Increased severity of liver injury and mortality of mice with hepsin deficiency during the early period after acetaminophen exposure

To investigate the potential physiological role of endogenous hepsin in the early response to APAP-induced liver injury, we analyzed hepsin expression in wild-type mouse liver lysates following APAP treatment using western blotting (Fig. S2). Hepsin levels remained stable at 0.5 h post-treatment with 400 mg/kg APAP, comparable to those in saline-treated control livers. However, a significant 70% reduction in hepsin expression was detected at the 1-h time point after APAP administration. Notably, histological analysis at this time point revealed no significant hepatocyte damage, suggesting that the rapid decrease in hepsin protein levels may primarily result from the activation and subsequent degradation of the hepsin serine protease in response to APAP-induced hepatotoxicity. Therefore, various APAP doses were tested to assess susceptibility to APAP-induced toxicity, aiming to investigate potential differences in tolerance between wild-type and hepsin−/− mice. After 8 h of exposure, the hepsin−/− mice exhibited lower survival rates compared with the wild-type mice at different APAP concentrations. This outcome displayed a dose-dependent negative correlation between the APAP doses and survival rate (Fig. 1A). Moreover, at an APAP dose of 400 mg/kg, hepsin−/− mice experienced rapid mortality within 8 to 10 h; in contrast, more than 80% of the wild-type mice survived beyond 30 h under the same conditions (Fig. 1B). For the hepsin−/− mice, the timing of mortality was such that it occurred prior the onset of liver repair and regeneration processes, thereby confirming the role of hepsin in determining the mechanism of APAP toxicity tolerance during the early phase of drug exposure.

Fig. 1

APAP induces aggravated liver injury and early lethality in hepsin−/− mice. (A) Survival rate of wild-type (WT) and hepsin–/– (KO) mice were assessed 8 h after APAP treatment at doses of 300, 400, or 600 mg/kg. (B) Survival rate after 400 mg/kg APAP treatment. (C) Measurement of serum ALT (alanine aminotransferase) and AST (aspartate aminotransferase) levels at the indicated time points after administering 400 mg/kg APAP. (D) Images of liver pathology, shown by hematoxylin and eosin staining and quantification of degenerated area as a percentage at the indicated time points after 400 mg/kg APAP treatment (n = 5–7 per group). (E) Levels of APAP-cysteine in mouse liver tissue at 1 and 2 h after 400 mg/kg APAP treatment (n = 6–8 per group). (F) Levels of total glutathione (GSH) in mouse liver tissue at the indicated time points after 400 mg/kg APAP treatment (n = 6–9 per group). (G) Western blot analysis of nitrotyrosine in mouse liver tissue at the indicated time points after 400 mg/kg APAP treatment (n = 6–10 per group). In the survival analysis, sample sizes for each group are indicated in brackets, and statistical significance was determined using the log-rank test with a p-value of < 0.0001. Data are presented as the mean ± SD in bar charts,with significance levels denoted by asterisks: *p < 0.05, **p < 0.01, ***p < 0.001

To understand the differences in APAP-induced toxicity between hepsin−/− and wild-type mice, the activities of various relevant enzymes were measured at 1, 2, 4, and 6 h after APAP administration. At the 4-h time point, serum levels of alanine transaminase and aspartate transaminase (AST) were markedly higher in the hepsin−/− mice, indicating more severe liver damage compared with their wild-type counterparts (Fig. 1C). Furthermore, quantification of hepatocellular vacuolation (a characteristic of cellular degeneration observed through tissue pathology staining) at the 2-h time point revealed a 1.5-fold greater area of liver degeneration in hepsin−/− mice compared with wild-type mice, which increased to ~ 2.5-fold greater by the 6-h time point (Fig. 1D). This correlation with the quantified degeneration area further supported the idea that, in the absence of hepsin, hepatic impairment is more severe during the early stages of APAP exposure.

