Pathological Microenvironment‐Remodeling Nanoparticles to Alleviate Liver Fibrosis: Reversing Hepatocytes‐Hepatic Stellate Cells Malignant Crosstalk

Abstract

During the onset and malignant development of liver fibrosis, the pernicious interplay between damaged hepatocytes and activated hepatic stellate cells (HSCs) induce a self‐perpetuating vicious cycle, deteriorating fibrosis progression and posing a grave threat to public health. The secretions released by damaged hepatocytes and activated HSCs interact through autocrine or paracrine mechanisms, involving multiple signaling pathways. This interaction creates a harsh microenvironment and weakens the therapeutic efficacy of single‐cell‐centric drugs. Herein, a malignant crosstalk‐blocking strategy is prompted to remodel vicious cellular interplay and reverse pathological microenvironment to put an end to liver fibrosis. Collagenases modified, bardoxolone and siTGF‐β co‐delivered nanoparticles (C‐NPs/BT) are designed to penetrate the deposited collagen barriers and further regulate the cellular interactions through upregulating anti‐oxidative stress capacity and eliminating the pro‐fibrogenic effects of TGF‐β. The C‐NPs/BT shows successful remodeling of vicious cellular crosstalk and significant disease regression in animal models. This study presents an innovative strategy to modulate cellular interactions for enhanced anti‐fibrotic therapy and suggests a promising approach for treating other chronic liver diseases.

Introduction

Liver fibrosis is an uncontrollable wound-healing response to repeated chronic liver injury.^[1] It is characterized by persistent hepatocyte necrosis, excessive proliferation and activation of hepatic stellate cells (HSCs), and continuous extracellular matrix (ECM) deposition.^[2,3] Moreover, liver fibrosis has a strong relationship with liver cancer and is an unavoidable stage in the progression of various chronic injuries to end-stage fatal liver diseases such as cirrhosis and liver cancer, evolving as a severe public health problem worldwide.^[4] Fortunately, clinical studies have shown that unlike cirrhosis, liver fibrosis is dynamic and reversible, presenting significant therapeutic potential.^[5] If the process of liver fibrosis can be controlled reasonably and effectively, it can not only maintain liver function and control disease progression but also reduce the prevalence and mortality of chronic liver disease to a certain extent.^[6] Therefore, effective treatment of liver fibrosis is significant for the prevention and intervention of various chronic or fatal liver diseases.

Till now, HSCs-targeting therapies have been a primary focus since HSCs are central to liver fibrosis.^[7] Due to off-target toxicity and delivery barriers, scientists have developed specific delivery systems to inhibit HSCs activation and induce HSCs senescence using targeting moieties such as hyaluronic acid, vitamin A, and chondroitin sulfate.^[8–10] Moreover, modifying collagenases to pass through ECM barriers and modulating liver sinusoids fenestrae to promote deep drug delivery have been shown to increase drug accumulation in fibrotic nidus and enhance the therapeutic effect.^[11,12] However, limited by the complex pathological microenvironment that is constructed by various signaling pathways and multiple injured cells, the previous studies achieved moderate therapeutic efficiency. Therefore, a deeper understanding of the pathological mechanisms and the development of more effective therapies are urgently needed to treat or even reverse liver fibrosis.^[13,14]

In our previous studies, we have proved that synergetic regulation of cellular crosstalk and interruption of vicious interplay could improve pathological microenvironment and enhance therapeutic effect, further demonstrating the applicability of HSCs-beyond therapy.^[15,16] Notably, hepatocytes, the most abundant cells in the liver, also play a crucial role in the initiation and development of liver fibrogenesis.^[17] In response to various injuries (such as excessive alcohol, toxic substances, metabolic disorders, and viral infections), the sharply increased oxidative stress injures hepatocytes, leading to necrosis or apoptosis by increasing mitochondrial permeability. Damaged hepatocytes release a large number of signaling molecules (hedgehog, nucleotides, and free fatty acids) and produce large amounts of reactive oxygen species (ROS), which stimulate the activation of quiescent HSCs.^[18–20] The uncontrollably generated ROS also upregulates the expression of key genes related to collagen formation, such as type I collagen, monocyte chemoattractant protein 1, and tissue inhibitor of metalloproteinase-1, leading to abnormal collagen deposition.^[21] Moreover, the excessive collagen secreted by activated HSCs would increase mechanical tension, affect the morphology and function of hepatocytes, and further aggravate hepatocytes necrosis.^[22] It is important to note that the malignant relationship between damaged hepatocytes and activated HSCs would create a pathological microenvironment, exacerbate cellular injury, and cause repeated stimulation, finally transforming into an uncontrollable and fatal disease.

