According to the human TOR2A gene sequence number (NM-001134430), the sequence of salusin-α was determined, and a short hairpin (sh)RNA design tool (BLOCK-iT™ RNAi Designer; Thermo Fisher Scientific, Inc.) was used to design three salusin-α interference sequences (Table I). The specificity of these sequences was then confirmed through BLAST analysis on NCBI database. These sequences were synthesized by Sangon Biotech Co., Ltd. Transfer plasmids [pHAGE (cat. no. 118692) and pLKO.1 (cat. no. 8453; both from Addgene)] and auxiliary plasmids [psPAX2 (cat. no. 12260) and pMD2G (cat. no. 12259; both from Addgene)] were provided by Tongji Medical College of Huazhong University of Science and Technology. SnapGene-designed salusin-α primers contained BamHI (cat. no. 1010A) and XhoI (cat no. 1094A; both from Takara Biotechnology Co., Ltd.) restriction sites (Table II), and pHAGE-Salusin-α was constructed according to the method of double-enzyme-digestion molecular cloning. Briefly, the PCR amplification product containing salusin-α and pHAGE vector were digested with BamHI and XhoI to produce the same sticky ends. Subsequently, the salusin-α gene sequence and linear Phage vector were purified and then joined using T4 DNA ligase (cat no. 2011A; Takara Biotechnology Co., Ltd.) in an overnight incubation at 16°C. pLKO.1-shSalusin-α was constructed according to the construction method for interference plasmids. The two synthesized interference sequences were annealed to form a double-stranded DNA, which was then ligated with the line vector of pLKO.1 that had been digested by AgeI (cat. no. R3552S; New England BioLabs, Inc.) and EcoRI (cat. no. R3101V; New England BioLabs, Inc.). The constructed pHAGE-Salusin-α and pLKO.1-shSalusin-α plasmids were respectively transfected into TOP10 competent cells. The transfection mixture was coated onto an LB plate and cultured overnight at 37°C in an incubator, prior to the selection of colonies on the plate. The constructed plasmids were confirmed by colony PCR. Table II shows the salusin-α primers and the primers designed for the interference sequence. Positive colonies of successfully cloned plasmid were inoculated into LB culture medium. Precisely following the instructions of the plasmid extraction kit (cat. no. 9760; Takara Biotechnology Co., Ltd.), the recombinant plasmids were extracted from the bacterial solution and sent to Sangon Biotech Co., Ltd. for sequencing. Finally, the recombinant plasmids were amplified and purified, and a large number of highly purified plasmids were obtained for subsequent experiments.

293T cells (cat. no. CRL-3216TM; American Type Culture Collection), which are also called human embryonic kidney cells (an epithelial-like cell that was isolated from the kidney of a patient), were obtained from the Tongji Medical College of Huazhong University of Science and Technology. The HepG2 cell line (has an epithelioid morphology and was isolated from male liver cancer) was purchased from ENOVA BIO (www.enovabio.com; cat. no. ECL0103). After STR identification, the aforementioned HepG2 cell line was found to have a similarity rate and of 98.51% with the HepG2 cell line in CELLOSAURUS database (accession no. CVCL_0027). Both cell lines were maintained in high-glucose DMEM (cat. no. PM150210; Procell Life Science & Technology Co., Ltd.) containing 10% (v/v) fetal bovine serum, 100 U/ml penicillin and 100 µg/ml streptomycin (cat. no. MA0110; Dalian Meilun Biology Technology Co., Ltd.) in a humidified incubator containing 5% CO₂ at 37°C. Mycoplasma detection was performed on the cells before the experiment, and HepG2 cells were identified by STR profiling to determine that there was no cross-contamination of the cells. To explore the manner in which salusin-α may regulate the PPARα/ApoA5/SREBP-1c pathway through AdipoR2, after having infected HepG2 cells with the lentivirus for 24 h, the cells were coincubated with Complete™ medium containing either the AdipoR2 inhibitor (1 µM thapsigargin; cat. no. T9033; MilliporeSigma) or the AdipoR2 agonist [20 µM 4-phenyl butyric acid (PBA); cat. no. 2312-73-4; Sigma-Aldrich; Merck KGaA] for 24 h (20).

