Hepatocyte-derived tissue extracellular vesicles safeguard liver regeneration and support regenerative therapy

Abstract

Tissue-derived extracellular vesicles (EVs) are emerging as pivotal players to maintain organ homeostasis, which show promise as a next-generation candidate for medical use with extensive source. However, the detailed function and therapeutic potential of tissue EVs remain insufficiently studied. Here, through bulk and single-cell RNA sequencing analyses combined with ultrastructural tissue examinations, we first reveal that in situ liver tissue EVs (LT-EVs) contribute to the intricate liver regenerative process after partial hepatectomy (PHx), and that hepatocytes are the primary source of tissue EVs in the regenerating liver. Nanoscale and proteomic profiling further identify that the hepatocyte-specific tissue EVs (Hep-EVs) are strengthened to release with carrying proliferative messages after PHx. Moreover, targeted inhibition of Hep-EV release via AAV-shRab27a in vivo confirms that Hep-EVs are required to orchestrate liver regeneration. Mechanistically, Hep-EVs from the regenerating liver reciprocally stimulate hepatocyte proliferation by promoting cell cycle progression through Cyclin-dependent kinase 1 (Cdk1) activity. Notably, supplementing with Hep-EVs from the regenerating liver demonstrates translational potential and ameliorates insufficient liver regeneration. This study provides a functional and mechanistic framework showing that the release of regenerative Hep-EVs governs rapid liver regeneration, thereby enriching our understanding of physiological and endogenous tissue EVs in organ regeneration and therapy.

Keywords: Cell cycle; Extracellular vesicles; Hepatocytes; Liver regeneration; Partial hepatectomy.

Fig. 1

LT-EVs are involved in the liver regenerative process after PHx. (A) Venn diagram of transcriptome of PHx and Sham livers. (B) Volcano plot of transcriptome of PHx and Sham livers. (C) GO enrichment analysis of DEGs in PHx over Sham livers. (D) GO terms of DEGs related to EVs enriched in PHx over Sham livers. (E) KEGG enrichment analysis of DEGs in PHx over Sham livers. (F) GSEA analysis of DEGs between PHx and Sham livers for the terms “Cell cycle”, “Extracellular exosome”, and “Extracellular vesicle”. (G) TEM analysis of Sham and PHx liver tissues. Hepatocytes were identified with featured morphologies. Red arrows indicating LT-EVs in the extracellular interstitial space. Bars: 1 μm (low magnification) and 200 nm (high magnification)

Fig. 2

Hepatocytes are the major origin of LT-EVs during liver regeneration. (A) tSNE plots displaying the distribution of different cell types in Sham and PHx livers. Source data were derived from the GSM4572241 and GSM4572243 series in the GEO database. (B) Cell type proportion in Sham and PHx livers. (C) Volcano plot showing DEGs between PHx and Sham hepatocytes. (D) GO enrichment analysis of DEGs in PHx over Sham hepatocytes. (E) Flow cytometric analysis of Sham and PHx LT-EVs with surface exposure of ASGPR, F4/80, CD11b, CD31, GFAP, CK19, CD3, and CD19. Corresponding isotype control groups were used to distinguish positively stained EVs. Mean ± SD. n = 3 per group. **, p < 0.01; ***, p < 0.001; ****, p < 0.0001; two-tailed Student’s unpaired t tests. (F) Pie charts illustrating the quantification of LT-EV constitution originated from different cell types using mean values of (E) normalized to percentages

