Treatment of Acute Kidney Injury Using a Dual Enzyme Embedded Metal-organic Framework-mediated Cascade That Catalyzes in Vivo ROS Scavenging

Background: Signicant advances have been made in recent years to utilize natural enzymes with antioxidant properties for the treatment of acute kidney injury (AKI). However, these enzymes have limited clinical utility due to their limited cellular uptake, poor pharmacokinetic properties, and suboptimal stability. We prepared a novel biomimetic mineralization approach to overcome these limitations. Results: Catalase (CAT) and superoxide dismutase (SOD) were encapsulated in a zeolitic imidazolate framework-8 (ZIF-8), then this SOD@CAT@ZIF-8 complex was anchored with MPEG 2000 -COOH to yield an MPEG 2000 -SOD@CAT@ZIF-8 (PSCZ) composite. This composite was then utilized as a stable tool with antioxidant properties for the integrated cascade-based treatment of AKI, remarkably improved intracellular enzyme delivery. The PSCZ composite showed a signicantly relieve ability of H 2 O 2 induced ROS and prevented apoptosis in HEK293 cells. In vivo, this composite showed a protective effect in the cisplatin-induced AKI mouse model. In addition, PSCZ treatment mitigated renal dysfunction and histological injury, like tubulointerstitial inltration. RNA sequencing results further conrmed the protective effect of PSCZ and the potential molecular mechanism was revealed. PSCZ treatment reduced renal inammation and inammatory gene expression and ameliorated increased apoptosis and inammatory inltration of renal tubules. Moreover, PSCZ administration regulated renal injury via the p53 signaling pathway. These renoprotective effects of PSCZ treatment were more potent in CP-AKI mice than those of SOD or CAT alone. This dual-enzyme-embedded metal-organic framework was able to scavenge reactive oxygen species synergistically and effectively. Conclusion: PSCZ can remarkably improve intracellular enzyme delivery, and it can be used as a stable tool with antioxidant properties for the integrated cascade-based treatment of AKI. The ZIF-8-based “armor plating” represents an effective means of shielding enzymes with improved therapeutic utility to guide the precision medicine-based of AKI.

