ABSTRACT
Partial hepatectomy (PH) is commonly used to treat patients with hepatocellular carcinoma. The recovery of patients from PH depends on the initiation of liver regeneration, a process that mainly relies on liver cell proliferation. As sex affects the human liver regeneration progress, we investigated sex disparity in PH-induced liver regeneration in adult zebrafish. We found that, after PH, males began liver regeneration earlier than females in terms of liver cell proliferation and liver mass recovery, and this was associated with earlier activation of Yap1 signaling in male than female livers. We also found that androgen receptors regulated the sex-biased liver regeneration in a Yap1-dependent manner and that activated estrogen receptors are responsible for the later onset of female hepatocyte proliferation. Furthermore, we identified that S100A1, a calcium-binding protein, regulates the sex disparity in liver regeneration, as heterozygous S100A1 knockout inhibited Yap1 activity in male livers and delayed hepatocyte proliferation in males following PH. Thus, multiple pathways and/or their interplays contribute to the sex disparity in liver regeneration, suggesting that sex-biased therapeutic strategies are required for patients who have received PH-based therapies.
INTRODUCTION
Partial hepatectomy (PH) has been widely used in the clinic for treating hepatocellular carcinoma to improve patients' overall survival (Labgaa et al., 2020). As mature liver cells can re-enter the cell cycle upon liver injury, patients can regenerate their lost liver mass after a major PH (Michalopoulos, 2007). To reduce the risk of compromised liver regeneration and improve the outcome of PH-based therapies, a comprehensive understanding of PH-induced liver regeneration is of great importance.
Clinical reports have shown that male patients have a higher level of liver damage, whereas female patients have better recovery, after receiving PH (Birrer et al., 2022; Dong et al., 2022). Furthermore, sex disparity in the speed of liver regeneration has been demonstrated in rodent PH models despite conflicting results regarding which sex has earlier regeneration (Batmunkh et al., 2017; Wang et al., 2013; Zhou et al., 2016). This sex-biased liver regeneration has often been attributed to the responses of estrogen receptors (ERs) because PH can induce ER activation in the liver and estrogen promotes hepatocyte proliferation (Batmunkh et al., 2017; Francavilla et al., 1990; Kao et al., 2017; Uebi et al., 2015). Nevertheless, sex disparity in ER activity can only explain a female-biased liver regeneration but not a male-biased liver regeneration. Apart from sex hormone receptors, only a handful of factors such as miR-17∼92 and histone deacetylase have been suggested to be responsible for sex-biased liver regeneration (Wang et al., 2013; Zhou et al., 2016). Overall, although the existence of sex disparity in PH-induced liver regeneration has been widely reported, the mechanisms underlying this disparity have not been sufficiently explored.
The Hippo-YAP1 signaling pathway is known to regulate ectopic organ growth including organ regeneration and tumorigenesis (Hansen et al., 2015; Kowalczyk et al., 2022). In a rat PH model, YAP1 was activated shortly after the surgery and switched off gradually when reaching the full recovery of liver mass, suggesting its correlation with liver regeneration (Grijalva et al., 2014). Knockout (KO) of YAP1 has also been found to attenuate PH-induced liver regeneration in rats (Lu et al., 2018). The major upstream regulators of Hippo-YAP1 signaling are tyrosine receptor kinases (RTKs), but many other factors have been found to interact with this pathway (Hansen et al., 2015). Recently, S100A1, a calcium-binding protein from the S100 family, was suggested to promote YAP1 activity in human papillary thyroid carcinoma cells in vitro and in vivo (Wang et al., 2021). Although YAP1 activity has been demonstrated to drive liver regeneration following PH, its role in the sex disparity of liver regeneration remains unclear.
Zebrafish (Danio rerio) have been widely used for regeneration research because of their high regenerative capacity (Gao and Peng, 2021). To study PH-induced liver regeneration, the zebrafish PH model was developed over 15 years ago and has improved continuously (Kan et al., 2009; Oderberg and Goessling, 2021; Sadler et al., 2007). It has been employed to investigate the mechanism underlying PH-induced liver regeneration in several studies (Chen et al., 2020; Dovey et al., 2009; Goessling et al., 2009; Kan et al., 2009; Zhu et al., 2014). Kan et al. (2009) showed that male zebrafish recovered liver mass earlier than female zebrafish after receiving PH, which is consistent with the results from studies using the mouse PH model (Wang et al., 2013; Zhou et al., 2016). However, sex disparity in zebrafish liver regeneration has yet to be fully understood.
In this study, we confirmed sex disparity in zebrafish liver regeneration following PH and further explored the underlying mechanism(s). In general, males showed earlier liver cell proliferation and earlier recovery of liver mass than females did after receiving PH, and this earlier initiation was associated with earlier activation of the Yap1 signaling pathway and androgen receptors (ARs) in their livers. Activated ERs were required for the relatively later initiation of hepatocyte proliferation in females. Using RNA-sequencing (RNA-Seq) analysis, we also identified S100A1 as being involved in sex-biased liver regeneration following PH. S100A1 knockdown inhibited Yap1 activity in larvae and delayed their liver regeneration following chemical-induced injury. Moreover, heterozygous KO of S100A1 inhibited Yap1 activity in adult male zebrafish livers and delayed their liver cell proliferation following PH. Together, these findings revealed the crucial roles played by sex hormone receptors, including ARs and ERs, and the S100A1-Yap1 signaling cascade in determining sex disparity in zebrafish liver cell proliferation following PH.
RESULTS
Male zebrafish recover liver mass earlier than female zebrafish following PH through compensatory liver regeneration
Previously, conflicting results have been reported regarding whether zebrafish can regrow the resected ventral lobe (Dovey et al., 2009; Goessling et al., 2009; Kan et al., 2009; Zhu et al., 2014). Oderberg and Goessling (2021) suggested that only some zebrafish can regrow the resected ventral lobe after 36 days. In the current study, we performed PH on Tg(fabp10a:LOXP-EGFP-LOXP-DsRed; TETRE:Cre-ERT2) zebrafish to visualize the shape of the regenerating liver (Li et al., 2019). As shown in Fig. 1A, the resected ventral lobes had no sign of regrowth within 7 days post-PH in both sexes, which is consistent with most previous reports (Kan et al., 2009; Oderberg and Goessling, 2021; Zhu et al., 2014). Later, a long-term experiment was conducted, and we found that neither female nor male zebrafish regrew their resected liver even after 1 month of recovery (Fig. 1B). The regrowth of ventral lobes was not observed in subsequent experiments, suggesting that the zebrafish used in this study underwent compensatory liver regeneration after PH.