To investigate the severe liver damage and increased mortality observed in hepsin−/− mice during the early stages after APAP administration, we employed ultra-performance liquid chromatography-tandem mass spectrometry to measure levels of APAP-cysteine adducts in mouse liver tissue, which indirectly assesses the generation of the toxic APAP metabolite, NAPQI (Hairin et al. 2013). The hepsin−/− mice had higher levels of APAP-cysteine adducts in the liver at 1 and 2 h post-APAP exposure compared with wild-type mice (Fig. 1E). Subsequently, we evaluated the levels of two crucial APAP-metabolizing enzymes, CYP2E1 and CYP1A2, revealing no significant differences in the protein expression levels and activities of CYP1A2 between hepsin−/− and wild-type mice either before or after APAP administration (Fig. S3).

At 2 h post-APAP treatment, hepsin−/− mice also exhibited a greater reduction in total GSH compared with wild-type mice (Fig. 1F), although GSH levels did not differ significantly between the hepsin−/− and wild-type controls treated with normal saline. Moreover, the restoration of GSH in hepsin−/− mice was less effective than in wild-type mice at 4 and 6 h following APAP treatment (Fig. 1F). In line with these observations, the nitrotyrosine level in the liver of hepsin−/− mice was approximately 2- to threefold higher than that measured for wild-type mice, with this difference becoming noticeable as early as 2 h post-APAP administration (Fig. 1G). Taken together, these findings indicated that hepsin−/− mice experience heightened oxidative stress in the early stages of APAP exposure, suggesting that endogenous hepsin might protect against toxicity during the early stages of APAP exposure, thereby significantly influencing survival outcomes in drug-induced toxicity cases.

Administering hepsin mitigates drug-induced liver damage by acetaminophen in hepsin−/− mice.

We employed an adeno-associated virus vector (AAV2/8) to deliver and express either human wild-type hepsin (hHPNWT; AAV-hHPNWT) or a loss-of-function mutant (hHPNRS; AAV-hHPNRS). The hHPNRS mutant was engineered to produce hepsin without functional activity by introducing two specific mutations in critical regions of its proteolytic activity. We validated the deficient cleavage process and the loss of protease function in this mutant form of HPN in our previous publication (Hsu et al. 2012), where it was used as a control in our experiments. An AAV encoding enhanced green fluorescent protein (EGFP; AAV-EGFP) was used as the vector control. The presence of human hepsin in hepsin−/− mouse serum and liver lysate was confirmed and quantified via enzyme-linked immunosorbent assay (ELISA) and western blot on Day 14 after administering the same dose of AAV-hHPNWT, AAV-hHPNRS or AAV-EGFP (Fig. 2A). These results confirmed that our AAV-hHPN vector design successfully enabled the overexpression of human hepsin in mouse liver. As expected, we observed that the loss-of-function mutant form of hHPNRS exhibited higher protein levels compared to the wild-type hHPNWT, both in serum and liver lysate. This difference may primarily be due to the auto-activation and subsequent degradation of hHPNWT (Vu et al. 1997; Wang et al. 2019), and it also provides evidence for the deficiency of protease activity in the loss-of-function mutant form of hHPNRS.

Fig. 2

AAV-mediated liver-specific administration of hepsin to hepsin−/− mice decreases their susceptibility to APAP, the area of liver degeneration, APAP-cysteine formation, and total glutathione in mouse liver. Hepsin−/− mice were administered AAV-hHPNWT, AAV-hHPNRS or AAV-EGFP for 3 weeks, followed by 400 mg/kg APAP treatment. (A) Human hepsin (hHPN) levels in serum and liver lysate prior to administering APAP (n = 3–5 per group). (B) Experimental timeline. (C) Images of liver pathology in sections assessed by hematoxylin and eosin staining and quantification as a percentage of the degenerated area at 2 h after 400 mg/kg APAP treatment. (D) Amounts of APAP-cysteine at 2 h after 400 mg/kg APAP treatment (n = 4–5 per group). (E) Total glutathione (GSH) at 2 h after 400 mg/kg APAP treatment (n = 4–5 per group). (F) Survival rate after 400 mg/kg APAP treatment. In the survival analysis, sample sizes for each group are indicated in brackets, and statistical significance was determined using the log-rank test, with significance levels represented by asterisks: **p < 0.01, ***p < 0.001. Data are presented as the mean ± SD in bar charts, with significance levels denoted by asterisks: *p < 0.05, **p < 0.01