293T cells in the exponential growth phase were selected for packaging of lentivirus. In this experiment, the second-generation lentivirus was packaged by liposome transfection. According to the Simple-Fect Transfection Reagent (cat. no. profect-01; Wuhan Signal Shuguang Biotechnology Co., Ltd.), 4 µg recombinant plasmid (pHAGE-Salusin-α or pLKO.1-shSalusin-α#1-3), 3 µg pMD2G plasmid and 1 µg psPAX2 plasmid were co-transfected into 293T cells to package lentivirus. The transfected 293T cells were placed into the 5%-CO₂ incubator at 37°C for culture. After 24 h of transfection, the culture medium was changed, and then continued to be cultured at 37°C. The peak period for 293T cells to produce lentivirus is 48-72 h after transfection; therefore, the cell supernatant was collected at 48 and 72 h respectively, and the cytopathic effect (CPE) was observed under an inverted optical microscope. After 72 h of transfection, the ability of cells to produce lentivirus rapidly decreased and a large number of cells died due to CPE, at which time, the culture could be terminated. The collected cell supernatants were centrifuged at 4°C, 1,500 × g for 10 min to remove the cell residue, subsequently filtered with 0.45 µm filter, sub-packaged and stored at -80°C for future experiments. In addition, 500 µl of the aforementioned filtered cell supernatant was collected, and the viral RNA was extracted using the TRIzol™ method (cat. no. 9108; Takara Biotechnology Co., Ltd.). Given that the pHAGE/pHAGE-Salusin-α and shMock/pLKO.1-shSalusin-α plasmids carried the specific genes woodchuck hepatitis virus post-transcriptional regulatory element (WPRE) and shSalusin-α respectively, the existence of WPRE and shSalusin-α was confirmed using semi-quantitative (SQ)-PCR (Table II) and 1.5% agarose gel electrophoresis (GoldView) to verify the production of this virus.

After 72 h of culture, precooled 2.5% glutaraldehyde (cat no. G916054; Shanghai Macklin Biochemical Co., Ltd.) was directly added to the 293T cells used for packaging without rinsing, and the cells were fixed for 1 h. The cells were quickly scooped out with a cell shovel and transferred to a centrifuge tube. Most of the cellular supernatant was removed by centrifugation at 4°C, 300 × g for 5 min, leaving ~1 ml of residual supernatant for gently blowing apart the cells, and then transferred to 1.5-ml EP tubes. After standing vertically and allowing the cells to settle naturally for 1 h, the supernatant was gently discarded, and 1 ml newly precooled 2.5% glutaraldehyde was slowly added along the tube wall. The cells were then stored in a refrigerator at 4°C for 2 h, 1% osmic acid fixed at 4°C for 2 h. Buffer washing and dehydration of 50% pyruvic acid for 15 min was performed. Epon 812 (cat no. 02662-AB; Structure Probe, Inc.) was used as a resin embedding agent to embed the sample at 60°C for 36 h, and then a slicer was used to cut the sample into 50 nm. Subsequently, at room temperature, the sample was dyed with uranium acetate for 30 min, then with lead citrate for 15 min, and finally sent to the People's Hospital of Wuhan University for subsequent TEM experiments.