The liver-to-body ratio (LBR) is commonly used to indicate changes in the liver mass (Batmunkh et al., 2017; Kan et al., 2009). Based on the LBR of EGFP+ zebrafish before and after receiving PH, the recovery of the LBR in zebrafish of both sexes during liver regeneration was measured (Fig. 1C). In male zebrafish, a significant increase in their LBR relative to that on Day 0 was observed on Day 3, which made their LBR comparable with the pre-PH level. By contrast, female zebrafish began their LBR recovery no earlier than 5 days post-PH. By Day 7, both male and female zebrafish achieved full recovery of their LBR. Therefore, male zebrafish recover liver mass earlier than female zebrafish after receiving PH.
Hepatocyte proliferation and hypertrophy initiate earlier in male zebrafish than in female zebrafish following PH
Liver regeneration following PH mainly relies on the proliferation of pre-existing hepatocytes unless the proliferating capability of hepatocytes is inhibited simultaneously (Luther, 2010; Wang et al., 2017). To examine the proliferation level of liver cells in zebrafish of both sexes following PH, we conducted immunohistochemical (IHC) staining of Pcna at 1, 3 and 7 days post-PH (Fig. 2A,B). We found that male zebrafish had significantly increased liver cell proliferation on Day 1 and Day 3, whereas female zebrafish had significantly increased liver cell proliferation on Day 3 and Day 7. Considering the possibility of hepatocyte proliferation initiating before Day 1 and the potential contribution of other cell types towards proliferation, we examined the level of hepatocyte proliferation in wild-type (WT; Singapore indigenous) zebrafish at 12, 24, 48 and 72 h post-PH by co-immunofluorescence (IF) staining of Pcna proliferation marker and Hnf4α hepatocyte marker (Fig. 2C,D). We found that male zebrafish showed significantly increased hepatocyte proliferation as early as 12 h post-PH, whereas female zebrafish did not show an increase until 48 h post-PH. Consistent with the results from IHC and IF imaging, PH induced earlier upregulation of cycle-related genes ccnd1 and ccng1 in male livers than in female livers based on real-time quantitative PCR (RT-qPCR) analysis (Fig. S1A,B). In general, the earlier initiation of liver cell proliferation during PH-induced liver regeneration matched the earlier recovery of LBR in male zebrafish.
In addition to hepatocyte proliferation, previous studies have suggested that the enlargement of hepatocytes, termed hepatocyte hypertrophy, contributes to the early recovery of liver mass following liver resection (Miyaoka et al., 2012; Marongiu et al., 2017). To evaluate the sex disparity in liver cell hypertrophy during zebrafish liver regeneration following PH, we performed IF staining of E-cadherin to outline the cell membrane in EGFP+ livers, and we measured the area of each hepatocyte based on the staining (Fig. S1C). On the next day of PH, the average size of hepatocytes in males, but not in females, was significantly increased (Fig. S1D). This PH-induced hypertrophy in male zebrafish did not persist but dropped below the baseline 3 days after PH. In female zebrafish, hepatocyte size increased steadily during PH-induced liver regeneration. By Day 7, their livers had significantly larger hepatocytes than those of females before PH. In summary, post-PH, male zebrafish livers showed significant hepatocyte hypertrophy at the early stage of regeneration, which was reversed later, whereas female zebrafish livers showed moderate hepatocyte hypertrophy over the course of regeneration. This earlier hepatocyte hypertrophy in male zebrafish livers could contribute to the earlier recovery of liver mass in male than in female zebrafish following PH.
Yap1 activity is responsible for the earlier initiation of liver regeneration in male zebrafish than in female zebrafish following PH
Because YAP1 activity is known to regulate liver regeneration, we suspected that it might play a role in determining the sex disparity in PH-induced zebrafish liver regeneration. To examine the activity of Yap1 in the zebrafish liver during PH-induced liver regeneration, PH and sham surgery were performed on WT (Tübingen) zebrafish of both sexes. By RT-qPCR analysis, we found that PH induced upregulation of yap1 in male and female livers within 24 h post-PH (Fig. 3A). Moreover, male zebrafish livers had significantly higher yap1 expression than female livers at 6 h post-PH. The expression of Yap1 downstream target genes – including ctgfa, cyr61 and ankrd1b – was upregulated in male livers at 6 h post-PH (Fig. 3B). Interestingly, at 24 h and 48 h post-PH, these Yap1 downstream target genes started to show significant upregulation in female zebrafish livers, suggesting later activation of Yap1 in females (Fig. 3C,D). Consistent with the results of RT-qPCR, IF staining of Yap1 protein showed that the nuclear Yap1 level in male livers was significantly increased at 6 h post-PH (Fig. 3E,F). By contrast, female zebrafish livers had a significantly increased level of nuclear Yap1 at 24 h post-PH, indicating later activation of Yap1 than in male zebrafish following PH. No increase in nuclear Yap1 was observed in male or female zebrafish livers 72 h after PH (Fig. S1E). Overall, the earlier initiation of liver cell proliferation in male zebrafish following PH was associated with earlier activation of Yap1 in male than in female livers.