In addition, transforming growth factor-β (TGF-β) is widely regarded as the most effective pro-fibrotic cytokine.^[23] TGF-β would induce the phosphorylation of drosophila mothers against decapentaplegic protein (SMAD) by binding to type I receptors, while activation of SMAD3 promotes the transcription of type I and type III collagen. Concurrently, TGF-β would activate mitogen-activating protein kinase signaling pathways, promoting HSCs activation. TGF-β also inhibits hepatocyte proliferation and hinders liver regeneration.^[24] Therefore, inhibiting TGF-β can reduce collagen fiber synthesis, inhibit HSCs activation, and promote hepatocyte regeneration. This suggests a potential synergetic therapeutic effect when combined with modulation of oxidative stress to reverse the malignant crosstalk between damaged hepatocytes and malignant HSCs, and to remodel the liver microenvironment.

Based on our previous studies and the interplay between these two types of pathological cells, we hypothesized that blocking the malignant interaction and improving the pathological microenvironment by synergistic modulating oxidative stress and the TGF-β signaling pathway could reverse the vicious cycle, normalize cellular interactions, and lead to complete disease resolution. Herein, a malignant crosstalk-blocking strategy was prompted by using chemogene therapy to remodel vicious cellular interplay and reverse pathological microenvironment to terminate liver fibrosis. Briefly, collagenases modified, bardoxolone, and siTGF-β co-encapsulated nanoparticles, nominated as C-NPs/BT, were constructed. The grafted collagenases are expected to help penetrate ECM barriers and promote cellular uptake. Bardoxolone, a potent antioxidant, is intended to repair damaged hepatocytes, relieve oxidative stress, and reduce HSC activation.^[25] The loaded siTGF-β is expected to prevent further HSC activation and promote hepatocyte proliferation using RNA interference strategies. By applying this integrative and straightforward strategy, the malignant interplay between damaged hepatocytes and activated HSCs would be interrupted, fostering a healthy liver microenvironment. This creates a positive feedback loop that amplifies the therapeutic effect (Scheme 1). This work pioneers the use of pathological microenvironment remodeling therapy to break the vicious cycle of liver fibrosis, offering a novel perspective for liver fibrosis treatment.

[IMAGE OMITTED. SEE PDF]

Results and Discussion

Malignant Crosstalk Between Hepatocytes and HSCs in Human Fibrotic Liver

Hepatocytes and HSCs are the two main components of hepatic cells. While hepatocytes are the most numerous, HSCs serve as the central hub during liver fibrogenesis. To obtain a deeper understanding of the pathological microenvironment in fibrotic liver, primary data from GSE126848 was re-analyzed. The results showed increased levels of oxidative stress, TGFβ1, and fibroblast activation in NAFLD patients (Figure 1A; Figure S1, Supporting Information).^[26] For further evaluation of the importance of hepatocytes and HSCs, we then analyzed primary data from NCBI database (GSE212837).^[27] The extracted cells were divided into five groups: hepatocyte, HSC, cholangiocytes, immune cells, and endothelial cells (Figure 1B). Oxidative stress and TGF-β-related signaling pathways were significantly upregulated in fibrotic liver (Figure 1C,D). Cell chat analysis revealed a sharp increase in cellular interactions between hepatocytes and HSCs during fibrogenesis (Figure 1E). Moreover, KEGG analysis showed gene enrichment related to these pathways, highlighting the significance of modulating oxidative stress and TGF-β signaling in damaged hepatocytes and activated HSCs (Figure 1F). Furthermore, analysis of human tissue from patients with liver cirrhosis or liver cancer revealed significant findings. H&E staining showed damaged hepatocytes and disrupted tissue structure, while Masson and Van Gieson staining exhibited excessive ECM deposition gradually replacing liver parenchyma. α-SMA staining revealed high activation of HSCs in fibrotic liver (Figure 1G).

