The animal study was approved by the Institutional Animal Care and Use Committee (approval no. IACUC 201903-138) of Ningbo University (Zhejiang, China) in March 2019. The experiment was performed in January 2020. A total of 70 male ICR mice (age, 6–8 weeks; weight, 30±2 g) were purchased from Shanghai SLAC Laboratory Animal Co., Ltd. The mice were housed at the Animal Center of Ningbo University and maintained under 12-h light/dark cycle at 24°C, with a relative humidity of 50–70%. The mice were provided with free access to commercial rodent chow and pure water. They were cared for in accordance with the principles of the Guide for Care and Use of Experimental Animals issued by Ningbo University.

The mice were randomly divided into seven groups as follows: The control group (control; n=10), the rAAV8-treated groups by IV injection (the rAAV8-TBG410-EGFP group, the rAAV8-TBG669-EGFP group and the rAAV8-CAG-EGFP group; n=10 per group) and the rAAV8-treated groups by IP injection (the rAAV8-TBG410-EGFP group, the rAAV8-TBG669-EGFP group and the rAAV8-CAG-EGFP group; n=10 per group). The vector constructs, rAAV-TBG410-EGFP, rAAV-TBG669-EGFP and rAAV-CAG-EGFP, encoding EGFP were designed and purchased from Guangzhou PackGene Biotech Co. Ltd. The viral particles were diluted in PBS (Thermo Fisher Scientific, Inc.) at 1×10¹² genome copies (GC)/ml immediately prior to injection and at a total of 1×10¹¹ GC in 100 µl PBS was administered to the mice in the aforementioned groups via either IV or IP injection. The infusion time for each mouse was ~30 sec. The mice in the control group were left untreated.

After the administration of rAAV8, the mental state, activity, eating, hair state of the mice was observed every day. When there were significant changes in the mental state, behavior, sharp decrease in activity, sparse hair, the mice would be euthanized to avoid greater pain. In the process of the experiment, no mice showed the aforementioned symptoms and none were found dead. After 4 weeks, blood was collected from all mice by retroorbital bleeding. Then, all of the mice were euthanized using CO₂ inhalation at a low flow rate (20% of the volume of the cage per minute), with ventilation maintained for 1–2 min. The mice were confirmed dead when no breathing, no corneal reflexes and body stiffness were examined. The blood samples were centrifuged at 670 × g for 20 min at 4°C and the serum was then stored at −80°C until further analysis. A section of the freshly isolated liver tissue was cut and immediately fixed with 4% paraformaldehyde (PFA) solution (Shanghai Guoyao Reagent Co. Ltd.) at 4°C for 24 h. The remaining liver tissues were kept at −80°C for future analysis.

Since only 13 of the 15 instrument lanes were available for western blot analysis, the livers of four mice in each group were randomly selected to compare the efficiency between IV and IP administration. When the transgene expression efficiency was compared among the three promoters in the IV and IP administration groups, three of the aforementioned four mouse livers in each group were used for western blotting. Total protein was extracted from 20 mg frozen liver tissues, stored at −80°C, using RIPA lysis buffer, supplemented with 1% protease inhibitors (both from Beijing Solarbio Science and Technology Co., Ltd.). The samples were then adequately homogenized at 960 × g for 30 sec at room temperature using a MagNA Lyser instrument (Roche Diagnostics). Tissue debris was removed by centrifugation at 2,400 × g at 4°C for 20 min. The protein concentration was determined using a BCA protein assay kit (Thermo Fisher Scientific, Inc.), then adjusted to 8 mg/ml. Loading buffer (5X; Beijing Solarbio Science and Technology Co., Ltd.) was then added to the sample (volume-volume, 1:4) and heated at 100°C for 5 min. Subsequently, 5 µl/lane of the protein extract was loaded and separated using SDS-PAGE (10% separating gel; 5% spacer gel) and transferred onto PVDF membranes following electrophoresis. The membranes were then blocked with 5% skimmed milk/TBS-0.1% Tween-20 for 3.5 h at room temperature. This was followed by incubation with EGFP (1:5,000; cat. no. ab184601) or GAPDH (1:5,000; cat. no. ab181602) (both from Abcam) primary antibodies overnight at 4°C. After washing with PB, the membranes were incubated with goat anti-mouse IgG antibody (1:5,000; cat. no. GAM001) or goat anti-rabbit IgG antibody (1:5,000; cat. no. GAR007) (both from MultiSciences) for 2 h at room temperature. Finally, the blotted membranes were exposed to ECL substrate (Advansta, Inc.), and the chemiluminescence imaging system, ChemiScope 6100 Touch, was used to capture the images. The measurement of the protein band density was performed using ImageJ software (version 1.8.0; National Institutes of Health).