To investigate whether the earlier initiation of liver regeneration in male zebrafish requires earlier Yap1 activation, we blocked Yap1 activity in WT (Tübingen) male zebrafish with the Yap1 inhibitor verteporfin (VP) prior to the PH/sham surgeries. Injecting VP into the male zebrafish inhibited the upregulation of Yap1 downstream target genes ctgfa (ccn2a) and cyr61 (ccn1) in the liver after PH, indicating the successful blockade of Yap1 activity (Fig. 4A). We then checked the levels of hepatocyte proliferation in dimethyl sulfoxide (DMSO)/VP-treated male zebrafish livers during PH-induced liver regeneration based on IF staining of Pcna/Hnf4α. We found that VP treatment inhibited the increase in hepatocyte proliferation at 24 h post-PH while enhancing hepatocyte proliferation at 72 h post-PH, which resulted in a regenerative pattern close to that of female zebrafish (Fig. 4B,C). To examine the effects of Yap1 inhibition on the recovery of LBR during PH-induced liver regeneration, we performed PH on DMSO/VP-treated Tg(fabp10a:DsRed; ela3l:EGFP) zebrafish (Korzh et al., 2008). Consistent with the inhibited hepatocyte proliferation, the recovery of LBR in VP-treated male LiPan zebrafish after PH was delayed compared with that in the control zebrafish (Fig. 4D). Therefore, Yap1 activation is responsible for the earlier initiation of hepatocyte proliferation in male zebrafish than in female zebrafish following PH.
ARs regulate the sex disparity in zebrafish liver regeneration following PH by promoting Yap1 activity
ARs are major sex hormone receptors that regulate the development and maintenance of male-associated phenotypes. In the rat PH model, inhibiting AR activity delayed liver regeneration (Tsukamoto and Kojo, 1990). Based on the results from RT-qPCR, the expression of AR-responsive genes – including cyp19a1b, sult2st3 and tmprss2 – increased significantly in male zebrafish livers, but not in female livers, at 6 h post-PH (Fig. 5A). To investigate the function of ARs in the sex disparity in zebrafish liver regeneration, we exposed WT (Tübingen) male zebrafish to flutamide, an AR antagonist, before and after the PH/sham surgeries. As shown in Fig. 5B, blocking AR activity with flutamide inhibited the upregulation of the Yap1 downstream targets ctgfa and cyr61 in male zebrafish livers at 24 h post-PH. The initiation of hepatocyte proliferation in the flutamide-treated male zebrafish was also delayed from 24 h post-PH to 72 h post-PH (Fig. 5C,D). In addition, blocking AR activity with flutamide delayed the recovery of LBR following PH in LiPan male zebrafish (Fig. 5E). Therefore, AR activation is required for the earlier initiation of hepatocyte proliferation in male zebrafish than in female zebrafish following PH.
To further examine how ARs regulate the sex disparity in zebrafish liver regeneration, we used an AR agonist, dihydrotestosterone (DHT), to treat female WT (Tübingen) zebrafish prior to PH. In contrast to the effect of flutamide, DHT treatment facilitated the initiation of hepatocyte proliferation in female zebrafish liver regeneration, with a significant increase in hepatocyte proliferation at 24 h post-PH (Fig. 5F,G). Intriguingly, this facilitating effect of DHT on female hepatocyte proliferation following PH could be abolished by blocking Yap1 activity with VP. Previous studies have reported that ARs can bind with YAP1 and promote its nuclear translocation in prostate cancer cells to enhance cell survival (Cinar et al., 2020; Kuser-Abali et al., 2015). In our experiments, DHT treatment induced upregulation of the Yap1 downstream targets ctgfa, amotl2b and cyr61 in female zebrafish livers at 24 h post-PH (Fig. 5H). This was consistent with the earlier observed inhibition of expression of Yap1 downstream targets following flutamide treatment. Together, these results suggest that AR activity can regulate the sex disparity in PH-induced zebrafish liver regeneration through promoting Yap1 activity.
ER activity regulates the pattern of female zebrafish liver regeneration following PH
Some studies have attributed the sex disparity in liver regeneration following PH to high ER activity in females (Batmunkh et al., 2017; Kao et al., 2017; Uebi et al., 2015). To confirm whether ERs are required for the relatively late initiation of female liver regeneration following PH, female WT (Tübingen) zebrafish were exposed to the ER antagonist ICI 182, 780 (ICI) from 48 h prior to the PH/sham surgeries. IF staining of Pcna proliferation marker and Bhmt hepatocyte marker was conducted to measure the level of hepatocyte proliferation (Fig. S2A). We found that inhibiting ERs with ICI advanced the initiation of hepatocyte proliferation in female zebrafish from 72 h post-PH to 24 h post-PH, which resulted in a pattern similar to that of liver regeneration in male zebrafish (Fig. S2B). In addition, the recovery of LBR in female LiPan zebrafish at 24 h post-PH was enhanced by ICI treatment (Fig. S2C). Notably, ICI treatment did not alter the expression of Yap1 downstream targets ctgfa and cyr61 in female livers at 24 h post-PH (Fig. S2D). This suggests that, unlike AR regulation of liver regeneration, ERs do not regulate liver regeneration by affecting Yap1 activity. To confirm whether the effect of ICI treatment relies on blocking the interaction between ER and estrogen, a nonsteroidal aromatase inhibitor, fadrozole, was used to block the production of estrogen in WT (Tübingen) female zebrafish during PH-induced liver regeneration. Fadrozole treatment facilitated the initiation of hepatocyte proliferation in female zebrafish following PH, which was consistent with the results from the ER blockade experiments (Fig. S2E,F). Therefore, the activity of ERs is responsible for the later initiation of hepatocyte proliferation following PH in female zebrafish than in male zebrafish.
PH induces transcriptomic changes differently between male livers and female livers
To identify potential regulators of sex disparity in PH-induced liver regeneration, we conducted RNA-Seq with the WT (Singapore indigenous) zebrafish liver tissues collected on Day 1 and Day 3 following PH. By comparing the transcriptome in the PH livers and the sham surgery-treated livers, 2531 differentially expressed genes (DEGs) during PH-induced liver regeneration were identified (Fig. 6A). As shown in Fig. 6B, the numbers of genes that were significantly deregulated by PH were markedly different between male and female liver. Female zebrafish had 2181 DEGs in the liver after PH on Day 1, whereas male zebrafish had only 197 DEGs. On Day 3, 157 DEGs were found in the male livers, and only 26 DEGS were found in the female livers. Moreover, DEGs in different sex groups at different time points during the PH-induced liver regeneration showed very limited overlap among each other (Fig. S3A,B).
