F4/80+ Kupffer Cell-Derived Oncostatin M Sustains the Progression Phase of Liver Regeneration through Inhibition of TGF-β2 Pathway

The role of Kupffer cells (KCs) in liver regeneration is complicated and controversial. To investigate the distinct role of F4/80+ KCs at the different stages of the regeneration process, two-thirds partial hepatectomy (PHx) was performed in mice to induce physiological liver regeneration. In pre- or post-PHx, the clearance of KCs by intraperitoneal injection of the anti-F4/80 antibody (α-F4/80) was performed to study the distinct role of F4/80+ KCs during the regenerative process. In RNA sequencing of isolated F4/80+ KCs, the initiation phase was compared with the progression phase. Immunohistochemistry and immunofluorescence staining of Ki67, HNF-4α, CD-31, and F4/80 and Western blot of the TGF-β2 pathway were performed. Depletion of F4/80+ KCs in pre-PHx delayed the peak of hepatocyte proliferation from 48 h to 120 h, whereas depletion in post-PHx unexpectedly led to persistent inhibition of hepatocyte proliferation, indicating the distinct role of F4/80+ KCs in the initiation and progression phases of liver regeneration. F4/80+ KC depletion in post-PHx could significantly increase TGF-β2 serum levels, while TGF-βRI partially rescued the impaired proliferation of hepatocytes. Additionally, F4/80+ KC depletion in post-PHx significantly lowered the expression of oncostatin M (OSM), a key downstream mediator of interleukin-6, which is required for hepatocyte proliferation during liver regeneration. In vivo, recombinant OSM (r-OSM) treatment alleviated the inhibitory effect of α-F4/80 on the regenerative progression. Collectively, F4/80+ KCs release OSM to inhibit TGF-β2 activation, sustaining hepatocyte proliferation by releasing a proliferative brake.


Introduction
Liver regeneration is the most significant reaction of the liver to an injury, which is super-orchestrated by intracellular crosstalk to recover the tissue lost and to maintain homeostasis and all the hepatic functions. Despite extensive research over the last two decades, the mechanism still remains unclear. Two-thirds partial hepatectomy and the chemical injury model are the two main models used to study liver regeneration. The partial hepatectomy (PHx) model, first reported in 1932, is a traditional and reliable model used to study regeneration with a very high reproducibility [1]. The process of liver regeneration can be divided into three phases: initiation, progression, and termination [2]. It remains unclear how these phases are highly interlinked and bounded. During the initiation of regeneration, the quiescent hepatocytes are primed by cytokines to enter the cell cycle within minutes to hours after PHx. Once initiated, the hepatocytes and other   As shown in Figure 2A, α-F4/80 post-PHx-treatment led to lower liver-body ratios than IgG-control-treatment did at 72 h, 96 h, and 120 h, especially at 120 h after PHx. At PHx day 5, the liver mass after α-F4/80 pre-PHx treatment recovered to almost 90% while that of post-PHx-treatment recovered only less than 60%. Given slightly higher serum aspartate transaminase (AST) and alanine transaminase (ALT) in the post-PHxtreated group than that in pre-PHx-treated one, F4/80 + KCs might affect liver resection recovery ( Figure 2B). Unexpectedly, the liver mass recovered to the baseline at day 8 in both treatments. In the control group, the peak of cells proliferation (Ki67 + cells) appeared 48 h after PHx ( Figure 2C-D). Additionally, depletion of F4/80 + KCs in pre-PHx significantly delayed the peak of cell proliferation from 48 to 120 h after PHx. Unexpectedly, depletion of them in post-PHx persistently lessened cell proliferation especially 2-5 days after PHx ( Figure 2E-F). Together, it indicated that F4/80 + KCs regulated the initiation and the progression of liver regeneration but not termination.
As shown in Figure 2A, α-F4/80 post-PHx-treatment led to lower liver-body ratios than IgG-control-treatment did at 72 h, 96 h, and 120 h, especially at 120 h after PHx. At PHx day 5, the liver mass after α-F4/80 pre-PHx treatment recovered to almost 90% while that of post-PHx-treatment recovered only less than 60%. Given slightly higher serum aspartate transaminase (AST) and alanine transaminase (ALT) in the post-PHx-treated group than that in pre-PHx-treated one, F4/80 + KCs might affect liver resection recovery ( Figure 2B). Unexpectedly, the liver mass recovered to the baseline at day 8 in both treatments. In the control group, the peak of cells proliferation (Ki67 + cells) appeared 48 h after PHx ( Figure 2C-D). Additionally, depletion of F4/80 + KCs in pre-PHx significantly delayed the peak of cell proliferation from 48 to 120 h after PHx. Unexpectedly, depletion of them in post-PHx persistently lessened cell proliferation especially 2-5 days after PHx ( Figure 2E-F). Together, it indicated that F4/80 + KCs regulated the initiation and the progression of liver regeneration but not termination.

