This study purchased 208 male SD rats (150–180 g; 5–6 weeks old; SPF grade) from Beijing HuaFuKang Bioscience Company (Beijing, China). The animal protocol was approved by the Institutional Animal Care and Use Committee of Kunming Medical University (Kunming, China; KMMU2020189). All animal experiments were performed in compliance with the National Institute of Health guidelines for animal experimentation.

The SD rat LT model was established using the “two-cuff” technique reported by Kamada et al. [16], which has been detailed in the Supplementary Materials. The study has been reported per the ARRIVE guidelines [17].

To explore the transfection efficiency of adeno-associated virus (AAV), 25 SD rats were randomly allocated into five groups: shRNA-NC, shRNA-HO-1, AAV-NC, AAV-HO-1, or normal groups. Rats in all groups were injected with 200 µL of AAV titer (2.5 $\times$ 10${}^9$ mg/µL; Hanbio Biotechnology, Shanghai, China) via the tail vein. The AAV-HO-1, shRNA-HO-1, and the respective blank control virus vectors were transduced into SD rats for 21 days.

Of the 208 rats, 88 were injected with AAV via the tail vein and were randomly divided into the following six groups: sham (n = 8, only laparotomy); DCD (n = 16, donors were injected with normal saline); DCD + shRNA-NC (n = 16, donors underwent transduction with HO-1-shRNA empty-loaded virus); DCD + shRNA-HO-1 (n = 16, donors underwent transduction with HO-1-shRNA-AAV); DCD + AAV-NC (n = 16, donors underwent transduction with HO-1-overexpressing empty-loaded virus); or DCD + AAV-HO-1 (n = 16, donors underwent transduction with HO-1-overexpressing AAV).

In addition, to explore whether HO-1 induces pyroptosis in hepatic IRI and whether this effect is observed in a time-dependent manner, 54 rats were randomly allocated into five groups: normal group (n = 6) and hepatic ischemia-reperfusion (IR) groups with reperfusion for 6 h (DCD + IR6h), 24 h (DCD + IR24h), 72 h (DCD + IR72h), or 168 h (DCD + IR168h) (each, n = 12).

Furthermore, 66 SD rats were randomly divided into the following six groups for assessing HO-1-related cell pyroptosis at 6 h (n = 12, i.e., 6 pairs). The groups were as follows: sham (n = 6, only underwent laparotomy), DCD (n = 12, the donor was injected with normal saline via the tail vein), DCD + shRNA-NC (n = 12, donor underwent transduction with HO-1 shRNA AAV via the tail vein); DCD + shRNA-HO-1 (n = 12, donor underwent transduction with HO-1 shRNA AAV via the tail vein); DCD + AAV-NC (n = 12, donor underwent transduction with HO-1-overexpressing AAV empty virus via the tail vein); and DCD + AAV-HO-1 (n = 12, donor underwent transduction with HO-1-overexpressing AAV via the tail vein).

The serum alanine transaminase (ALT) and aspartate aminotransferase (AST) levels in SD rats were measured using a fully automatic biochemical analyzer (BIO-RAD, Hercules, CA, USA), expressed in U/L. In addition, serum IL-1$\beta$ and IL-18 levels in blood from the coronary artery (ERC007.96, ERC010.96, Neobioscience Technology Company, Shenzhen, China) were assessed using a commercially available enzyme-linked immunosorbent assay kit, according to the manufacturer’s instructions. Optical density was measured at 450 nm.

Initially, 50 µg of fresh SD rat liver tissue was weighed and then homogenized and filtered into a cell suspension. This suspension was centrifuged, and then the obtained pellet was mixed with an appropriate volume of DCFH-DA (S0033S, Beyotime Biotechnology, Shanghai, China), which was diluted with serum-free culture medium (1:1000) to make a final concentration of 10 µmol/L. Subsequently, the medium was incubated at 37 °C for 20 min. The cells were then washed thrice with a serum-free cell culture medium to remove the DCFH-DA that had not entered the cells. The ROS-positive control group was treated with Rosup for 20–30 min. Then, flow cytometry (excitation wavelength: 488 nm, emission wavelength: 525 nm) was used to detect the fluorescence intensity before and after stimulation.