293T cells were inoculated into 12-well plates for 2×10⁵ cells per well, and cultured overnight at 37°C. On the second day, the cells proliferated to 20-30% fusion density, and the viral solution stored in the refrigerator at -80°C was melted in the ice bath, and the cell culture solution containing DMEM (without serum) was used for gradient dilution as follows: i) First diluent: 10 µl virus solution + 90 µl culture medium for virus dilution; ii) Second diluent: 10 µl of first diluent + 90 µl culture medium for virus dilution; iii) Third diluent: 10 µl of second diluent + 90 µl culture medium for virus dilution; iiii) Fourth diluent: 10 µl of third diluent + 90 µl culture medium for virus dilution. The desired cell hole was selected, the lentivirus diluent in each tube was gently mixed, 90 µl was added into each cell hole, and placed into a cell incubator at 37°C for overnight culture. The control virus with known titer was added with No. 1-4 diluent, and the sample to be tested was added with first and second diluent. On the fifth day, the cell state was observed, the cell density was >80%, and RNA was extracted for reverse transcription-quantitative PCR (RT-qPCR). After the lentivirus infected cells successfully, 500 µl TRIzol solution was added to each well of 12-well plate to extract the total RNA. After that, RT-qPCR were used to detect the target gene WPRE and internal reference gene GAPDH (primer sequences are listed in the Table II). PCR system is presented in Table III. The thermocycling conditions for the PCR program were as follows: 95°C pre-denaturation for 30 sec; denaturation at 95°C for 15 sec; annealing at 60°C for 60 sec; 40 cycles. Finally, the standard curve was constructed based on the CT value of WPRE and GAPDH in the control group and the known virus titer, and then the relative virus titer was calculated according to the CT value of the experimental group. The virus titer equals to: The relative virus titer x dilution multiple.

Multiple data have shown that the MOI of HepG2 is 5-10, thus the six MOI gradients of 2, 5, 7, 10, 15 and 20 were set. Cells in each well of the six-well plate were inoculated at a density of 8×10⁵. When the density reached 80% the following day (the number of cells was ~2×10⁶), according to the formula MOI=(virus titer x added virus volume)/the number of cells transfected, the corresponding added virus volumes were calculated to be 0.2, 0.5, 0.7, 1, 1.5 and 2 ml respectively. Lentivirus was added one by one, and the fresh culture medium was changed after 12 h. Antibiotics were used for screening and puromycin (cat no. ST551; Beyotime Institute of Biotechnology) was diluted to 100 µg/ml (from an initial concentration of 10 mg/ml). Then, according to the manufacturer's instructions, puromycin was used at a concentration of 3 µg/ml for HepG2 cells and 2 µg/ml for 293T cells (60 µl of diluted puromycin was added to the six-well plate (2 ml) of HepG2 and 40 µl of diluted puromycin to the six-well plate (2 ml) of 293T cells]. After 3-5 days of culture, under the microscope, it was found that the pore cells with 1 ml of virus had a higher survival rate and favorable cell state, and the pores with >1 ml of virus had different degrees of black spots, thus the appropriate MOI value of HepG2 was 10. MOI value of 293T cells was 1.

Before the formal experiment, the collected pHAGE, shMock and pLKO.1 lentiviruses were used to infect the cultured 293T cells respectively. At the same time, a group of 293T cells without virus (NC group) was set up, and the expression of salusin-α in each control group was detected at 48 h after infection. After that, the formal experiment began. 293T cells in favorable growth condition were inoculated in sixwell plates at a density of 1.2×10⁵ cells/well. The next day, after the cell confluence reached 70%, lentivirus infection was initiated. According to the titer and MOI detected by the aforementioned experiments, it was detected that 1 ml of viral stock solution was the most favorable. Therefore, after the medium in the six-well plate had been removed, 1 ml of the aforementioned viral solution and 1 ml fresh complete medium were added to each well, in addition to polybrene (cat no. BL628A; Biosharp Life Sciences) at a final concentration of 8 µg/ml in each well. Subsequently, the six-well plate was placed into the 5%-CO₂ incubator at 37°C for culture overnight. After 12 h incubation, the medium was replaced with fresh culture medium. Transiently infected 293T cells were collected 24 h after infection, and RT-qPCR (Table II) and WB analyses were then used to analyze the expression level changes of the target genes and proteins. The identical method of lentiviral infection was used for the HepG2 cells as for the 293T cells, with the only difference that, whereas 293T cells are easily infected and changes in gene and protein expression can be detected after only 24 h of infection, for the HepG2 cells the rapid collection of infected cells 48 h after infection to detect changes in target proteins via WB analysis was required. Moreover, based on the aforementioned experimental procedures, in order to explore the manner in which salusin-α may regulate the PPARα/ApoA5/SREBP-1c pathway through AdipoR2, after having infected HepG2 cells with the lentivirus for 24 h, the agonist (PBA) and inhibitor (thapsigargin) of AdipoR2 were respectively added, then the expression levels of proteins associated with lipid metabolism were detected.

