To investigate the expression of FUT8 in LUAD, we performed a meta-analysis based on data from GEO (http://www.ncbi.nlm.nih.gov/geo/). The following search terms were used to retrieve eligible datasets from GEO: (lung OR pulmonary) AND (cancer OR carcinoma OR adenocarcinoma OR tumor OR malignanc^* OR neoplas^*). The inclusion criteria were as follows: 1) human species; 2) FUT8 expression was examined in both LUAD tissues and non-cancerous tissues; 3) at least three samples were included in each group. The pooled standard mean difference (SMD) with a 95% confidence interval (CI) was calculated to assess the overall expression level of FUT8 in LUAD. Chi-square test and I ² statistic were used to evaluate the heterogeneity among included studies. A random-effects model was selected if there was heterogeneity (I ² > 50% or p < .05). Otherwise, a fixed-effects model would be selected (19). The stability of the pooled SMD was assessed by sensitivity analysis. Publication bias was ascertained using Begg’s funnel plot and Egger’s test. A summary receiver operating characteristic (sROC) curve was generated to illustrate the diagnostic ability of FUT8 expression for LUAD.

LUAD RNA-Seq datasets of tumor and nontumor tissues was downloaded from TCGA Data Portal (https://portal.gdc.cancer.gov/). The differences in FUT8 expression between tumor and nontumor groups were analyzed using R language. The prognostic value of FUT8 was analyzed by the meta-analysis tool Kaplan-Meier plotter (20). Univariate Cox regression analysis, Kaplan-Meier survival plot with hazard ratio (HR) and log-rank p-value were calculated and plotted.

A total of 92 patients with primary LUAD who underwent surgical resection at the First Affiliated Hospital of Xi’an Jiaotong University between January 2008 and July 2013 were enrolled in this study for IHC analysis. They had received no prior radiotherapy and/or chemotherapy, and had available paraffin-embedded specimens. The corresponding adjacent normal tissues were collected from 88 of the 92 LUAD patients. We examined the clinicopathologic features as follows: age at surgical resection, sex, stage, lymph node metastases, differentiation, EGFR mutation status, and overall survival. The study was approved by the ethic committee of the First Affiliated Hospital of Xi’an Jiaotong University (No. XJTU1AF2020LSK-199) and informed consent was obtained from every participant before the initiation of any study-related procedure.

The method used for immunostaining was the streptavidin-biotin amplified system. Briefly, after deparaffinization and rehydration, antigen retrieval was done by heating in Tris-EDTA buffer (pH 9.0) at 97°C followed by cooling at room temperature. Slides were then incubated with 1000-fold diluted anti-human FUT8 monoclonal antibody (Proteintech, Wuhan, China) at 4°C overnight. Subsequently, the biotinylated secondary antibody (Vector Laboratories, Burlingame, CA, United States) was applied to all the sections for 30 min at room temperature, and the sections were washed thrice with PBS, incubated with avidin-biotin-peroxidase complex (Vector Laboratories, United States) for 30 min. Following 3-3′-diaminobenzidine color development, slides were counterstained with hematoxylin.

FUT8 expression was evaluated using the intensity of staining (0 = negative, 1 = weak, 2 = intermediate, and 3 = strong) and the proportion of stained tumor cells per slide (0–100%) as previously reported (21). FUT8 immunostaining was considered to indicate high expression when ≥30% of tumor cells with intermediate or strong staining. Cases with negative, weak, and focal staining with 2 or 3 intensity were considered to indicate low expression. The levels of FUT8 were independently evaluated by two pathologists.

Human lung adenocarcinoma cell line A549 was obtained from the Chinese Academy of Sciences Cell Bank (Shanghai, China), and cultured in RPMI-1640 medium (Invitrogen, Carlsbad, CA, United States) supplemented with 10% fetal bovine serum (FBS; Sigma-Aldrich, Darmstadt, Germany) and 1% penicillin/streptomycin (Invitrogen). A549 Cells were transfected with FUT8 siRNA or negative control (NC) siRNA using Lipofectamine 2000 reagent (Invitrogen) according to the manufacturer’s instruction. The sequences were si-FUT8 (sense: 5′-GUG​GAG​UGA​UCC​UGG​AUA​UTT-3′, Anti-sense: 5′-AUA​UCC​AGG​AUC​ACU​CCA​CTT-3′), and si-NC (sense: 5′-UUCUCCGAACGUG UCACGUdTdT-3′, Anti-sense: 5′-ACGUGACACGUUCGGAGAAdTdT-3′). After 48 h, the transfection efficiency was verified via RT-PCR and western blotting.

