Introduction
The FAT cadherin family, which includes FAT1, FAT2, FAT3, and FAT4, is a large atypical calmodulin superfamily in mammals [
1].
Fat4 knockout mice die at birth and exhibit defective PCP in tissues including the kidney, inner ear, neural tube [
2], and cerebral cortex [
3]. FAT4 is the closest vertebrate homolog of
Drosophila Fat (ft) and plays a role in controlling planar cell polarity (PCP) [
4] and regulating the Hippo signaling [
5]. FAT4 protein is located to the cell membrane in mammalian cells and is thought to have tumor-suppressive effects; double allele FAT4 inactivation promotes tumorigenicity in mouse mammary epithelial cell lines [
6]; however, its role in tumor immunity is unknown.
Aberrant activation of the Wnt/β-catenin pathway drives the progression of a variety of human malignancies, including cervical cancer [
7,
8]. Cadherin-related proteins interact with β-catenin in normal epithelial cells and sequester it in the cell periphery, regulating epithelial cell development and intracellular adhesion [
9,
10]. In the Wnt/β-catenin pathway, β-catenin is a key effector controlled by Wnt ligands and regulates gene transcription. Wnt ligands activate cytoplasmic protein disheveled (DVL), which inhibits degradation complexes (AXIN, GSK3, CK1, and APC). Non-phospho-β-catenin then translocates into the nucleus and interacts with T cell-specific factor (TCF)/lymphoid enhancer-binding factor (LEF) transcription factors to activate Wnt target genes [
11,
12]. Notably, β-catenin has recently been identified as a transcription factor for the
CD274 and
STT3 gene that promotes the N-glycosylation and stabilization of PD-L1 [
13,
14]. The tumor microenvironment (TME) is a source of both standard and non-standard Wnt ligands that can induce aberrant activation of Wnt signaling in cancer cells, leading to epithelial-mesenchymal transition (EMT) and altered immune responses [
15,
16]. Furthermore, β-catenin activation modulates regulatory T cells (Treg) survival as well as tumor cell escape from immune surveillance [
13,
17]. Since the Wnt/β-catenin pathway is essential for cancer cell survival and immune editing, it is a profitable target for antitumor immunity.
Immune checkpoint blockade (ICB) therapies have shown to be effective in metastatic melanoma and non-small cell lung cancer [
18,
19], and clinical trials in cervical cancer and other cancers are currently underway [
20], however drug resistance and recurrence remain common [
21]. The programmed cell death 1 (PD-1) receptor is found on activated T cells, and its ligand PD-L1, is a major co-inhibitory checkpoint signal used by tumors to block the cytotoxic T lymphocyte (CTL) activity, promote T cell apoptosis, and allow cancer cells to escape immune surveillance [
22]. Recent research has discovered that tumor PD-L1 expression levels can be used to predict clinical response to anti-PD-L1/PD-1 therapy [
23]. The extracellular domain of PD-L1 has four N-X-T/S motifs (N35, N192, N200, and N219), all of which can be N-glycosylated [
24]. N-glycosylation occurs mainly in the endoplasmic reticulum (ER) and Golgi apparatus, where glycosyltransferases transfer N-acetylglucosamine to the N-X-T/S motif in the ER lumen, followed by PD-L1 transport to the Golgi apparatus for further maturation and folding by glycosidases, which translocate to the cell membrane to promote immune escape [
25,
26]. Importantly, the oligosaccharyltransferase complex (OST) catalytic subunit STT3 is required for PD-L1 glycosylation and protein stabilization, while β-catenin binding to the transcription factor TCF7L2 is essential for the induction of STT3A/B transcription[
14]. Glycosylation stabilizes PD-L1 and protects PD-L1 from GSK3β-mediated 26S proteasome-dependent ubiquitination degradation [
24]. In general, glycosylation and ubiquitination modifications of PD-L1 can cross-regulate each other and promote PD-L1 stability. As a result, expanding our understanding of the regulation of PD-L1 expression is critical for developing anti-PD-L1/PD-1 therapy and promoting cancer immunotherapy [
27].
In this study, we discovered that FAT4 inhibited Wnt/β-catenin signaling by antagonizing nuclear localization of β-catenin and promoting phosphorylation and degradation of β-catenin by the “destruction degradation complex (AXIN1, APC, GSK3, CK1)”. Furthermore, FAT4 inhibits PD-L1 and STT3A transcription in a β-catenin-dependent manner and induces aberrant PD-L1 glycosylation; specifically, FAT4 overexpression inhibited PD-L1 ER-Golgi transport and cell membrane localization, attenuated STT3A-PD-L1 interaction, and thus induced polyubiquitination-dependent degradation of PD-L1. Importantly, FAT4 overexpression was able to promote complete tumor regression in C57BL/6 mouse models of cervical cancer, and we collected tumors prior to tumor regression and found that FAT4 overexpression promoted cytotoxic T lymphocyte (CTL) infiltration and activation. Our findings suggest that FAT4 enhances antitumor immune responses by inhibiting the β-catenin/STT3/PD-L1 signaling pathway and that FAT4 is promising as a new target for combination immunotherapy.
