Cell culture and clinical hepatocellular carcinoma samples

The HHL5 normal human hepatic cells and HCC cell lines (Huh7, QGY7701, MHCC97H, QGY, as well as 7404) were procured from the Cell Bank of the Chinese Academy of Sciences in Shanghai, China. The HCC cell line PVTT was acquired from Shanghai Institute of Nutrition and Health, which is part of the Chinese Academy of Sciences. As for culture medium, 7404 cells were cultivated in RPMI-1640 medium supplied with 10% FBS, whereas remainder group of cells were cultured in DMEM medium supplied with 10% FBS. Both medium contained penicillin and streptomycin with a concentration of 100 U/mL, 100 μg/mL, respectively. In addition, the cultivation environment was sterile incubator at a temperature of 37 ℃ containing 5% CO2.

Hepatocellular carcinoma samples and adjacent non-tumor tissues were acquired from Shanghai Eastern Hepatobiliary Surgery Hospital, affiliated to the Second Military Medical University, after patient consent. This cohort consists of 65 HCC tissues and corresponding adjacent non-tumor tissues, which were used to evaluating the mRNA levels of tyrosine hydroxylase. The study received the endorsement of the Ethics Committee of the Second Military Medical University for all experimental procedures.

Plasmids and transfection

To induce TH overexpression, its coding sequence was cloned into the pLVX–IRES-puro vector. Sequences targeting TH for RNA interference were selected out of the Merck site (https://www.sigmaaldrich.cn/CN/zh/product/sigma/shrna) and inserted to the pLKO.1-puro vector. These targeted shRNA sequences included:

shTH #1 F: 5ʹ-CCGGGGTGTTTGAGACGTTTGAAGCCTCGAGGCTTCAAACGTCTCAAACACCTTTTTG-3ʹ,

shTH #1 R: 5ʹ-AATTCAAAAAGGTGTTTGAGACGTTTGAAGCCTCGAGGCTTCAAACGTCTCAAACACC-3ʹ;

shTH #2 F: 5ʹ-CGGGCTGGACAAGTGTCATCACCTCTCGAGAGGTGATGACACTTGTCCAGCTTTTTG-3ʹ,

shTH #2 R: 5ʹ-AATTCAAAAAGCTGGACAAGTGTCATCACCTCTCGAGAGGTGATGACACTTGTCCAGC-3ʹ.

The Smad2 coding sequence was inserted into the pGEX-4T1 vector to stimulate the production of a fusion protein called GST–Smad2 in prokaryotic cells with IPTG.

The psPAX2 and pMD2.G, served as lentiviral packaging plasmids, were co-transfected with plasmids inserted into specific sequences into HEK293T cells by employing lipofectamine 8000 transfection Reagent. After a period of 24 h, The liquid portion of the mixture was gathered at a period of 48 h and 72 h after transfection. PEG8000 was introduced to this liquid portion, and it was subjected to centrifugation for 1 h at a temperature of 4 ℃ and a centrifugal pull of 1600 × g. This process was carried out to cleanse the lentivirus. Following the centrifugation process, the liquid portion above the sediment was removed, and the concentrated viral was overnight dissolved in 2 mL of basal medium. Next, 4 × 105 cells were infected with 400 μL of lentivirus solution. Following a 24 h incubation, the cells were passaged and subjected to puromycin screening with a concentration of 2 μg/mL. Expression of the target protein was subsequently measured by Western blot analysis.

Extraction of RNA with a quantitative polymerase chain reaction

The RNA was isolated using TRIzol and specific mass of RNA was converted into complementary DNA by using the PrimeScript RT reagent kit (Takara). The qPCR examination was conducted using the SYBR Green Kit and the CFX96 real-time fluorescence qPCR system from BIO-RAD. Actin was utilized as an internal control. The 2−ΔΔCt of target genes were calculated. The primer sequences carried out for TH in this section are the following:

TH F: 5ʹ-CATGGTAAGAGGGCAGGGC-3ʹ,

TH R: 5ʹ-GTGTAGGATGCAGCTGGGG-3ʹ.

