Objective To investigate the in vitro anti-hepatitis B virus(HBV)effects of icariside Ⅱ(ICS Ⅱ)and its impact on mitochondrial fission.Methods HBV-positive hepatocellular carcinoma HepAD38 cells were used as the cellular model.The cytotoxicity of ICS Ⅱ was assessed via CCK8 assay.The secretion levels of HBV surface antigen(HBsAg)and HBV e antigen(HBeAg),as well as HBV DNA copy numbers,were measured by ELISA and qPCR after treatment with ICS Ⅱ alone or ICS Ⅱ in combination with entecavir(ENT).The effects of ICS Ⅱ on mitochondrial morphology and motility were observed using confocal laser scanning microscopy and transmission electron microscopy(TEM).After ICS Ⅱ treatment,Western blot was performed to analyze the expression levels of key proteins involved in mitochondrial dynamics.Additionally,intracellular reactive oxygen species(ROS)production was evaluated via fluorescence staining.Results The CCK8 assay results showed that ICS Ⅱ treatment at 25 μmol/L had no significant effect on cell proliferation after 72 h.ICS Ⅱ significantly inhibited the secretion levels of HBsAg and HBeAg,with the respective inhibition rates reaching 54.90％and 39.65％(P＜0.05).Additionally,ICS Ⅱ alone reduced HBV DNA copy numbers by 15.19％,while ENT alone achieved a 34.11％inhibition rate.Notably,ICS Ⅱ in combination with ENT reduced HBV DNA copy numbers by 55.81％(P＜0.05).Furthermore,ICS Ⅱ induced mitochondrial shortening and enhanced mitochondrial motility in HepAD38 cells(P＜0.05).ICS Ⅱ significantly increased the expression levels of mitochondrial motility-related proteins,including Mfn1,Fis1,and phosphorylated Drp1(ser 616)(P＜0.05),while no significant changes were observed in the expression levels of Mfn2,total Drp1,or Drp1(ser 637)(P＞0.05).Additionally,ICS Ⅱ significantly suppressed the production of intracellular ROS in HepAD38 cells(P＜0.05).Conclusion ICS Ⅱ inhibits HBV replication in HepAD38 cells,and the underlying mechanism may be associated with the promotion of mitochondrial fission and suppression of ROS production.

Objective:To investigate the correlation between impulse oscillometry system examination indicators and conventional pulmonary ventilation function.Methods:The pulmonary ventilation function data of 10 883 patients from January 1, 2020 to December 31, 2022 at Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology were included. The one-second rate [ratio of forced expiratory volume in the first second (FEV 1) to forced vital capacity (FVC)] measured as a percentage of the predicted value was ≥92% for the control group ( n=3 478) and <92% for the pulmonary obstruction group ( n=7 405). The obstruction group was subdivided into five groups according to the degree of pulmonary dysfunction: mild group ( n=3 938),moderate group ( n=1 142),oderate-severe group ( n=917),severe group ( n=737),and extremely severe group ( n=671). Conventional pulmonary ventilatory function FVC, FEV 1, one-second rate, and forced expired flow at 50% of FVC (MEF50%), forced expired flow at 75% FVC (MEF25%), maximal mid-expiratory flow (MMEF), peak expiratory flow (PEF), and pulsed oscillation pulmonary function test were detected in both groups of patients. Impedance at 5 Hz (Z5) means total respiratory resistance, resistance at 5 Hz (R5) means total airway resistance, reactance at 5 Hz (X5) indicates the elastic recoil of the peripheral airways, and resistance at 20 Hz (R20) represents resistance of the central airways. R5-R20 reflects resistance in the small airways. Additionally, peripheral resistance (Rp), respiratory resonance frequency (Frex), and area under the reactance curve (Ax) were also measured. Correlation between the indicators of the two groups and the sensitivity and specificity of the impulse oscillometry system parameters for the diagnosis of obstructive pulmonary ventilation dysfunction were analyzed. Results:Pulmonary function force expiratory volume in the first second as a percentage of predicted value (FEV 1%Pre) [80.10 (54.95,97.10)%],one-second rate [62.43(48.67, 67.02)%],MEF50% [1.33 (0.62,1.97)L/s],MEF25% [0.28 (0.17,0.41)L/s], MMEF [0.85 (0.43,1.29)L/s],and PEF [5.64 (3.73,7.50)]L/s in the obstruction group were significantly lower than those in the control group ( P<0.05). The differences within the subgroups of the obstruction group were also significant ( P<0.05). Pulsed oscillation Z5 [0.42 (0.33,0.55)kPa·L -1·s -1],Rp [0.25 (0.20,0.45)kPa·L -1·s -1], R5 [0.39 (0.31,0.49)kPa·L -1·s -1], R20 [0.28 (0.24,0.34)kPa·L -1·s -1], R5-R20 [0.09 (0.05,0.17)kPa·L -1·s -1],Frex [16.32 (13.07,20.84)Hz], and Ax [0.67 (0.28,1.64)] indices in the obstruction group were significantly