Following the experimental setup, each group of hepsin−/− mice were administered APAP at 400 mg/kg, after which the survival rates were closely monitored (Fig. 2B). Overexpression of hHPNWT in hepsin−/− mice significantly increased their tolerance to APAP, as evidenced by a marked reduction in liver degeneration area and APAP-cysteine adduct levels at 2 h post-APAP injection compared to the two control groups (Fig. 2C, D). Furthermore, wild-type hepsin expression appeared to mitigate the depletion of GSH, a protective effect not observed in hepsin−/− mice expressing hHPNRS or in the vector-control group (Fig. 2E). Consistently, overexpression of hHPNWT in hepsin−/− mice resulted in a significant increase in APAP tolerance, with survival rates exceeding 80% even after 80 h (Fig. 2F). In contrast, survival rates were similar between vector-control mice and those expressing the loss-of-function mutated hepsin, with all mice in these groups succumbing within 30 h (Fig. 2F). This clearly indicated that the expression of hHPNWT could notably reduce the early symptoms of APAP-induced hepatotoxicity in hepsin−/− mice, i.e., it effectively diminished the severity of these symptoms to levels comparable to those observed in wild-type mice. These results established a crucial connection between the serine protease function of hepsin and protection against APAP-induced hepatotoxicity.

Transcriptome analysis of hepsin−/− and wild-type mice during the early period after APAP exposure

To clarify the function of hepsin in the early stages of APAP-induced hepatotoxicity, a transcriptome analysis employing RNA sequencing was conducted within the first 2 h post-APAP administration. Our results demonstrated a significant downregulation of the PI3K/AKT pathway (p = 0.049) at one hour, and of mTOR (p = 0.033 at one hour and p = 0.021 at two hours) following APAP treatment in wild-type mice (Fig. S4). These findings align with previous reports indicating that downstream pathways, particularly PI3K/AKT and mTOR, are responsive to hepsin downregulation (Fig. S2) (Li et al. 2020b). Furthermore, the analysis indicated that, at 1 and 2 h following APAP administration, hepsin−/− mice displayed 156 and 141 differentially expressed genes, respectively, when compared with wild-type mice at the same time points. Notably, 17 of these genes were differentially expressed at both 1 and 2 h, underscoring their significance in the early stage after APAP administration (Fig. 3A). These 17 genes were further subjected to Gene Ontology analysis, and temporal data are depicted in a heat map. This analysis revealed significant differences in the liver transcriptomes of wild-type and hepsin−/− mice post-APAP exposure. Particularly, genes involved in lipid metabolism (Abcg5), drug metabolism (Cyp2a4), GSH metabolism (Gstm2 and Gstm6), and oxidative phosphorylation (Lhpp) were expressed at higher levels in the liver of hepsin−/− mice compared with wild-type mice. These results further suggest that hepsin−/− mice experienced increased oxidative stress in the early hours following APAP administration, leading to more severe drug-induced liver damage (Fig. 3B).

