Human liver samples were collected from HCC patients undergoing surgical resection or liver transplantation, and healthy donors for liver transplantation in the form of histologic section at Shulan (Hangzhou) Hospital, China. Protocols for patient tissue collection were reviewed and approved by the ethics committees of Shulan (Hangzhou) Hospital, China (№ 20170207). Written informed consent was obtained. All experiments were done in accordance with the governmental and institutional guidelines of the ICH-GCP (according to the principles of the Declaration of Helsinki) and were performed under the written approval by the First Affiliated Hospital of Zhejiang University, China (№ 2018-327) and the Human Ethics Committee of Shulan (Hangzhou) Hospital. Histopathological grading was performed at the department of pathology in Shulan (Hangzhou) hospitals. All the participants that donated liver tissues are anonymized and summarized in Suppl. Table S1.
Canine protocols were approved by the First Affiliated Hospital of Zhejiang University, College of Medicine, Zhejiang University, China (№ 2018-396) and all procedures were performed in accordance with the institutional guidelines. Ten-months-old male beagle canines were purchased from the ANNIMO Science and Technology and maintained in a specific pathogen-free environment.
AAI (purity > 98%, HPLC, Delta) was mixed with filler and filled into capsules. Canines were randomly assigned into two groups and given capsules with filler AAI filler (3 mg/kg/day ) or control for 10 days. Canines were sacrificed 11 days after initiation of the treatment. Livers were excised immediately after sacrifice. Part of the liver was fixed in 4% (wt/vol) neutral buffered formalin (pH 7.4) and embedded in paraffin for histologic analyses; the remaining liver was immediately snap-frozen in liquid nitrogen and kept at -80℃ until use.
Immunohistochemistry: Anti-FLAP (sc-28815), mGST2 (sc-65130), LTC4S (sc-22564) were purchased from Santa Cruz Biotechnology (California, USA). CysLTR2 (ab32536) was purchased from Multisciences Biotech (Hangzhou, China). CYLD (#8462) was purchased from Cell Signaling Technology (Danvers, USA). 4′,6′-diamidino-2-phenylindole (DAPI) (ZLI-9557) was purchased from Zhongshan Jinqiao Biotechnology (Beijing, China).
Western blot: Anti-FLAP, mGST2, LTC4S, CysLTR2 and CYLD were the same as mentioned above. 5-LO (sc-8885) was purchased from Santa Cruz Biotechnology. LTB4 receptor 1 (BLT1) (ab131041) and BLT2 (ab84600), were purchased from Abcam (Cambridge, England). CysLTR1 (ab32534) was purchased from Multisciences Biotech (Shanghai, China). Antibodies of c-Jun N-terminal kinase (JNK) (#9252), phospho-JNK (p-JNK) (#9251), eukaryotic translation initiation factor 2 alpha (eIF2a) (#2103), p-eIF2a (#3398), 78-kDa glucose-regulated protein (GRP78) (#3177), and 94 kDa Glucose-Regulated Protein (GRP94) (#2104) were purchased from Cell Signaling Technology. Microsomal prostaglandin E synthase-1 (mPGES-1) (#160140) was purchased from Cayman Chemical (Ann Arbor, USA). β-Actin (Mab1445), GAPDH (Mab5465-040), α-Tubulin (ab36864), β-Tubulin (ab012) and the HRP-conjugated secondary antibodies (goat anti-mouse, LK-GAM007; goat anti-rabbit, LK-GAR007; and rabbit anti-goat, LK-RAG007) were purchased from Multisciences Biotech.
Histological assessments of liver tissues
Human liver paraffin-embedded consecutive 4 μm-thick sections were performed for immunohistochemical analysis. Anti-Ki-67 (IR098) and CD34 (IM034) were purchased from LBP (Guangzhou, China). Canine liver sections were prepared as described previously  for hematoxylin and eosin (H&E) or immunohistochemical stains, and consecutive 8 µm-thick frozen sections (cut by a freezing microtome, CM1950, Leica, Wetzlar, Germany) for immunofluorescent staining. Immunofluorescence analysis was performed using confocal microscopy (TCS SP8 MP, Leica). Both human and canine liver sections were stained with indicating antibodies to distinguish the immunohistochemical or immunofluorescence changes in expression and distribution in situ. For Ki-67 tissue evaluation, section was graded based on the percentage of Ki-67 positively stained nuclei, using the range 0~100 %.