[IMAGE OMITTED. SEE PDF]

Preparation, Characterization and Biosafety of C-NPs/BT

To reverse the malignant crosstalk between damaged hepatocytes and activated HSCs by synergistically modulating related cell injury, C-NPs/BT were prepared using T-tubes for chemogene therapy. The optimal molar ratio of ALC-0315: Cholesterol: DSPC: DSPE-PEG-Mal was determined to be 45: 40: 10: 5 (Table S1, Supporting Information). TEM images revealed that C-NPs/BT had a spherical and uniform appearance (Figure 2A). The average particle size was 143.74 ± 3.20 nm, and the zeta potential was -9.80 ± 2.10 mV. C-NPs/BT exhibited good stability in 10% FBS or 5% glucose, with no significant changes in particle size over 72 h (Figure 2B). The drug loading and encapsulation efficiency were measured to be 3.38 ± 0.08% and 74.00 ± 4.91%, respectively. The agarose gel retardation study showed that the cationic lipids ALC-0315 and siRNA were completely complexed at a weight ratio of 10:1 (Figure 2C).

[IMAGE OMITTED. SEE PDF]

To verify the gene-protecting function of the nanoparticles, a lysosomal escape assay was performed on HSC-T6 cells. The results showed that the cationic lipids facilitated lysosomal escape through the nephew sponge effect (Figure 2D). Collagenases were modified via click chemistry reactions between the maleimide group and the thiol group. Visual inspection using Prussian blue revealed that C-NPs could ablate deposited collagen and achieve deeper penetration (Figure 2E).

Due to the potential toxicity limiting the application of nano-delivery system, the biosafety of C-NPs/BT was tested on cells and healthy mice. As shown in Figure S2 (Supporting Information), C-NPs/BT showed no obvious cell toxicity on three dominant hepatic cells, hepatocytes (L02), HSCs (HSC-T6), and macrophages (RAW 264.7) over 24 and 48 h, with cell viability above 80%. The blood compatibility of the nanoparticles was also tested. Similar to PBS, C-NPs/BT,C-NPs/B, C-NPs/T, and B+T (bardoxolone and siTGF-β) showed good blood compatibility (Figure S3, Supporting Information).

Next, the acute toxicity of nanoparticles was evaluated. Different preparations were injected intravenously into healthy mice for three continuous days and the blood sample was analyzed. There was no significant difference in ALT and AST levels between the C-NPs/BT group and the PBS group, demonstrating biosafety in liver (Figure S4, Supporting Information). Moreover, H&E staining revealed no significant inflammation in major tissues, such as liver, spleen, kidney, lung, and heart after repeated administration, further confirming the biosafety of C-NPs (Figure S5, Supporting Information).

Anti-fibrotic Effects of C-NPs/BT In Vivo

To explore the therapeutic effect of C-NPs/BT in vivo, a carbon tetrachloride (CCl₄)-induced liver fibrosis model was established. Mice were divided into six groups and were injected with CCl₄ intraperitoneally twice a week for six weeks. The groups included normal mice and fibrotic mice treated with C-NPs/BT,C-NPs/B, C-NPs/T, B+T, and PBS (Control group) to compare the efficacy (Figure S6, Supporting Information). Sirius Red and Masson staining showed reduced ECM deposition in the C-NPs/BT treated group, which was significantly different from the Control group. (Figure 2F,G) The semi-quantification further confirmed the substantial ECM ablation and fibrosis resolution in the C-NPs/BT treated group. (Figure 2H,I). In contrast, heavy ECM deposition was observed in the C-NPs/B, C-NPs/T, and B + T groups, demonstrating the limited effect of single-target therapy for hepatocytes or HSCs. Serum ALT and AST activity indicated that chemogene therapy with C-NPs/BT could reduce hepatocyte injury and protect liver function, mainly by remodeling the malignant interplay between damaged hepatocytes and activated HSCs (Figures 2J,K).