Fig. 3

Liver transcriptomes after 1 and 2 h after 400 mg/kg APAP treatment. (A) Comparison between transcriptomes of hepsin−/− and wild-type mice. The Venn diagram shows the differential expression of transcripts between these two mouse groups assessed at 1 and 2 h after 400 mg/kg APAP treatment. The overlap of the differentially expressed transcripts is shown. (B) Unsupervised clustering analysis of the 17 differentially expressed transcripts that were common between the 1- and 2- hour time points after 400 mg/kg APAP treatment. (C) Functional enrichment analysis of differentially expressed transcripts at 1 and 2 h after 400 mg/kg APAP treatment. (D) Results from a gene set enrichment analysis for hepsin−/− and wild-type mice at 1 h after 400 mg/kg APAP treatment. APAP is indicated as AP. Each group consisted of three mice

Next, a functional enrichment analysis was conducted on the genes that were differentially expressed at 1 and 2 h post-APAP administration in hepsin−/− mice. The analysis of data for both time points revealed significant upregulation of genes in various pathways, especially those related to general metabolism, GSH metabolism, and oxidative stress, such as oxidative phosphorylation and ROS, in hepsin−/− mice compared with wild-type mice (Fig. 3C). Gene set enrichment analysis highlighted the early activation of oxidative stress-related pathways, including oxidative phosphorylation and ROS generation, as early as 1-h post-APAP exposure. This upregulation aligned with Gene Ontology analysis data and our findings on the early stages of APAP-induced hepatotoxicity in hepsin−/− mice (Fig. 3D).

Notably, the results from our functional enrichment analysis highlighted significant distinctions in the GJ pathway transcriptome between hepsin−/− and wild-type mice at 1-h post-APAP administration (Fig. 3C, Fig. S5). This phenomenon was especially noteworthy given the body of research linking the expression of Cx32, which is an essential structural protein in hepatocyte GJs, with the underlying mechanisms of liver toxicity induced by APAP. Previous research has suggested that hepatic GJ proteins might facilitate the spread of free radicals, thereby exacerbating the severity of drug-induced liver damage (Patel et al. 2012). The notable increase in Cx32 level in hepatocytes of hepsin−/− mice, along with our previous work establishing hepsin as a regulator of the cellular abundance of hepatic GJ proteins (Hsu et al. 2012), led us to propose that hepsin could significantly diminish the severity of APAP-induced hepatotoxicity by regulating the abundance of hepatic GJ proteins, which may alter the distribution of free radicals between neighboring hepatocytes, thereby potentially increasing oxidative stress.

Inhibiting the elevated expression of GJ components in hepsin−/− mice reduces APAP toxicity by limiting oxidative stress in the liver

To investigate GJ expression variations after APAP administration in wild-type and hepsin−/− mice, immunofluorescence staining for Cx32 was assessed at different time points. In wild-type mice, Cx32 abundance in hepatocytes decreased substantially as early as 1-h post-APAP treatment, reaching a 20% reduction compared to the steady-state or control group. This decrease in GJ expression was consistently evident at later time points, 2 to 6 h post-APAP exposure (Fig. 4A, Fig. S6). Considering that hepatocyte GJs aid the transmission of oxidative stress molecules in mice (Igarashi et al. 2014; Patel et al. 2012), our observation that APAP decreased GJ abundance suggested that hepatocytes mount a defense against the spread of oxidative stress molecules within liver tissue during the early stages of APAP exposure by reducing the expression of Cx32 (and consequently GJs).

Fig. 4

Hepsin−/− mice exhibit delayed downregulation of connexin 32 (Cx32) after injection with APAP, and blockade of GJ intercellular communication formed by Cx32 alleviates liver injury and decreases the diffusion of reactive oxygen species. (A) Representative immunofluorescence photomicrographs of mouse liver sections stained for Cx32, quantified as relative fluorescence intensity (RFI) at the indicated time points after 400 mg/kg APAP treatment. NS, normal saline control group (n = 4–8 per group). (B) An incision loading/dye transfer test was conducted to evaluate functional GJ intercellular communication in liver tissue 3 h after injection with 1 mg/kg 2APB. Representative immunofluorescence photomicrographs were used to quantify the permeability of Lucifer yellow relative to rhodamine-dextran, representing GJ connectivity between hepatocytes. Veh, vehicle (0.1% DMSO) (n = 3–4 per group). (C) Representative immunofluorescence photomicrographs of mouse liver sections stained with dihydroethidium (DHE) show the amount of ROS 6 h after injection of mice with 500 mg/kg thioacetamide (TAA), along with 1 mg/kg 2APB or vehicle control (Veh) (n = 4–8 per group). (D) Experimental timeline. (E, F) Survival rate of wild-type and hepsin−/− mice treated with 20 mg/kg or 1 mg/kg 2APB, followed by administration with 500 mg/kg APAP. In the survival analysis, sample sizes for each group are indicated in brackets, and statistical significance was determined using the log-rank test, with significance levels represented by asterisks: ***p < 0.001, ****p < 0.0001. Data are presented as the mean ± SD in bar charts, with significance levels denoted by asterisks: *p < 0.05, **p < 0.01