Considering the earlier initiation of liver regeneration in male than in female zebrafish following PH, we focused on genes that were significantly upregulated in the liver of one sex, but not upregulated [log2 fold change (FC)<1] in the liver of the other sex, on Day 1. Table S1 lists the top 20 genes that were most significantly upregulated in the male and female liver on Day 1. Of these genes, BX511067.1 had the strongest upregulation in the PH male livers. However, BX511067.1 is a gene identified recently, and its function is completely unknown. This made it difficult to predict its potential role in liver regeneration. Another gene of interest was s100a1, which was the only gene that was significantly upregulated in the male zebrafish liver while being downregulated in the female zebrafish liver on Day 1 (Fig. 6C). To verify the male-predominant upregulation of s100a1 in the zebrafish liver at the early stage of liver regeneration, we performed RT-qPCR analysis with WT (Tübingen) zebrafish liver samples collected at 6 h, 24 h and 48 h post-PH. We found that male zebrafish livers showed s100a1 upregulation since 6 h post-PH, whereas female zebrafish livers showed this upregulation only after 24 h (Fig. 6D). In addition, the level of s100a1 expression in the PH male zebrafish livers was significantly higher than that in the PH female livers. The upregulation of s100a1 in female and male livers disappeared by 48 h post-PH. The expression of S100A1 can be induced by SOX9, which is related to sex determination and reversal, and another SOX family member, SOX5, can further enhance the SOX9-induced upregulation of S100A1 (Saito et al., 2007). In our experiments, both sox5 and sox9 were found to be upregulated only in male livers at 6 h post-PH (Fig. 6E). Because the sex disparity in the upregulation of s100a1, sox5 and sox9 matched that of liver regeneration, we hypothesized that S100A1 plays a role in the sex disparity in PH-induced zebrafish liver regeneration.
To investigate the sex disparity in the regulation of signaling pathways and other biological functions during PH-induced zebrafish liver regeneration, we performed gene set enrichment analysis (GSEA) of hallmark gene sets with the RNA-Seq results of the regenerating livers of female and male zebrafish. The results were categorized based on the previous study by Liberzon et al. (2015). We found that a major proportion of hallmark pathways were affected differentially by PH between males and females (Fig. S3C,D). For example, the pathways of unfolded protein response and mTORC1 signaling were upregulated in male zebrafish livers on Day 1 but were not upregulated in female livers until Day 3, which matched the sex disparity in liver regeneration after PH. The activation of the unfolded protein response might be caused by increased protein production associated with cell proliferation, while mTORC1 signaling has been reported to regulate hepatocyte hypertrophy during PH-induced liver regeneration (Fouraschen et al., 2013; Haga et al., 2005). Many pathways, such as those in the immune, metabolic and proliferation categories, were regulated differently by PH on Day 3 relative to Day 1 in zebrafish livers. This suggests that activation of those pathways could play a temporal role in liver regeneration and that their regulation could represent regeneration progresses. Most pathways under the immune category and proliferation category were significantly upregulated in the female zebrafish liver on Day 1 and Day 3, respectively. The upregulation of proliferation-related pathways in the female zebrafish liver on Day 3 matched the significantly increased hepatocyte proliferation in females at the same time point following PH. By contrast, we did not find an overall activation of proliferation-related pathways in the male zebrafish liver on Day 1, which was inconsistent with the observed earlier hepatocyte proliferation in males than in females during PH-induced liver regeneration. Considering that the initiation of male hepatocyte proliferation can be traced as early as 12 h post-PH, these pathways could possibly be activated in the male liver before the examined time points. To support that, early E2 response was upregulated in the female livers before Day 1, whereas male livers had upregulation of the late E2 response at that time point, implying that activation of the early E2 response in male livers could have already finished by Day 1. In addition to the sex disparity, PH-induced deregulation of hallmark pathways in zebrafish livers changed over time. Many pathways, such as the pathways under the categories of immune, metabolic and proliferation, were regulated reversely by PH on Day 3 compared with their regulation on Day 1 in male and female zebrafish. This phenomenon suggested that those pathways could play a temporal role in liver regeneration and that their activities might represent the progress of regeneration (e.g. initiation or termination). Overall, the differential regulation of genes and pathways between male livers and female livers following PH provided us with candidates potentially responsible for the observed sex disparity in PH-induced zebrafish liver regeneration.
s100a1 knockdown delays liver expansion during zebrafish larvae development and liver regeneration via inhibiting Yap1 activity
Because the function of S100A1 in the normal liver has not been studied before, we used the morpholino (MO) targeting the start codon of s100a1 (MO-s100a1-ATG) to investigate whether s100a1 knockdown affects liver development and regeneration in Tg(fabp10a:CFP-NTR) zebrafish larvae (Choi et al., 2014). As shown in Fig. S4A, knockdown of s100a1 significantly reduced the size of the liver in nitroreductase (NTR)+ larvae at 5 days post fertilization (dpf). By 6 dpf, the size of the livers in the s100a1-knockdown larvae became comparable with that of the control larvae probably due to the loss of MO activity (Fig. S4B). Consistent with the impaired liver expansion, reduced expression of cell cycle-related genes ccnd1 and cdk1 was found in the s100a1-knockdown group at 4.5 dpf (Fig. S4C).
With NTR+ zebrafish, we were able to induce hepatocyte ablation with metronidazole (MTZ) for studying the effect of s100a1-knockdown on chemical-induced liver regeneration (Fig. 7A). Twenty-four hours of MTZ treatment significantly reduced the liver size of NTR+ larvae, and then the liver sizes in both the s100a1-knockdown larvae and control larvae gradually recovered over time (Fig. 7B,C). The s100a1-knockdown larvae regenerated significantly smaller livers than did the control larvae within 48 h after MTZ-induced injury (Fig. 7D). Therefore, the expansion of the liver during embryonic development and liver regeneration in zebrafish larvae required the participation of S100A1. To examine whether s100a1 knockdown affects Yap1 activity, we performed RT-qPCR analysis of the expression of Yap1 downstream targets in NTR+ larvae after 24 h of MTZ treatment. We found that knockdown of s100a1 significantly reduced the expression of Yap1 downstream targets, including ctgfa and cyr61, in NTR+ larvae (Fig. 7E). This result suggests that S100A1 can regulate liver expansion in zebrafish larvae by promoting Yap1 activity.
