The expression profiles of 33 distinct tumor types were analyzed, as cataloged by the Cancer Genome Atlas (TCGA, https://www.cancer.gov/ccg/research/genome-sequencing/tcga) initiative. It is important to note that this research adhered to the guidelines outlined by TCGA, which resulted in the waiver of ethical approval and the need for informed consent from the patients involved.

The mRNA expression of IFITM10 were compared between normal tissues and tumor tissues (data were acquired from TCGA, https://www.cancer.gov/ccg/research/genome-sequencing/tcga).

GSEA method in conjunction with the “clusterProfiler” package (version 3.17.0) were performed to run the set enrichment analysis [18]. The analysis involved the utilization of Hallmark MSigDB gene sets as the reference gene sets for the enrichment analysis.

In this study, all animal procedure was approved by the Ethics committee of Chongqing University of Arts and Sciences (CQWU2022001 and 14/03/2022). BALB/c-Nude mice (6–8 weeks) were bought from Huafukang Bio-Technology (Beijing, China). Xenografts were constructed by subcutaneously injecting with 1 $\times$ 10⁶ indicated cells, including HCT-116-shNC, HCT-116-shIFITM10#1 and HCT-116-shIFITM10#2 cells. After 28 days, mice were sacrificed by cervical dislocation and xenografts were collected and analyzed.

HEK293 were purchased from Procell Life Science&Technology Co., Ltd. (CL-0001, Wuhan, China), SW-480 and HCT-116 cells were obtained from Nanjing Cobioer Biosciences Co., Ltd. (Nanjing, China). HEK293, SW-480, NCM356, Lovo, RKO cells were cultured in DMEM (PM150210A, Pricella, Wuhan, China) medium with 10% fetal bovine serum (FBS, 1645210, Pricella, Wuhan, China) under 37 °C and 5% CO₂ conditions, HCT-116 were cultured in McCoy’s-5A (PM150710, Pricella, Wuhan, China) with 10% FBS under 37 °C and 5% CO₂ conditions. Human Umbilical Vein Endothelial Cells (HUVECs) (CTCC-0804-PC, Meisen, Jinhua, China) were cultured in Endothelial Cell Growth Medium 2 (CTCC-002-031, Meisen, Jinhua, China) under 37 °C and 5% CO₂ conditions. For all the cell lines that were used in this study has been validated through STR profiling. Meanwhile, no mycoplasma was tested out among all cell lines.

The IFITM10 shRNA was applied to inhibit the IFITM10 expression in HCT-116 and SW480 cell line. We used the following sequences: shIFITM10#1 (5${}^{\prime }$-GGGAACCTGTCACAGACAATGTTCAAGAGACATTGTCTGTGACAGGTTCCC-3${}^{\prime }$), shIFITM10#2 (5${}^{\prime }$-CATTGGTGATCATCTTCAATCTTCAAGAGAGATTGAAGATGATCACCAATG-3${}^{\prime }$). The HEK293 cells were used to package lentiviral vector expressing a IFITM10 shRNA or a control shRNA. HCT-116 and SW480 cells were inoculated into six-well culture plates. The culture medium was replaced by lentiviral particles for 24 hours. Subsequently, puromycin (540411, Merck, Nantong, China) were used to select the stable integration 2 days.

For protein analysis preparation, the cells were lysed utilizing a commercial lysis kit (P0013B, Beyotime, Shanghai, China) supplemented with a protease cocktail inhibitor (HY-K0013, MCE, Shanghai, China). Protein levels were quantified by BCA assay (P0009, Beyotime, Shanghai, China). Proteins were separated by sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and then were transferred onto polyvinylidene difluoride (PVDF) membranes (IPVH00010, EMD Millipore, Shenzhen, China). The PVDF membranes were then blocked using 5% non-fat milk in TBS buffer at room temperature (RT) for 30 minutes. Following this, the membranes were incubated overnight with primary antibodies, including anti-IFITM10 (1:1000, A22196, ABclonal, Wuhan, China), anti-p-STAT3 (1:1000, AP0070, ABclonal, Wuhan, China), anti-STAT3 (1:1000, 60199-1-Ig, Proteintech, Wuhan, China), anti-VEGF (1:1000, A12303, ABclonal, Wuhan, China) and anti-GAPDH (1:2000, GB11002, Servicebio, Wuhan, China). After three washes with TBST, the membranes were further incubated with secondary antibody (the HRP Goat Anti-Rabbit IgG (1:5000, AS014, ABclonal, Wuhan, China) and Anti-Mouse IgG (1:5000, AS003, ABclonal, Wuhan, China) at RT for 1.5 hour. Finally, the blots were imaged using Tanon 5200 Multi imager (Tanon, Shanghai, China).