Compared with the natural protective mechanism of wild-type mice, hepsin−/− mice had a different response to Cx32 expression after APAP exposure. In the normal saline control group, hepsin−/− mice showed a 1.5-fold increase in hepatocyte Cx32 expression relative to the wild-type control group. After APAP administration, however, the decrease in GJ expression in hepatocytes of hepsin−/− mice was more gradual compared with that in wild-type mice. Consequently, there was an approximate threefold increase in Cx32 expression in wild-type mice observed at 1-h post-APAP treatment. In contrast, a significant reduction in Cx32 expression in hepsin−/− mice was not evident until 2 h post-APAP administration (Fig. 4A, Fig. S6). Therefore, we hypothesized that the excessive expression of GJs post-APAP exposure in hepsin−/− mice could be a key factor contributing to the increased severity of liver damage observed following APAP administration.

To further substantiate that inhibiting GJs is an effective means of protecting against APAP-induced hepatotoxicity, we utilized the GJ functional inhibitor 2-aminoethoxydipenyl borate (2APB) (Leytus et al. 1988). To confirm the inhibitory effects of 2APB, an incision loading/dye transfer experiment was conducted. Gap junctional transmission was substantially greater in hepsin−/− mice, likely due to the elevated expression of Cx32; however, a 3-h treatment with 2APB led to a substantial decrease in GJ transmission in both wild-type and hepsin−/− mice (Fig. 4B). Given the characteristics of short-lived free radical species, which are challenging to confirm as transferring between cells through gap junctions after APAP overdose, we employed a thioacetamide-induced hepatotoxicity model. This model, known for producing a higher amount of ROS, was used to assess oxidative stress in mouse liver using the ROS probe dihydroethidium, as demonstrated in previous studies (Patel et al. 2012). A 6-h treatment with thioacetamide led to a significant increase in ROS levels compared with the control group, but pre-treatment with 2APB effectively mitigated this oxidative stress in the liver of both wild-type and hepsin−/− mice. These results indicated that the administration of 2APB effectively inhibited the transmission of oxidative stress molecules by hepatocyte GJs and impacted oxidative stress in the liver. Notably, the hepsin−/− mice exhibited a more pronounced increase in ROS levels after thioacetamide treatment compared with wild-type mice, which aligned with our hypothesis that hepsin−/− mice experience heightened ROS stress in the early stages of APAP exposure due to their elevated expression of GJ proteins (Fig. 4C).