F4/80+ KCs Maintain Hepatocyte Proliferation during the Progression Phase of Liver Regeneration
The liver mass recovery and physiological function after PHx mainly depends on the rapid proliferation of hepatocytes and liver endothelial cells (LECs) [16][17][18][19][20]. Confocal images showed that the depletion of F4/80 + KCs in pre-PHx significantly enhanced the hepatocyte proliferation (Ki67 + /HNF-4α + ) from 72 to 120 h after PHx. Unexpectedly, the depletion of them in post-PHx persistently lessened hepatocyte proliferation, especially 72-120 h after PHx. Contrastively, the depletion of F4/80 + KCs in pre-or post-PHx treatment had little effect on the number of Ki67 + /CD31 + LECs ( Figure 3A-C). The protein levels of HNF-4α, PCNA, and cyclin D1 were upregulated in the pre-PHx group ( Figure 3D) but downregulated in the post-PHx one ( Figure 3E). In contrast, no significant alteration in CD31 and VEGFR2 occurred in the pre-or post-PHx groups. These data indicated F4/80 + KC maintenance of hepatocyte proliferation in the progression stage of liver regeneration.

F4/80+ KCs Maintain Hepatocyte Proliferation during the Progression Phase of Liver Regeneration
The liver mass recovery and physiological function after PHx mainly depends on the rapid proliferation of hepatocytes and liver endothelial cells (LECs) [16][17][18][19][20]. Confocal images showed that the depletion of F4/80 + KCs in pre-PHx significantly enhanced the hepatocyte proliferation (Ki67 + /HNF-4α + ) from 72 to 120 h after PHx. Unexpectedly, the depletion of them in post-PHx persistently lessened hepatocyte proliferation, especially 72-120 h after PHx. Contrastively, the depletion of F4/80 + KCs in pre-or post-PHx treatment had little effect on the number of Ki67 + /CD31 + LECs ( Figure 3A-C). The protein levels of HNF-4α, PCNA, and cyclin D1 were upregulated in the pre-PHx group ( Figure 3D) but downregulated in the post-PHx one ( Figure 3E). In contrast, no significant alteration in CD31 and VEGFR2 occurred in the pre-or post-PHx groups. These data indicated F4/80 + KC maintenance of hepatocyte proliferation in the progression stage of liver regeneration.

Different Transcriptional Alterations Occur in F4/80 + KCs between the Initiation and Progression Stages of Liver Regeneration
Transcriptional analysis of isolated KCs from hepatectomized mice at the initiation or progression phases of liver regeneration. Notably, it was quite different between the initiation (4 and 6 h after PHx) and progression phases (72 and 120 h after PHx). As is shown in Figure 4A,B, there were a large number of differentially expressed genes (DEGs) between the two phases. Gene Ontology (GO) enrichment analysis mainly focused on the cytokine biosynthetic process ( Figure 4C) and chemokine activity ( Figure 4D). Apparently, the constitutive differences further indicated the different functions of KCs in the initiation and progression phases of liver regeneration.

Different Transcriptional Alterations Occur in F4/80 + KCs between the Initiation and Progression Stages of Liver Regeneration
Transcriptional analysis of isolated KCs from hepatectomized mice at the initiation or progression phases of liver regeneration. Notably, it was quite different between the initiation (4 and 6 h after PHx) and progression phases (72 and 120 h after PHx). As is shown in Figure 4A-B, there were a large number of differentially expressed genes (DEGs) between the two phases. Gene Ontology (GO) enrichment analysis mainly focused on the cytokine biosynthetic process ( Figure 4C) and chemokine activity ( Figure 4D). Apparently, the constitutive differences further indicated the different functions of KCs in the initiation and progression phases of liver regeneration.