Methods
Clinical specimens
All immunohistochemical specimens, including 50 pairs of cervical squamous cell carcinoma and paracancerous tissues and 20 pairs of normal cervical specimens, were obtained from patients who underwent surgical resection in the Second Hospital of Jilin University from 2013 to 2017, and all patients did not receive radiotherapy or chemotherapy before operation. In addition, for immunoblotting, we collected 12 pairs of fresh surgically removed cervical squamous cell carcinoma and paraneoplastic tissues. The study was authorized by the Institute Research Ethics Committee (No.2020-146) before written consent from all patients.
Datasets involved were downloaded from the TCGA (The Cancer Genome Atlas) Dataset of CESC (Cervical squamous cell carcinoma and endocervical adenocarcinoma).
Cell culture
Human cervical cancer cell lines Hela, Caski, SiHa, C33A, and ME180 were obtained from the Chinese Academy of Sciences Shanghai Institute for Biological Sciences Cell Resource Center. Mouse cervical cancer cell line U14 was obtained from the National Infrastructure of Cell Line Resource (Beijing). Hela, Caski, SiHa cells were cultured in Roswell Park Memorial Institute (RPMI) 1640 medium (SH30255.FS, Hyclone, Logan, UT, USA), C33A and U14 cells were cultured in Dulbecco modified essential medium (DMEM, SH30285. FS, Hyclone), while ME180 cells were cultured in McCoy’s 5A medium (SH30270.01, Hyclone). All medium was supplemented with 10% fetal bovine serum (FBS, FB15015, Clark), 100 U/ml penicillin, and streptomycin antibiotics (P1400, Solarbio, Beijing, China) at 37℃ in 5% CO2. All cell lines have been tested for mycoplasma contamination and were validated by short tandem repeat (STR) polymorphism analysis performed by the Genetica DNA Laboratories.
Vectors
The deficient Cas9-synergistic activation mediator (dCas9-SAM) system was used to endogenously enhance FAT4/Fat4 mRNA expression. The dCas9-VP64-Puro was used to express wild-type dCas9, and sgRNA-MS2-P65-HSF1-Neo was used to target the FAT4/Fat4 promoter region. For sgRNA sequences: sgFAT4 #H1: GGCTGTAGGCGGTCTGGTGT; #H2: TAGCATCCCGAGAAGCCAGT; sgFat4 #M1: GATTATGCAGCTGACTGCCA; sgFat4 #M2: GTGGAAGAGAACATTGGAGA.
Establishment of FAT4 overexpressing cell lines
The day before lentivirus transduction, the specified cell lines were grown in 6-well plates at a density of 5 × 104 cells/well (20–30% density). The media containing the dCas9-VP64-Puro lentivirus (MOI = 10) was cultured for another 12 h. 48 h later, the appropriate concentration of puromycin was added to the cell cultures to obtain a mixed clonal cell line stably expressing dCAS9-VP64. And then one or more sgRNA-MS2-P65-HSF1 lentiviruses were infected, G418 screened, and western blotting and qPCR were performed to identify the overexpressed clones.
Mice
This study was approved and reviewed by the Institutional Animal Care and Use Committee (IACUC) of Jilin University (KT202102006). Mice used in all experiments were obtained from Charles River, and the animals’ health and immune status were normal. Animal facilities that house mice are regularly checked for standard pathogens, and mice are housed on a 12/12-h light/dark cycle with a maximum of 5 per cage, with free access to food and water. For the immunodeficient mouse models, sgFAT4 human cervical cancer ME180 cells, sgFat4 mouse cervical cancer U14 cells, and their CTRL groups (1 × 106 cells suspended in 50 µl PBS, 1:1 mixed with Matrigel Matrix) were injected into the left dorsal region of 6-week-old female BALB/c nude mice (Charles River).
For the immune-competent mouse models, sgFat4 mouse cervical cancer U14 cells and CTRL group (1 × 106 cells suspended in 50 µl PBS, 1:1 mixed with Matrigel Matrix) were injected subcutaneously into the left dorsal region of 6-week-old female C57BL/6 female mice (Charles River). Tumor diameters were measured and recorded every 3 days. Tumor volume (mm3) was calculated by measuring the longest diameter and shortest diameter of the tumor: Volume = (shortest diameter)2 × longest diameter × 0.5. After 2 weeks, 5 mice were randomly selected for euthanasia, and tumors were collected, weighed, and processed to prepare frozen sections, flow cytometry, and paraffin-embedded tissue. The remaining mice were used to monitor and record survival to 60–90 days. Animals were euthanized in advance when the tumor growth exceeded 10% of the original body weight of the animal, the average tumor diameter exceeded 20 mm (not more than 2000m3), or the tumor metastasized or rapidly grew to ulceration, resulting in infection or necrosis and other adverse indicators.