The core of SQ-PCR is to determine the number of cycles of the target gene at the exponential amplification period. In order to determine the number of SQ-PCR cycles of the target gene, total RNA was extracted from 293T cells that had not been transfected with the virus using TRIzol™ (cat no. 9108; Takara Biotechnology Co., Ltd.), and following the instructions of the reverse transcription kit (cat no. D7170M; Beyotime Institute of Biotechnology), 1 µg template RNA was reverse-transcribed into cDNA. PCR Thermocycler (Biometra Tone 96 G; Analytik Jena AG) was used to PCR-amplify salusin-α (Table II) for multiple rounds, and the PCR cycle numbers were set to 20, 22, 24, 26, 28, 30 and 32, respectively. The thermocycling conditions for PCR were as follows: 94°C pre-denaturation for 3 min; denaturation at 94°C for 30 sec; annealing at 64°C for 30 sec; and extension at 72°C for 30 sec, completed the cycle number of times, and extended for 5 min at the end. The method used to find the number of cycles used for AdipoR2 SQ-PCR was consistent with that for salusin-α, except that the annealing temperature for AdipoR2 was set at 60°C. Subsequently, the PCR products were electrophoresed (1.5% agarose gels), and the electrophoretic bands were analyzed using Tanon 1600 Gel Imaging System (Tanon Science and Technology Co., Ltd.) to determine the index stage and platform stage of PCR amplification. Subsequently, PCR was performed for the target gene with the number of cycles at the exponential stage, and the same number of cycles of PCR was used for GAPDH (Table II) as the internal control to eliminate any error caused by sample addition. Finally, agarose electrophoresis with subsequent imaging in the gel image system was performed. The brightness of the bands was observed, and the optical density (OD) of each group of bands was measured using the Tianneng gel analysis software (biotanon.com). The ratios of the target gene OD value/GAPDH OD value were used to express the relative expression level of mRNA, to compare the expression levels of the target genes in each group.

Control and treatment cells were washed with PBS. After the cells were lysed on ice with RIPA lysate (cat. no. MAO151; Dalian Meilun Biology Technology Co., Ltd.) containing PMSF (cat. no. G2008-1ML; Wuhan Servicebio Technology Co., Ltd.) for 30 min, the cells in each group were respectively collected, and centrifuged at 4°C (4,500 × g for 10 min) to remove the cell fragments. Subsequently, the total protein concentration in the supernatant was measured using a bicinchoninic acid (BCA) kit (cat no. E-BC-K318-M; Elabscience Biotechnology, Inc.), and the protein content of each sample was adjusted to the same concentration at 5 µg/µl with PBS. The diluted protein samples were then mixed with 1X SDS loading buffer solution and denatured at 95°C for 5 min, and SDS-PAGE (either 8 or 10% gels) was used to separate the proteins, which were subsequently transferred to PVDF membranes (cat no. IPVH00010l; EMD Millipore). Non-specific binding to the membranes was blocked with 5% skimmed milk powder (cat no. 1172GR100; BioFRoxx; neoFroxx) for 2 h at room temperature, and the membranes were subsequently incubated overnight at 4°C with different primary antibodies, as follows: Anti-salusin-α rabbit polyclonal antibodies (1:1,000; cat. no. ab232928; Abcam), anti-AdipoR2 mouse monoclonal antibodies (1:1,000; cat. no. sc-514045; Santa Cruz Biotechnology, Inc.), anti-PPARα rabbit polyclonal antibodies (1:1,000; cat. no. ab227074; Abcam), anti-ApoA5 mouse monoclonal antibodies (1:1,000; cat. no. ab115772; Abcam), anti-SREBP-1c rabbit polyclonal antibodies (1:1,000; cat. no. ab28481, Abcam) and anti-GAPDH rabbit polyclonal antibodies (1:10,000; cat. no. ab9485; Abcam). Subsequently, diluted horseradish peroxide-labeled secondary antibodies were added as follows: Goat anti-rabbit IgG (1:50,000; cat. no. ab205718; Abcam) and goat anti-mouse IgG (1:10,000; cat. no. ab205719; Abcam). The membranes were incubated on a shaking table at room temperature for 2 h, prior to washing three times using TBST (0.1% Tween). WBs were visualized using ECL chemiluminescence kit (cat. no. MA0186; Dalian Meilun Biology Technology Co., Ltd.), and a chemiluminescence imaging system was used for exposure imaging. Furthermore, the gray values of the strips were analyzed with ImageJ (V1.8.0.112; National Institutes of Health), and the gray value ratios of each group of target strips and internal reference strips were recorded.