F4/80 + KCs Maintain the Progression of Liver Regeneration through Inhibition of the TGF-β2/smad2 Pathway
As shown in Figure 5A, the depletion of F4/80 + KCs in pre-PHx significantly decreases the expression of IL-6 and TNF-α, which are reported to play important roles in the initial of liver regeneration [14]. Thus, these data further confirmed that F4/80 + KCs

F4/80 + KCs Maintain the Progression of Liver Regeneration through Inhibition of the TGF-β2/smad2 Pathway
As shown in Figure 5A, the depletion of F4/80 + KCs in pre-PHx significantly decreases the expression of IL-6 and TNF-α, which are reported to play important roles in the initial of liver regeneration [14]. Thus, these data further confirmed that F4/80 + KCs are involved in the initial stage of liver regeneration. To further reveal the role of KCs in the progression phase of liver regeneration, quantitative real-time PCR (qRT-PCR) was analyzed for the proliferation and anti-proliferation factors [21] between IgG treatment and post-PHx F4/80 + KC clearance groups. Despite the fluctuation in proliferation factors between the two groups ( Figure 5B), extraordinarily, the anti-proliferation factor TGF-β2 in the post-PHx F4/80 + KC clearance group was consistently expressed higher than that in IgG treatment group at 72 h, 96 h, and 120 h after PHx ( Figure 5C). Coincidently, KEGG enrichment analysis of the isolated F4/80 + KCs between the two phases also implied that the TGF-β signaling pathway was involved in the progression phase of liver regeneration ( Figure 6A). ELISA assays confirmed higher levels of TGF-β2 in the post-PHx group than that in the IgG treatment group ( Figure 6B). In addition, phosphorylation of Smad2 or Smad3 also increased in post-PHx F4/80 + KC clearance ( Figure 6C). Thus, TGF-β2/Smad2 pathway is possibly involved in F4/80 + KC-regulated progression of liver regeneration.
Molecules 2021, 26, x FOR PEER REVIEW 9 are involved in the initial stage of liver regeneration. To further reveal the role of KC the progression phase of liver regeneration, quantitative real-time PCR (qRT-PCR) analyzed for the proliferation and anti-proliferation factors [21] between IgG treatm and post-PHx F4/80 + KC clearance groups. Despite the fluctuation in proliferation fac between the two groups ( Figure 5B), extraordinarily, the anti-proliferation factor TG in the post-PHx F4/80 + KC clearance group was consistently expressed higher than th IgG treatment group at 72 h, 96 h, and 120 h after PHx ( Figure 5C). Coincidently, KE enrichment analysis of the isolated F4/80 + KCs between the two phases also implied the TGF-β signaling pathway was involved in the progression phase of liver regenera ( Figure 6A). ELISA assays confirmed higher levels of TGF-β2 in the post-PHx group t that in the IgG treatment group ( Figure 6B). In addition, phosphorylation of Smad Smad3 also increased in post-PHx F4/80 + KC clearance ( Figure 6C). Thus, TGF-β2/Sm pathway is possibly involved in F4/80 + KC-regulated progression of liver regeneratio   To further explore this possibility, the TGF-βR inhibitor (TGF-βRI) was injected into the post-PHx α-F4/80-treated mice. As shown in Figure 7A, TGF-βRI treatment obviously inhibited the phosphorylation of smad2 and smad3 induced by post-PHx α-F4/80 injection. Moreover, TGF-βRI treatment rescued the diminishment in the liver recovery mass ( Figure 7B); the number of Ki67 + hepatocytes ( Figure 7C); and the protein levels of HNF-4α, PCNA, and Cyclin D1 ( Figure 7D) in post-PHx α-F4/80-treated mice. Taken together, these data indicated that F4/80 + KCs sustained the progression phase of liver regeneration, likely through inhibition of the TGF-β2/smad2 pathway. To further explore this possibility, the TGF-βR inhibitor (TGF-βRI) was injected into the post-PHx α-F4/80-treated mice. As shown in Figure 7A, TGF-βRI treatment obviously inhibited the phosphorylation of smad2 and smad3 induced by post-PHx α-F4/80 injection. Moreover, TGF-βRI treatment rescued the diminishment in the liver recovery mass ( Figure 7B); the number of Ki67 + hepatocytes ( Figure 7C); and the protein levels of HNF-4α, PCNA, and Cyclin D1 ( Figure 7D) in post-PHx α-F4/80-treated mice. Taken together, these data indicated that F4/80 + KCs sustained the progression phase of liver regeneration, likely through inhibition of the TGF-β2/smad2 pathway. In addition, we preliminarily explored the cell source of TGF-β2 in the progression phase of liver regeneration. Firstly, both parenchymal and non-parenchymal cells were isolated after PHx and detected by their biomarkers, respectively ( Figure 8A-C). TGF-β2 mRNA extraordinarily fluctuated in HSCs and peaked at 24 h after PHx ( Figure 8D). Combined with the TGF-β2 increase after depletion of KCs in post-PHx ( Figure 6B), F4/80 + KCs maintain hepatocyte proliferation, possibly by affecting HSC-derived TGF-β2 during the progression phase of liver regeneration. In addition, we preliminarily explored the cell source of TGF-β2 in the progression phase of liver regeneration. Firstly, both parenchymal and non-parenchymal cells were isolated after PHx and detected by their biomarkers, respectively ( Figure 8A-C). TGF-β2 mRNA extraordinarily fluctuated in HSCs and peaked at 24 h after PHx ( Figure 8D). Combined with the TGF-β2 increase after depletion of KCs in post-PHx ( Figure 6B), F4/80 + KCs maintain hepatocyte proliferation, possibly by affecting HSC-derived TGF-β2 during the progression phase of liver regeneration.