Reduced s100a1 expression delays PH-induced liver regeneration in male zebrafish by inhibiting Yap1 activity
To confirm whether the sex-biased liver regeneration in zebrafish following PH requires the expression of s100a1, we generated an s100a1-KO zebrafish strain with CRISPR/Cas9-mediated large-fragment deletion. Two guide RNAs (gRNAs) were designed to target the 5′ and 3′ terminals of the s100a1 coding region and microinjected into WT (AB) zebrafish embryos together with the Cas9 protein (Fig. S5A). This resulted in the deletion of the whole translated region of s100a1. By PCR and agarose gel electrophoresis, we confirmed that heterozygous and homozygous s100a1-KO zebrafish have the short allele resulting from the deletion in their liver genomes (Fig. S5B). In addition, full-length s100a1 was not present in the liver genome of homozygous KO fish, indicating complete loss of the full-length allele. After sequencing the short deletion allele, we confirmed that the targeted region was deleted completely in the genome of s100a1-KO zebrafish (Fig. S5C). In addition, the expression level of s100a1 in the s100a1+/− male livers was significantly lower than that in the livers of their WT siblings, suggesting that the heterozygous KO induced haploinsufficiency of s100a1 (Fig. S5D). However, it seems that liver growth in adult zebrafish was not affected by s100a1 haploinsufficiency as the LBR in s100a1+/− males was comparable with that in WT males (Fig. S5E).
With this newly generated zebrafish mutant strain, we compared PH-induced liver regeneration between s100a1+/− male zebrafish and WT (AB) male zebrafish. Based on the results from IF imaging, s100a1+/− male zebrafish showed inhibited hepatocyte proliferation compared with their WT siblings at 24 h post-PH, which indicates delayed initiation of liver regeneration (Fig. 8A,B). Furthermore, the recovery of LBR was delayed in s100a1+/− males compared with WT males following PH (Fig. 8C). Because we proposed that S100A1 regulates zebrafish liver regeneration by regulating Yap1 activity, we examined the expression level of s100a1 and Yap1 downstream target genes in the s100a1+/− male livers and WT male livers after PH. As expected, the upregulation of s100a1 at the early phase of PH-induced liver regeneration was inhibited in the s100a1+/− male livers (Fig. 8D). The expression levels of Yap1 downstream targets ctgfa, amotl2b and cyr61 were upregulated in the WT livers but not in the s100a1+/− livers at 6 h post-PH. The inhibition of the upregulation of ctgfa in the s100a1+/− livers continued until 24 h post-PH. In addition, s100a1+/− male zebrafish showed no hepatocyte hypertrophy within 72 h following PH, indicating that hepatocyte hypertrophy at the early stage of regeneration requires S100A1 (Fig. S6A,B). Therefore, S100A1 drives the earlier initiation of liver regeneration in male zebrafish following PH via regulating hepatocyte hypertrophy and promoting Yap1-mediated hepatocyte proliferation.
DISCUSSION
Zebrafish have been widely used in modeling human diseases including tissue regeneration owing to their physiological and molecular similarities with humans, high regenerative capability and amenability for in vivo imaging. The existence of sex disparity in liver regeneration following PH has been well demonstrated in the clinic and in animal models (Batmunkh et al., 2017; Birrer et al., 2022; Wang et al., 2013). In the current study, we found that male zebrafish began liver regeneration earlier than female zebrafish following PH in terms of liver cell proliferation and LBR recovery. The earlier liver cell proliferation observed in male zebrafish is consistent with previous reports of male zebrafish recovering LBR earlier than female zebrafish following PH (Kan et al., 2009). Compared with the results from previous studies using rodent PH models, the male-biased liver regeneration observed in zebrafish is consistent with that in mice (Wang et al., 2013; Zhou et al., 2016) but contradictory to that in rats (Batmunkh et al., 2017). Currently, there is an absence of clinical data on the sex disparity in liver regeneration following PH. A recent study showed that female patients have faster liver regeneration than male patients, but the data were obtained from patients who received liver partition and portal vein ligation rather than liver resection (Birrer et al., 2022). Therefore, data from the current study conducted in zebrafish, together with the earlier works conducted in rodents, may provide valuable references for improving the clinical application of PH-based therapies.
After liver resection, mammal livers enlarged the remnant to compensate the lost liver mass rather than regenerating the resected liver lobes. Nevertheless, the terms ‘liver regeneration’ and ‘compensatory liver regeneration’ have traditionally been used to describe this process in the literature (Abu Rmilah et al., 2019; Michalopoulos, 2007). In the current study, we found that both male and female zebrafish recovered their LBR by 7 days after PH without regrowing their resected ventral lobes. This compensatory regeneration in the short term is consistent with mammalian liver regeneration and many previous reports on zebrafish liver regeneration (Kan et al., 2009; Sadler et al., 2007; Zhu et al., 2014). For long-term recovery, Kan et al. (2009) suggested no regrowth of ventral lobes in zebrafish up to 6 months post-PH, while Zhu et al. (2014) observed regrowth of ventral lobes within 60 days. The most recent study by Oderberg and Goessling (2021) showed that zebrafish livers have the capability for both epimorphic and compensatory regeneration. In their study, only one transgenic zebrafish line was used, but only some of the fish could regrow their resected liver lobes within 36 days post-PH. It is unclear whether this variation in the capability of regeneration is due to subtle differences in the genetics. In the current study, we did not observe any sign of regrowth even after 30 days.