A range of antioxidant enzymes, including superoxide dismutase (SOD), catalase (CAT), and reduced glutathione (GSH), function as ROS scavengers that can mitigate oxidative stress within a given cellular environment. Given that they are enzymes rather than small molecule therapeutics, these enzymes can catalyze the sustained conversion of toxic ROS species into less damaging byproducts, with SOD facilitating the conversion of •O 2 -into H 2 O 2 and CAT then converting H 2 O 2 into water and O 2 . While they can function in concert with one another, CAT is primarily located within peroxisomes, whereas SOD is typically located in the cytosol or mitochondria, so additional transportation is necessary to facilitate H 2 O 2 degradation. These endogenous enzyme systems show promise for in vivo therapeutic ROS scavenging and associated AKI treatment as effective and well-controlled antioxidants. Such enzymatic cascade reactions have advantages over conventional multi-stage reactions, reducing barriers to diffusion while increasing local intermediate concentrations to enhance overall reaction e ciency at a molecular level. These promising therapeutic properties have led many research groups to design enzymatic scaffold systems to mimic these e cient cascade systems in clinical settings. The in vivo application of these enzyme-based platforms would represent a key breakthrough with the potential to improve patient outcomes. However, several signi cant barriers currently limit the in vivo deployment of these enzyme-based systems, including the fact that they can readily undergo enzymatic degradation, are prone to enzyme aggregate formation, tend to lose catalytic activity throughout prolonged storage, and exhibit poor uptake by target cells owing to their negatively charged surfaces. Embedding or encapsulating these enzymes in a solid carrier framework thus offers an opportunity to overcome these delivery and stability limitations.
Metal-organic frameworks (MOFs) are novel biocompatible materials that exhibit tunable porosity and a high surface area, making them ideal for the packaging of candidate therapeutic biomolecules [1][2][3].
However, these approaches do not allow for precise control of the size of the generated biomacromolecular-MOF composites and are often highly ine cient. A recent biomimetic mineralization strategy has been developed to overcome these limitations [8], leveraging the fact that biomacromolecules exhibit an a nity for metal ions to conduct rapid macromolecule embedding a MOF through the stimulation of zeolitic imidazolate framework-8 (ZIF-8) growth. This approach can shield the enclosed macromolecules within a porous ZIF-8 exoskeleton capable of protecting against denaturation while permitting selective transportation through this porous network structure. ZIF-8 exoskeletal constructs are highly stable and biocompatible under physiological conditions, while they break down when exposed to the more acidic conditions found within solid tumors. Most studies of such MOF biocomposites to date have explored their use as biosensors or catalytic agents in aqueous or organic solutions [9][10][11]. The use of these MOFs to deliver proteins or nucleic acids is, in contrast, a relatively novel approach [12,13]. To our knowledge, no prior studies described the biomimetic mineralization-based encapsulation of dual-enzyme within MOF crystals to enhance ROS scavenging. We hypothesized that a ZIF-8 exoskeleton would be su cient to protect SOD and CAT (SOD@CAT@ZIF-8), improve their stability, and thereby overcome the current barriers to alleviate ROS-mediated renal damage.
In the present study, we aim to develop a MOF encapsulating key antioxidant enzymes using a nanoscale ZIF-8 exoskeleton scaffold containing SOD and CAT (SOD@CAT@ZIF-8). This platform was further functionalized by anchoring MPEG 2000 -COOH to the surface of the MOF complex, yielding more stable and biocompatible composites (Scheme 1). Relative to free CAT and SOD, this integrated MPEG 2000 -SOD@CAT@ZIF-8 (PSCZ) platform exhibited greater SOD and CAT enzymatic e ciency in vitro and was found to synergistically enhance ROS scavenging in vivo, protecting mice against AKI-associated oxidative renal tissue damage. Together, our data indicate that MOF platforms incorporating integrated enzyme cascades offer great promise for the in vivo biomedical treatment of AKI and a range of other in ammatory conditions. Oxygen Species Assay Kit were purchased from Beyotime Company (China). All the aqueous solutions were prepared using puri ed deionized (DI) water puri ed with a puri cation system (Direct-Q3, Millipore, USA). The other solvents used in this work were purchased from Sinopharm Chemical Reagent (China) and Aladdin-Reagent (China).

Construction of SOD@CAT@ZIF-8 through biomimetic mineralization
Typically, 4 mL aqueous solution containing HMeIM (990 mg, 12 mM) was mixed with 4 mL of Zn(NO 3 ) 2 •6(H 2 O) solution (52 mg, 0.175 mM), 400 µL of SOD (10 mg) and CAT (10 mg) solution. The mixture was incubated at room temperature for 5 min, and then the solids were collected and washed with deionized water. The blue solids were subsequently dried at room temperature to obtain the composite SOD@CAT@ZIF-8. The SOD and CAT loading in ZIF-8 was calculated as follows:

Protein loading (%) = [(A -B) / A] × 100%
A and B represented the amount of protein in the initial solution and washing solution, measured based on the BCA assay.
Synthesis of PSCZ MPEG 2000 -COOH was connected to the surface of SOD@CAT@ZIF-8 through a coordination bond between -COOH group and the Zn ion to prepare PMs. Brie y, pre-formed CAT@SOD@ZIF-8 was dispersed in an MPEG 2000 -COOH solution of 10 mg/mL using deionized water as the solvent. The mixture was sonicated for 10 min and stirred at room temperature for 48 h to form a PSCZ suspension. The white product was collected by centrifugation, followed by washing more than three times with deionized water and drying at 35 °C under vacuum.

Superoxide anion scavenging with PSCZ
The superoxide anion scavenging activity was assessed with a SOD assay kit (Solarbio, China). The concentrations (100 μg/mL) test was performed according to the instructions provided with the kit. The absorbance at 560 nm was measured using a quartz colorimetric dish reader after incubating at 37 °C for 30 min.