For histological analyses, the liver tissues of rats were fixed with 4% paraformaldehyde for 48 h. The fixed liver samples were then dehydrated, embedded in paraffin, and cut into 3 mm thick sections (n = 3 for each liver). The sections were stained with hematoxylin-eosin (HE) for pathological evaluation.

In addition, fresh liver tissues were embedded in optimum cutting temperature compound, frozen, sectioned, and observed under an inverted fluorescence microscope.

For immunohistochemical analyses, liver tissues were embedded in paraffin and sliced, as mentioned above. The sections obtained were incubated at 65 ℃ for 20 min and were deparaffinized using a gradient of xylene and ethanol solution, followed by treatment with 0.01 mol/L (pH 6.0) citrate buffer for antigenic repair. Subsequently, the sections were treated with 3% hydrogen peroxide to quench endogenous peroxidase activity and, after 15 min, were blocked with 5% bovine serum albumin for 30 min. The sections were then incubated with NLRP3 antibody (dilution ratio: 1:200; 19771-1-AP, ProteinTech, Wuhan, Hubei, China) overnight at 4 °C. Following this, the sections were treated with secondary antibody for 50 min and incubated with DAB. The sections obtained were observed under an optical microscope (Olympus, cymml-3ptxcj-004, Hamburg, Germany). All images shown in the results represent at least three images of each liver. The Liver Suzuki scoring standard was used [18, 19]. ImageJ software (Rasband WS, ImageJ, V 1.8.0, National Institutes of Health, Bethesda, MD, USA) was used to quantify the stained areas in sections.

As reported previously [20], the membrane was blocked with 5% skim milk for 2 h and then incubated with rabbit anti-HO-1 (dilution ratio 1:1000; E6Z5G, Cell Signaling Technology, Danvers, MA, USA), anti-caspase-1 (dilution ratio 1:1000; EPR19672, Abcam, Cambridge, UK), anti-GSDMD (dilution ratio 1:1000; EPR20859, Abcam), and NLRP3 (dilution ratio 1:1000; 19771-1-AP, Proteintech) overnight at 4 °C. After washing thrice with Tris-Buffered Salineand Tween 20 (TBST), the membrane was incubated with IgG antibody at room temperature for 2 h. The membrane was completely covered with the developing liquid and observed using the gel imaging system (Monad, GD50202, Suzhou, China).

RT-qPCR was performed as described previously using the SuperReal PreMix Plus (SYBR Green, Roche, Mannheim, Germany) kit and the Mx3005 P Real-Time PCR System (Agilent, Santa Clara, CA, USA) [21]. Briefly, total RNA from liver tissues was extracted using TRIzol (Invitrogen, Waltham, MA, USA). cDNA was extracted using SureScript™ First-Strand cDNA Synthesis Kit (GeneCopoeia, Guangzhou, China). mRNAs amplification was performed using the following specific primers: HO-1 (102 bp) forward 5${}^{\prime }$-CCATCCCTTACACACCAGCC-3${}^{\prime }$, reverse 5${}^{\prime }$-GGTAGCGGGTATATGCGTGG-3${}^{\prime }$; $\beta$-tubulin (152 bp) forward 5${}^{\prime }$-GAGGCAGATGGCAGTGACAG-3${}^{\prime }$, reverse 5${}^{\prime }$-TGGTTGGGGAACACGGAGTA-3${}^{\prime }$ (Sangon Biotech, Shanghai, China).

Quantitative data are presented as mean $\pm$ standard error. Measurement data were presented as $\overline{x}$$\pm$ s. Multiple groups of measurement data were compared using a single-factor analysis of variance and an advanced homogeneity of variance test. In the case of uniform distribution, the least significant difference test was used, and in the case of skewed distribution, Dunnett’s T3 test was used. Survival and statistical data were assessed using Kaplan–Meier analysis. Differences were considered statistically significant when p 0.05. All statistical graphs were created using the GraphPad Prism software (GraphPad Software 9, Inc., San Diego, CA, USA), and all data were analyzed using SPSS 24.0 (IBM Corp., Armonk, NY, USA).