F4/80 + KC-Derived OSM Inhibits the TGF-β2/smad2 Pathway to Drive the Progression Phase of Liver Regeneration
Analysis of the differentially expressed genes (DEGs) in isolated KCs from hepatectomized mice and the pairwise comparisons were performed between the initial (PHx 4 h and 6 h) and progression phases (PHx 72 h and 120 h). Some significantly upregulated and downregulated genes ( Figure 9A) were further analyzed by the KEGG pathway and were possibly enriched in functions related to the TNF, PI3K/Akt, cytokine-cytokine receptor interaction, Jak-STAT, chemokine, and TGF-beta signaling pathways ( Figure 6A). Among of them, OSM function involves these pathways [22][23][24][25][26]. Moreover, OSM acts as a regeneration-promoting cytokine to initiate liver regeneration [27][28][29]. We therefore examined the possible involvement of OSM in F4/80 + KCs regulating the progression phase of liver regeneration. The OSM mRNA levels at 72 or 120 h were significantly higher than that at 4 or 6 h ( Figure 9B). Moreover, α-F4/80 treatment in post-PHx led to a decrease in serum OSM ( Figure 9C). However, impaired liver regeneration in the post-PHx-α-F4/80 group could be rescued by the administration of r-OSM ( Figure 9D); increased the number of Ki67 + hepatocytes ( Figure 10A); and augmented the expressions of HNF-4α, PCNA, and Cyclin D1 during the progression of regeneration ( Figure 10B). Moreover, α-F4/80mediated TGF-β2 activation was significantly attenuated by r-OSM rescues in vivo ( Figure 10C). These results suggested that F4/80 + KC-derived OSM sustained the progression phase of liver regeneration through inhibition of the TGF-β2/smad2 pathway.

F4/80 + KC-Derived OSM Inhibits the TGF-β2/smad2 Pathway to Drive the Progression Phase of Liver Regeneration
Analysis of the differentially expressed genes (DEGs) in isolated KCs from hepatectomized mice and the pairwise comparisons were performed between the initial (PHx 4 h and 6 h) and progression phases (PHx 72 h and 120 h). Some significantly upregulated and downregulated genes ( Figure 9A) were further analyzed by the KEGG pathway and were possibly enriched in functions related to the TNF, PI3K/Akt, cytokine-cytokine receptor interaction, Jak-STAT, chemokine, and TGF-beta signaling pathways ( Figure 6A). Among of them, OSM function involves these pathways [22][23][24][25][26]. Moreover, OSM acts as a regeneration-promoting cytokine to initiate liver regeneration [27][28][29]. We therefore examined the possible involvement of OSM in F4/80 + KCs regulating the progression phase of liver regeneration. The OSM mRNA levels at 72 or 120 h were significantly higher than that at 4 or 6 h ( Figure 9B). Moreover, α-F4/80 treatment in post-PHx led to a decrease in serum OSM ( Figure 9C). However, impaired liver regeneration in the post-PHx-α-F4/80 group could be rescued by the administration of r-OSM ( Figure 9D); increased the number of Ki67 + hepatocytes ( Figure 10A); and augmented the expressions of HNF-4α, PCNA, and Cyclin D1 during the progression of regeneration ( Figure 10B). Moreover, α-F4/80-mediated TGF-β2 activation was significantly attenuated by r-OSM rescues in vivo ( Figure 10C). These results suggested that F4/80 + KC-derived OSM sustained the progression phase of liver regeneration through inhibition of the TGF-β2/smad2 pathway.