PSCZ scavenger hydrogen peroxide
The superoxide anion scavenging activity was assessed with a CAT assay kit (Solarbio, China). The PSCZ, PSZ, PCZ concentrations (100 μg/mL) test was performed according to the instructions provided with the kit. The absorbance at 405 nm was measured using a quartz colorimetric dish reader after incubating at room temperature for 10 min. PSCZ scavenging of reactive oxygen species in vitro 2',7'-Dichloro uorescin diacetate (DCFH-DA, D6883, Sigma-Aldrich, USA), an oxidation-sensitive uorescent dye, was used to detect the intracellular ROS level according to the literature. Brie y, DCFH-DA is a non-uorescent chemical compound that could freely diffuse through the cell membrane and be hydrolyzed by intracellular esterase to DCFH. The intracellular ROS could oxidize the non-uorescent DCFH to uorescent DCF. Therefore, the quantity of intracellular ROS is correlated with the uorescent intensity of DCF. After incubation with H 2 O 2 for 24 h, cells were gently rinsed thrice with a serum-free medium to remove the free PSCZ. Then, a nal concentration of 10 μM of DCFH-DA in serum-free medium was added to the cells and incubated in the dark at 37 °C for 30 min. Afterward, the cells were washed with serum-free medium thrice to remove the unloaded DCFH-DA probe, then were scanned using a laser confocal microscope (Zeiss LSM780, Germany). Flow cytometry analysis was used to quantify the intracellular ROS levels.
Furthermore, cells seeded in 6-well plates were stained with Annexin V-FITC apoptosis detection kit (A211-01, Vazyme, China) to detect the ratio of apoptotic and necrotic cells. Brie y, HEK293 cells in a well were collected, washed with cold PBS, and re-suspended in 195 μL binding buffer after incubation with H 2 O 2 for 24 h. Then, 5 μL Annexin V-FITC and 10 μL PI were sequentially added to the cell suspension and incubated at room temperature in the dark for 15 min. After that, cells were analyzed by the Flow cytometer (NovoCyte, Agilent, USA). At least 50,000 cells were analyzed in each sample.
Finally, we used the mitochondrial membrane potential detection kit JC-1 (C2006, Beyoyime, China) probe method to detect mitochondrial membrane potential changes. HEK293 cells were seeded separately into confocal laser Petri dishes. Follow the instructions provided with the kit. After incubation with H 2 O 2 for 24 h, cells were gently rinsed thrice with PBS to remove the free PSCZ. Then add 1 mL of JC-1 dye solution and mix well. Then the cells were incubated in an incubator at 37℃ for 20 min. An appropriate amount of JC-1 staining buffer (1×) was prepared by adding 1 mL of JC-1 staining buffer (5×) to 4 mL of distilled water and placed in an ice bath during incubation. At the end of incubation at 37℃, the supernatant was aspirated, washed twice with JC-1 staining buffer (1×), and then 2 mL cell culture solution was added, which could contain serum and phenol red. Finally, using a laser confocal microscope (Zeiss LSM780, Germany).

PSCZ scavenging of reactive oxygen species in vitro
Cisplatin-Induced AKI model Female BALB/c mice aged 6-8 weeks were selected, all mice were received intraperitoneal injections of cisplatin (20 mg/kg). 24 h after injections, the mice in each group were sacri ced to monitor the model development. The blood samples and renal tissues were obtained and analyzed. The mice were intravenously injected with PSCZ (15 mg/kg in 100 μL PBS) at 1 h after intraperitoneal injections of cisplatin for the treatment group.

Treatment of CP-AKI Mice
One hour after the AKI model induction, different treatments were performed on AKI model mice: group 1 was healthy mice (n = 5); group 2 was AKI mice (PBS, n = 5); group 3 was AKI mice treated with 1× PBS (n = 5); group 4 was AKI mice treated with PSCZ (15 mg/kg in 100 μL PBS, n = 5). After 24 h p.i, blood samples and tissues of each organ were collected, the treatment group was compared with healthy mice to evaluate kidney function.