The fluorescence intensity in the liver tissue was assessed 21 days after AAV transduction (Fig. 1A–D). HO-1 transcription levels were significantly lower in the shRNA-HO-1 group than in the shRNA-NC group. The gene transcription level was higher in the AAV-HO-1 group than in the AAV-NC (normal control), shRNA-HO-1, shRNA-NC, and normal groups (Fig. 1E). The fluorescence intensity and HO-1 mRNA transcription levels in the four DCD groups are shown in Table 1 and at four-time points are shown in Table 2.

Fig. 1.

Efficiency of AAV transfection and HO-1 mRNA transcription level in rat liver tissue. (A) AAV2/9-EGFP interference control. 400$\times$. (B) AAV2/9-r-Hmox1 shRNA-EGFP interferes with expression. 400$\times$. (C) AAV2/9-ZsGreen overexpression control. 400$\times$. (D) AAV2/9-CMV-r-Hmox1-3xflag-ZsGreen is overexpressed. 400$\times$. (E) The change in HO-1 gene transcription level mRNA after AAV transfection in SD rats. *, compared with the normal group, the difference is statistically significant, p 0.05; #, compared with the shRNA-HO-1 group, the difference is statistically significant, p 0.05; &, compared with the AAV-NC group, the difference is statistically significant, p 0.05. (F) Analysis of postoperative survival time of SD rats between groups after successful DCD liver transplantation modeling. AAV, adenoassociated virus; HO-1, Heme oxygenase-1; DCD, Donors after circulatory death; shRNA, short hairpin RNA; NC, normal control.

The overall comparison result was $\chi$${}^2$ = 14.449 (degree of freedom $\nu$= 5; p 0.05). In addition, paired comparisons revealed significant survival (p 0.05) among the sham, DCD + shRNA-HO-1, and DCD + AAV-HO-1 groups (Fig. 1F).

The effects of ischemia on the transplanted liver were assessed at different durations of reperfusion (DCD + IR6h, DCD + IR24h, DCD + IR72h, and DCD + IR168h groups). HE staining revealed rapid progression of tissue damage in the DCD + IR6h group, along with obvious necrosis. In the DCD + IR24h group, the liver tissue damage was greater than in the DCD + IR6h group, with extensive necrosis. However, in the DCD + IR72h and DCD + IR168h groups, necrosis was not as obvious as in the DCD + IR24h group, and the degree of damage gradually decreased. Particularly, the normal lobules had been destructured, and the sinus space was still wider than normal. Compared with the normal group, the DCD + IR6h, DCD + IR24h, DCD + IR72h, and DCD + IR168h groups demonstrated significantly greater liver tissue damage (p 0.05). Compared with the DCD + IR6h group, the normal, DCD + IR24h, and DCD + IR168h groups showed significant differences in the degree of injury (p 0.05) (Fig. 2).

Fig. 2.

HE staining and Suzuki score of the donor livers after liver transplantation. (A–E) HE results of liver tissue after LT, (A) normal, (B) DCD + IR6h, (C) DCD + IR24h, (D) DCD + IR72h, and (E) DCD + IR168h. (F) The histogram based on the Suzuki score, *, p 0.05 compared with normal; #, p 0.05 compared with DCD + IR6h. LT, liver transplantation; IR, ischemia-reperfusion.

Over time, the ROS levels in the liver tissues of each group decreased after reperfusion. Compared with the normal group, the DCD + IR6h, DCD + IR24h, DCD + IR72h, and DCD + IR168h groups had significantly lower ROS levels (p 0.05). Compared with the DCD + IR6h group, the normal, DCD + IR72h, and DCD + IR168h groups were significantly different (p 0.05) (Fig. 3).

Fig. 3.

ROS level changes after LT. (A–E) After DCD LT, the donor liver of ischemia was followed by different durations of reperfusion for 6 h, 24 h, 72 h, and 168 h. (F) Flow cytometry was used to detect the expression level of ROS in the liver tissue. *, compared with normal group, p 0.05; #, compared with DCD + IR6h group, p 0.05.

The liver function and cytokine levels would change accordingly after reperfusion. The results revealed that ALT and AST levels increased significantly at 6 and 24 h but decreased at 72 and 168 h. Compared with the normal group, ALT and AST levels were significantly different in the DCD + IR6h, DCD + IR24h, DCD + IR72h, and DCD + IR168h groups (p 0.05). The normal, DCD + IR72h, and DCD + IR168h groups differed significantly from the DCD + IR6h group (p 0.05) (Fig. 4A,B).