Discussion
KCs serve as the liver's immune sentry, sense diverse stimulants, and warn other hepatic cells through cytokines and cell communication [30,31]. KC activation is beneficial to liver regeneration and provides the initial priming force for hepatocyte proliferation. Hepatocytes also produce granulocyte macrophage colony stimulating factor (GM-CSF) and activate KCs during liver regeneration [14,32]. KCs promote liver regeneration after partial liver transplantation through the inhibition of apoptosis and activation of the IL-6/p-STAT3 pathway [15]. KCs initiate liver regeneration via IL-6 and TNF-α after PHx

Discussion
KCs serve as the liver's immune sentry, sense diverse stimulants, and warn other hepatic cells through cytokines and cell communication [30,31]. KC activation is beneficial to liver regeneration and provides the initial priming force for hepatocyte proliferation. Hepatocytes also produce granulocyte macrophage colony stimulating factor (GM-CSF) and activate KCs during liver regeneration [14,32]. KCs promote liver regeneration after partial liver transplantation through the inhibition of apoptosis and activation of the IL-6/p-STAT3 pathway [15]. KCs initiate liver regeneration via IL-6 and TNF-α after PHx [33], and thus, the elimination of KCs delays liver regeneration and diminishes the survival of hepatectomized mice [34,35]. However, KCs have been also reported to negatively regulate liver regeneration through IL-1β, which is a negative regulator of hepatocyte proliferation [8]. Thus, the role of KCs in liver regeneration is complex and related to different experimental conditions. Intraperitoneal injection of α-F4/80 can remove almost 70% of F4/80 + KCs in mice [36]. Although the α-F4/80-eliminated efficiency of KCs is not higher than GdCl3 or liposome, it is operationally simple and fast and is able to remove F4/80 + KCs in different stages of liver regeneration. In our current study, intraperitoneal injection of α-F4/80 before PHx (pre-PHx) removed F4/80 + KCs justly in the initial phase of liver regeneration because its number recovered to the level of IgG treatment 48 h later after PHx ( Figure 1C). In contrast, injection at 8 h after PHx (post-PHx) indeed eliminated them during the progression phase of regeneration ( Figure 1G). Interestingly, α-F4/80 intervention at the different phases of regeneration yielded unexpected results. The latter intervention increased TGF-β2 from 72 h to 120 h after PHx, which possibly inhibited hepatocyte proliferation during the progression of liver regeneration.
The administration of high doses of TGF-β significantly delayed DNA synthesis in rats after PHx, involved in the termination of regeneration [37]. Nevertheless, Oe et al. [38] reported that the disruption of TGF-β signaling enhanced the proliferative response of hepatocytes after PHx but did not affect the termination of liver regeneration. In our study, the administration of Ly2109761 in vivo, a selective dual inhibitor of TGF-βR I/II, indicated the involvement of the TGF-β2/Smad2 signal pathway in the progression phage of liver regeneration. TGF-β is predominately expressed in HSCs [4]. Given that the serum TGF-β2 increased in the post-PHx α-F4/80-eliminated group, we speculate that KCs communicate with HSCs to regulate the progression phase of liver regeneration.
OSM, a member of the interleukin-6 family, shares the gp130 receptor subunit as a signal transducer and acts as a key downstream mediator of IL-6 in liver regeneration. It has been demonstrated to specifically promote differentiation of fetal hepatocytes and to control liver regeneration. OSM is mainly produced by immune cells, such as activated T lymphocytes, monocytes, and macrophages. Additionally, OSM elicits diverse biological functions in various cell types, among which the ability to regulate cell proliferation and apoptosis is most notable [38][39][40][41]. OSM decreased after α-F4/80 administration in post-PHx, while r-OSM replenishment promoted hepatocyte proliferation via inhibition of the TGF-β2/smad2 pathway, removing an endogenous growth inhibitory action. It partially explained TGF-β2 upregulation but no impact on the progression of liver regeneration.
The genome sequencing results were very different between the initiation and the progression phases during regeneration. In addition to OSM, it remains to be further studied whether other genes are involved in liver regeneration. Although KC depletion has no obvious effect on TGF-β1 expression, the role of TGF-β1 in the regeneration process could not be completely excluded because of Ly2109761 as a dual inhibitor of TGF-βRI and TGF-β-RII.
Collectively, F4/80 + KCs are involved not only in initiating liver regeneration but also in maintaining progression ( Figure 11). KC-derived OSM inhibits TGF-β2 to sustain hepatocyte proliferation during the progression phase of liver regeneration. les 2021, 26, x FOR PEER REVIEW 16 of 21 Figure 11. Schematic representation of the distinct roles of KCs during liver regeneration. In the process of liver regeneration, F4/80+ KCs participate in the initial stage of liver regeneration by regulating the expression of IL-6 and TNF-α. Besides, F4/80+ KCs sustain the progression phase of liver regeneration by releasing OSM to block TGFβ2 activation.