Fig. 3

Hep-EVs reveals enhanced release with proliferative information during liver regeneration. (A) Flow cytometric analysis showing ASGPR surface expression on LT-EVs before and after immunomagnetic sorting. (B) Size distribution of LT-EVs analyzed by NTA. (C) Quantification of particle number of LT-EVs by NTA and protein amount of LT-EVs determined by the BCA assay. Mean ± SD. n = 3 per group. *, p < 0.05; ***, p < 0.001; one-way ANOVA with Turkey’s post-hoc tests. (D) Representative TEM image of LT-EVs from Sham and PHx livers. Bars = 250 nm. (E) Size distribution of sorted Hep-EVs analyzed by NTA. (F) Quantification of particle number of sorted Hep-EVs by NTA and protein amount determined by the BCA assay. Mean ± SD. n = 3 per group. ***, p < 0.001; ****, p < 0.0001; one-way ANOVA with Turkey’s post-hoc tests. (G) Venn diagram of proteome of PHx and Sham Hep-EVs. (H) Volcano plot of proteome of PHx and Sham Hep-EVs. (I) GO terms in the Cellular Component category of DEPs enriched in PHx over Sham Hep-EVs. (J) Western blot analysis of protein marker expression of LT-EVs and Hep-EVs. (K) Top-ranked DEPs of interest in Hep-EVs and Western blot validation of expression

Fig. 4

Hep-EV release is indispensable for orchestrating liver regeneration. (A) Schematic diagram of the study design using AAV8-shRNA to inhibit hepatocyte Rab27a via i.v. injection at 4 weeks prior to PHx. (B) Quantification of particle number of LT-EVs and Hep-EVs determined by NTA. Hep-EVs were immunomagetically sorted from LT-EVs. Mean ± SD. n = 3 per group. **, p < 0.01; ****, p < 0.0001; one-way ANOVA with Turkey’s post-hoc tests. (C) Kaplan-Meier survival analysis of mice. n = 8 per group. Log-rank test was used. (D) The gross view images and H&E staining of liver tissues. Bars = 500 mm (gross view) and 100 mm (H&E). (E) Quantification of liver weight over body weight ratio. Mean ± SD. n = 3 per group. *, p < 0.05; one-way ANOVA with Turkey’s post-hoc tests. (F) IF staining of hepatocyte proliferation, macrophage inflammation, LSEC and HSC activation in the liver. Bars = 50 μm. (G) Quantification of percentages of proliferative hepatocytes, total and inflammatory macrophages, total and activated LSECs and HSCs in the liver. Mean ± SD. n = 3 per group. *, p < 0.05; one-way ANOVA with Turkey’s post-hoc tests

Fig. 5

Hep-EVs from the regenerating liver reciprocally promote proliferation of hepatocytes via Cdk1 activity. (A) Nine-quadrant diagram showing integrated analysis of hepatocyte transcriptome from scRNA-seq and Hep-EV proteome. (B) GO enrichment analysis of the P3 quadrant showing consistent upregulation of hepatocyte genes and Hep-EV proteins. (C) EdU assay and IF staining of PHH3 with Ki67 for proliferative hepatocytes treated with Hep-EVs and the Cdk1 inhibitor RO-3306. Bars = 100 μm. (D) Quantification of percentages of EdU⁺, Ki67⁺, PHH3⁺ hepatocytes and PHH3⁺Ki67⁺ hepatocytes among Ki67⁺ hepatocytes. n = 3 per group. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001; one-way ANOVA with Turkey’s post-hoc tests. (E) Representative images of wound closure assay of hepatocytes treated with Hep-EVs and the Cdk1 inhibitor RO-3306. Bars = 250 μm. (F) Quantification of wound closure rates of hepatocytes. n = 3 per group. *, p < 0.05; **, p < 0.01; one-way ANOVA with Turkey’s post-hoc tests. (G) Schematic diagram showing the mechanistic findings of this study

Fig. 6

Replenishment of Hep-EVs from the regenerating liver rescues insufficient liver regeneration. (A) Schematic diagram of the study design using Hep-EV infusion as a replenishment into AAV8-shRab27a-treated mice before liver regeneration analysis. (B) Biodistribution of DiR-labeled Hep-EVs at 24 h after infusion. (C) Kaplan-Meier survival analysis of mice. n = 8 per group. Log-rank test was used. (D) Gross view liver images, H&E staining of liver tissues, and IF staining of hepatocyte proliferation. Bars = 500 mm (gross view), 100 mm (H&E), and 50 μm (IF). (E) Quantification of liver weight over body weight ratio and percentages of proliferative hepatocytes. Mean ± SD. n = 3 per group. *, p < 0.05; **, p < 0.01; one-way ANOVA with Turkey’s post-hoc tests