higher than those in the control group. X5 [-0.14 (-0.23, -0.10)kPa·L -1·s -1] was significantly lower than that in the control group ( P<0.05). Z5, Rp, X5, R5, R5-R20, Frex, and Ax were statistically significant between different degrees of obstruction in the obstruction group ( P<0.05). The impulse oscillometry system parameters Z5, Rp, R5, R20, R5-20, Frex, and Ax were negatively correlated with the indices of conventional pulmonary ventilation ( r=-0.21-0.68, P<0.05), and the parameter X5 was positively correlated with the indices of conventional pulmonary ventilation ( r=0.41-0.68, P<0.05). The pulsed oscillation pulmonary function test parameters X5 (58.60%-95.68%) and Ax (57.08%-98.06%) presented the best sensitivity; X5 (86.29%-98.82%), Frex (86.69%-94.71%), and Ax (88.10%-98.53%) displayed the best specificity; and R20 presented the worst sensitivity and specificity. The sensitivity and specificity were slightly better in female patients than in male patients. Conclusion:The technical parameters of the impulse oscillometry system showed significant correlation with relevant indices of conventional pulmonary ventilation function detection. These well reflect the changes of different degrees of pulmonary ventilation function and have greater significance for reference in evaluating the degree of pulmonary function impairment.

Objective:This study aimed to investigate the physical properties, biocompatibility, and 3D printing performance of a novel hybrid bioink composed of gelatin methacrylated (GelMA) and chitin nanocrystal (ChiNC).Methods:The study was conducted from May 2021 to December 2022, four different bioinks were prepared by adding varying amounts of ChiNC to GelMA bioink. The GelMA concentration in all four bioinks was 100 mg/ml, while the ChiNC concentrations were 0 mg/ml (no ChiNC added), 5 mg/ml, 10 mg/ml, and 20 mg/ml, respectively, named as GC0, GC5, GC10, and GC20 bioinks. The cross-sectional morphology of the hydrogels formed after photocuring the four bioinks was observed using scanning electron microscopy, and the porosity was calculated. Weighing the hydrogels before and after swelling, and then calculate the equilibrium swelling rate. HUVECs were seeded on the surfaces of the hydrogels prepared from the four bioinks and cultured in medium. Cell proliferation was assessed using CCK-8 assays at 1d, 3d, and 7d to compare the proliferation rates of cells on the four hydrogels. HUVECs were added to the four bioinks, and grid-like scaffolds were printed and cultured in medium. Live-Dead staining was performed at 1d and 7d to observe cell viability. Compare the printing effect of bioinks by observing its forming continuous threads properties during extrusion. Finally, tissue-engineered bladder patches simulating the mucosal layer, submucosal layer, and muscular layer anatomical structures of the bladder wall were 3D bioprinted using the optimized bioink composition, and the stability and fidelity of the printed structures were observed to further validate the feasibility of printing multi-layered complex structures with the bioink.Results:Scanning electron microscopy revealed that the porosity of the GC0, GC5, GC10, and GC20 hydrogels were (51.43±6.23)%, (51.85±6.47)%, (50.55±4.59)%, and (42.49±2.20)%, respectively. The differences in porosity between the GC0 group and the other three groups were not statistically significant ( P=0.9994, P=0.9948, P=0.1200). The equilibrium swelling ratio of the other three groups [(8.81±0.41), (7.95±0.19), (7.71±0.14)] was significantly lower than that of the GC0 group (9.37 ± 0.49), and the differences were statistically significant ( P=0.0457, P<0.01, P<0.01). CCK-8 assay showed no significant difference in absorbance value between the GC10 group (0.360±0.009) and the GC0 group (0.357±0.007), GC5 group (0.350±0.012), and GC20 group (0.345±0.018) on the first day ( P=0.9332, P=0.5464, P=0.4937). However, on the third day, the absorbance value of the GC10 group (0.755±0.012) was significantly higher than that of the GC0 group (0.634±0.010), GC5 group (0.704±0.009), and GC20 group (0.653±0.015) ( P<0.01, P=0.0033, P=0.0002). On the seventh day, the absorbance value of the GC10 group (1.001±0.031) was significantly higher than that of the GC0 group (0.846±0.026), GC5 group (0.930±0.043), and GC20 group (0.841±0.024)( P=0.0012, P=0.1390, P=0.0010). The addition of human umbilical vein endothelial cells (HUVECs) into the four groups of hydrogels enabled the printing of grid-like scaffolds, and Live-Dead staining was performed on day 1 and day 7. The cell viability of HUVECs in the four groups on day 1 was (90.13±1.63)%, (90.6±2.45)%, (92.58±2.15)%, and (91.40±3.17)%, respectively. There were no statistically significant differences between the GC0 group and the other three groups ( P=0.9869, P=0.3093, P=0.8008). On day 7, the cell viability was (89.97±3.10)%, (92.18±2.21)%, (92.05±2.25)%, and (90.12±1.97)% for the four groups, respectively. There were no statistically significant differences between the GC0 group and the other three groups ( P=0.3965, P=0.4511, P=0.9995). Bioink extrusion test showed that the GC0 hydrogel could be extruded continuously and form threads at temperatures between 24℃ and 25℃, while the GC10 hydrogel could be extruded continuously and form threads at temperatures between 24℃ and 27℃. Printing tissue engineered bladder patches simulating the anatomical structure of the bladder mucosal layer, submucosal layer, and muscular layer using GC10 bioink, and the printed patches were stable, without collapse, and had high fidelity. Conclusions:Adding ChiNC to GelMA promotes cell adhesion, proliferation, and expands the printing window of GelMA bioink. The biocompatibility of the mixed bioink prepared by adding 10 mg/ml ChiNC in GelMA is good, capable of printing tissue-engineered bladder patches that mimic the anatomical structure of natural bladder walls.

Objective:To prepare the whole bladder acellular matrix (BAM) using the self-designed perfusion decellularization system, and evaluate the feasibility of constructing the tissue engineering bladder with the adipose-derived stem cells (ADSCs).Methods:This study was conducted from October 2020 to April 2021. The self-designed perfusion decellularization system was used, and four different decellularization protocols (group A, group B, group C and group D) were formulated, according to the flow direction of the perfusate and the action time of different decellularization solutions. Among them, the urethral orifice of the bladder tissue was used as the outflow tract of the perfusion fluid in groups A and B. The top of the bladder was cut off and used as the outflow tract of the perfusion fluid in groups C and D. In groups A and C, 1% Triton X-100 was treated for 6 h, and 1% sodium dodecyl sulfate (SDS) was treated for 2 h. In groups B and D, 1% Triton X-100 was treated for 7 h, and 1% sodium dodecyl sulfate (SDS) was treated for 1 h. In addition, the tissue in the normal bladder group was directly obtained from the natural bladder tissue, which did not require perfusion, cryopreservation and thawing. The fast and efficient decellularization protocol was screened out through HE, DAPI, Masson trichrome and Alcian Blue staining and quantitative analyses to prepare the whole bladder scaffold. The prepared BAM was used as the scaffold material, and the ADSCs were used as the seeding cells to construct the tissue engineering bladder. HE and DAPI staining were used to observe the distribution of ADSCs on the BAM.Results:HE and DAPI staining showed that there was no obvious nuclear residue in the group C. Masson trichrome and Alcian Blue staining showed that the collagen structure and glycosaminoglycan were well preserved in the group C. There was no significant difference in bladder wall thickness between the group C and the normal bladder group [(975.44±158.62)μm vs.(1 064.49±168.52)μm, P > 0.05]. The DNA content in the group C [(43.59 ±4.59) ng/mg] was lower than that in the normal bladder group, group A, group B and group D [(532.50±26.69), (135.17±6.99), (182.49±13.69) and(84.00±4.38)ng/mg], and the difference was statistically significant ( P<0.05). The collagen content [(10.98 ± 0.29)μg/mg] and glycosaminoglycan content [(2.30±0.18)μg/mg] in group C were not significantly different with those in the normal bladder group [(11.69±0.49) and (2.36±0.09)μg/mg, P>0.05]. Scanning electron microscopy showed that a large number of pore structures could be observed on the surface of the prepared BAM in groups A-D and were facilitated to cell adhesion. The isolated and cultured ADSCs were identified by flow cytometry to confirm the positive expression of CD90 and CD29, and the negative expression of CD45 and CD106. Live/dead staining and CCK-8 detection confirmed that the prepared BAM in the group C had no cytotoxicity. HE and DAPI staining showed that a large number of ADSCs were distributed on the surface and inside of the tissue engineering bladder. Conclusions:The whole bladder shape BAM prepared by the self-designed perfusion decellularization system could be used as the scaffold material for bladder tissue engineering, and the constructed tissue engineering bladder could be used for bladder repair and reconstruction.