Kidney function test
After euthanizing the mice, blood samples were collected in lithium heparin tubes. The blood serum was collected after centrifuging the blood samples at 2000 g for 15 min at 4℃. Finally, the samples were sent to the Laboratory Department of the First A liated Hospital of Zhengzhou University for analysis of blood urea nitrogen (BUN) levels and blood creatinine (CRE) levels.

Haemotoxylin and Eosin (H&E) Staining of Kidney Sections
In the AKI model and time-point of interest, mice were euthanized, and the kidneys were collected and xed with paraformaldehyde (4% in PBS). Laboratory Animal Center of Zhengzhou University embedded the samples in para n wax, sliced, and then stained with H&E stain.

Analysis of the therapeutic effect in renal tissue
Kidneys from each group were snap-frozen and stored at 80℃ until use. Kidney homogenates were prepared according to the protocols of different assays. SOD level and CAT level were assessed with a SOD assay kit and CAT assay kit (Sigma-Aldrich, USA). To evaluate ROS's scavenging effect histologically, each group's kidney tissues were removed and stored in a refrigerator at -80 ℃ until use. Kidney tissues of mice in each group were sliced using a frozen micro-slicer. Frozen kidney tissue slides (about 5μm thickness) were washed with PBS and stained with 1 mM DCFH-DA for 30 min to detect superoxide formation. Then a cover glass was applied to each slide using Vectashield mounting medium. Finally, a laser confocal microscope was used to observe the uorescence changes. The other part of the tissue was frozen and sliced in the same way, then stained with TNFα-and IL1β antibodies and DAPI. After staining, the tissue was observed under the confocal microscope.

RNA sequencing
Mice were sacri ced to harvest the kidneys. The kidneys were washed thrice with saline and snap-frozen in liquid nitrogen. Total RNA was extracted by using the TIANGEN Animal Tissue Total RNA Extraction Kit (DP419, TIANGEN, China). RNA integrity was veri ed by Agilent 4200 Bioanalyzer and quanti ed using ND-2000 (Nanodrop Technology). Cell rRNA was removed by using MGIEasy rRNA removal kit. Library construction was carried out using the MGIEasy RNA Library Prep Set (96 RXN) according to the manufacturer's protocol. Libraries were visualized on the Agilent 4200 Bioanalyzer to check insert size and quanti ed by using the Qubit Fluorometers to determine the concentration. Libraries were pooled and loaded on the ow cell to run on MGISEQ-2000 sequencer as paired-end read for 150 cycles on each side.

Bioinformatics analysis
Transcriptomics sequencing and mapping Clean reads were mapped to Mus musculus reference genome (GRCm38) with STAR aligner (version 2.6.1d). [34] Transcript-level gene expression was estimated by Salmon (version 0.14.1). [35] Transcript per million (TPM) values were calculated using tximport. A community-developed pipeline called bcbionextgen was used to automate the upstream bioinformatic analysis. [36] Identi cation and functional annotation of differentially expressed genes (DEGs) The R package bcbioRNASeq can take the bcbio-nextgen output as input and generate count matrices, identify DEGs, conduct functional annotations and visualize the results. [37] DEGs were identi ed for each group compared with the control group using DESeq2,[38] with a fold-change threshold of 1 and P values cutoff of 0.05. Functional annotation of DEGs was performed using the clusterPro ler R package to identify associated pathways in AKI. [39] We further explored the mechanism of AKI and whether PSCZ can reverse the unfavorable changes in transcriptional level.