Although male zebrafish showed a significant increase in hepatocyte proliferation as early as 12 h post-PH, the absolute level of proliferation was only ∼5%. In the current study, the LBR of male zebrafish showed full recovery as early as 24 h after 30% of PH. Unless the speed of liver cell mitosis in adult zebrafish is comparable with that in embryonic development, it would be difficult for male zebrafish to recover their LBR in such a short period with only a low level of hepatocyte proliferation. One explanation could be that male zebrafish compensate for the lost liver mass rapidly not only by hepatocyte proliferation but also by hepatocyte hypertrophy at the early stage of liver regeneration. Hepatocyte hypertrophy has been suggested to contribute to liver regeneration together with hepatocyte proliferation in rodent PH models (Miyaoka et al., 2012; Marongiu et al., 2017). In our experiments, significant hepatocyte hypertrophy occurred in the male zebrafish liver at 1 day post-PH but regressed at the later time points. The hepatocyte hypertrophy during PH-induced liver regeneration is mediated by mTORC1 signaling activation (Fouraschen et al., 2013; Haga et al., 2005). Consistently, our GSEA results also showed that the mTORC1 pathway was significantly upregulated in the male zebrafish liver at Day 1 post-PH. Whether hepatocyte hypertrophy is required for the earlier LBR recovery of male livers could be a focus for future studies.
The Hippo-YAP1 signaling pathway controls organ regeneration and tumorigenesis by regulating cell differentiation and proliferation (Hansen et al., 2015; Kowalczyk et al., 2022). Blocking YAP1 activity impaired hepatocyte proliferation and liver mass recovery in mice following PH, demonstrating that YAP1 activity is important to PH-induced liver regeneration (Liu et al., 2019; Lu et al., 2018). In the current study, we have found that the earlier hepatocyte proliferation and LBR recovery in male zebrafish following PH required earlier activation of Yap1 in male livers. This result suggests, for the first time, the role played by Yap1 activity in regulating the sex disparity in liver regeneration. The uptake of testosterone in the rat liver after PH suggests that ARs participate in regulating liver regeneration (Yamaguchi et al., 2001). However, the specific roles played by ARs in liver regeneration are poorly understood. In the current study, we found that ARs regulated the earlier initiation of hepatocyte proliferation and LBR recovery in male than in female zebrafish after PH. ARs have been found to bind with YAP1 in prostate cancer cells and enhance its activity by facilitating its nuclear translocation (Cinar et al., 2020; Kuser-Abali et al., 2015). In our experiments, AR activity regulated the initiation of hepatocyte proliferation following PH in a Yap1-dependent manner. Thus, ARs regulate the sex disparity in zebrafish liver regeneration, at least partially, through interacting with Hippo-Yap1 signaling.
In zebrafish, estrogen promotes hepatocyte proliferation and regulates the sex disparity in adult liver size during development (Chaturantabut et al., 2019). Notably, the promoting effect of estrogen on zebrafish liver growth is mediated by G protein-coupled estrogen receptors (GPERs) rather than ERs. In a previous study, inhibiting GPER activity, but not ER activity, during estrogen treatment abolished the estrogen-induced liver growth in zebrafish, whereas activating GPER alone induced a comparable level of hepatocyte proliferation as estrogen treatment (Chaturantabut et al., 2019). In the current study, inhibiting ER activity in female zebrafish livers with ICI promoted their hepatocyte proliferation and LBR recovery at the early stage following PH. As ICI is an inhibitor targeting ERs, this result does not necessarily conflict with the promoting effect of estrogen on hepatocyte proliferation mediated by GPER. However, ICI treatment has been reported to attenuate estrogen-induced liver cell proliferation in rats, which conflicts with the idea that estrogen promotes hepatocyte proliferation not through activating ERs (Batmunkh et al., 2017). Future studies may need to confirm the specificity of ICI on ERs and the effect of ER activation alone on liver regeneration without potential interference from GPER activity.
According to our RNA-Seq analysis of the regenerating zebrafish livers, the regulation of genes during PH-induced liver regeneration was very different between male zebrafish and female zebrafish. We found a relatively small number of significantly deregulated genes in male livers compared with female livers on Day 1 after PH. This result seems inconsistent with the earlier responses of male zebrafish than female zebrafish towards PH. It is likely that the time point we chose to collect samples was not early enough to catch rapid changes in RNA expression. We identified S100A1 as a potential regulator responsible for sex disparity in zebrafish liver regeneration after PH because it shows a male-biased expression significantly at the early stage of regeneration. As a calcium-binding protein, the main function of S100A1 is regulating the release and uptake of calcium in the smooth reticulum in a calcium level-dependent manner (Scott and Kekenes-Huskey, 2016). Recently, Wang et al. (2021) have demonstrated that S100A1 can regulate the Hippo-YAP1 signaling pathway in papillary thyroid carcinoma cells to promote cell proliferation. According to their study, knockdown of S100A1 can promote the phosphorylation of YAP1, which inactivates YAP1 and results in reduced nuclear YAP1 level and inhibited cell growth. In the current study, deficiency in S100A1 reduced Yap1 activity and impaired liver regeneration in both larval and adult zebrafish. Furthermore, the direct transcription regulators of S100A1, including Sox5 and Sox9, showed male-biased upregulation in zebrafish livers at the early stage of PH-induced regeneration. Based on these results, we propose a signaling cascade regulating the sex disparity of zebrafish liver regeneration following PH: PH induces expression of Sox5 and Sox9 in male zebrafish livers; the increased level of Sox5 and Sox9 further induce the transcription of S100A1, resulting in earlier upregulation of S100A1 in male livers than females; subsequently, S100A1 promotes the activation of Yap1 through inhibiting its phosphorylation; and the activated Yap1 then induces the expression of downstream genes and initiates liver cell proliferation. Exactly how S100A1 regulates Yap1 phosphorylation requires further investigation.
Previous studies have demonstrated the function of S100A1 and ARs in regulating YAP1 activity (Cinar et al., 2020; Wang et al., 2021). In the current study, we confirmed that both ARs and S100A1 regulate sex disparity in liver regeneration by regulating Yap1 activity. These results seem to suggest a potential interaction between sex hormone receptors and S100A1. However, there are not yet any studies on this topic. In addition, s100a1 has not been identified as an estrogen-responsive gene or an androgen-responsive gene (Jin et al., 2013; Nishi et al., 2022). Therefore, no direct links can be established between sex hormone receptors and S100A1 except that both regulate Yap1 signaling activity.