A subsequent experiment was conducted to explore the correlation between the observed overexpression of GJs and the ensuing heightened liver damage in hepsin−/− mice following APAP administration. Both wild-type and hepsin−/− mice received an intraperitoneal injection of 2APB (20 mg/kg), followed by a dose of APAP (500 mg/kg) 2 h later (Fig. 4D). For both the wild-type and hepsin−/− mice treated with 2APB, the survival rate increased significantly compared with the vehicle control group at 72 h post-APAP administration (Fig. 4E). This indicated that the functional inhibition of GJ proteins not only protects against APAP-induced hepatotoxicity but also effectively mitigates the increased sensitivity to APAP as seen in hepsin−/− mice. However, when administering a lower dose of 2APB, i.e., 1 mg/kg, the protective effect on APAP tolerance was observed only in wild-type mice, for which survival rate was 100% at 60 h post-2APB administration; in contrast, the survival rate was only 8% for hepsin−/− mice. At 80 h post-administration, the survival rate was 90% for wild-type mice but dropped to 0% for hepsin−/− mice (Fig. 4F). This difference was likely attributable to the insufficient inhibition of excessive GJ proteins in hepsin−/− mice in response to the lower 2APB dose of 1 mg/kg, lending further support to the hypothesis that hepsin may help reduce APAP toxicity by regulating the expression of hepatic GJ proteins. Collectively, these results affirmed that hepsin plays a crucial role in regulating GJ communication, which in turn impacts oxidative stress. This regulation by hepsin contributed to early protection against APAP-induced liver damage.

Administering hepsin to wild-type mice increases APAP tolerance and downregulates GJ expression

Considering the potential clinical and medical applications, we evaluated a therapeutic strategy involving the administration of hepsin to wild-type adult mice and assessed its potential to mitigate APAP-induced liver injury. Three weeks after transduction with AAV, the serum levels and liver lysate of human hepsin were confirmed and quantified via ELISA and western blot following the administration of the same dose of AAV-hHPNWT, AAV-hHPNRS or AAV-EGFP (Fig. 5A). Additionally, in wild-type mice that overexpressed hHPNWT, there was a significant reduction in Cx32 expression, i.e., approximately 50 ± 9% compared with both the AAV-hHPNRS and the vector-control (AAV-EGFP) groups (Fig. 5B). These findings suggest that administering hepsin can effectively regulate GJ expression in wild-type mice, highlighting its potential as a therapeutic target to protect against APAP-induced hepatotoxicity, potentially in human patients.

Fig. 5

AAV-mediated liver-specific administration of hepsin to adult wild-type mice alleviates APAP-induced liver injury. Adult wild-type mice were administered AAV-hHPNWT, AAV-hHPNRS or AAV-EGFP for 3 weeks and then treated with 600 mg/kg APAP. (A) Human hepsin (hHPN) levels in serum and liver lysate prior to administering APAP (n > 15 per group for serum level detection and n = 3 per group for liver lysate detection). (B) Representative immunofluorescence photomicrographs of mouse liver sections stained for Cx32 after liver-specific administration of hepsin by AAV. Quantification is represented by relative fluorescence intensity (RFI) (n = 4–7 per group). (C) Experimental timeline. (D) Measurement of serum AST and ALT levels at 10 h after injection with 600 mg/kg APAP (n = 5 per group). (E) Images of liver pathology in sections assessed by hematoxylin and eosin staining at 10 h after 600 mg/kg APAP treatment, show the degenerated area indicated by a yellow dashed line. (F) Survival rate. In the survival analysis, sample sizes for each group are indicated in brackets, and statistical significance was determined using the log-rank test, with significance levels represented by asterisks: ****p < 0.0001. Data are presented as the mean ± SD in bar charts, with significance levels denoted by asterisks: *p < 0.05, **p < 0.01

To assess the ability of hepsin overexpression to protect against APAP-induced hepatotoxicity, a lethal dose of 600 mg/kg APAP was first administered to mice in each experimental group (Fig. 5C). Serum aspartate aminotransferase and alanine transaminase levels were measured at 10 h post-APAP administration. The group of wild-type mice expressing hHPNWT exhibited significantly lower levels of both enzymes, which reflected reduced liver damage, compared to the two control groups (Fig. 5D). Consistently, there was a marked reduction in liver degeneration area at the same time point in wild-type mice expressing hHPNWT, compared to the two control groups (Fig. 5E). Additionally, a survival-rate analysis revealed a significant increase in APAP tolerance in mice expressing hHPNWT, with more than 80% surviving at the 80-h post-administration time point. In contrast, the survival of mice expressing a hHPNRS was similar to that of the vector-control group (AAV-EGFP), with all mice succumbing within approximately 60 h (Fig. 5F). These results demonstrated that administering hepsin afforded protection against APAP-induced liver damage in adult wild-type mice and ultimately prolonged their survival. This protective effect is achieved through the downregulation of GJ expression, suggesting that hepsin could be a viable therapeutic agent for APAP-induced hepatotoxicity.