Results And Discussion
Synthesis and characterization of PSCZ As shown in Scheme 1, the native enzyme could be embedded in a porous nanomaterial ZIF-8, in which Zn(OAc) 2 , 2-HMeIM, SOD, and CAT were incubated for 5 min at room temperature. The morphology of composite SOD@CAT@ZIF-8 and PSCZ was characterized by SEM ( Figure 1A). The SEM image showed that SOD@CAT@ZIF-8 and PSCZ both possessed uniform rhombic dodecahedral morphology similar to blank ZIF-8. As shown in Figure 1B, the PXRD pattern of PSZ, PCZ, and PSCZ both exhibited peaks matching well with the simulated ZIF-8 suggesting that ZIF-8 frameworks were formed in the biomimetic mineralization, and the structural integrity of ZIF-8 could be retained in the presence of SOD and CAT. FT-IR spectra of PSZ, PCZ, and PSCZ characteristic with MPEG 2000 -COOH peaks at 2865-2875 cm −1 that were respectively attributable to -COOH and -OCH 3 , indicating the successful conjunction of MPEG 2000 .
We further investigated the antioxidant activities of PSCZ, PSZ, and PCZ. Among biologically relevant ROS, H 2 O 2 is of greatest importance because of its membrane permeability, longer half-life than O 2− and ·OH, and consequently highest intracellular concentration. As shown in Figure 1E and 1F, we investigated the cascading enzyme scavenging ability of PSCZ, a remarkable difference in the activity was observed. PSCZ showed enhanced antioxidant activities in both SOD-and CAT-like activities.
Assessment of the ability of PSCZ to scavenge ROS in vitro Given the promising properties of PSCZ particles as stable, biocompatible antioxidant agents, we next assessed their ROS scavenging activity in vitro relative to that of free CAT and SOD. Given that AKI pathogenesis is associated with oxidative damage to the renal tubules, we utilized the HEK293 cell line model to assess the ability of PSCZ to protect against ROS-mediated injury [14]. We additionally utilized an xCELLigence (RTCA) instrument to monitor the survival and growth of these cells in a quantitative, real-time, dynamic manner. This assay approach is superior to traditional endpoint-based viability assays that assess membrane permeability [15], enabling the assessment of cell survival in a label-free and automated manner throughout a given experimental period [16]. RTCA approach was used to measure the survival and growth of cells over a 24 h period following treatment with H 2 O 2 and these different nanoenzymes preparations. PSCZ showed superior antioxidant capacity at 5 μg/mL than PSZ and PCZ even at a higher (20 μg/mL) concentration (Figure 2A). We then treated HEK293 cells with H 2 O 2 to induce oxidative stress to evaluate the ability of PSCZ to protect against oxidative damage ( Figure 2B). While the viability of H 2 O 2 -treated cells was ~50%, PSCZ treatment was associated with a signi cant increase in these cells' survival due to the observed reductions in intracellular ROS production and the consequent preservation of mitochondrial functionality ( Figure 2B).
We also utilized the DCFH-DA uorescent probe to assess intracellular ROS levels to further con rm whether PSCZ could suppress H 2 O 2 -induced ROS generation in HEK293 cells. While a signi cant increase in DCFH-DA uorescence was evident in H 2 O 2 -treated cells relative to untreated controls ( Figure 2D), this uorescence intensity was signi cantly reduced in cells that had additionally been treated with PSCZ. This nding was further con rmed via ow cytometry ( Figure 2C), revealing that PSCZ could reliably suppress H 2 O 2 -mediated free radical generation.
Mitochondrial oxidative phosphorylation is the primary mechanism whereby most cells produce ATP [17]. Given that renal PTECs lack glycolytic activity and are under high metabolic energy demands, these cells exist under conditions of ischemia and hypoxia [18]. This increases the susceptibility of these renal tubular epithelial cells to acute renal injury. Mitochondria are also central regulators of intracellular apoptotic signaling, with the loss of mitochondrial membrane potential (Δψm) being an early apoptotic indicator [19]. Therefore, we assessed the impact of PSCZ on HEK293 cell Δψm using a JC-1 staining approach, with a higher red/green JC-1 ratio indicative of lower levels of mitochondrial dysfunction. We observed a signi cant decrease in the Δψm of these cells following H 2 O 2 treatment ( Figure S1), whereas PSCZ treatment yielded a JC-1 uorescence ratio comparable to that in control cells, suggesting that PSCZ can preserve mitochondrial membrane integrity.
Excess ROS generation can initiate apoptotic signaling cascades within cells [20,21]. Given the acute sensitivity of renal tubule cells to oxidative damage, we next assessed the ability of PSCZ to facilitate in vitro ROS scavenging activity in our HEK293 cell model system. When the death of these cells was assessed via ow cytometry following Annexin V/PI staining, we found that H 2 O 2 treatment (200 μm) increased the frequency of apoptotic HEK293 cells to 36.43%, relative to control cells (3.28%) (Figure 3). In contrast, cells treated with H 2 O 2 and PSCZ exhibited signi cantly lower rates of apoptotic death (6.43%). These ndings thus con rmed that PSCZ could protect against oxidative damage in HEK293 cells in vitro.