Total RNA were isolated from harvested cells using TRIzol reagent (Invitrogen). The cDNA was synthesized using PrimeScript RT master mix (Perfect Real Time) (TaKaRa, Shiga, Japan). RT-PCR analysis of FUT8 and β-actin was performed in a StepOnePlus real-time PCR system (Applied Biosystems Inc., Foster City, CA, United States) using SYBR Premix Ex Taq II (Perfect Real Time) (TaKaRa). PCR amplifications were conducted using the following protocol: 95°C for 30 s, for 40 cycles at 95°C for 5s (denaturation step), at 60°C for 30 s (annealing/extension steps). The 2^−ΔΔCT method was used to determine differences in FUT8 expression levels between specimens. The primers used in this work were obtained from Sangon Biotech (Shanghai, China) and their sequences were included in Supplementary Table S1.

After the indicated treatment, cells were lysed in RIPA buffer (Beyotime, China) containing proteinase inhibitor cocktail (Sigma-Aldrich). The Pierce Coomassie Plus colorimetric protein assay (Thermo Fisher Scientific, Waltham, MA, United States) was used to measure the protein concentration, and approximately 30 μg of total protein was separated by 8% SDS-PAGE and transferred onto nitrocellulose. The membrane was blocked in 5% skim milk for 1 h at room temperature and then overnight at 4°C with primary antibodies against FUT8 (Abcam, Waltham, MA, United States) and β-actin (Sigma-Aldrich). After incubated with HRP-conjugated secondary antibody (Bio-Rad, Hercules, CA, United States), the blots were visualized using the Enhanced Chemiluminescence Detection Kit (GE Healthcare).

Cell proliferation was detected by the cell count kit-8 (CCK-8, Dojindo, Japan), according to the manufacturer’s instruction. Briefly, 1 × 10⁴ transfected cells per well were seeded in 96-well plates. After 24, 48, or 72 h of cell growth, 10 μl CCK-8 was added to each well and incubated for 2 h at 37°C. The absorbance value was measured in a microplate reader at 450 nm.

For apoptosis measurement, 1 × 10⁵ transfected cells were seeded in 6-well plates with RPMI-1640 medium containing 10% FBS for 48 h. The cells were then harvested, washed twice with ice-cold phosphate-buffered saline, and re-suspended with binding buffer at a concentration of 1 × 10⁶ cells/ml. Apoptosis was determined by flow cytometry after staining with 7-AAD and annexin V-APC (Keygen Biotech, Nanjing, China).

The relationship between FUT8 expression and clinicopathological characteristics was analyzed using χ ² test for categorical variables and using unpaired t test for continuous variables. Overall survival (OS) was defined as the time interval from date of surgery to the date of death, and was censored at last follow-up. Survival data were estimated by the Kaplan-Meier method and compared using the log-rank test. Univariate and multivariate Cox regression analyses for OS were used to identify the significant prognostic factors in LUAD. p-values of <.05 were considered statistically significant on the basis of 2-sided testing. All statistical analyses were performed using SPSS 22.0 and GraphPad Prism v5.0.

A total of 15 records with FUT8 expression detected in both LUAD samples and control samples were collected in this study. The basic information on these datasets is included in Table 1. The expression of FUT8 in each dataset is presented as scatter plots in Figure 1. Due to the divergences in individual studies, we then combined all of the included datasets. The pooled SMD of FUT8 was 1.40 (95% CI = .95–1.85) by the random-effects model (Figure 2A). The I-squared value was 82.8%, and the p-value was less than .001. Sensitivity analysis suggested that the pooled results were stable (Figure 2B). No publication bias was detected in these studies as indicated by Begg’s funnel plot (p = .235) and Egger’s test (p = .289) (Figure 2C). Simultaneously, sROC curve was also generated to illustrate the ability of FUT8 expression to differentiate LUAD patients from normal controls. As shown in Figure 2D, the area under the curve (AUC) for FUT8 was .95 (95% CI = .93–.97), with a sensitivity and specificity of .72 (95% CI = .62–.80) and .95 (95% CI = .92–.98), respectively.