Animal Treatments and 2/3 PHx
All experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health; 6-8-week old male mice were subjected to standard 70% PHx under chloral hydrate anesthesia [1].

Isolation of Primary Hepatocytes and Nonparenchymal Cells (NPCs)
Primary hepatocytes and NPCs were isolated from mice. Briefly, mice were anaesthetized by 5 ml/Kg-5% chloral hydrate and their portal veins were cannulated with detaining needle; then, the inferior vena cava was cut and the liver was perfused at 5 mL/min through the inferior vena cava with D-Hanks (B430, basalmedia, Shanghai, China) with 0.5 mM EDTA and 0.5% BSA at 37 °C for 5 min; D-Hanks (B410, basalmedia) containing 0.5% collagen-IV (C5138, Sigma, St Louis, MO, USA) was applied for digestion of the liver; and after perfusion for an appropriate time, the liver was dissociated in a suspension buffer and filtered with a 100 μm pore cell strainer and collected by centrifugation at 50 g for 5 min. Hepatocytes were pelleted and enriched using a percoll cushion (45%) and pelleted. The supernatant containing NPCs were pelleted by centrifugation at 300 g for 10 min.

Library Construction for RNA-Seq and Sequencing Procedures
For RNA-seq, F4/80 + KCs were isolated from NPCs by staining with APC-conjugated anti-F4/80 Ab (Cat#123116, BioLgend, San Diego, CA, USA) by flow cytometry at different time points (4, 6, 72, and 120 h after PHx). Flow cytometry was carried out with a MoFlo™ Figure 11. Schematic representation of the distinct roles of KCs during liver regeneration. In the process of liver regeneration, F4/80+ KCs participate in the initial stage of liver regeneration by regulating the expression of IL-6 and TNF-α. Besides, F4/80+ KCs sustain the progression phase of liver regeneration by releasing OSM to block TGFβ2 activation.

Animal Treatments and 2/3 PHx
All experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health; 6-8-week old male mice were subjected to standard 70% PHx under chloral hydrate anesthesia [1].

Isolation of Primary Hepatocytes and Nonparenchymal Cells (NPCs)
Primary hepatocytes and NPCs were isolated from mice. Briefly, mice were anaesthetized by 5 mL/Kg-5% chloral hydrate and their portal veins were cannulated with detaining needle; then, the inferior vena cava was cut and the liver was perfused at 5 mL/min through the inferior vena cava with D-Hanks (B430, basalmedia, Shanghai, China) with 0.5 mM EDTA and 0.5% BSA at 37 • C for 5 min; D-Hanks (B410, basalmedia) containing 0.5% collagen-IV (C5138, Sigma, St Louis, MO, USA) was applied for digestion of the liver; and after perfusion for an appropriate time, the liver was dissociated in a suspension buffer and filtered with a 100 µm pore cell strainer and collected by centrifugation at 50× g for 5 min. Hepatocytes were pelleted and enriched using a percoll cushion (45%) and pelleted. The supernatant containing NPCs were pelleted by centrifugation at 300× g for 10 min.