Fig. 4.

Liver function and protein changes after LT. (A) Biochemical analysis results of ALT. (B) Biochemical analysis results of AST. (C) The level of IL-1$\beta$ in serum. (D) The level of IL-18 in serum. (E) Western blot method to detect the protein expression of HO-1, Caspase-1, p22, full-GSDMD, and cleaved-N-GSDMD in each group. (F–H) the statistical results of HO-1/$\beta$-tubulin, p22/$\beta$-tubulin, and cleaved-N-GSDMD/$\beta$-tubulin in each group, respectively. *, compared with normal, p 0.05; #, compared with DCD + IR6h, p 0.05.

Over time, IL-1$\beta$ and IL-18 levels in the liver tissues increased and then decreased, which was consistent. Compared with those in the normal group, cytokine levels were significantly different in the DCD + IR6h, DCD + IR24h, DCD + IR72h, and DCD + IR168h groups (p 0.05). Moreover, the normal, DCD + IR72h, and DCD + IR168h groups differed significantly from the DCD + IR6h group (p 0.05) (Fig. 4C,D).

Over time, proteins showed a trend with the prolongation of the postoperative reperfusion time after LT. Results of western blotting revealed that (Fig. 4E) HO-1 expression was significantly increased in the DCD + IR6h group, but it decreased in the other long-term groups. Moreover, HO-1 expression in the DCD + IR6h group was significantly different from the normal group, DCD + IR72h, and DCD + IR168h groups (p 0.05). However, it was not significantly different from the DCD + IR24h group. In addition, compared with the normal group, the DCD + IR6h, DCD + IR24h, and DCD + IR72h groups had significantly different HO-1 expression (p 0.05) (Fig. 4F).

The p22 of pro-caspase1 was significantly increased in the short term but gradually decreased over the long term. Compared with the normal group, p22 expression in the DCD + IR6h, DCD + IR24h, and DCD + IR72h groups was significantly different (p 0.05). Although p22 expression in the DCD + IR168h group was higher than the normal group, the difference was not statistically significant. Compared with the DCD + IR6h group, p22 expression in the normal, DCD + IR72h, and DCD + IR168h groups was statistically different (p 0.05) (Fig. 4G).

The expression of cleaved-N-GSDMD was the highest in the short-term groups but decreased in the long-term groups. Compared with the normal group, the expression of cleaved-N-GSDMD in the DCD + IR6h, DCD + IR24h, and DCD + IR72h groups were significantly different (p 0.05). Although the expression in the DCD + IR168h group was higher than in the normal group, it was not statistically significant. Compared with the DCD + IR6h group, the normal, DCD + IR72h, and DCD + IR168h had significantly different expression levels (p 0.05) (Fig. 4H).

Immunohistochemistry revealed that NLRP3 was not expressed in the normal group. Notably, the DCD + IR6h and DCD + IR24h groups demonstrated extensive necrosis. However, NLRP3 expression in the DCD + IR6h group was the lowest among the other groups. Over time, NLRP3 expression increased in the short term but decreased over the long term. Compared with the DCD + IR6h group, the normal group, DCD + IR24h, DCD + IR72h, and DCD + IR168h groups differed significantly in terms of NLRP3 expression (p 0.05) (Fig. 5).

Fig. 5.

Change in NLRP3 after LT. (A–E) Immunohistochemistry of NLRP3 results of liver tissue after LT. (A) Normal, (B) DCD + IR6h, (C) DCD + IR24h, (D) DCD + IR72h and (E) DCD + IR168h. (F) The expression level of NLRP3 in the liver tissue. *, compared with normal, p 0.05; #, compared with DCD + IR6h, p 0.05.

Based on previous experiments, the time point of reperfusion was chosen as 6 h in this study. After 21 days of AAV pretreatment, the donor’s liver was reperfused for 6 h after DCD LT. Compared with the control group, the DCD + shRNA-HO-1 group exhibited greater damage in the liver tissue, with larger areas of necrosis, severe congestion, and sinus space. The necrotic area in the DCD + AAV-HO-1 group was significantly reduced, the hepatic cord was intact, the sinus space was approximately the same as the sham group, and the degree of congestion was significantly lower. Compared with the DCD group, the sham, DCD + shRNA-HO-1, and DCD + AAV-HO-1 groups showed statistically significant differences in the degree of damage (p 0.05). The damage was not significantly different in the DCD + shRNA-NC and DCD + AAV-NC groups. Moreover, compared with the degree of damage in the DCD + shRNA-HO-1 group, the DCD, DCD + shRNA-NC, DCD + AAV-NC, and DCD + AAV-HO-1 groups was statistically significant (p 0.05). In addition, the DCD + AAV-NC and DCD + AAV-HO-1 groups differed significantly in the degree of damage (p 0.05) (Fig. 6).