Tube formation was performed as previously described [19]. In brief, 50 µL Matrigel (BD Biosciences, FranklinLakes, NJ, USA) was added into a prechilled 96-well plate before being placed in a 37 °C CO₂ incubator for 30 minutes. P5 HUVECs were seeded on Matrigel with 2 $\times$ 10⁴ cells per well in 100 µL of culture medium which was mixed with 50 µL of conditioned medium from clutuerd HCT-116 cells. After 24 hours incubation, phase contrast photos of each well were taken by microscope (IX53/DP80, Olympus Corporation, Tokyo, Japan). The number of junctions was quantified by ImageJ1.46r (NIH, Bethesda, MD, USA).

HCT-116 cells with or without IFITM10 inhibition were seeded at a density of 5 $\times$ 10³ cells per well in 96-well plates. After indicated time of incubation, 10 µL of CCK-8 solution (HY-K0301, MCE, Shanghai, China) was added to each well. Following a 2-hour incubation period, the absorbance at 450 nm was measured using a microplate reader (Agilent BioTek Cytation 5, Winooski, VT, USA).

CRC cells (with or without IFITM10 inhibition) were seeded onto coverslips and subsequently fixed with 4% Paraformaldehyde. Following fixation, the cells were blocked with 1% FBS for 1 hour, and incubated with indicated primary antibodies (anti-IFITM10, 1:200, ABclonal, Wuhan, China; anti-STAT3, 1:200, Proteintech, Wuhan, China) at 4 °C overnight. After the primary antibody incubation, the cells were treated with secondary antibodies (Beyotine, Shanghai, China) and incubated for 1 hour. Finally, the cell nuclei were stained with DAPI (RM02978, ABclonal, Wuhan, China) for 15 minutes at room temperature. Cells were imaged using a confocal laser scanning microscope (BX53, Olympus Corporation, Tokyo, Japan).

Total RNA was extracted using RNA Easy Fast Tissue/Cell kit (DP451, TIANGEN, Beijing, China). The synthesized cDNA was obtained using ReverAid First Strand cDNA Synthesis Kit (K1622, Thermo scientific, Shanghai, China). cDNA was then mixed with Genious 2X SYBR Green Fast qPCR Mix (RK21206, Abclonal Technology, Wuhan, China) and the following primers: VEGFA forward: 5${}^{\prime }$-CCTGGCTTTCTGCTGTACCT-3${}^{\prime }$, reverse: 5${}^{\prime }$-GCTGGTAGACGTCCATGAACT-3${}^{\prime }$; $\beta$-actin forward: 5${}^{\prime }$-CACTGTCGAGTCGCGTCC-3${}^{\prime }$, reverse: 5${}^{\prime }$-TCATCCATGGCGAACTGGTG-3${}^{\prime }$. To quantify the mRNA levels of the specific genes, ABI 7300 QuantStudio3 PCR (RT-PCR) System (Thermo Fisher, Shanghai, China) was used for mRNA quantify.

Kaplan-Meier analysis (log-rank test) was performed to examine the correlation between the expression level of IFITM10 and indicated outcomes in the TCGA dataset, including overall survival (OS), progression-free interval (PFI), and disease-specific survival (DSS). Survival curves with a significance threshold of *p 0.05. Furthermore, receiver operating characteristic (ROC) curves in tumors were conducted to reveal the influence of IFITM10 on prognosis.

Statistical analyses were conducted using Prism V.9.0 (GraphPad Software, Inc., San Diego, CA, USA). Data are presented as mean $\pm$ SD (unless indicated). T test, One-way ANOVA, or Two-way ANOVA was used for data process (unless indicated). Statistical significance was defined as *p 0.05, **p 0.01, ***p 0.001, ****p 0.0001.

We analyzed the levels of IFITM10 mRNA in various types of cancer using data from the TCGA database (Fig. 1A). Among the paired 23 tumor types examined, IFITM10 was found to be significantly overexpressed in 8 of them (Fig. 1B). Interestingly, it is worth mentioning that only the prostate adenocarcinoma (PRAD) exhibited a significant decline in IFITM10 expression. Additionally, when comparing unmatched groups (Fig. 1C) and paired samples from the same patients (Fig. 1D), we observed that the expression of IFITM10 was considerably higher in CRC tumors compared to pericancerous tissues. Further, colorectal cancer cell lines and normal colon gland cells were applied to investigate the protein levels of IFITM10. Consistent with previous results, IFITM10 were higher in CRC cells compared to normal colon gland cell line (Fig. 1E). These results support the notion that IFITM10 may play a crucial role in tumorigenesis and tumor development of colorectal cancer.