Western blot

The PBS buffer was used for the purpose of washing the cells. And then, cells were lysed on ice by using RIPA lysis buffer containing protease inhibitors along with phosphatase inhibitors. The cellular extract was collected and subjected to centrifugation at 4 ℃ and 14,000 rpm for a duration of 20 min. The protein concentration was measured with the BCA kit. The protein samples were combined with protein loading buffer and subjected to heat at 100 ℃ for a duration of 5 min. The protein was transfer printed to a polyvinylidene difluoride membrane and then incubated with a 5% bovine serum albumin solution to block nonspecific binding. Subsequently, the membrane was subjected to incubation with targeted primary antibodies at a temperature of 4 ℃ for the duration of 8–10 h. On the following day, the membrane was cleansed employing TBST and then exposed to HRP-coupled secondary antibodies for a duration of 1–2 h. The membrane was cleansed with TBST, the signals were identified employing a chemiluminescent reagent (Millipore, WBKLS0050), and protein bands were examined employing Image Lab (BIO-RAD).

The antibodies utilized for such investigation were listed: TH (Brand &Cat no: Proteintech, 25859-1-AP; Dilution: 1:2000), Flag-tag (Brand &Cat no: Proteintech, 20543-1-AP; Dilution: 1:3000), HA-tag (Brand &Cat no: Proteintech, 66006-Ig; Dilution: 1:3000), GST-tag (Brand &Cat no: Proteintech, 10000-0-AP; Dilution: 1:3000), β-Tubulin (Brand &Cat no: Proteintech, 10068-1-AP; Dilution: 1:3000), GAPDH (Brand &Cat no: Proteintech, 10494-1-AP; Dilution: 1:10,000), Actin (Brand &Cat no: Santa Cruz, SC-8432; Dilution: 1:3000), HSP90 (Brand &Cat no: Proteintech, 13171-1-AP; Dilution: 1:3000), Smad2 (Brand &Cat no: Proteintech, 12570-1-AP; Dilution: 1:3000), and phospho-Smad2 (ser465/ser467) (Brand &Cat no: CST, 18338S; Dilution: 1:1000).

Immunohistochemistry

The dewaxed tissue sections were immersed in an EDTA solution and subjected to boiling at a temperature of 100 ℃ for a duration of 30 min to facilitate the recovery of antigens. Once the tissues reached room temperature through ambient cooling. The endogenous peroxidase activity was inhibited by employing endogenous peroxidase blockers. Subsequently, the sections underwent three rounds of washing in PBS, followed by incubation with three percent normal goat serum for a duration of 30 min at room temperature. The TH antibody (Brand &Cat no: Abcam, EP1532Y; Dilution: 1:500) was incubated with the sections overnight at a temperature of 4 ℃. The next day, the sections were washed three times with PBS and then incubated with secondary antibodies for 1 h at 37 ℃. Immunohistochemical staining was conducted utilizing 3,3ʹ-diaminobenzidine, while the nuclei were stained with hematoxylin. The Inform 2.4.0 system, developed by PerkinElmer, was utilized to assess the H-scores of immunohistochemical staining. The Kaplan–Meier technique was utilized to draw survival curves, and survival analysis was conducted employing the log-rank test.

Immunoprecipitation

To identify the binding between externally introduced TH and Smad2, the Flag-TH plasmid and the HA–Smad2 plasmid were simultaneously introduced into HEK293T cells employing transfection. After 48 h of transfection, the HEK293T cells were lysed employing IP lysis buffer, including protease inhibitors and phosphatase inhibitors. The IP lysis buffer composition included a solution containing 50 mM Tris–HCl (pH 7.4), 150 mM NaCl, and 5 mM EDTA, and 1% NP-40. The supernatants were collected after centrifugation at 4 ℃ and 14,000 rpm for 20 min. Add Flag-beads (Brand &Cat no: Sigma, A2220) or HA-beads (Brand &Cat no: Thermo Fisher, 88837) to cell supernatants incubate in rotate at a temperature of 4 ℃ for a duration of 3 h. The beads underwent three rounds of washing with wash buffer. Subsequently, the beads were mixed with protein loading buffer and subjected to heating at a temperature of 100 ℃ for a duration of 5 min. Western blot analysis was following conducted to detect the interaction.

To confirm the binding between endogenous TH and endogenous Smad2 in liver cancer cells, Huh7 cells were lysed employing an IP lysis buffer contains protease inhibitors and phosphatase inhibitors. Cell supernatants were following have an incubation with 2 μg of anti-TH antibody or 2 μg of anti-Smad2 antibody at a temperature of 4 ℃ for the duration of 8 h. The next day, 25 μL Protein A/G beads (Brand &Cat no: Bimake, B23202) were rotated incubated with the supernatant at a temperature of 4 ℃ for a duration of 2 h. The beads underwent three rounds of washing with IP lysis buffer. Samples were mixed with protein loading buffer, heated at 100 ℃ for 5 min, and then subjected to Western blot.

