watanabe, H. Wada, Purification of l-dopa decarboXylase from rat liver and production of polyclonal
and monoclonal antibodies against it, J. Biochem. 101 (1987) 405–414, https://
[34] P. Beard, Biomedical photoacoustic imaging, Interface Focus 1 (2011) 602–631, https://doi.org/10.1098/rsfs.2011.0028.
[35] J. Weber, P.C. Beard, S.E. Bohndiek, Contrast agents for molecular photoacoustic imaging, Nat. Methods 13 (2016) 639–650, https://doi.org/10.1038/nmeth.3929.
[36] D. Razansky, M. Distel, C. Vinegoni, R. Ma, N. Perrimon, R.W. Ko¨ster,
V. Ntziachristos, Multispectral opto-acoustic tomography of deep-seated fluorescent proteins in vivo, Nat. Photon. 3 (2009) 412–417, https://doi.org/ 10.1038/nphoton.2009.98.
[37] Y. Wu, S. Huang, J. Wang, L. Sun, F. Zeng, S. Wu, Activatable probes for diagnosing and positioning liver injury and metastatic tumors by multispectral optoacoustic tomography, Nat. Commun. 9 (2018) 3983, https://doi.org/10.1038/s41467-018- 06499-1.
[38] Y. Wu, J. Chen, L. Sun, F. Zeng, S. Wu, A nanoprobe for diagnosing and mapping lymphatic metastasis of tumor using 3D multispectral optoacoustic tomography owing to aggregation/deaggregation induced spectral change, Adv. Funct. Mater. 29 (2019) 1807960, https://doi.org/10.1002/adfm.201807960.
[39] J. Ouyang, L. Sun, Z. Zeng, C. Zeng, F. Zeng, S. Wu, Nanoaggregate probe for breast cancer metastasis through multispectral optoacoustic tomography and
aggregation-induced NIR-I/II fluorescence imaging, Angew Chem. Int. Ed. Engl. 59 (2020) 10111–10121, https://doi.org/10.1002/anie.201913149.
[40] R.C. Coll, A.A. Robertson, J.J. Chae, S.C. Higgins, R. Mun˜oz-Planillo, M.C. Inserra,
I. Vetter, L.S. Dungan, B.G. Monks, A. Stutz, D.E. Croker, M.S. Butler,
M. Haneklaus, C.E. Sutton, G. Nún˜ez, E. Latz, D.L. Kastner, K.H.G. Mills, S.
L. Masters, K. Schroder, M.A. Cooper, L.A.J. O’Neill, A small-molecule inhibitor of the NLRP3 inflammasome for the treatment of inflammatory diseases, Nat. Med. 21 (2015) 248–255, https://doi.org/10.1038/nm.3806.
[41] E. Ricciotti, G.A. FitzGerald, Prostaglandins and inflammation, arterioscler, thromb. vasc. Biol 31 (2011) 986–1000, https://doi.org/10.1161/ ATVBAHA.110.207449.
[42] D.A. Rudnick, D.H. Perlmutter, L.J. Muglia, Prostaglandins are required for CREB activation and cellular proliferation during liver regeneration, Proc. Natl. Acad.
Sci. U. S. A 98 (2001) 8885–8890, https://doi.org/10.1073/pnas.151217998.
[43] W. Goessling, T.E. North, S. Loewer, A.M. Lord, S. Lee, C.L. Stoick-Cooper,
G. Weidinger, M. Puder, G.Q. Daley, R.T. Moon, L.I. Zon, Genetic interaction of
PGE2 and wnt signaling regulates developmental specification of stem cells and regeneration, Cell 136 (2009) 1136–1147, https://doi.org/10.1016/j. cell.2009.01.015.
[44] J. Wang, P. Kubes, A reservoir of mature cavity macrophages that can rapidly invade visceral organs to affect tissue repair, Cell 165 (2016) 668–678, https://doi. org/10.1016/j.cell.2016.03.009.
[45] Y. Zhang, W. Poon, A.J. Tavares, I.D. McGilvray, W.C. Chan, Nanoparticle–liver interactions: cellular uptake and hepatobiliary elimination, J. Contr. Release 240 (2016) 332–348, https://doi.org/10.1016/j.jconrel.2016.01.020.
[46] M. Bartneck, K.M. Scheyda, K.T. Warzecha, L.Y. Rizzo, K. Hittatiya, T. Luedde,
G. Storm, C. Trautwein, T.G.M. Lammers, F. Tacke, Fluorescent cell-traceable dexamethasone-loaded liposomes for the treatment of inflammatory liver diseases,
Biomaterials 37 (2015) 367–382, https://doi.org/10.1016/j.
[47] G. Kanuri, I. Bergheim, In vitro and in vivo models of non-alcoholic fatty liver disease (NAFLD), Int. J. Mol. Sci. 14 (2013) 11963–11980, https://doi.org/ 10.3390/ijms140611963.
[48] W. Tang, M. Shen, J. Xie, D. Liu, M. Du, L. Lin, H. Gao, B.R. Hamaker, M. Xie, Physicochemical characterization, antioXidant activity of polysaccharides from Mesona chinensis Benth and their protective effect on injured NCTC-1469 cells
induced by H2O2, Carbohydr. Polym. 18 (2017) 538–546, https://doi.org/
[49] Y.Y. Sun, X.F. Li, X.M. Meng, C. Huang, L. Zhang, J. Li, Macrophage phenotype in liver injury and repair, Scand. J. Immunol. 85 (2017) 166–174, https://doi.org/ 10.1111/sji.12468.
[50] P.J. Leenen, M.F. de Bruijn, J.S. Voerman, P.A. Campbell, W. van Ewijk, Markers of mouse macrophage development detected by monoclonal antibodies, J. Immunol. Methods 174 (1994) 5–19, https://doi.org/10.1016/0022-1759(94)90005-1.
[51] E. Melloul, D.A. Raptis, A. Boss, T. Pfammater, C. Tschuor, Y. Tian, R. Graf, P.
A. Clavien, M. Lesurtel, Small animal magnetic resonance imaging: an efficient tool to assess liver volume and intrahepatic vascular anatomy, J. Surg. Res. 187 (2014)
458–465, https://doi.org/10.1016/j.jss.2013.11.1079.
[52] N. Beziere, N. Lozano, A. Nunes, J. Salichs, D. Queiros, K. Kostarelos,
V. Ntziachristos, Dynamic imaging of pegylated indocyanine green (ICG) liposomes
within the tumor microenvironment using multi-spectral optoacoustic tomography (MSOT), Biomaterials 37 (2015) 415–424, https://doi.org/10.1016/j. biomaterials.2014.10.014.

[53] U. Merle, O. Sieg, W. Stremmel, J. Encke, C. Eisenbach, Sensitivity and specificity of plasma disappearance rate of indocyanine green as a prognostic indicator in acute liver failure, BMC Gastroenterol. 9 (2009) 91, https://doi.org/10.1186/ 1471-230X-9-91.
[54] H. Feng, Q. Li, L. Wang, G. Yuan, W. Cao, Indocyanine green clearance test combined with meld score in predicting the short-term prognosis of patients with

acute liver failure, Hepatob. Pancreat. Dis. 13 (2014) 271–275, https://doi.org/ 10.1016/S1499-3872(14)60040-0.
[55] J.S. Souris, C. Lee, S. Cheng, C. Chen, C. Yang, J.A. Ho, C. Mou, L. Lo, Surface
charge-mediated rapid hepatobiliary excretion of mesoporous silica nanoparticles, Biomaterials 31 (2010) 5564–5574, https://doi.org/10.1016/j. biomaterials.2010.03.048.