As a main component of ECM, hydroxyproline content indicates the severity of fibrosis. The hydroxyproline content in the C-NPs/BT treated group significantly decreased and was comparable to that of normal group, unlike the B+T treated group (Figure 2L). Moreover, the H&E stained sections of other major organs (spleen, kidneys, lung, and heart) showed no obvious difference from those of normal mice, suggesting the safety of this chemogene therapy (Figure S7, Supporting Information). These results comprehensively support the potent therapeutic effect of this combined therapy, suggesting its potential for the clinical application in remodeling the harmful crosstalk between damaged hepatocytes and activated HSCs.

Hepatocytes Injury Relieving Effects and Mechanism Verification

Hepatocytes injury and increased membrane permeability are inevitable problems in liver fibrosis caused by various etiologies. For further exploration of the effects of C-NPs/BT on damaged hepatocytes, the cellular uptake was first investigated. To simulate the deposited ECM in the Disse space, collagen I (150 µg mL⁻¹) was pre-layered on L02 cells for 5 h. The fluorescence in the free drug (coumarin 6, Cor6) treated group was almost invisible, suggesting minimal uptake by hepatocytes (Figure 3A). In contrast, the NPs/Cor6 treated group showed slightly higher fluorescence, while those treated with C-NPs/Cor6 was much brighter green fluorescence. This increase is mainly due to the modification with collagenases, which could ablate deposited collagen and enhance uptake. Flow cytometry quantitation showed similar results (Figure 3B). The DCFH-DA probe demonstrated that, compared to other nanoparticles, the final preparation effectively downregulated the level of oxidative stress (Figure 3C). Moreover, to assess the ability of C-NPs/BT to promote hepatocytes proliferation, Edu staining was applied (Figure 3D). The bright green fluorescence in C-NPs/BT treated group indicated that the preparation could mitigate the cell damage caused by H₂O₂ and maintain cell activity.

[IMAGE OMITTED. SEE PDF]

For lipid nanoparticles, the distribution is mainly dependent on the compositions and the proteins absorbed. C-NPs were composed of commonly-used materials, namely DSPC, ALC-0315, DSPE-PEG-Mal, and cholesterol. In our previous study and other references, the biodistribution of such kind of lipid nanoparticles is mainly distribute to the liver and the modification of collagenase would enhance liver accumulation by penetrating the collagen barrier.^[8,11,28] Therefore, following the 3R principle of animal experiments, we omitted the in vivo biodistribution of nanoparticles in this project. To test whether C-NPs/BT could fully function in vivo, the biodistribution in liver was then examined. Albumin was used as the marker of hepatocytes. As shown in Figure 3E, the red fluorescence of C-NPs/DiI was significantly brighter than that of NPs/DiI and free DiI. This observation could be attributed to the fact that collagenase could ablate excessive collagen, thereby promoting deeper accumulation. Next, the liver sections of CCl₄-induced fibrotic mice were stained with ki67 (Figure 3F). The red fluorescence in the treated groups showed enhanced hepatocytes proliferation compared to the control group. Moreover, the combination of bardoxolone and siTGF-β (C-NPs/BT) exhibited more red fluorescence than C-NPs/B and C-NPs/T treated group. This result suggests that eliminating oxidative stress and inhibiting TGF-β could synergistically promote cell viability and growth.