Fig. 1.

Interferon-$\gamma$ inducible protein 10 (IFITM10) is overexpressed in altered carcinomas. (A,B) In The Cancer Genome Atlas (TCGA) database, mRNA expression of IFITM10 was analyzed in 33 tumors and in paired samples of 18 tumors. (C,D) IFITM10 expression in colorectal cancer tumors compared to peri-carcinous tissue in both the paired and unpaired experimental groups. (E) Expression level of IFITM10 protein in colorectal cancer cells and normal colon gland cells. ns, p $\ge$ 0.05; *p 0.05; **p 0.01; ***p 0.001.

We found that IFITM10 is upregulated in CRC, however, little was known about the role of IFITM10 in CRC. Next, we performed GSEA with a public dataset and found that upregulation of IFITM10 was associated with angiogenesis (Fig. 2A). We further utilized IFITM10-targeted shRNA to knockdown IFITM10 in CRC cells (HCT-116 and SW480). Western blot showed the efficacy of IFITM10-1 knockdown (Fig. 2B). A mouse transplanted tumor model was then constructed using indicated cells with 5 mice per group, tumor growth curves was depicted and data showed slower tumor growth in the shIFITM10 groups, compared to control group (Fig. 2C). After 28 days, the mice were executed, and xenografts were collected and weighed. Xenografts formed by shIFITM10 cells are smaller, pale in the appearance, and lighter in weight (Fig. 2D,E). IHC staining showed that IFITM10 knockdown inhibited VEGFA expression in xenograft (Fig. 2F). CD31 immunohistochemical staining showed a decrease in tumor vascular density after IFITM10 inhibition (Fig. 2G). Furthermore, we detected the expression level of VEGFA in indicated xenografts. And data showed that VEGFA were reduced while IFITM10 was inhibited (Fig. 2H).

Fig. 2.

Inhibition of IFITM10 restrained tumor-induced angiogenesis in vivo. (A) Gene Set Enrichment Analysis (GSEA) indicated enrichment of IFITM10-associated Differentially Expressed Genes (DEGs) in the angiogenesis signaling pathway. (B) HCT-116 and SW480 cells were transfected with indicated shRNA, and the IFITM10 protein expression was analyzed. (C) The growth curve of mice xenografts formed by shCON, shIFITM10#1 and shIFITM10#2 cells (n = 5). (D) Mice xenografts formed by HCT-116 with or without IFITM10 intervention. (E) Tumor weights after sacrificing the mice (n = 5). (F) Immunohistochemistry (IHC) staining of xenografts using antibodies against IFITM10, VEGFA, and CD31. Scale bar: 100 µm (G) Microvascular densities of the xenografts in indicated groups, quantified by CD31 (n = 5). (H) Western blot results of indicated xenografts. ****p 0.0001. VEGF, vascular endothelial growth factor; VEGFA, vascular endothelial growth factor A; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; shNC, control group; shIFITM10, IFITM10 interference.

We next explored the role of IFITM10 during angiogenesis in vitro. The single gene co-expression heat map (from the TCGA database) revealed a positive correlation between IFITM10 and VEGFA (Fig. 3A). Meanwhile, we detected VEGFA secretion with IFITM10 inhibition by ELISA assay. Results showed that IFITM10 knockdown reduced VEGFA secretion levels in both HCT116 and SW480 cells (Fig. 3B,C). Then, tube formation test was performed using HUVECs, which were incubated in specified conditioned medium (CMs). Results showed that less tubular formation was observed in the shIFITM10 group (Fig. 3D–F). In addition, CCK8 assay showed that, the proliferation capacity of HUVECs were restrained with IFITM10 inhibition CMs compared to control CMs (Fig. 3G,H). Taken together, inhibition of IFITM10 restrained tumor-induced angiogenesis both in vivo and in vitro.

Fig. 3.

Inhibition of IFITM10 restrained tumor-induced angiogenesis in vitro. (A) The single-gene co-expression heatmap in the TCGA database shows a positive correlation between the expression of IFITM10 and VEGF in CRC. (B,C) Detection of VEGFA concentration of conditioned medium obtained from indicated cells by enzyme-linked immunosorbent assay (ELISA) (n = 3). (D–F) Tube-formation assay was performed with Human Umbilical Vein Endothelial Cells (HUVECs) incubating with indicated conditioned mediums (CMs) (n = 3). (G,H) The proliferation capacity of HUVECs detected by CCK-8 assay. Scale bar: 250 µm. **p 0.01; ***p 0.001; ****p 0.0001, statistical analysis was performed with one-way ANOVA with Dunn’s test for multiple comparisons.