The liver sections were fixed in 4% PFA fix solution at 4°C for 24 h. They were then sequentially dehydrated in 15 and 30% sucrose solution overnight at 4°C, until the samples sunk. The samples were embedded in Optimal cutting temperature medium (OCT; Sakura Finetek Japan Co., Ltd.) and stored at −80°C. The liver tissues were cut into 10-µm-thick cryosections using a Leica cryostat (Leica Microsystems GmbH). The sections were then incubated with mouse monoclonal anti-EGFP antibody (1:100; cat. no. ab184601; Abcam) overnight at 4°C. The sections were subsequently incubated with Alexa Fluor^® 488 goat anti-mouse IgG (1:1,000; cat. 4408S; Cell Signaling Technology, Inc.) for 2 h at room temperature, followed by staining with DAPI (Sigma-Aldrich; Merck KGaA) for 10 min in the dark. The tissue sections were visualized using a Leica immunofluorescent confocal microscope (Leica Microsystems GmbH) at ×200 magnification. The measurement of the fluorescence intensity was performed using Leica LAS X software (version 3.4.1; Leica Microsystems GmbH).

To determine whether the treatment induced hepatotoxicity, the serum samples were thawed at 4°C. The alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bile acid (TBA) and total bilirubin (TBIL) activity in the serum samples was then measured using enzymatic colorimetry with the Multiskan GO plate reader (Thermo Fisher Scientific, Inc). The procedures for the analysis were performed according to the instructions provided by the kits (ALT, cat. no. H001; AST, cat. no. H002; TBA, cat. no. H101T; TBIL, cat. no. H115; all from Ningbo Medical System Biotechnology Co., Ltd.).

The formalin-fixed liver tissues were dehydrated in a gradient ethanol series (70, 80, 90 and 100%) and washed with xylene. They were then embedded in paraffin at 56°C and cut into 4-µm-thick sections using the Leica RM2235 Manual Rotary Microtome (Leica Microsystems GmbH) and stained with hematoxylin (cat. no. G1140) for 3 min and eosin (cat. no. G1100) (both from Beijing Solarbio Science & Technology Co., Ltd.) for 2 min at room temperature. The observation of the stained liver tissue sections was performed using an Olympus BX41 microscope (Olympus Corporation) at ×100 and ×400 magnifications.

The frozen liver tissues (20 mg) of five mice in each group randomly selected for RT-qPCR analysis. The liver tissues were lysed with TRIzol^® (Invitrogen; Thermo Fisher Scientific, Inc.) and homogenized at 960 × g for 30 sec at room temperature using a MagNA Lyser instrument (Roche Diagnostics). Pure chloroform was then added for 5 min to extract total RNA at room temperature. This was followed by centrifugation at 3, 200 × g for 15 min at 4°C and precipitation with 75% ethanol. The RNA concentration was quantified using the Multiskan GO plate reader (Thermo Fisher Scientific, Inc) and its purity was determined using the OD260/OD280 calculation. Total RNA was reverse transcribed into cDNA using RT (20 µl; cat. no. CW2569M; CoWin Biosciences Co., Ltd.) as previously described (20). The RT temperature protocol was as follows: 25°C for 5 min, 42°C for 1 h, inactivation at 70°C for 5 min and chilling at 4°C for holding. The primer sequences are presented in Table SI. qPCR was performed in 96-well plates using a 5-µl system containing 1 µl total cDNA, 2.2 µl UltraSYBR Mixture (cat. no. CW0957H; CoWin Biosciences Co., Ltd.), 0.1 µl forward and reverse primer, and 1.6 µl RNase-free water using the LightCycler 480 II system (Roche Diagnostics). The following thermocycling conditions were used: Initial denaturation at 95°C for 10 sec, 55°C for 10 sec and 72°C for 15 sec. The 2^−ΔΔCq formula was used to quantify the expression levels of target genes (21). The measured mRNA abundance was normalized to 18S rRNA. The expression levels in the control group were set to 1, and the data of the other six groups were normalized and expressed as relative expression.