Library Construction for RNA-Seq and Sequencing Procedures
For RNA-seq, F4/80 + KCs were isolated from NPCs by staining with APC-conjugated anti-F4/80 Ab (Cat#123116, BioLgend, San Diego, CA, USA) by flow cytometry at different time points (4,6,72, and 120 h after PHx). Flow cytometry was carried out with a MoFlo™ XDP Cell Sorter (BECKMAN Coulter, Miami, FL, USA). Data analysis was conducted using Summit 5.2 software (Beckman, Miami, FL, USA). The total RNA of KCs was isolated using RNeasy mini kit (Qiagen, Valencia, CA, USA). Strand-specific libraries were prepared using the TruSeq Stranded Total RNA Sample Preparation kit (Illumina, San Diego, CA, USA) following the manufacturer's instructions. Briefly, mRNA was enriched with oligo(dT) beads. Following purification, the mRNA was fragmented into small pieces using divalent cations under 94 • C for 8 min. The cleaved RNA fragments were copied into first-strand cDNA using reverse transcriptase and random primers. This was followed by secondstrand cDNA synthesis using DNA polymerase I and RNase H. These cDNA fragments then went through an end repair process, the addition of a single "A" base, and then ligation of the adapters. The products were then purified and enriched with PCR to create the final cDNA library. The purified libraries were quantified by Qubit 2.0 Fluorometer (Life Technologies, Carlsbad, CA, USA) and validated by Agilent 2100 bioanalyzer (Agilent Technologies, Palo Alto, CA, USA) to confirm the insert size and to calculate the mole concentration. A cluster was generated by cBot with the library diluted to 10 pM and then was sequenced on the Illumina HiSeq X ten (Illumina, San Diego, CA, USA). The library construction and sequencing were performed at Shanghai Biotechnology Corporation. The Differentially Expressed Genes (DEGs) in KCs of different stages of liver regeneration and the related functional clusters were compared and analyzed.

Immunofluorescence and Immunohistochemical (IHC) Staining
The liver samples were fixed in 10% buffered formalin overnight, dehydrated, and embedded in paraffin; 5 µm paraffin sections of the liver tissue samples were used for immunofluorescence and IHC staining.
The immunohistochemistry procedures were as follows. Paraffin sections were incubated with the following antibodies: Ki67 (1:100, Abcam, Cambridge, MA, USA), and anti-rabbit or anti-mouse peroxidase-conjugated secondary antibodies (1:50, Santa Cruz Biotechnology, Santa Cruz, CA, USA) were applied, and the DAB Substrate (Dako, Hamburg, Germany) was used for coloration. Counterstaining was performed with hematoxylin (Sigma, St Louis, MO, USA). Photographs of the three representative fields were captured randomly by the Image-Pro Plus v6.0 software (Media Cybernetics Inc, Bethesda, MD, USA), and analysis of the digital images were accomplished using Image-Pro Plus v6.0 software.

Digital Image Analysis of Ki67 with Virtual Dual Staining
HALO platform version 3.1.1076 (Indica Labs, Corrales, NM, USA) was used for digital image analysis of Ki67. Furthermore, hepatocytes were distinguished by setting the nuclear contrast threshold, minimum nuclear OD, nuclear size, and minimum nuclear roundness.

Real-Time PCR
The trizol method was used to isolate total RNA from liver tissues or cell lysates according to the manufacturer's instructions (Gibco, Carlsbad, CA, USA). qRT-PCR was performed using a SYBR Green Premix Ex Taq (Takara Bio inc., Otsu, Shiga, Japan) on Light Cycler ® 96 (Roche, Basel, Switzerland). qRT-PCR data were normalized to actin. Furthermore, qRT-PCR analysis of the biomarkers of hepatocytes (c-Met and OSMR), HSCs (GFAP and Desmin), or KCs (F4/80 and CD11b) isolated from mice was conducted. The primer sequences are shown in Table 1.

Measurement of the Serum ALT, AST, TGF-β2, and OSM
Blood samples were withdrawn from the inferior vena cava of mice under anesthesia. The serum levels of aspartate transaminase (AST) and alanine transaminase (ALT) were determined by an automatic biochemical analyzer (MGC 240, Thermo Scientific™ Indiko™). In addition, the serum levels of TGF-β2 (CUSABIO Biotech, Wuhan, China) and OSM (CUSABIO Biotech, Wuhan, China) were measured by ELISA Kits according to the manufacturer's protocols.

Statistical Analysis
Statistical analysis was carried out using GraphPad Prism 5 software (GraphPad Software, La Jolla, CA, USA). Data are presented as means ± standard deviations (SD). Statistical significance was calculated using Student's t test. p < 0.05 was considered significant. Means ± SDs are shown in Figures where applicable.