All data are expressed as the mean ± standard deviation. The statistical analysis was conducted using GraphPad Prism 8 (Dotmatics). All experiments were performed three times. Prior to the statistical analysis, the normality of data was examined using the Shapiro-Wilk test. ANOVA was used for normally distributed data, whereas the Kruskal-Wallis H rank sum test was used for non-normally distributed data. Depending on the design of the experiment, the data were analyzed using one-way ANOVA followed by Tukey's post hoc test. P<0.05 was considered to indicate a statistically significant difference.

The construction process for the recombinant plasmid is revealed in Fig. 1A. Positive clones of the TOP10 E. coli competent cells transfected with the recombinant plasmids pHAGE-Salusin-α and pLKO.1-shSalusin-α were selected. The electrophoretic results of the colony PCR products revealed that there were bands at ~100 and 300 bp, respectively (Fig. 1B), which was consistent with the anticipated molecular weights of 101 bp for salusin-α and 311 bp for shSalusin-α. Subsequently, the sequencing results of the recombinant plasmids were submitted for BLAST analysis on NCBI. These data showed that the inserted DNA sequences of the recombinant plasmids pHAGE-Salusin-α and pLKO.1-shSalusin-α were completely consistent with salusin-α and the designed shSalusin-α#1-3, respectively (Fig. 1C), indicating that the shRNA of salusin-α and salusin-α were accurately inserted into the vectors, and the recombinant plasmids were successfully constructed.

pHAGE-Salusin-α and shSalusin-α recombinant plasmids were co-transfected with the helper plasmids psPAX2 and pMD2G, respectively, into 70% confluent 293T cells to complete the packaging of the lentivirus (Fig. 2A). The peak period of packaging and releasing the virus was 48-72 h after transfection. During this period, it was observed that the 293T cells used to package the lentivirus appeared to clearly exhibit the CPE, whereas the CPE did not appear in the cells of the negative control (NC) group (Fig. 2B), indicating that the packaging of the virus was successful. Similarly, the TEM photos confirmed that lentivirus particles could be observed in the pHAGE, pHAGE-Salusin-α, shMock and pLKO.1-shSalusin-α#1-3 groups, but they were not observed in the NC group (Fig. 2C). Furthermore, according to the PCR electrophoretic results, there were bands at ~100 and 300 bp respectively, which were consistent with the expected sizes of 93 and 311 bp for salusin-α and shSalusin-α respectively, suggesting that the lentivirus packaging was successful (Fig. 2D).

PCR products with the PCR cycle numbers 20-32 were detected via electrophoresis. When the cycle number was 22, the band of salusin-α began to appear, although after cycle numbers 30 and 32, the brightness of the band did not change (data not shown), indicating that this was the beginning of the plateauing-out period. Based on these results, 28 cycles were selected as the number of cycles of SQ-PCR in these experiments for salusin-α. With the same analytical method, the results revealed that, for AdipoR2, 26 was selected as the number of cycles of SQ-PCR (data not shown). Subsequently, 293T cells were infected with the synthetic shMock and pLKO.1-shSalusin-α#1-3 viruses. After 24 h, SQ-PCR was used to detect gene expression and to evaluate the interference efficiency of pLKO.1-shSalusin-α#1-3. In the present study, the electrophoretic band brightness of salusin-α in the pLKO.1-shSalusin-α#1, 2 and 3 groups was lower compared with that in the shMock and NC groups, indicating that the designed three interfering sequences could all inhibit the expression of the salusin-α gene, but under the same treatment conditions, compared with the NC and shMock groups, the mRNA level of salusin-α in the 293T cells decreased most significantly in the pLKO.1-shSalusin-α#1 group (P<0.05) (Fig. 2E and F). Therefore, the pLKO.1-shSalusin-α#1 group was chosen for subsequent experiments. Furthermore, comparison results of each control group demonstrated that there were no significant difference in the expression level of salusin-α mRNA and protein among the control groups (Fig. 3, P>0.05). Therefore, in the next interference experiment, only shMock was used as the control.