Additionally, the molecular mechanisms were verified by measuring related protein levels. Western blot analysis showed that the expression of phosphorylated nuclear factor E2-related factor 2 (p-Nrf2), Smad7, and heme oxygenase 1 (HO-1) in the C-NPs/BT treated group was significantly upregulated in hepatocytes and fibrotic mice (Figure 3G; Figure S8, Supporting Information). This is primarily because bardoxolone could stimulate Nrf2, an important transcription factor that regulates antioxidant responses in hepatocytes. Under physiological conditions, Nrf2 forms a complex with Keap1 and is degraded by ubiquitination. In response to oxidative stress, Nrf2 separates from Keap1, accumulates in the cytoplasm, and translocates to the nucleus to initiate transcription of antioxidant protective genes. Upregulated Nrf2 would further upregulate HO-1 transcription to eliminate excessive ROS. Bardoxolone also increases the level of Smad7 and reduced the expression of TGF-β, achieving the anti-fibrotic effect. It was observed that the expression of Smad7 in the control group was higher than that in the normal group, which might be explained as compensatory increase. Compared to normal mice, those treated with PBS exhibited swollen hepatocytes and destroyed tissue structures (Figure 3H). However, the C-NPs/BT treated group showed recovered hepatocytes and less inflammatory cell infiltration than C-NPs/B, C-NPs/T, B+T treated groups, preliminarily proving the great effect of this chemogene therapy to remodel the vicious crosstalk between damaged hepatocytes and activated HSCs.

HSCs Activation Inhibiting Effects and Mechanism Verification

During the pathological process, activated HSCs secrete higher levels of tissue inhibitors of metalloproteinases-1 and collagen I, which lead to excessive extracellular matrix and hinder the drug delivery. The activated HSCs caused an imbalance between collagen generation and degradation, resulting in aggravated liver fibrosis. Therefore, the therapeutic effect of C-NPs/BT in inhibiting HSC activation was further explored. Cor6 was also used as a model drug to investigate the cellular uptake of C-NPs/BT in the presence of collagen barrier. It was demonstrated that the collagenases helped improve the uptake efficiency by about two folds, greatly higher than that of free Cor6 (Figure 4A,B). Next, TGF-β (10 ng/mL) was used to stimulate HSC activation in vitro. As shown in Figure 4C and Figure S9 (Supporting Information), the red fluorescence of α-SMA and collagen I represents the activation degree of HSCs. After stimulation, the control group exhibited heavily deposited collagen, while the other groups could counteract its stimulating effect to a certain extent and C-NPs/BT showed the optimal effect. C-NPs/BT work in two ways: i) the loaded bardoxolone eliminate ROS through Nrf2-HO-1 pathways to inhibit HSCs phenotype transformation; ii) the loaded siTGF-β exert gene-silencing effect to inhibit HSCs activation.

[IMAGE OMITTED. SEE PDF]

From our preliminary study, we found that endowing nanocarriers with a proteolytic surface could ablate the collagen barrier, promote their permeation in fibrotic extracellular matrix, and finally help delivery of carriers to activated HSCs in vivo. To investigate the cellular localization of the nanoparticles, tissue immunofluorescent staining was performed. The hydrophobic fluorochrome, DiI, was intravenously administered into mice that had been treated with CCl₄ for 4 weeks (once a day, three days), either alone or complexed with different nanoparticles. The green fluorescence of α-SMA was highly merged with the red fluorescence of DiI in C-NPs/DiI group, demonstrating that C-NPs/DiI were taken up by HSCs (Figure 4D).

Moreover, α-SMA-stained liver sections showed that, compared to C-NPs/B, C-NPs/T, B+T, and the control group, C-NPs/BT could effectively inhibit activation of HSCs and liver fibrogenesis in animal models, similar to the normal group (Figure 4E). The enhanced therapeutic effect might be due to the ability of C-NPs/BT to relieve hepatocyte injury and promote cell proliferation, interrupting the stimulation from hepatocytes to HSCs.

