Tissue Microarray Construction and HCC Samples Analysis

Cytochrome P450 3A5 Manipulates the Progression of Hepatocellular Carcinoma via Selectively Targeting ROS/mTORC2/p-AKT (S473) Signaling

Supplementary data

Supplementary Materials and Methods

Tissue microarray construction and HCC samples analysis

Tissue microarray slides were constructed in collaboration with Shanghai Biochip Company (Shanghai, China). Tumor malignantgrade was graded according to the Edmondson–Steiner grading system. The TNM stage was defined according to the Barcelona Clinic Liver Cancer (BCLC) staging system.

The additional HCC patients were recruited between January 1 and October 30, 2011, and the resected samples were subjected to RNA extraction for qRT-PCR and western blot verification. In addition, 28 pairs of HCC samples with portal vein tumor thrombus diagnosed by pathologists were obtained from the Eastern Hepatobiliary Surgery Hospital and were used for further qRT-PCR analysis.

In vivo metastasis assay

Ten 6-week-old male nude mice were randomized into two groups (n=5), and either SMMC-7721-CYP3A5 or SMMC-7721-GFP cells (1×106) were injected into the tail veins for the establishments of pulmonary metastatic model. Mice were sacrificed 10 weeks postinjection and examined microscopically by H&E staining for the development of lung metastatic foci. Animals were housed in cages under standard conditions, following the requirements of the Second Military Medical University Animal Care Facility and the National Institutes of Health guidelines. Another twenty 6-week-old male nude mice were randomized into two groups (n=10), and injected into the tail veins with either SMMC-7721-CYP3A5 or SMMC-7721-GFP cells (1×106). These mice were monitored for survival analysis.

Antibodies andreagents

The primary antibodies specific for CYP3A5 (ab108624) or CYP3A4(#13384) were purchased from Abcamor CST Company. Antibodies for MMP2, MMP7, MMP9, TIMP1 and TIMP2 were from Epitomics Company. Antibodies for phosphorylated AKT (p-AKT), total AKT (t-AKT), phosphorylated ERK(p-ERK), total ERK(t-ERK), phosphorylated JNK(p-JNK), total JNK(t-JNK), phosphorylated p38(p-p38), total p38(t-p38), E-Cadherin, N-Cadherin, Vimentin, Snail, Rictor, mTOR, β-Actin and GAPDH were from Cell Signaling Technology (Beverly, MA). Antibody for Rpl26(ab59652) was purchased from Abcam (Cambridge, MA).Crystal violet and 4',6'-Diamidino-2-phenylindole (DAPI) , GSH and NAC were from Sigma-Aldrich.Two AKT inhibitors, AZD5363 and MK-2206, were from Selleckchem (Houston, TX). The inhibitor of NADPH oxidase and nitric oxide synthase, Diphenyleneiodonium (DPI,CAS 4673-26-1), was purchased from Merck (USA). HGF and EGF were from PeproTech. Protein A/G agarose resin beads were from Thermo Scientific.Inactive AKT1 (#14-279) was from Millipore.

Cell lines, lentivirus, smallinterfering RNA

HCC cell linesSMMC-7721, and HCC-LM3 were purchased from Cell Bank of Type Culture Collection of Chinese Academy of Sciences. Cell lines were maintained at 37℃ in an atmosphere containing 5% CO2 in DMEM supplemented with 10% fetal bovine serum. Lentiviral vectors containing the human CYP3A5 gene or green fluorescent protein (GFP) were constructed and designated as “LV-CYP3A5” or “LV-GFP”. Lentivirus production was completed by the GeneChem Company (Shanghai, China). SMMC-7721 and HCC-LM3 cells were infected with the concentrated virus at a multiplicity of infection (MOI) of 20in the presence of polybrene (8 μg/ml) for 8 hours. After 24 hours the supernatant was replaced with fresh medium. The expression of CYP3A5 in the infected cells was validated by qRT-PCR and western blotting.