To further investigated the molecular mechanism of IFITM10 during angiogenesis, GSEA enrichment analysis was performed. Results showed that IFITM10 upregulation may trigger IL-6/STAT3 signaling (Fig. 4A). Next, we found a positive correlation between IFITM10 and STAT3 with single-gene co-expression analysis (Fig. 4B). Thus, we hypothesized that IFITM10 may activate angiogenesis process through STAT3 activation. To validate this hypothesis, western blot assays were performed. Our data showed that with IFITM10 intervention, the phosphorylation of STAT3 was impaired (Fig. 4C). The extent of nuclear translocation of STAT3 reflects its functional status and its transcriptional activity. Next, we conducted a protein fractionation assay to separate nuclear and cytosolic proteins and found that IFITM10 facilitated the movement of STAT3 from the cytoplasm into the nucleus (Fig. 4D). Furthermore, an immunofluorescence assay demonstrated that IFITM10 knockdown resulted in a reduced aggregation of STAT3 in the nucleus (Fig. 4E). Together, these results indicate that IFITM10 promotes the activation of STAT3 signaling pathway.

Fig. 4.

IFITM10 promotes the activation of STAT3 signaling pathway. (A) GSEA analysis indicated an association between IFITM10 expression and activation of the IL-6/JAK/STAT3 pathway. (B) The single gene co-expression heat map revealed significant relationships between IFITM10 and STAT3. (C) Western blotting was used to detect changes in STAT3 phosphorylation upon IFITM10 intervention in HT-116 and SW480 cell lines. (D) Nuclear and cytoplasmic proteins were separated and analyzed by western blotting after IFITM10 intervention in HT-116 and SW480 cell lines. (E) An immunofluorescence assay was conducted to assess changes in the nuclear localization of STAT3 in response to IFITM10 intervention. Scale bar: 25 µm.

The transcriptional activity of STAT3 plays critical roles in both VEGFA secretion and angiogenesis. We have showed that IFITM10 promoted angiogenesis, VEGFA expression, and STAT3 activation. We then determined whether IFITM10-mediated angiogenesis was dependent on the STAT3 activation. By western blot assay, we found that IFITM10 overexpression increased STAT3 phosphorylation and upregulated VEGFA, but those phenomena were reversed with STATTIC application (a pharmacological inhibitor of STAT3 activation) (Fig. 5A). Further, consistent results were obtained through VEGFA secretion detection by ELISA assay (Fig. 5B,C). We next incubated HUVECs with indicated CMs. Results showed that, compared to control group, culturing with CMs that obtained from IFITM10-overexpressing cells, HUVECs exhibited better migratory ability and this effect was inhibited with STAT3 inhibition (Fig. 5D–F). Therefore, we concluded that the angiogenic function of IFITM10 is mediated by STAT3 activation.

Fig. 5.

IFITM10 mediated angiogenesis depend on STAT3 activation. (A) The expression level of indicated targets with/without IFITM10 intervene or STATTIC. (B,C) The section level of VEGFA was detected by ELISA assay. (D–F) Tube-formation assay was performed with HUVECs incubating with indicated CMs through indicated treatment (n = 3). Scale bar: 250 µm, **p 0.01, ***p 0.001, ****p 0.0001, statistical analysis was performed with one-way ANOVA with Dunn’s test for multiple comparisons.

To evaluate the diagnostic and prognostic significance of IFITM10 in CRC, we conducted several analyses. Firstly, we utilized receiver operating characteristic (ROC) curves to assess the ability of IFITM10 to differentiate CRC diagnosis. The results showed that IFITM10 had a sensitivity and specificity for diagnosing CRC, as indicated by the area under the curve (AUC) of 0.622 (Fig. 6A). Additionally, we performed Kaplan-Meier analysis to determine the predictive value of IFITM10 for clinical outcomes. The analysis revealed that high IFITM10 expression was associated with worse overall survival (hazard ratio [HR]: 2.18, p 0.001), disease-specific survival (HR: 2.83, p 0.001), and progression-free interval (HR: 1.58, p 0.001) compared to the low IFITM10 group (Fig. 6B–D). These findings suggest that elevated IFITM10 expression is indicative of an unfavorable prognosis in CRC patients. These findings emphasize the potential utility of IFITM10 as a biomarker for the diagnosis and prognosis of CRC.

Fig. 6.

Predictive value of IFITM10 in colorectal cancer (CRC) diagnosis and prognosis. (A) Receiver operating characteristic curve analysis evaluated the diagnostic performance of IFITM10 in CRC. (B–D) Kaplan-Meier analyses compared overall survival, disease-specific survival, and progression-free interval between CRC patients with high and low IFITM10 expression.