Interruption of Vicious Interplay Between Damaged Hepatocytes and Activated HSCs

The interactions between damaged hepatocytes and activated HSCs are hypothesized to aggravate cellular damage and repeatedly stimulate HSC activation, accelerating disease progression. To investigate whether C-NPs/BT could interrupt the vicious interplay and reverse pathological environment, a cell coculture model was established. Briefly, L02 cells were pre-treated with H₂O₂ (1 mm) for 4 h and then were seeded onto the upper plate, while quiescent HSCs were seeded onto the lower plate (Figure 4F). Upon cell adhesion, the medium was replaced with fresh medium, and the coculture began. Confocal laser scanning microscopy showed that with the treatment of TGF-β and the abnormal-L02, HSCs were activated and secreted excessive collagen. However, Treated-L02 treated HSCs showed almost no collagen deposition and was even lower than that of the control group, possibly because HSCs cultured with 10% FBS were partially activated.

The detrimental effect of HSCs on hepatocytes is thought to result from the collagen deposited by activated HSCs, which can cause hepatocyte injury through mechanical pressure. As liver fibrosis progresses, the nature and quantity of liver collagen gradually change. Normally ordered and loose collagen fibers become thicker, more disorganized, and increasingly interconnected. Consequently, the mechanical force in the Disse space would increase significantly. It is proposed that after the incubation of C-NPs/BT, the modified collagenases could ablate deposited collagen and induce HSCs to adopt a collagen-degrading phenotype, thus alleviating hepatocytes injury. Moreover, the re-quiescent HSCs would reduce oxidative stress level, further mitigating the harmful effects on hepatocytes. We cocultured damaged hepatocytes and activated HSCs in vitro to test the efficacy of C-NPs/BT. As shown in Figure S10 (Supporting Information), C-NPs/BT treatment could obviously decrease the green fluorescence of H₂O₂ and the red fluorescence of α-SMA, compared with PBS-treated group. Therefore, the strategy to recover activated HSCs and damaged hepatocytes is expected to achieve greater therapeutic efficacy by remodeling cellular crosstalk and reversing the pathological microenvironment. This suggests the great clinical potential of C-NPs/BT for treating liver fibrosis.

Conclusion

With a deeper understanding of the mechanisms of liver fibrosis, the need to go beyond the previous single-cell-centric therapies has become more urgent. In summary, this study proposed a malignant crosstalk-remodeling strategy to interrupt the malignant interplay between damaged hepatocytes and activated HSCs by synergistically regulating oxidative stress and TGF-β signaling pathway. This approach achieved promising results in a CCl₄-induced fibrotic mice model. After effectively interrupting the vicious cycle, it was hypothesized that a virtuous loop would gradually form, primarily because the recovered hepatocytes and re-quiescent HSCs would benefit each other. Building on our previous study, this study offers a new approach to treat liver fibrosis that goes beyond targeting HSCs alone. Moreover, the preparation of C-NPs/BT is simple and reproducible, enhancing the clinical feasibility and controllability.

It is worth noting that other major organs may also affect the liver. During fibrogenesis, it has been proved that at least the lung, the spleen, and the adipose tissue would affect the process of liver fibrosis. Specifically, disruption of the gut barrier and the translocation of harmful gut microbes to the liver would promote macrophages polarization, and thus aggravate fibrosis progression.^[29] Moreover, spleen derived-monocytes would migrate to the liver and function as macrophages, triggering liver inflammation and finally exacerbate fibrosis.^[30] Besides, the lipid metabolism was disturbed in fibrotic liver and adipose tissue, and the accumulated excessive lipid would lead to vacuolization and ballooning degeneration of hepatocytes. The adipose tissue would release inflammatory factors and excessive fatty acid to the liver and therefore worsen the disease.^[31] Therefore, it may be a useful mean to develop multiple organ-targeted therapy to treat liver fibrosis.

Experimental Section

Ethics Statement

Experimentation on animals was conducted in accordance with the guidelines approved by the regional ethics committee of China Pharmaceutical University (2023-03-002). Human studies were performed in compliance with the ethical standards of the 1964 Declaration of Helsinki and its later amendments, and had been approved by the local Ethics Committee of Zhongda Hospital affiliated with Southeast University (Ethics number: 2018ZDSYLL070-P01). All participants gave their informed consent prior to inclusion in the study.