SMMC-7721-CYP3A5 and SMMC-7721-GFP cells in 6-well plate were transfected with siRNAs(RiboBio, Guangzhou, China) by INTERFERinTM transfection reagent (Polyplus, France) according to the manufacturer's instructions. The expression of Rictor was detected 48 hours later.The siRNA sequences used were included in thefollowing sense and antisense oligonucleotides:

si-Rictor-01, sense-5’-ACUUGUGAAGAAUCGUAUCdTdT-3’, antisense-3’dTdTUGAACACUUCUUAGCAUAG-5’;

si-Rictor-02, sense-5’-GGCCAGACCUCAUGGAUAAdTdT-3’, antisense-3’-dTdTCCGGUCUGGAGUACCUAUU-5’;

si-Rictor-03, sense-5’-CAAACAAGGCUGUGAUAUUdTdT-3’, antisense-3’-dTdTGUUUGUUCCGACACUAUAA -5.’

Si-CYP3A4-01

si-Rpl7-01, sense-5’-ACAAGGAAUAUAGGCAGAUdTdT-3’

Antisense-3’-dTdTUGUUCCUUAUAUCCGUCU-5’

si-Rpl7-02, sense-5’- AACAUGCUGAGGAUUGUAGdTdT-3’

Antisense-3’-dTdTUUGUACGACUCCUAACAUC-5’

si-Rpl7-03, sense-5’- GAUUCAUGAGAUCUAUACUdTdT-3’

Antisense-3’-dTdTCUAAGUACUCUAGAUAUGA-5’.

Plasmids preparation and transient transfection

The full-length CYP3A5 plasmid (pcDNA3.1-CYP3A5) was constructed by inserting a PCR-amplified full-length CYP3A5 fragment into the Xbal/KpnI sites of pcDNA3.1 vector (Invitrogen), using the primers, 5'-CTAGTAGCATGGACCTCATCCCAAATT -3'(forward), and 5'- CGGGGTACCTCTCCACTTAGGGTTCCATC -3' (reverse).For plasmid transfection, SMMC-7721 cells were transiently transfected using PEI (Polyplus; AFAQ) according to the manufacturer's instructions.

Quantitative Real-Time PCR and western blot

Total RNA from different cell lines and HCC samples was isolated using Trizol reagent (Invitrogen) following the manufacturer’s instructions. The cDNA template was prepared using random primers and Moloney murine leukemia virus reverse transcriptase (Promega) according to the manufacturer’s protocol. After the reverse transcription (RT) reaction, the cDNA template was quantitatively measured by using Real-time PCR technology. Real-time PCR was performed using an ABI PRISM 7300 sequence detector (Applied Biosystems) and SYBR Green PCR kit (Applied TaKaRa, Japan). The primers used in this study were as follows:

for CYP3A5, 5'-ACAGCAGCACTCAGCTAAAAG-3' (forward)

and 5'-GTGCTCTCCACAAAGGGGTC-3' (reverse);

for CYP3A4, 5’-GCTCTCTCATTGTCTGTGTAG-3’ (forward)

and 5’-GAAATAGTCCCGTGAGAAGCA-3’ (reverse);

for β-actin, 5'-GCGAGAAGATGACCCAGATCAT-3' (forward)

and 5'-GCTCAGGAGGAGCAATGATCTT -3' (reverse);

and for 18S rRNA, 5'-CGGCTACCACATCCAAGGAA-3' (forward)

and 5'-GCTGGAATTACCGCGGCT-3'(reverse).

β-Actin or 18S rRNA was used as a control. Every sample was tested in duplicate. Extracts of cell and tissue lysates were analyzed by western blotting using specific primary antibodies and appropriate fluorescein-conjugated secondary antibody. Immunocomplexes were then detected using Odyssey fluorescence scanner (Li-Cor, Lincoln, NE).

Supplementary Table

Supplementary Table 1. Clinical Pathological Factors and CYP3A5 Expression Level Based on Immunohistochemisry in HCC Tissue Microarray