Objective:To investigate the effect of tissue engineered bladder patch constructed by double-layer silk scaffold and adipose-derived stem cells (ADSCs) in the repair and reconstruction of bladder.Methods:This study was conducted from May 2020 to March 2021. The silk fibroin (SF) aqueous solution was obtained from silkworm cocoons, and a double-layer silk scaffold composed of silk fibroin film and silk fibroin sponge was further prepared. The rat ADSCs were isolated, cultured, and the ADSCs surface markers (CD29, CD90, CD45, CD106) were identified by flow cytometry. The ADSCs were planted on a double-layer silk scaffold to construct a tissue-engineered bladder patch. Thirty-six male SD rats were randomly divided into three groups: tissue engineered bladder patch group (SF-ADSCs group, n=15), double-layer silk scaffold group (SF group, n=15), control group ( n=6). The tissue engineered bladder patch (SF-ADSCs group) and double-layer silk scaffold (SF group) were wrapped on the omentum to promote vascularization. The vascularization was evaluated by HE and immunofluorescence staining. The wrapped tissue engineered bladder patch and double-layer silk scaffold were used to repair the defective bladder. In the control group (six rats), the incision was closed immediately after the bladder tissue fully exposed. At 4 weeks and 12 weeks after operation, the general morphology of bladder tissue and cystography were performed to evaluate the recovery of bladder morphology. After the graft was harvested, HE and Masson's trichrome staining and immunofluorescence staining were used to observe the regeneration of bladder wall tissue. Urodynamics was used to assess the recovery of bladder function at 12 weeks after operation. Results:The flow cytometry results confirmed that the isolated cells positively expressed CD29 and CD90, and there was no significant expression of CD45 and CD106. Gross observation and scanning electron microscope confirmed that the preparation of double-layer silk scaffold not only had a pore structure that was conducive to cell planting, but also had good toughness and was facilitated to surgical suture. The number (43.50±2.66) and area (0.73±0.03)% of vascular-like structures in the SF-ADSCs group after the omentum encapsulation was significantly higher than that in the SF group [(24.50±3.51), (0.55±0.05)%], and the difference was statistically significant ( P<0.05). At 4 weeks after bladder repair, the histological staining of the grafts in the SF-ADSCs and SF groups showed a large number of degraded fragments of double-layer silk scaffold. At 12 weeks, the morphology of the graft in the SF-ADSCs group showed uniform bladder morphology, which was similar to that of normal bladder tissue. Immunofluorescence staining showed that the continuous urothelial layer, abundant smooth muscle tissue, vascular structure and regenerated neurons could be observed in the SF-ADSCs group. Urodynamic test showed that the bladder maximum volume (0.74±0.03)ml and compliance (16.68±0.44)μl/cm H 2O in the SF-ADSCs group, which were better than that in the SF group [(0.47±0.05)ml, (14.89±0.37)μl/cm H 2O], but lower than that in the control group [(1.12±0.08)ml, (19.34±0.45)μl/cm H 2O], and the difference was statistically significant ( P<0.05). Conclusions:The tissue engineered bladder patch constructed with double-layer silk scaffolds and ADSCs could promote the morphological repair of bladder tissue, the regeneration of bladder wall structure and the recovery of bladder physiological function.