PSCZ effectively facilitates in vivo ROS scavenging
Given the promising in vitro results detailed above, we next explored the in vivo utility of our MOF antioxidant platform using a murine cisplatin-induced AKI (CP-AKI) model system [22]. Cisplatin (CDDP) is a widely used chemotherapeutic drug that accumulates within renal proximal tubule epithelial cells (PTECs), resulting in localized renal in ammation, damage, apoptotic death, and AKI that can manifest in the form of CDDP-associated nephrotoxicity [23]. The pathogenesis of CP-AKI has been linked to ROS production and consequent acute tubular apoptosis [24]. Thus, this CP-AKI model system was selected for further analyses of the relative safety and in vivo antioxidant activity of PSCZ.
CP-AKI mice were established as in prior reports by intraperitoneally injecting these animals with CDDP (20 mg/kg) and then measuring serum BUN and CRE levels after 24 h to gauge renal function [25], given that these compounds accumulate in the blood in the context of renal damage. Mice in corresponding treatment groups were intravenously treated with PSCZ 1 h before CDDP administration ( Figure 4A). CRE and BUN levels were signi cantly higher in CP-AKI mice than healthy controls ( Figure 4B-C), consistent with the successful induction of CDDP-mediated kidney damage in these animals. Saline infusions are commonly administered to patients before and during CDDP treatment [25], as such infusions exhibit poorly understood renoprotective activity against CDDP-induced nephrotoxicity. However, CP-AKI will manifest in roughly 30% of treated patients [26]. Our mouse model system observed comparable ndings, as only slight reductions in CRE and BUN levels were observed in CP-AKI mice treated with PBS, with these levels being signi cantly higher than those in control animals. In contrast, the treatment of CP-AKI mice with PSCZ resulted in signi cant reductions in CRE and BUN levels to concentrations comparable to those in healthy control mice. These data underscored the therapeutic utility of PSCZ as a safe and effective tool for treating CP-AKI in mice.
We additionally isolated renal tissue sections from these treated mice and stained them with H&E stain to assess treatment-related changes in tissue pathology. Renal casts can arise due to denatured protein precipitation in kidney tubules and are often considered to be an indicator of kidney disease. Similarly, in ammatory cell in ltration and the formation of vacuoles within renal tubules can be visualized to detect and gauge the severity of the renal in ammatory injury. While many casts were evident in the renal tissue samples from CP-AKI model mice ( Figure 4D), relatively few were detectable in AKI model animals that had been administered PSCZ, consistent with the ability of these nanoparticles to preserve the integrity of renal tissues better.
To further assess the ability of PSCZ to suppress renal ROS generation, we next used the uorescent ROS probe DCFH-DA to stain kidney tissue sections from CP-AKI mice in our different treatment groups and then imaged these tissues via laser scanning confocal microscopy. This approach revealed that PSCZ administration signi cantly inhibited renal ROS accumulation in our AKI model mice ( Figure S2).
Tumor necrosis factor-alpha (TNF-α) and interleukin-1β (IL-1β) are key in ammatory cytokines and drivers of apoptotic cell death in the context of cisplatin-induced renal injury [27,28]. Excess ROS production by cells under in ammatory conditions can induce additional proin ammatory cytokine production, thereby promoting further immune cell in ltration and renal damage. IL-1β can further amplify these in ammatory processes through feedback mechanisms [24]. Renal tissue samples from treated mice were collected and analyzed to reveal the underlying mechanism of PSCZ. Immuno uorescence results showed that IL-1β and TNF-α levels were comparable in the PSCZ-treated and control groups, suggesting