Variable / CYP3A5 expression in HCC
Relative low (n=60) / Relative high (n=15) / P Value
Age (years)＊ / 52 (33-69) / 56 (32-70) / 0.1845＊
Sex, M: F / 49:11:00 / 15:00 / 0.0726
Diameter (cm), (<3cm:3-5cm:>5cm) / 16:19:25 / 7:06:02 / 0.1059
Differentiation grade, (Well:Moderate:Poor) / 1:40:19 / 4:07:04 / 0.0023†
Liver cirrhosis, yes:no / 22:38 / 8:07 / 0.2386
Tumor number (single:multiple) / 27:33:00 / 13:02 / 0.0038†
PVTT (yes:no) / 13:47 / 0:15 / 0.0474†
Microvascular invasion (yes:no) / 13:47 / 0:15 / 0.0474†
Metastatic lymph node (yes:no) / 4:56 / 0:15 / 0.3041
TNM stage (I:II:III) / 11:23:26 / 10:04:01 / 0.0006†

Data are expressed as ratios or as the median (range)

＊: Mann-Whitney test

†:p<0.05 by chi-square test or Student t test

Supplementary Table 2 The summary demographics of patients participated in overall survival and tumor free survival analysis

Characteristics / Number of patients
Patients / 159
Age(years) / ≤50 / 94
>50 / 65
Gender / Male / 141
Female / 18
Liver cirrhosis / Yes / 110
No / 49
AFP (ng/mL) / ≤20 / 47
>20 / 112
Tumor size(cm) / ≤5 / 49
>5 / 110
Micro-vascular invasion / Yes / 57
No / 102
Pathological satellite / Yes / 59
No / 100
Tumor number / Single / 137
Multiple / 22
HBe antigen / Positive / 119
Negative / 40
HBs antigen / Positive / 130
Negative / 29
Supplementary Table3. Univariate Analysis of Clinical Variables Associated With Recurrence and Survival
Variables / p value
TFS / OS
Tissue microarrayassay(n=159)
Liver cirrhosis( no vs. yes) / 0.001 / 0.043
AFP level, ng/mL(≤20 vs. >20) / <0.001 / 0.056
Age(≤50 vs. >50) / 0.316 / 0.455
Tumor diameter, cm(≤5 vs. >5) / <0.001 / 0.030
Tumor number(single vs. multiple) / 0.260 / 0.002
Micro-vascular invasion(no vs. yes) / 0.056 / 0.260
Pathological satellite(no vs. yes) / 0.001 / 0.034
Sex(female vs. male) / 0.356 / 0.763
HBe antigen(positive vs. negative) / 0.230 / 0.417
HBs antigen(positive vs. negative) / 0.002 / 0.019
CYP3A5(high vs. low) / 0.034 / 0.046

Note: Univariate analysis was calculated by the Kaplan-Meier method

(p<0.05 by Log-rank test).

Supplementary Figure legends

Figure S1.

(A)Representative image of immunohistochemical staining of CYP3A5 expression in HCC tissue microarray (75 pairs).

(B)Comparison of CYP3A5 expression in L02, SMMC-7721 and CSQT-1 cell lines.

Figure S2.

(A)Representative image of immunohistochemical staining of CYP3A5 expression in HCC tissue microarray (159 pairs).

Figure S3.

(A)CYP3A5 mRNA expression levels were screened in SMMC-7721, HCC-LM3, HepG2, MHCC97H, MHCC97L, Huh7, PLC and QGY-7701 HCC cell lines, which were commonly used in our lab.

(B)Proliferation of indicated cell lines was evaluated by determining the cell viability with Cell Counting Kit 8(CCK8) Assay according to the manufacturer’s instructions.

Figure S4.

(A)SMMC-7721-GFP and SMMC-7721-CYP3A5 cell lines were serum-starved overnight and then stimulated with EGF (100ng/ml) for indicated time periods. The expression t-AKT and p-AKT (S473) were analyzed by western blotting.

(B)The HCCLM3 cells were starved with serum-free DMEM overnight and exposed to HGF (40ng/ml) for 0, 5, 15 and 30 minutes. The total expression levels and phosphorylation levels of AKT (p-AKT), ERK1/2 (p-ERK1/2), JNK1/2 (p-JNK1/2) and P38 (p-P38) were analyzed by western blotting.

(C)SMMC-7721 cells were transiently transfected with a control vector (pcDNA3.1), or pcDNA-CYP3A5, or pcDNA-CYP3A5 plus pmyr-AKT. The cells were subjected to the transwell migration assay and the Matrigel invasion assay after 24 hours of transfection. Representative images were shown. Magnification, ×200;

Figure S5.