Fig. 6.

HE staining results and Suzuki score of different groups after AAV pretreatment. (A–F) Postoperative liver tissue HE staining results. (A) Sham, (B) DCD, (C) DCD + shRNA-NC, (D) DCD + shRNA-HO-1, (E) DCD + AAV-NC, and (F) DCD + AAV-HO-1. (G) A histogram based on the Suzuki score. *, compared with the DCD group, p 0.05; #, compared with the DCD + shRNA-HO-1 group, p 0.05; &, compared with the DCD + AAV-NC group, p 0.05.

The hepatic ROS level in the DCD + shRNA-HO-1 group was significantly higher than that in other groups but was significantly lower than the DCD + AAV-HO-1 group. The ROS level differed significantly between the DCD + shRNA-HO-1 group and the DCD, DCD + shRNA-NC, DCD + AAV-NC, and DCD + AAV-HO-1 groups (p 0.05). Similarly, the DCD + AAV-NC and DCD + AAV-HO-1 groups had significantly different ROS levels (p 0.05) (Fig. 7).

Fig. 7.

After AAV pretreatment, the donor’s liver was reperfused for 6 h after DCD LT and the level of ROS was measured in liver tissues. *, compared with DCD group, p 0.05; #, compared with DCD + shRNA-HO-1 group, p 0.05; &, compared with DCD + AAV-NC group, p 0.05. (A–F) The result of ROS level in flow cytometry. (A) Sham; (B) DCD; (C) DCD + shRNA-NC; (D) DCD + shRNA-HO-1; (E) DCD + AAV-NC, (F) DCD + AAV-HO-1 groups. (G) The bar chart shown the percentage of ROS cells in each group.

Following DCD LT, the donor liver was pretreated with AAV for 21 days and was then reperfused for 6 h. Following this, serum ALT and AST levels in the DCD + shRNA-HO-1 group increased significantly. However, these levels in the DCD + AAV-HO-1 group showed a downward trend compared with those in the control, but they were still higher than those in the sham group.

Compared with the DCD + shRNA-HO-1 group, the DCD, DCD + shRNA-NC, DCD + AAV-NC, and DCD + AAV-HO-1 groups had significantly different ALT levels (p 0.05). Moreover, the ALT level differed significantly between the DCD + AAV-NC and DCD + AAV-HO-1 groups (p 0.05) (Fig. 8A).

Fig. 8.

With AAV pretreatment, the donor liver was reperfused for 6 hours after DCD LT. (A) Biochemical analysis results of ALT. (B) Biochemical analysis results of AST. (C) The expression level of IL-1$\beta$ in serum. (D) The expression level of IL-18 in serum. (E) The mRNA level of HO-1 in the liver tissue. (F) Western blot method to detect the protein expression of HO-1, Caspase-1, p22, full-GSDMD, and cleaved-N-GSDMD in each group. (G) Statistical results of (G) HO-1/$\beta$-tubulin, (H) p22/$\beta$-tubulin, and (I) cleaved-N-GSDMD/$\beta$-tubulin in each group. *, compared with DCD group, p 0.05; #, compared with DCD + shRNA-HO-1 group, p 0.05; &, compared with DCD + AAV-NC group, p 0.05.

Compared with the DCD + shRNA-HO-1 group, the DCD, DCD + shRNA-NC, DCD + AAV-NC, and DCD + AAV-HO-1 groups had significantly different AST levels (p 0.05). Moreover, the AST level differed significantly between the DCD + AAV-NC and DCD + AAV-HO-1 levels (p 0.05) (Fig. 8B).