Objective:To analyze the clinical effects of the pre-expanded internal mammary artery perforator flap in large chest keloids surgical treatment.Methods:Patients with large chest keloid were treated with the pre-expanded internal mammary artery perforator flap between January 2017 and September 2021. The surgical treatment was divided into two different phases. In the first phase, a tissue expander was implanted beneath the skin within the angiosome of the internal mammary artery perforator. The expander was injected with normal saline once a week. In the second phase, the expander and the keloid tissue were removed, and a pre-expanded internal mammary artery perforator flap was designed to cover the wound. Radiotherapy and hyperbaric oxygen therapy were performed in the postoperative period. The treatment effect was followed up. The postoperative complications were analyzed, and the recurrence and patient satisfaction rates were recorded.Results:A total of 41 patients were enrolled, including 20 male and 21 female patients. The patients’ age ranged from 24 to 64, with a mean disease history of 11.9 years. The mean size of the keloid was 9 cm × 8 cm. Some patients were treated with one expander, but four expanders were needed in some extensive cases. The volume of the expander ranged from 80 to 600 ml. The mean volume was 300 ml, with a mean expansion time of 3 months. The mean flap size was 9 cm × 8 cm. Two cases with distal necrosis were observed. Five cases suffered from partial incision scar hyperplasia. No recurrence occurred during the followed-up period. Thirty-six patients (87.8%) were satisfied with the operation effect, and five (12.2%) thought the effect was acceptable.Conclusions:The pre-expanded internal mammary artery perforator flap is an effective treatment for the large chest keloid. It can provide sufficient skin tissue for wound repair, with a stable blood supply and an excellent curative effect.

Objective:To analyze the clinical effects of the pre-expanded internal mammary artery perforator flap in large chest keloids surgical treatment.Methods:Patients with large chest keloid were treated with the pre-expanded internal mammary artery perforator flap between January 2017 and September 2021. The surgical treatment was divided into two different phases. In the first phase, a tissue expander was implanted beneath the skin within the angiosome of the internal mammary artery perforator. The expander was injected with normal saline once a week. In the second phase, the expander and the keloid tissue were removed, and a pre-expanded internal mammary artery perforator flap was designed to cover the wound. Radiotherapy and hyperbaric oxygen therapy were performed in the postoperative period. The treatment effect was followed up. The postoperative complications were analyzed, and the recurrence and patient satisfaction rates were recorded.Results:A total of 41 patients were enrolled, including 20 male and 21 female patients. The patients’ age ranged from 24 to 64, with a mean disease history of 11.9 years. The mean size of the keloid was 9 cm × 8 cm. Some patients were treated with one expander, but four expanders were needed in some extensive cases. The volume of the expander ranged from 80 to 600 ml. The mean volume was 300 ml, with a mean expansion time of 3 months. The mean flap size was 9 cm × 8 cm. Two cases with distal necrosis were observed. Five cases suffered from partial incision scar hyperplasia. No recurrence occurred during the followed-up period. Thirty-six patients (87.8%) were satisfied with the operation effect, and five (12.2%) thought the effect was acceptable.Conclusions:The pre-expanded internal mammary artery perforator flap is an effective treatment for the large chest keloid. It can provide sufficient skin tissue for wound repair, with a stable blood supply and an excellent curative effect.

The pathogenesis of endometriosis is not well understood at the moment, and the lack of effective biomarkers often leads to delayed diagnosis of the disease. Lipidomics provides a new approach for the diagnosis and prediction of endometriosis. Sphingomyelin, phosphatidylcholine and phosphatidylserine in peripheral blood, endometrial fluid, peritoneal fluid and follicular fluid have good diagnostic value for endometriosis and disease classification; the lipid metabolites in the eutopic endometrium tissue are expected to be biomarkers of early endometriosis; and the lipid metabolites in peripheral blood are also of great value for predicting endometriosis-related infertility. The development of lipidomics technique will further advance the progress on the pathogenesis, prediction, diagnosis and treatment of endometriosis.