The mRNA and protein expression of salusin-α and AdipoR2 in the 293T cells were detected by SQ-PCR and WB analysis. It was clearly revealed that, in the cells infected with pHAGE-Salusin-α virus, the expression levels of salusin-α mRNA and protein were significantly increased (P<0.05), as well as the expression levels of AdipoR2 mRNA and protein (Fig. 4A and B). On the other hand, in the cells infected with pLKO.1-shSalusin-α virus, the mRNA and protein levels of salusin-α were significantly reduced (P<0.05), resulting in a significant inhibition of the expression of AdipoR2 (Fig. 4A and C). However, these changes were not observed in the NC, pHAGE or shMock groups (Fig. 4). Collectively, these results demonstrated that salusin-α has a positive regulatory effect on AdipoR2.

Subsequently, given the aforementioned results wherein the association between salusin-α and AdipoR2 in 293T cells was studied, and since hepatocytes have an important role in regulating lipid metabolism and salusin-α and AdipoR2 are both involved in lipid metabolism, HepG2 cells were chosen to explore further the association between salusin-α and AdipoR2. The cultured HepG2 cells were infected with pHAGE-Salusin-α, pLKO.1-shSalusin-α#1 and the control virus. After culturing for 48 h, the cells were collected for WB detection. As expected, in this experiment, compared with the non-infected NC group and the infected pHAGE or shMock virus groups, the expression level of AdipoR2 in the HepG2 cells infected with pHAGE-Salusin-α increased significantly (Fig. 5A), whereas the opposite results were obtained in the pLKO.1-shSalusin-α#1 group (Fig. 5B) (all P<0.05). Furthermore, no significant differences were observed among the NC, pHAGE and shMock groups for AdipoR2 expression in HepG2 cells (P>0.05).

In order to further explore whether overexpression of salusin-α can activate the PPARα/ApoA5/SREBP-1c signaling pathway via the regulation of AdipoR2, thereby inhibiting lipid metabolism in HepG2 cells, the effects of overexpression or interference of salusin-α on the expression levels of PPARα, ApoA5 and SREBP-1c by were detected using WB analysis. As demonstrated in Fig. 6A, compared with the NC group, the overexpression of salusin-α led to a significant increase in the expression levels of PPARα and ApoA5in HepG2 cells, although the expression level of SREBP-1c was opposed to PPARα and ApoA5. However, in the pLKO.1-shSalusin-α#1 group, the expression levels of PPARα and ApoA5 were clearly inhibited, whereas the expression of SREBP-1c was moderately increased (Fig. 6B). No significant differences were observed in the expression levels of these proteins in the NC, pHAGE or shMock groups (Fig. 6). Furthermore, it is noteworthy that the changes in the intracellular protein levels after adding thapsigargin and PBA conformed with the anticipated results. After adding thapsigargin to the pHAGE-Salusin-α group, the expression of AdipoR2 was significantly decreased compared with the pHAGE-Salusin-α group without thapsigargin treatment, as well as the protein levels of PPARα and ApoA5, although the SREBP-1c protein level increased, reversing the changes of these protein levels caused by the increase in salusin-α protein level (Fig. 6A). However, co-incubation of PBA with HepG2 cells infected with pLKO.1-shSalusin-α#1 led to a clear increase in the expression levels of AdipoR2, PPARα and ApoA5, although the levels of SREBP-1c protein decreased, which were the opposite of the changes of these protein levels caused by the decrease of salusin-α protein level (Fig. 6B). Taken together, these data clearly confirmed that overexpression of salusin-α could inhibit lipid metabolism in HepG2 cells through upregulating AdipoR2 expression to activate the PPARα-ApoA5/SREBP-1c signaling pathway.