The hepatic levels of IL-1$\beta$ and IL-18 in the DCD + shRNA-HO-1 group were higher than in the other groups. However, in the DCD + AAV-HO-1 group, the hepatic IL-1$\beta$ and IL-18 levels showed a downward trend compared with the control group, were significantly lower than those in the DCD + shRNA-HO-1 group, and were slightly higher than those in the sham group. Moreover, the levels of IL-1$\beta$ and IL-18 in each group correlated positively.

Compared with the DCD + shRNA-HO-1 group, the IL-1$\beta$ level differed significantly in the DCD, DCD + shRNA-NC, DCD + AAV-NC, and DCD + AAV-HO-1 groups (p 0.05). Compared with the DCD + AAV-NC group, the IL-1$\beta$ level in the sham and DCD + AAV-HO-1 groups was significantly different (p 0.05) (Fig. 8C).

Compared with the IL-18 level in the DCD + shRNA-HO-1 group, the DCD, DCD + shRNA-NC, DCD + AAV-NC, and DCD + AAV-HO-1 groups was significantly different (p 0.05). Compared with the DCD + AAV-NC group, the sham and DCD + AAV-HO-1 groups had significantly different IL-18 levels (p 0.05) (Fig. 8D).

The HO-1 mRNA level in the transplanted liver in each group was in line with our expectations. The mRNA transcription level of the HO-1 interference expression group was significantly lower than that of the control but was slightly higher than that of the sham group. In contrast, the mRNA transcription level of the DCD + AAV-HO-1 was significantly higher. The mRNA transcription levels in the DCD, DCD + shRNA-NC, DCD + shRNA-HO-1, DCD + AAV-NC, and DCD + AAV-HO-1 groups were significantly different (p 0.05). Moreover, at the protein level, the trend in HO-1 expression was consistent with that at the transcription level. HO-1 expression was significantly different between the DCD + shRNA-HO-1 group and all other groups except for DCD (p 0.05). At the same time, HO-1 expression was significantly different between the DCD + AAV-HO-1 and DCD + AAV-NC groups (p 0.05) (Fig. 8E–G).

The expression of p22 of pro-caspase1 was increased significantly in the DCD + shRNA-HO-1 group but was decreased in the DCD + AAV-HO-1 group. In addition, the expression was significantly different in the DCD + shRNA-HO-1 group compared with the DCD, DCD + shRNA-NC, DCD + AAV-NC, and DCD + AAV-HO-1 groups (p 0.05). Moreover, the DCD + AAV-NC and DCD + AAV-HO-1 groups differed significantly in terms of p22 expression (p 0.05) (Fig. 8H).

The trend in the expression of spliced cleaved-N-GSDMD of full-GSDMD was similar to that of p22. The expression of spliced cleaved-N-GSDMD was increased significantly in the DCD + shRNA-HO-1 group and was decreased significantly in the DCD + AAV-HO-1 group. Compared with spliced cleaved-N-GSDMD expression in the DCD + shRNA-HO-1 group, the DCD, DCD + shRNA-NC, DCD + AAV-NC, and DCD + AAV-HO-1 groups was significantly different (p 0.05) (Fig. 8I).

In addition, treatment altering the expression of HO-1 immediately affected the NLRP3 level, which in turn affected the level and activation of caspase-1 and GSDMD.

Immunohistochemical analysis revealed that NLRP3 expression in the DCD + shRNA-HO-1 group was significantly decreased; the group also had large areas of tissue necrosis. The expression differed significantly among DCD, DCD + shRNA-NC, and DCD + AAV-NC groups (p 0.05). Compared with the DCD + AAV-NC group, the DCD + AAV-HO-1 group had significantly lower NLRP3 expression (p 0.05) (Fig. 9).

Fig. 9.

Following AAV pretreatment, the donor liver was reperfused for 6 hours after DCD LT, and the expression of NLRP3 was observed. (A–F) Immunohistochemistry results of NLRP3 coloration in each group (200$\times$). (A) Sham, (B) DCD, (C) DCD + shRNA-NC, (D) DCD + shRNA-HO-1, (E) DCD + AAV-NC, and (F) DCD + AAV-HO-1. (G) The statistical results of the relative expression levels of NLRP3 in each group relative to the sham. *, compared with DCD group, p 0.05; #, compared with DCD + shRNA-HO-1 group, p 0.05; &, compared with DCD + AAV-NC group, p 0.05.