Statistical Analysis

All data were expressed as mean ± standard deviation (SD) and analyzed by Prism 10. Statistical significance was calculated using one-way ANOVA test with the Tukey's multiple comparison analysis or two-tailed t-test and defined as p < 0.05 (ns: p > 0.05, ^*p < 0.05, ^**p < 0.01, ^***p < 0.001, and ^****p < 0.0001).

Acknowledgements

L.-F.Z., W.-Q.D., and X.-H.W. contributed equally to this work. This work was financially supported by the National Natural Science Foundation of China (82020108029, 82473867, 82073398, 82302367, and 82304416) and the National Key R&D Program of China (2022YFE0198400). This work was also supported by the Natural Science Foundation of Jiangsu Basic Research Program-Project jointly funded by the province and city (No. BK20232035) and the Natural Science Foundation of Jiangsu Province (BK20231019 and BK20231016), the Priority Academic Program Development of Jiangsu Higher Education Institutions and the “111” Project from the Ministry of Education of China, the State Administration of Foreign Experts Affairs of China (B16046), the China Postdoctoral Science Foundation (2022M720173), Jiangsu Funding Program for Excellent Postdoctoral Talent (2023, China), the Fundamental Research Funds for the Central Universities (2632023GR20). This research was funded by grants from the National Research Foundation (NRF) of Korea, Ministry of Science and ICT, Republic of Korea (NRF-2018R1A5A2024425). Authors would like to thank Xiao-nan Ma for providing technical assistance with confocal microscopy work on the Public Laboratory Platform of China Pharmaceutical University.

Conflict of Interest

The authors declare no conflict of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

V. Hernandez‐Gea, S. L. Friedman, Annu. Rev. Pathol. Mech. Dis. 2011, 6, 425.

S. L. Friedman, B. A. Neuschwander‐Tetri, M. Rinella, A. J. Sanyal, Nat. Med. 2018, 24, 908.

C. Trautwein, S. L. Friedman, D. Schuppan, M. Pinzani, J. Hepatol. 2015, 62, S15.

H. Devarbhavi, S. K. Asrani, J. P. Arab, Y. A. Nartey, E. Pose, P. S. Kamath, J. Hepatol. 2023, 79, 516.

R. Vuppalanchi, M. Noureddin, N. Alkhouri, A. J. Sanyal, Nat. Rev. Gastroenterol. Hepatol. 2021, 18, 373.

S. L. Friedman, M. Pinzani, Hepatol. 2022, 75, 473.

J. Lee, J. Byun, G. Shim, Y.‐K. Oh, Nat. Commun. 2022, 13, 1516.

J.‐B. Qiao, Q.‐Q. Fan, C.‐L. Zhang, J. Lee, J. Byun, L. Xing, X.‐D. Gao, Y.‐K. Oh, H.‐L. Jiang, J. Control. Release 2020, 321, 629.

J.‐B. Qiao, Q.‐Q. Fan, L. Xing, P.‐F. Cui, Y.‐J. He, J.‐C. Zhu, L. Wang, T. Pang, Y.‐K. Oh, C. Zhang, H.‐L. Jiang, J. Controlled Release 2018, 283, 113.

J. Luo, P. Zhang, T. Zhao, M. Jia, P. Yin, W. Li, Z.‐R. Zhang, Y. Fu, T. Gong, ACS Nano 2019, 13, 3910.

Q.‐Q. Fan, C.‐L. Zhang, J.‐B. Qiao, P.‐F. Cui, L. Xing, Y.‐K. Oh, H.‐L. Jiang, Biomaterials 2020, 230, [eLocator: 119616].

L.‐F. Zhang, X.‐H. Wang, C.‐L. Zhang, J. Lee, B.‐W. Duan, L. Xing, L. Li, Y.‐K. Oh, H.‐L. Jiang, ACS Nano 2022, 16, [eLocator: 14029].

R. Loomba, S. L. Friedman, G. I. Shulman, Cell 2021, 184, 2537.

N. C. Henderson, F. Rieder, T. A. Wynn, Nature 2020, 587, 555.

L.‐F. Zhang, W.‐Q. Deng, Q.‐W. Huang, J.‐J. Zhang, Y. Wang, T.‐J. Zhou, L. Xing, H.‐L. Jiang, Adv. Mater. 2024, 36, [eLocator: 2311474].