Western blot analysis and ELISA
Western blot assay was performed as described previously . Briefly, samples were homogenized with RIPA buffer (P0013K, Beyotime, Shanghai, China) for protein extraction. Whole blotting gel quantified with a fluorescence scanner imaging system (Bio-Rad, Hercules, California, USA) and densitometry was performed using Image J software (https://imagej.nih.gov/ij, NIH, Bethesda, MD, USA).
Supernatant containing CysLTs was extracted from liver tissue homogenate of canines (1:10 W:V, 50 mmol/L Tris-HCl, pH=7.5). CysLTs level was measured by ELISA Kit (Cayman, USA) on a DTX 880 Multimode Detector (Beckman Coulter, USA).
Bioinformatics analysis of gene transcriptions
MRNA sequencing data of HCC tissues and clinical follow-up record of HCC cohort were downloaded from The Cancer Genome Atlas (TCGA) in accordance with the data usage policy. MRNA expression of indicating genes was retrieved . Patients were allocated into dichotomized subgroups based on processed expression for the survival analysis. Kaplan-Meier plots were used for comparison of survival curves, and log-rank tests were applied for P-values calculation.
Real-time qRT-PCR analysis
Real-time qRT-PCR analysis was conducted according to the protocol . Briefly, total RNAs from human HCC cell lines were isolated with TRIzol reagent (Gibco BRL, USA). After 2 min of initial denaturation at 95°C, amplification used 40 cycles following 15 s at 95°C and 1 min at 60°C. The primer sequences (5’-3’) for FLAP, Lta4h (LTA4 hydrolase) and GAPDH (in human HCC cells) were documented in Supplementary Table S2.
Cell culture, genetic-handling of FLAP in human HCC cells
The human HCC cell line HepG2, Hep3B2.1-7 and PLC/PRF/5 were purchased from Shanghai Institute of Cell Bank, Chinese Academy of Sciences (№ 22008). Cells were maintained in DMEM (12800-017, Life Technologies, USA) supplemented with 10% fetal bovine serum (FBS; 10099-141, Gibco, Germany).
Overexpression of FLAP in HepG2 cells was achieved by lentiviral infection. Lentivirus (lv) were constructed, concentrated and purified by Genechem (Shanghai, China). HepG2 cells were infected with lv-Control (lv-Con) or lv-FLAP (HepG2lv-Con, HepG2lv-FLAP) (MOI=10), and then selected for 3 days in the presence of 2.5 μg/ml puromycin (Gibco, Germany). Knockdown of FLAP in HepG2 cells was conducted by short hairpin RNA (shRNA). Sh-FLAP varietas and sh-Con plasmids (Sangon Biotech, Shanghai, China) were transfected with Lipofectamine 2000 transfection agent (HepG2sh-FLAP, HepG2sh-Con) (Invitrogen, Life Technologies) according to the manufacturer’s instructions. Sequences for shRNA were shown in Supplementary Table S3. Lv-RNA or shRNA interfering efficiency was validated by qRT-PCR.
HepG2lv-FLAP and HepG2sh-FLAP, and wild type (WT), HepG2lv-Con or HepG2sh-Con cells were further incubated with AAI (final concentration 1.25 μM) for another 7 days. Cells were harvested on day 10 after initiation of infection or transfection. The expression of FLAP in cells was examined by Flow cytometry according to commercial instruction. Finally, cells were analyzed by a BD FACSCount II Flow Cytometer (BD Biosciences, San Jose, USA).
Pairwise comparisons between continuous data were analyzed using an unpaired two-tailed Student t test, and multiple comparisons were analyzed by one-way ANOVA. All data were expressed as mean ± SD, and P < 0.05 were considered statistically significant. Significant P values are indicated by asterisks in the individual figures.