these nanoparticles did not induce direct in ammation in mice at the utilized concentrations and can reduce in ammation-induced kidney damage ( Figure 5A-5D). Notably, in AKI model mice, PSCZ administration suppressed TNF-α and IL-1β expression relative to untreated model controls. We additionally assessed SOD and CAT levels in renal homogenates from these mice ( Figure 5E and 5F), to con rm that these antioxidant enzymes were successfully delivered to damaged kidney tissue [29]. Excessive ROS levels can result in SOD and CAT depletion, as was evident in CP-AKI and PBS-treated mice. In contrast, PSCZ resulted in renal SOD and CAT levels similar to those in healthy control animals. It was con rmed that PSCZ was successfully delivered to the injured renal tissue and alleviated renal in ammatory injury.
To further test the therapeutic utility of PSCZ, we next measured malondialdehyde (MDA) levels and blood biochemical parameters to evaluate renal excretory function. MDA analysis results revealed that both CP-AKI mice and PBS-treated AKI mice exhibited elevated MDA levels consistent with renal failure, while PSCZ administration lowered these levels consistent with the alleviation of AKI-related renal damage ( Figure 5G). Analyses of blood biochemical parameters from these mice indicated that PSCZ treatment was not associated with obvious renal toxicity ( Figure 5H-5L), and hematological parameters in PSCZ-treated animals were comparable to those in healthy controls. These ndings suggested that our PSCZ particles were highly biocompatible and useful as a tool for AKI treatment. However, it is essential to note that we only tested for acute toxicity associated with this therapeutic platform, and further research will be necessary to establish whether PSCZ is retained in the kidney beyond the 24 h time point, whether it is eliminated via renal metabolic processing, and whether it induces any long-term toxicity in vivo.
RNAs-seq analysis con rmed the successful construction of the AKI model and PSCZ showed a superior protective effect in the AKI mouse model In the current study, we set four different groups: control group without any treatment (brief as the control group), cisplatin-induced AKI group (brief as CP-AKI group), AKI group pretreated with PSCZ (brief as the PSCZ+AKIgroup), and AKI group pretreated with PBS (brief as the PBS+CP-AKI group). More detailed information is shown in the M&M section.
An average of 63.34 (range from 58.51 to 66.33) million raw reads were obtained for all samples. After removing low-quality reads, an average of 60.52 (range from 56.05 to 63.16) million clean reads were retained for further analysis. The raw data were uploaded to National Omics Data Encyclopedia (NODE) database (https://www.biosino.org/node) with the following accession number OEP000285.
To further elucidate the underlining therapeutic mechanisms, CP-AKI was chosen as the representative disease model for further transcriptomic analysis. DESeq2 identi ed a total of 601, 157, and 88 DEGs for the CP-AKI group, CP-AKI+PSCZ group, and CP-AKI+PBS group compared with the control group, respectively ( Figure 6D). An unguided principal component analysis (PCA) of the data revealed different transcriptomic pro les between PSCZ and PBS-treated AKI mice kidneys ( Figure 6C). The Venn diagram in Figure 6B showed many differentially expressed genes were produced in the CP-AKI group compared with the control group, and the differentially expressed genes were decreased in the PSCZ treatment group.
We also investigated the impact of PSCZ on gene expression related to growth arrest and DNA damage response, apoptosis, oxidation-reduction, etc. ( Figure 6G). In brief, the PSCZ treatment could reverse the impact of cisplatin-induced AKI in the mouse model.
DEGs in the CP-AKI group were enriched in the p53 signaling pathway, FoxO signaling pathway, apoptosis-related pathways, etc ( Figure 6E and F, Figure S3). This is consistent with previous research.