In addition to s100a1, some other genes that can promote cell proliferation – such as anxa2a, fgf13b and postnb – were also upregulated only in the male zebrafish livers on Day 1 following PH (Li et al., 2021; Lu et al., 2020; Wu et al., 2018). Of them, postnb encodes an isoform of periostin, and periostin has been found to be required for competent liver regeneration after PH in mice (Wu et al., 2018). Knockout of Postn resulted in impaired liver regeneration in terms of hepatocyte proliferation, LBR recovery and immune responses following PH in mice. Because periostin was mainly produced by hepatic stellate cells (HSCs) in the mammalian liver, the expression pattern of postnb following PH suggests that HSCs could play a role in triggering the early initiation of liver regeneration in male zebrafish (Kumar et al., 2018). For the genes upregulated in the female zebrafish on Day 1, some of the most significantly upregulated genes (e.g. rnd1b and calm1b) encode inhibitors of cell proliferation, which may contribute to the relatively slow response of hepatocyte proliferation in females compared with males after PH (Orellana et al., 2010; Qin et al., 2018).
In conclusion, the current study confirms that male zebrafish undergo earlier liver cell proliferation and liver mass recovery than female zebrafish during PH-induced liver regeneration. This sex disparity is regulated by sex hormone receptors, including ARs and ERs, and the S100A1-Yap1 signaling cascade. These findings could enhance understanding of how sex affects liver regeneration following PH. Moreover, the current study has identified sex hormone receptors and S100A1 as potential targets for facilitating hepatocyte proliferation during PH-induced liver regeneration, which provides clues for developing novel therapeutic strategies.
MATERIALS AND METHODS
Zebrafish husbandry
The maintenance of zebrafish used in the current study follows the Institutional Animal Care and Use Committee guidelines from the Southern University of Science and Technology, China, and the National University of Singapore. Transgenic zebrafish lines including Tg(fabp10a:LOXP-EGFP-LOXP-DsRed; TETRE:Cre-ERT2) (gz38Tg) with EGFP-labeled hepatocytes (Li et al., 2019), Tg(fabp10a:DsRed; ela3l:EGFP) (gz15Tg) with DsRed-labeled hepatocytes and GFP-labeled exocrine pancreas (Korzh et al., 2008), and Tg(fabp10a:CFP-NTR) (s931Tg) with liver-specific expression of CFP-linked NTR (Choi et al., 2014) were used in the current study and designated as EGFP+, LiPan and NTR+, respectively. The genetic backgrounds of the WT zebrafish used in the current study included AB, Tübingen and Singapore indigenous strain, which was initially purchased from a local ornamental fish farm in Singapore and has been maintained in our laboratory for more than 20 years. The exact strains of WT zebrafish used in each experiment are indicated in the text.
Mutant generation
To generate s100a1-KO zebrafish with CRISPR/Cas9-mediated large-fragment deletion, two gRNAs targeting the 5′ UTR and 3′ UTR of the s100a1 gene were designed with CRISPRscan and synthesized by Genscript Biotech (Nanjing, Jiangsu, China). gRNAs were microinjected together with EnGen Spy Cas9 NLS (M0646, NEB) protein into zebrafish embryos (AB strain) at the one-cell stage to induce deletion of the coding region of s100a1. The sequences of the 5′ gRNA and 3′ gRNA are 5′-GGGGGAATTAGGCAAATAGG-3′ and 5′-TATGGGGGTGTGTTGTCATA-3′, respectively. The primers used for PCR validation of s100a1-KO are 5′-GTTAAATCACGCAAGTCCAGTCT-3′ (forward) and 5′-AGGGCAAATCAAGTCAGCCAA-3′ (reverse). Gel electrophoresis of PCR products was performed using a standard protocol. Sequencing of PCR products was conducted by Bio Basic Asia Pacific, Singapore.
Partial hepatectomy and LBR measurement
Adult zebrafish (>12 weeks old) of both sexes were used in the PH experiments. The procedure for performing PH on the zebrafish was derived from a published protocol (Oderberg and Goessling, 2021). In general, after anesthesia with 150 mg l−1 buffered tricaine (MS-222), a 2-3 mm incision was made on the zebrafish abdomen with a micro stab knife and precise scissors to expose the liver. Subsequently, the ventral lobe of the liver was removed with tweezers, resulting in ∼30% of the liver being resected. Sham surgery was performed as a control by making an incision on the abdomen without removing the liver tissue. To measure the LBR, euthanized zebrafish were placed on KimWipe papers (Kimberley-Clark, Irving, TX, USA) to remove attached water and then weighed on an analytical balance. Subsequently, zebrafish were dissected, and the livers were collected in pre-weighed Eppendorf tubes separately and weighed again. The liver weight was calculated by subtracting the weight of empty tubes.
Chemical treatment
Chemicals used in the current study include DHT (S4757, Selleck), fadrozole (HY-14247; MedChemExpress), flutamide (S1908, Selleck), ICI (Fulvestrant) (S1191, Selleck) and VP (HY-B0146, MedChemExpress). Stocks of chemicals were prepared at first by dissolving chemicals in DMSO (D4540, Sigma-Aldrich). Fadrozole, flutamide and ICI were used for treating adult zebrafish from 48 h before the surgery at 100 μg l−1, 5 μM and 1 mg l−1, respectively, through exposure. The control group for chemical exposure received 0.1% DMSO treatment. For VP and DHT treatments, chemical stocks were diluted to the working concentration in a solvent consisting of 10% DMSO, 45% PBS (P6748, Sigma-Aldrich), 40% PEG300 (HY-Y0873, MedChemExpress) and 5% Tween 80 (HY-Y1891, MedChemExpress). Twelve hours before the surgery, 25 ng VP and/or 200 nM DHT was delivered to adult zebrafish in 5 μl of the solvent by intraperitoneal injection. The control group for VP and DHT treatment was injected with only the solvent. Intraperitoneal injection was conducted following a published protocol (Kinkel et al., 2010) except MS-222 was used for anesthesia.