Novel combination therapy with hepsin and low doses of NAC improves therapeutic effectiveness and extends survival in APAP-induced lethality

The standard clinical approach for treating APAP-induced liver injury involves administering NAC. Given that the effectiveness of NAC for treating APAP-induced liver injury is confined to modest improvements within a brief period after onset and considering that high doses of NAC can potentially negatively impact hepatocyte metabolism and liver regeneration (Jaeschke et al. 2020), we pursued a combination therapy strategy involving the administration of hepsin, aiming to amplify the therapeutic efficacy while employing lower doses of NAC. We first established a therapy model of NAC in our animal model. In an experimental model using a lethal dose of 600 mg/kg APAP in adult wild-type mice, administering a high dose of 300 mg/kg NAC at 1-h post-APAP exposure resulted in significant therapeutic effectiveness, with 100% survival. In contrast, mice in the control group died within 50 h. Moreover, this high dose of NAC continued to show therapeutic benefits for nearly 130 h post-APAP exposure, with a survival rate of 50–60%. In contrast, administering a lower dose of 200 mg/kg NAC was insufficient to prevent mortality (Fig. 6A).

Fig. 6

Combination of AAV-mediated liver-specific administration of hepsin and post-injection with N-acetyl-cysteine has a superior therapeutic effect compared with N-acetyl-cysteine treatment alone. (A) The survival rate after 600 mg/kg APAP administration, followed by different doses of NAC therapy provided 1 h post-APAP treatment. (B) Experimental timeline. Adult wild-type mice were administered AAV-hHPNWT, AAV-hHPNRS, or AAV-EGFP for 3 weeks, followed by 600 mg/kg APAP treatment for 1 h, after which the mice were treated with NAC. (C) Measurement of serum AST and ALT at 20 h after APAP treatment (n = 3–5 per group). (D) Images of liver pathology in sections assessed by hematoxylin and eosin staining at 20 h after APAP treatment. (E) Survival rate. In the survival analysis, sample sizes for each group are indicated in brackets, and statistical significance was determined using the log-rank test, with significance levels represented by asterisks: *p < 0.05, **p < 0.01, ***p < 0.01. Data are presented as the mean ± SD in bar charts, with significance levels denoted by asterisks: **p < 0.01

To assess the therapeutic effectiveness of combining hepsin with a reduced dose of 200 mg/kg NAC (Fig. 6B), we measured serum AST and ALT levels in the early stage post-APAP administration to evaluate protection against 600 mg/kg APAP-induced hepatotoxicity. The group of wild-type mice treated with hepsin and a reduced dose of 200 mg/kg NAC exhibited significantly lower levels of both enzymes, indicating reduced liver damage compared to the other groups (Fig. 6C). This finding was consistent with the marked reduction in liver degeneration observed in the combination therapy group at the same time point (Fig. 6D). Additionally, survival-rate analysis revealed a notable increase in therapeutic effectiveness, with 100% survival for wild-type mice over a 100-h period following a lethal dose of APAP when treated with the combination of hepsin and 200 mg/kg NAC. This level of improvement was not observed in the combination therapy groups administered with AAV-hHPNRS or the vector control (AAV-EGFP) (Fig. 6E). Thus, the combined administration of hepsin with NAC optimized the therapeutic outcome, enhancing effectiveness beyond what was achievable with NAC alone. This result highlights the potential of combined therapy with hepsin and NAC to treat APAP-induced liver injury, paving the way for promising future research.