Cisplatin can interfere with DNA replication and DNA repair mechanisms, cause DNA damage, and induce apoptosis in cells. [30] Cisplatin can also increase cellular ROS and induce endoplasmic reticulum (ER) stress, contributing to cisplatin toxicity. Cell death and ER stress are characteristic of cisplatin-induced AKI. [26,31] P53 signaling is an early-stage response to cisplatin toxicity in renal cells. Cisplatin caused renal cells to initiated the p53 dependent DNA repair pathway. .If ROS continues to be overproduced in renal cells, the intrinsic apoptotic pathway will be activated. In the early stage, when cisplatin interferes with DNA replication and DNA repair mechanism, nano inhibits the excessive production of ROS caused by cisplatin, promotes the initiation of DNA repair pathway, and alleviates renal cell damage( Figure 6A, Figure S4).
In the current study, the kidney proximal nephron tubule segment injury marker Havcr1 (also known as Kim1) slightly changed (log 2 FC = 0.42, adj. P = 0.437) CP-AKI group compared with the control group. This is probably because the kidneys were harvested in the very early stage of AKI. This is consistent with a previous study that cisplatin-induced AKI mainly caused proximal nephron tubule segment injury in the very early stage. [32] Bax, Bak1 are important signals in the apoptosis cascade and normally act on the mitochondrial membrane to promote permeabilization and release of cytochrome C and ROS. In CP-AKI group, Bax and Bak1 signi cantly increased (log 2 FC = 1.16, adj. P = 0.024; log 2 FC = 0.99, adj. P = 0.048, respectively). This means that ROS level increased by cisplatin. Aco1 and Cyp2e1, which are important oxidation-reduction enzymes, signi cantly decreased (log 2 FC = -0.74, adj. P = 0.007; log 2 FC = -1.03, adj. P = 0.026, respectively). This indicated that the normal function of oxidation-reduction was compromised. Gadd45a and Gadd45b, which are growth arrest and DNA damage response genes, also increased (log 2 FC = 1.13, adj. P = 0.045; log 2 FC = 1.29, adj. P = 0.024, respectively). This indicated that kidney cells responded to cisplatin-induced DNA damage by cell cycle arrest and initiated the DNA repair system, in which case, the p53 signaling pathway. Tnfrsf12a and Tnfrsf10b (also known as Dr5), which belong to tumor necrosis factor receptor superfamily and transduces apoptosis signals, signi cantly increased in CP-AKI group (log 2 FC = 1.50, adj. P =0.003; log 2 FC = 1.45, adj. P =0.006, respectively). Caspase8 also signi cantly increased in CP-AKI group (log 2 FC = 0.74, adj. P =0.044). This indicated that cisplatin-induced intrinsic apoptosis through Caspase8 by upregulating Tnfrsf10b in the CP-AKI group. The results indicated that the apoptosis process was initiated in the CP-AKI group even after 24 h being treated with cisplatin. Bcl10 contains a caspase recruitment domain (CARD) and has been shown to induce apoptosis and activate NF-κB. Bcl10 slightly increased in CP-AKI group (log 2 FC = 0.70, adj. P =0.159).

Ethics approval
The Ethics Committee of Experimental Animal Center of Zhengzhou University approved the animal experiment in this study, and the Ethics Approval Certi cate number is ZZU-LAC20200703.

Consent for publication
All authors have reviewed the manuscript and approved the submission of this manuscript.

Availability of data and materials
All data associated with this study are present in the paper or the Supplementary Materials. RNA sequencing data can be accessed in the NODE database with the following URL (http://www.biosino.org/node/project/detail/OEP000285).    Assessment of the biocompatibility of PSCZ in vivo. A) CP-AKI modeling approach. B) CRE levels in CP-AKI mice were measured at 24 h following the indicated treatments. C) BUN levels in CP-AKI mice were measured at 24 h following the indicated treatments. D) Tissues from the indicated treatment groups were assessed following H&E staining.