GST-pull down

The induction of GST as well as GST–Smad2 fusion protein expression in BL21 codon competence cells was achieved by employing IPTG (Brand &Cat no: Sangon, A100487) at a dose of 1 μM. The bacteria precipitations were subsequently lysed employing an IP lysis solution supplemented with protease and phosphatase inhibitors. Above lysate was subjected into incubation with GST beads at a temperature of 4 ℃ for a duration of 2 h.

The Flag-TH plasmids were introduced into HEK293T cells by transfection, and after 48 h, the cells were lysed employing an IP lysis solution. Enriched GST–Smad2 proteins and GST coupling with GST beads, were incubated with the cell lysate supernatant for 3 h at 4 ℃. Following three times of washes with IP wish buffer, the beads were mixed with 50 μL protein loading buffer and denatured at 100 ℃ for 5 min. The protein directly binding involving TH and Smad2 was examined employing Western blot assessment.

In vivo metastasis assay

This experiment utilized BALB/c-Nude mice (Brand: Gempharmatech), specifically male mice aged 5–6 weeks. Huh7 cells were subjected to lentivirus-mediated luciferase expression and subsequently screened upon treatment of Hygromycin B. Subsequently, each nude mice received an injection of 1 × 106 cells through the tail vein. Following a duration of 4 weeks, the nude mice were administered an intraperitoneal injection of 3 mg of D-fluorescein potassium salt (Brand &Cat no: Beyotime, ST196) and subsequently sedated with isoflurane. After a duration of 10 min, the bioluminescence of the mice was examined utilizing small animal bioluminescence imaging.

Soft agar assay

Upon reaching a cell confluence of 70–80%, the cells underwent digestion, and cells were resuspended in basal medium. The bottom gel with a volume of 300 μL per well, consisting of 20% FBS, 40% 2 × MEM, and 40% 1.25% agar. The culture layer with a volume of 400 μL per well, with a mixture of 25% FBS, 37.5% 2 × MEM, 37.5% 1% agar, 0.8% 2 mM L-glutamine and 800 cells. This assay was performed on 24-well plates, with PBS surrounded to sustain a moist culture environment. There were 4 replicate wells per cell. The colony was cultured at a temperature of 37 ℃ with a 5% concentration of carbon dioxide for a duration of 10–14 days. Five samples were chosen at random from per well for cell colony counting employing a microscope.

Transwell invasion assay

Growth factor-free Matrigel was diluted with basal DMEM medium at a ratio of 3:100, gently mixed, and promptly applied to the surface of the upper compartment. After 30 min, the liquid portion was removed. The lower chamber was replaced with 500 μL of DMEM medium with 30% FBS. The cells were resuspended with basic DMEM media at a concentration of 1 × 105 cells per milliliter. Following the mixing process, 200 μL cell solution containing 2 × 104 cells was placed within the upper chamber. After incubated at 37 ℃ with 5% CO2 for a duration of 48 h. The cells were fixed with a 4% paraformaldehyde solution and then subjected to staining with crystal violet. Five fields were chosen from each transwell chamber, and photographs were captured employing a microscope.

CCK8 assay

Every well of 96-well plates was seeded with 1 × 103 cells and then cultivated within the incubator at a temperature of 37 ℃ with a CO2 concentration of 5%. The following 5 days, the fresh basic medium with a concentration of 10% Cell Counting Kit-8 solution was added into the wells after using the negative pressure aspirator to suck the old medium away. After a 2-h incubation, the optical density at 450 nm was measured. Proliferation rate was drawn by a curve of OD450.

Wound healing assay

The HCC cell lines were distributed into culture dishes with a diameter of 3.5 cm with the same number of cells. Upon reaching full cell confluence, a linear wounding was created employing a pipette tip. Healing progression of wound was subsequently recorded employing microscopy at various time intervals. The assessment of wound healing rate of each cell was conducted using the Image J program.

Statistical analysis

The experiment results were analyzed by using SPSS Version 23.0 and GraphPad Prism Version 8.0.2 (GraphPad software, La Jolla, CA, US). Data were assessed using log-rank tests, two-sided χ2 tests, and two-tailed Student's t tests. The experiment results were considered to be statistically significant when the p value was less than 0.05.