The data are presented as the mean ± SEM. Data analysis was performed using SPSS version 23 software (IBM Corp.) and column charts were generated using GraphPad Prism software (version 8.0; GraphPad Software, Inc.). Statistically significant differences were determined using one-way ANOVA followed by Tukey's post hoc test for multiple comparisons. When data were not normally distributed, the Kruskal-Wallis test was used for the determination of differences among groups and Bonferroni's post hoc test was used for multiple comparisons. P<0.05 was considered to indicate a statistically significant difference.

The schematic presentation of the three different rAAV8 vectors used in the present study was presented in Fig. 1. rAAV8 vector constructs encoding EGFP were transduced, driven by the CAG, TBG669 and TBG410 promoters. An equal concentration of 1×10¹¹ GC in 100 µl PBS was delivered to the ICR mice via either IP or IV injection. At 4 weeks after the injection, the mice were euthanized, and blood and liver tissues were collected for analysis.

Western blot analysis was performed to determine the protein expression level of EGFP among the three promoters, under the same administration routes. As shown in Fig. 2, the protein expression level of EGFP was the highest in the rAAV8-CAG-EGFP group, by both administration routes. With IV administration, significant differences were observed between the control, TBG410, TBG669 and CAG groups., EGFP protein expression level induced by the CAG promoter was 67-fold higher compared with that in the rAAV8-TBG410-EGFP group and 26-fold higher compared with that in the TBG669 group. Gene transduction induced by the TBG669 promoter was almost 3-fold higher than that induced by the TBG410 promoter (Fig. 2A). Similarly, with the IP administration, the EGFP expression level in the rAAV8-CAG-EGFP group was 75-fold higher compared with that in the TBG410 group, and 41-fold higher compared with that in the TBG669 group (Fig. 2B).

The comparison of the two administration routes also revealed notable results. Driven by the CAG promoter, the abundance of EGFP was significantly increased; the ratio of EGFP/GAPDH in the control, IV and IP groups was 0.008, 0.791 and 0.513, respectively. The protein expression level of EGFP with the IV and IP injections was 99- and 64-fold higher compared with that in the control group, respectively (Fig. 3A). In the rAAV8-TBG669-EGFP group, the ratio of EGFP/GAPDH in the control, IV and IP groups was 0.142, 1.027 and 0.446, respectively. EGFP protein expression with the IV injection was 2-fold higher compared with that in the IP injection group (Fig. 3B). Generally, EGFP protein expression via the IV route was superior than that via the IP route, with all vectors. However, driven by the CAG promoter and compared with that in the other two promoters, the ratio of EGFP/GAPDH in the control, IV and IP groups was 0.097, 0.922 and 0.853, respectively. EGFP protein expression was similar between both delivery routes (Fig. 3C).

The EGFP expression patterns were assessed using a fluorescence microscope to confirm the quantification of the densitometric analysis. The representative fluorescent micrographs of the CAG promoter with IV and IP administrations are presented in Fig. 4. In the rAAV8-CAG-EGFP group, a robust transduction was observed, and gene expression mainly transduced in the nucleus around the central veins. In addition, the EGFP intensity was close to 4-fold higher via the IV injection compared with that via IP injection (Fig. S1). With respect to the TBG669 and TBG410 promoters, their transgene efficiency was in accordance with that observed with western blot analysis described above (data not shown).

A total of three different analyses were performed to evaluate the safety of the three rAAV8 vectors used in the present study. Biochemically, the serum ALT, AST, TBA and TBIL levels exhibited no significant changes among the three rAAV8 vectors (Fig. 5). In addition, no liver injury or inflammation was observed in any of the groups (Fig. 6). The high magnification images revealed no evidence of cellular damage and an inflammatory response (Fig. S2), supporting the aforementioned and biochemical data. Compared with that in the control group, the mRNA expression levels of chemokine ligand 2, Tnf-α, suppressor of cytokine signaling 3 and fibrinogen α chain in the other six groups exhibited no significant differences (Fig. 7). These results suggested that the three rAAV8 vectors did not cause inflammatory responses.