The data on 533 LUAD cases (533 cancer tissues and 59 normal lung tissues) with FUT8 expression was extracted from TCGA database. As shown in Figure 3A, the expression level of FUT8 in LUAD tissues was significantly elevated compared with normal tissues in TCGA (log₂FC = 1.38, p = 9.78 × 10⁻²⁷), which was consistent with the result of meta-analysis by GEO datasets. ROC curve was generated to assess the ability of FUT8 differentiating LUAD from normal lung tissues (Figure 3B). The AUC was .919 (95% CI = .897–.941, p < .001) with a specificity of 98.3% at a sensitivity of 79.2%.

The Kaplan-Meier plotter was used to evaluate the association between FUT8 expression and survival using GEO and TCGA datasets. A total of 672 LUAD patients with FUT8 expression and survival data were categorized into high expression group and low expression group according to the median of all samples. We found that FUT8 overexpression was correlated with worse OS in LUAD patients (p = .00015; HR = 1.62; 95% CI = 1.26–2.08) (Figure 3C). The median survival of LUAD patients with low FUT8 expression was 136.33 months, while high FUT8 expression cohort had a median survival of 75.43 months.

To verify the results from public databases, we examined FUT8 expression in LUAD tissues and their matched adjacent normal tissues by IHC. As shown in Table 2, the overexpression rate of FUT8 protein was significantly higher in LUAD (52/92, 56.5%) compared to normal tissues (14/88, 15.9%) (p = 1.585 × 10⁻⁸). The representative pictures of FUT8 immunohistochemical expression in LUAD and normal samples are shown in Figure 4. The correlations between FUT8 protein expression and clinicopathological features was further investigated in 92 patients with LUAD. The result showed that FUT8 expression was not associated with age, sex, differentiation, tumor status, clinical stage, lymph node metastases and EGFR mutation (all p > .05; Table 3).

Figure 5A shows the Kaplan-Meier survival curve for OS for high and low FUT8 expression groups. The LUAD patients with low FUT8 expression showed significantly better OS than those with high FUT8 expression (p = .0002). We also assessed the prognostic significance of clinicopathological features for survival of patients with LUAD (Figure 5). The results revealed that clinical stage and lymph node metastasis correlated to OS of LUAD patients. The patients with advanced stage LUAD demonstrated significantly worse prognosis than those with early stage LUAD (p = .000005; Figure 5D). The lymph node-positive patients showed significantly worse OS than those with negative lymph node metastasis (p = .001; Figure 5E).

Univariate analysis revealed that high FUT8 expression (HR = 3.27; 95% CI = 1.70–6.30; p = .0004), advanced stage (HR = 3.45; 95% CI = 1.95–6.08; p = .00002), and positive lymph node metastasis (HR = 2.55; 95% CI = 1.40–4.65; p = .002) were associated with shorter OS (Table 4). After accounting for the effects of age, sex, differentiation, tumor status, stage and lymph node metastasis, multivariate Cox regression analysis indicated that the level of FUT8 expression was an independent predictor of poor OS (HR = 2.87; 95% CI = 1.42–5.77; p = .003; Table 4). Additionally, clinical stage was also an independent prognostic factor (HR = 2.81; 95% CI = 1.36–5.83; p = .005; Table 4).

To determine the biological role of FUT8 in LUAD cell proliferation and apoptosis, we established FUT8-knockdown cell lines (A549) using siRNA. RT-PCR and western blot analysis indicated that the expression of FUT8 was dramatically down-regulated on the mRNA and protein level by FUT8 siRNA (Figure 6A). As shown in Figure 6B, silencing the expression of FUT8 in vitro markedly arrested cell proliferation in LUAD cells. In addition, LUAD cells with down-regulated expression of FUT8 showed a significantly increased percentage of apoptotic cells (p < .001; Figure 6C).