(A)Four stable cell lines, SMMC-7721-GFP, SMMC-7721-CYP3A5, HCCLM3-GFP, and HCCLM3-CYP3A5 were starved with AZD5363(10uM) or MK-2206 (2uM) DMEM overnight and exposed to HGF (40ng/ml) for 10 minutes. The phosphorylation levels of AKT (p-AKT) were analyzed by western blotting.

(B)Proliferationof indicated cell lines was evaluatedby determining the cell viability withCell Counting Kit 8 Assay accordingto manufacturer’s instructions in the presence or absence of AZD5363 or MK2206.

(C&D) the capacities of migration or invasion of HCCLM3-GFP and HCCLM3-CYP3A5 were examined. Representative results are shown. Magnification: ×200.

(E) Left Panel: the correlation of CYP3A5 level with those of p-AKT (s473) in 159 HCC patients. Correlation between two variables was calculated bySpearman rank correlation coefficient (r); Right Panel: the representative picture of immunochemical staining of serial HCC sections for CYP3A5 and p-AKT (s473).

Figure S6.

(A&B) SMMC-7721-GFP and SMMC-7721-CYP3A5 cells were transfected with siRNA duplexes targeted against human Rictor (si-Rictor) or control siRNA (si-NC). 48h after transfection the indicated cells were subjected to the transwell migration assay (A) and the Matrigel invasion assay (B). Representative images were shown. Magnification, ×200.

(C) Phosphatidic acid concentration in indicated cell lines were measured utilizing Total Phosphatidic Acid Assay Kit.

Figure S7.

(A&B) The intracellular GSH level was examined. left panel: the GSH level in SMMC-7721-CYP3A5 vs. SMMC-7721-GFP cell lines were measured with DMEM medium containing 10% FBS in the presence or absence of DPI; right panel: we calculated the GSH level of serum-starved SMMC-7721-CYP3A5 and GFP cells in the absence or presence of HGF stimulation. DPI was also applied to block the effect of HGF as indicated. Three independent experiments were performed (*: P<0.05; **:P<0.01).

(C)The level of AKT ser473 phosphorylation was examined in HGF-stimulated stable cell lines with or without DPI treatment.

(D)The interaction between Rpl26 and Rictor or mTOR was detected by IP method with specific antibodies against Rictor, mTOR or Rpl26.

(E)The level of AKT ser473 phosphorylation was examined in HGF-stimulated stable cell linesafter siRNA-Rpl7-02 treatment.

Figur S8.

(A)The gene chip data of HCC were obtained from TCGA (The Cancer Genome Atlas) and GEO (Gene Expression Omnibus) databases, and the expression levels of CYP3A4 mRNA in HCC and adjacent non-tumor tissues were statistically analyzed (*: P<0.05 and **: P<0.01).

(B)Expression levels of CYP3A4 mRNA in 60 paired HCC and adjacent non-tumor tissues were evaluated by qRT-PCR.

(C)The effect of cells weretransfected with scrambled smallinterfering RNA (siRNA) or specificsiRNA against CYP3A4 (1#) (40nM) for 48 hours and levels of CYP3A4 protein were determined byWestern blot assay.

(D) Correlation of clinicopathologic characteristics with CYP3A5 expression. Blue Row: the clinic factors with the p value lower than 0.05.

(E)Proliferation of indicated cell lines was evaluated by determining the cell viability with Cell Counting Kit 8(CCK8) Assay according to the manufacturer’s instructions.

(FG) SMMC-7721 andHCCLM3 cells were transfected with siRNA duplexes targeted against human CYP3A4 (si-CYP3A4) or control siRNA (si-NC). 48h after transfection the indicated cells were subjected to the transwell migration assay (F) and the Matrigel invasion assay (G). Representative images were shown. Magnification, ×200;

(H) The intracellular GSH level was examined. Left panel: the GSH content in SMMC-7721-si-CYP3A4 and HCCLM3-si-CYP3A4 vs. SMMC-7721-si-NC and HCCLM3-si-NC cell lines were measured with DMEM medium containing 10% FBS; lower panel: we calculated the GSH level of serum-starved SMMC-7721-si-CYP3A4 and GFP cells in the absence or presence of HGF stimulation. Three independent experiments were performed.