RNA extraction and RT-qPCR
Total RNA extraction from tissues and embryos was performed using TRIzol reagent (Invitrogen), followed by reversed transcription using a Transcriptor First Strand cDNA Synthesis Kit (Roche). The synthesized cDNA was used for RT-qPCR with SsoAdvanced Universal Supermixes (Bio-Rad) and GoTaq qPCR Master Mix (Promega) in a CFX96 Touch Real-Time PCR System (Bio-Rad) and ABI 7500 Fast Dx Real-Time PCR Instrument (Thermo Fisher Scientific), respectively. Genes of interest were amplified by 40 cycles using a standard three-step protocol (95°C, 20 s; 65°C, 15 s; 72°C, 30 s). All primers used in RT-qPCR are listed in Table S2.
RNA-Seq and bioinformatic analysis
Liver tissues collected from WT adult zebrafish after surgery were used for RNA extraction with TRIzol reagent. For RNAs extracted from PH/sham surgery-treated zebrafish, library preparation and Illumina sequencing were conducted by Novogen, Hong Kong. Each sample used for RNA-Seq contained RNAs from three individual zebrafish livers, and each biological group contained two pooled samples. After sequencing, reads were mapped to zebrafish reference genome assembly GRCz11 with TopHat2. The raw counts of sequences mapped to each gene were quantified using StringTie, followed by differential expression analysis using DESeq2 (Love et al., 2014). The criteria of significantly deregulated genes were adjusted P-value (padj)≤0.05 and FC≥2. To identify enriched pathways, pre-ranked GSEA was performed using the fgsea package (Korotkevich et al., 2016 preprint). The Hallmark pathway gene sets used for GSEA were downloaded from the Molecular Signature Database (Subramanian et al., 2005). The DEGs were ranked by their values of Wald statistic generated from DESeq2.
MO knockdown
For knockdown of s100a1, the MO targeting the start codon of s100a1 gene, MO-s100a1-ATG (5′-GTGACGACAGCTTGAAAAGATGTT-3′) (Gene Tools, Philomath, OR, USA), was synthesized and applied. The effectiveness of MO-s100a1-ATG inhibiting S100A1 expression was validated previously by western blotting (Zhu et al., 2024). Aliquots of MO (50 nM) were injected into zebrafish embryos at the one-cell stage. MTZ treatment was conducted by exposing 3.5 dpf larvae to 10 mM MTZ (S1907, Selleck) in 0.1% DMSO for 24 h. The control larvae were exposed to 0.1% DMSO only.
Histological analyses
Adult zebrafish liver and gut tissues were harvested together and fixed in 10% formalin and 4% paraformaldehyde in PBS overnight for subsequent paraffin embedding and cryo-embedding, respectively. Paraffin-embedded samples were sectioned at 5 μm with a microtome, and frozen samples were sectioned at 8 μm with a cryostat. All frozen samples were used for IF staining, while paraffin-embedded samples were primarily used for IHC staining. Sections were incubated with primary antibodies overnight at 4°C.. For IHC staining, a DAKO Real Envision Detection System (K500711, Agilent) was used for incubation of horseradish peroxidase (HRP)-conjugated secondary antibody and color development. For IF staining, sections were incubated with fluorescent-dye conjugated secondary antibodies (Alexa Fluor™, Invitrogen) for 2 h at room temperature. The primary antibodies used in IF and IHC include mouse anti-Pcna (sc-56, Santa Cruz Biotechnology) at a dilution of 1:200, mouse anti-E-cadherin (610182, BD Biosciences) at a dilution of 1:200, rabbit anti-Bhmt (ab211119, Abcam) at a dilution of 1:200, rabbit anti-Hnf4α (ab201460, Abcam) at a dilution of 1:2000, and rabbit anti-Yap1 (4912S, Cell Signaling Technology) at a dilution of 1:100. During immunostaining of proteins except for Yap1, tissue sections underwent antigen retrieval by placing in citrate buffer (C9999, Sigma-Aldrich) in 95°C water baths.
Photography and image analysis
Side views of dissected adult zebrafish were photographed individually with a stereo fluorescence microscope (SZX16, Olympus, Tokyo, Japan). Larvae photography was performed with an Axio Imager 2 light microscope (Carl Zeiss AG, Oberkochen, Germany) after immobilizing larvae in 3% methylcellulose. For IHC staining, imaging was conducted using the Axio Imager 2 light microscope and a NanoZoomer S60 digital slide scanner (Hamamatsu Photonics, Shizuoka, Japan). For imaging IHC staining, at least two high-power fields of view per sample were captured. Tissue sections of IF staining were imaged with an LSM 900 confocal laser scanning microscope (Carl Zeiss AG). For imaging IF staining, at least three high-power fields of view per sample were captured and analyzed. All image analyses were performed with ImageJ (Schneider et al., 2012).
Statistical analysis
For statistical analysis, Prism 8.0 (GraphPad, San Diego, CA, USA) was applied. Statistical significance was determined by two-tailed unpaired Student’s t-test unless stated otherwise.
Acknowledgements
The authors acknowledge the fish facilities and the Confocal Microscopy Laboratories from Department of Biological Science at National University of Singapore and School of Life Sciences at Southern University of Science and Technology. We also acknowledge the research scholarship of the Joint PhD Program between Southern University of Science and Technology and National University of Singapore.
Footnotes
Author contributions
Conceptualization: M.Z., Y.L., Z.G., D.L.; Methodology: M.Z., Y.L.; Software: M.Z.; Validation: M.Z.; Formal analysis: M.Z.; Investigation: M.Z., Y.L.; Resources: M.Z., Y.L., Z.G., D.L.; Data curation: M.Z.; Writing - original draft: M.Z.; Writing - review & editing: M.Z., Q.S., Z.G., D.L.; Visualization: M.Z.; Supervision: Z.G., D.L.; Project administration: Z.G.; Funding acquisition: Z.G., D.L.
Funding
This work was supported by the National Key Research and Development Program of China (2018YFA0801006 to D.L.), Ministry of Education - Singapore (MOE2019-T2-2-018 to Z.G.), National Natural Science Foundation of China (31771618 to D.L.) and Natural Science Foundation of Guangdong Province (2022A1515011878 to D.L.). Open Access funding provided by Southern University of Science and Technology. Deposited in PMC for immediate release.
Data availability
All relevant data can be found within the article and its supplementary information.
References
Competing interests
The authors declare no competing or financial interests.