Biomarkers/blood* ; Blood Chemical Analysis/trends* ; Body Fluids/chemistry* ; Endometriosis/diagnosis* ; Female ; Humans ; Lipidomics/trends*

Objective To investigate the clinical,imaging,intestinal pathological characteristics and prognosis of gluten ataxia (GA).Methods The clinical data,treatment and prognosis in a patient with GA that was confirmed by pathology and hospitalized in the Department of Neurology,China-Japan Friendship Hospital in July 2018,were analyzed retrospectively.The related literature was reviewed and the clinical feature was summarized.Results The patient is a 41-year old man.He suffered from progressive cerebellar ataxia,and the brain magnetic resonance imaging exhibited diffused cerebellar atrophy.Serum human leukocyte antigen (HLA) tests showed that the patient carried HLA-DQ2 genotype.IgA type anti-gliadin antibody was positive (39.39 RU/ml).Duodenoscopy biopsy revealed mild villus atrophy and lymphocytic infiltration,indicating celiac disease.The diagnosis of GA was established then and the patient was administered gluten-free diet combined with intravenous immunoglobulin,which markedly improved the cerebellar symptoms and signs of cerebellar speech,walk capability and daily living activities.He could do long distance driving independently two months later.Conclusions GA is one of immune-mediated reversible acquired cerebellar ataxia caused by gluten sensitivity.The genotype,serologic features,and clinical phenotype of GA in Chinese mainland population might be similar with those in European and American countries.

Objective: To investigate the effect of long-chain non-coding RNA Fez family zinc finger protein 1 antisense RNA1 (lncRNA FEZF1-AS1) on the biological function of hepatocellular carcinoma (HCC). Methods: SMMC771 and BEL-7402 cells were transfected with sh-FEZF1-AS1 and OE-FEZF1-AS1, respectively. The expression of lncRNA FEZF1-AS1 was detected by real-time quantitative PCR. Cell proliferation was detected by Cell Counting Kit-8 (CCK-8), and apoptosis was detected by flow cytometry. The effects of lncRNA FEZF1-AS1 on invasion and migration were detected by Transwell and wound healing assays. The expression levels of adhesion molecules were detected by Western blot. The effect of lncRNA FEZF1-AS1 on the in vivo growth was verified by nude mice xenograft experiments. Results: The silencing or ectopic expression of lncRNA FEZF1-AS1 inhibited or promoted the proliferation of hepatocellular carcinoma cells. CCK-8 assay showed that the proliferation abilities of SMMC7721 and BEL-7402 cells in sh-FEZF1-AS1 transfection group significantly decreased, achieving (35.43±4.06)% and (34.68±3.97)%, respectively, on the fifth day. There were significant differences between sh-FEZF1-AS1 group and sh-NC group [52.21±8.46)% and (53.76±7.64)%] (all P<0.05). In contrast, the proliferation ability of SMMC7721 and BEL-7402 cells transfected with OE-FEZF1-AS1 was significantly increased, achieving (83.49±6.92)% and (80.31±3.13)%, respectively, on the fifth day. There were significant differences between OE-FEZF1-AS1 and OE-NC group [53.03±8.84)% and (55.11±7.09)%] (all P<0.05). The subsequent flow cytometry results showed that cell apoptotic rates of SMMC7721 and BEL-7402 cells transfected with sh-FEZF1-AS1 were (13.02±1.38)% and (11.88±1.29)%, respectively, which were significantly higher than those in sh-NC groups [(5.57±1.46)% and (8.06±1.42)%, respectively, all P<0.05]. In contrast, the apoptotic rates of SMMC7721 and BEL-7402 cells transfected with OE-FEZF1-AS1 were (3.01±0.39)% and (3.22±0.43)%, which were significantly lower than those in OE-NC groups [(6.68±0.96)% and (6.63±0.45)%, all P<0.05]. In addition, knockdown or overexpression of lncRNA FEZF1-AS1 expression inhibited or enhanced the migration and invasion abilities as well as the levels of adhesion molecules in hepatocellular carcinoma cells. After 30 days of feeding under the same conditions, the tumor volumes of sh-FEZF1-AS1 and sh-NC SMMC7721 cells xenograft mice models were (0.26±0.03) cm³ and (0.63±0.06) cm³, respectively, showing significant difference (P<0.05). The tumor volumes of sh-FEZF1-AS1 and sh-NC BEL-7402 cells were (0.31±0.02) cm³ and (0.72±0.08) cm³, and the difference was statistically significant (P<0.05). Conclusion lncRNA FEZF1-AS1 may strengthen the growth, migration and invasion of hepatocellular carcinoma cells.