B.‐W. Duan, Y.‐J. Liu, X.‐N. Li, M.‐M. Han, H.‐Y. Yu, H.‐Y. Hong, L.‐F. Zhang, L. Xing, H.‐L. Jiang, Adv. Sci. 2023, 11, [eLocator: 2306899].

S. H. Ibrahim, P. Hirsova, G. J. Gores, Gut 2018, 67, 963.

Y. Xiao, K. Batmanov, W. Hu, K. Zhu, A. Y. Tom, D. Guan, C. Jiang, L. Cheng, S. J. McCright, E. C. Yang, M. R. Lanza, Y. Liu, D. A. Hill, M. A. Lazar, Sci. Transl. Med. 2023, 15, [eLocator: eadc9653].

L. Rui, J. D. Lin, Annu. Rev. Nutr. 2022, 42, 91.

K. Jin, Y. Shi, H. Zhang, G. Zhangyuan, F. Wang, S. Li, C. Chen, J. Zhang, H. Wang, W. Zhang, B. Sun, J. Hepatol. 2023, 79, 403.

R. Bataller, D. A. Brenner, J. Clin. Invest. 2005, 115, 209.

Y. Long, Y. Niu, K. Liang, Y. Du, Trends Cell Biol. 2022, 32, 70.

T. Tsuchida, S. L. Friedman, Nat. Rev. Gastroenterol. Hepatol. 2017, 14, 397.

L. W. Schrum, M. A. Bird, O. Salcher, E.‐R. Burchardt, J. W. Grisham, D. A. Brenner, R. A. Rippe, K. E. Behrns, Am. J. Physiol. Gastrointest. Liver Physiol. 2001, 280, G139.

P. E. Pergola, C. J. Meyer, M. Krauth, H. Christ‐Schmidt, N. Engl. J. Med. 2011, 365, 327.

M. P. Suppli, K. T. G. Rigbolt, S. S. Veidal, S. Heebøll, P. L. Eriksen, M. Demant, J. I. Bagger, J. C. Nielsen, D. Oró, S. W. Thrane, A. Lund, C. Strandberg, M. J. Kønig, T. Vilsbøll, N. Vrang, K. L. Thomsen, H. Grønbæk, J. Jelsing, H. H. Hansen, F. K. Knop, Am. J. Physiol. Gastrointest. Liver Physiol. 2019, 316, G462.

S. Wang, K. Li, E. Pickholz, R. Dobie, K. P. Matchett, N. C. Henderson, C. Carrico, I. Driver, M. Borch Jensen, L. Chen, M. Petitjean, D. Bhattacharya, M. I. Fiel, X. Liu, T. Kisseleva, U. Alon, M. Adler, R. Medzhitov, S. L. Friedman, Sci. Transl. Med. 2023, 15, [eLocator: eadd3949].

X. Shan, Z. Zhao, P. Lai, Y. Liu, B. Li, Y. Ke, H. Jiang, Y. Zhou, W. Li, Q. Wang, P. Qin, Y. Xue, Z. Zhang, C. Wei, B. Ma, W. Liu, C. Luo, X. Lu, J. Lin, L. Shu, Y. Jie, X. Xian, D. Delcassian, Y. Ge, L. Miao, Nat. Commun. 2024, 15, 7263.

G. Kim, Z. Chen, J. Li, J. Luo, F. Castro‐Martinez, J. Wisniewski, K. Cui, Y. Wang, J. Sun, X. Ren, S. E. Crawford, S. P. Becerra, J. Zhu, T. Liu, S. Wang, K. Zhao, C. Wu, Cell 2024, 187, 914.

G. Tarantino, V. Citro, C. Balsano, Expert Rev. Gastroenterol. Hepatol. 2021, 15, 759.

L. Scheja, J. Heeren, J. Hepatol. 2016, 64, 1176.