FACS analysis of CTL profiles
All flow cytometry antibodies used in this study were listed in Supplementary Table
1. U14 xenograft tumors were exfoliated quickly and gently, and single-cell suspensions were produced after physical grinding and filtering. After blocking with Anti-Mouse CD16/CD32 (553140, BD Biosciences) antibody, the cells were incubated in a medium containing PMA/Ionomycin mixture (1x) and BFA/Monensin Mixture (1x) for 5 h at 37 °C and protected from light. Cells were stained with CD45-Brilliant Violet 510™ (30-F11), CD3-PE (17A2), CD4-FITC (GK1.5), CD8-FITC (53-6.7), PD-1-APC (29F.1A12) in FACS Buffer. After fixation and permeabilization by Foxp3/Transcription Factor Staining Buffer Set (00-5521-00, eBioscience, Thermo Scientific), intracellular GZMB and IFN-γ were stained using IFN-γ-PECY7 (4S.B3), Granzyme B-APC (QA16A02) antibody. Stained cells were analyzed by BD FACSCanto II (BD Biosciences) cytometer. Data were processed by the FlowJo 10.8 software.
Immunofluorescence
Fresh mouse tumor mass was isolated and embedded in OCT blocks frozen to prepare 6 mm thick frozen sections. Cells were inoculated on coverslips at the bottom of a 48-well plate (2–4 × 104 cells/well). Cells and tissues were fixed in 4% paraformaldehyde for 30 min at room temperature (RT) and then permeabilized in 0.5% Triton X-100 solution for 10 min. Non-specific sites were blocked with 5% bovine serum albumin for 30 min at RT and incubated with primary antibody overnight at 4 °C. After washing with PBS, CoraLite488 or 594 secondary antibodies (Proteintech) were incubated for 1 h at RT, then stained with Hoechst (Solarbio) for 10 min at RT. Coverslips with the anti-fading solution were mounted on glass slides and observed under a Zeiss LSM 880 confocal microscope.
Cells or tissues are lysed in lysis buffer and the supernatant fractions are collected. SDS-PAGE is used to separate proteins, which are then transferred to polyvinylidene difluoride (PVDF) membranes. The PVDF membranes were blocked in 5% skimmed milk for 1 h at RT before being treated with specified primary antibodies overnight at 4 °C. The membranes were then incubated with horseradish peroxidase‐conjugated secondary antibodies. For immunoprecipitations, the specified antibody was precleared with protein A/G beads (MedChemExpress, Shanghai, China) for 2 h and then rotational incubated with cell lysates overnight at 4 °C. The protein A/G beads were rinsed four times with phosphate-buffered saline containing Tween 20 after incubation (PBST), then the samples were analyzed by immunoblotting. Enhanced chemiluminescence was used to detect immunoreactive bands. Supplementary Table
1 has detailed antibody information.
After euthanasia of mice, spleens were ground in a 70 μm cell sieve, and lymphocytes suspended in culture medium were collected. After centrifugation, erythrocyte lysate (R1010, Solarbio) was added and resuspension. The extracted mouse lymphocytes were cultured in Dynabeads® Mouse T-Activator CD3/CD28 (11452D; Life Technologies, Thermo Scientific) and Recombinant Mouse IL-2 (1000 U/mL, HZ-1015, Proteintech) for one week according to the manufacturer’s protocol. After allowing the cancer cells to attach to the culture dish overnight, they were treated with activated T cells for 48 h (1:3). Cells and cell debris were removed by PBS washing, and then live cancer cells were quantified by spectrophotometry at OD (570 nm), followed by crystalline violet staining.
PD-1 and PD-L1 binding assay
Cells were fixed in 4% paraformaldehyde for 30 min at RT and then incubated with recombinant human PD-1 Fc protein (0.1 µg/mL, 1086-PD, R&D Systems) or recombinant mouse PD-1 Fc protein (1 µg/mL, 1021-PD, R&D Systems) for 1 h. After washing with PBS, cells were incubated with Alexa Fluor 488-labeled Goat anti-Human/Mouse IgG (H + L) Cross-Adsorbed Secondary Antibody (A-11013, A-31561; Life Technologies, Thermo Scientific) for 1 h at RT, followed by staining with Hoechst (C0030, Solarbio) for 10 min at RT. Alexa Fluor 488 fluorescence intensity was measured using a microplate reader Synergy Neo (BioTeK, Winooski, VT, USA), and cells were visualized using a confocal microscope (LSM880, Carl Zeiss). Five different microscope images were randomly selected for quantitative analysis.
Immunohistochemistry
All of the specimens were formalin-fixed and embedded in paraffin wax. For IHC assays, paraffin-embedded tissues are dewaxed and antigens are repaired at 115 °C for 20 min using a Tris-EDTA (PH = 9.0) based unmasking solution. After blocking endogenous peroxidase activity with 3% hydrogen peroxide, the non-specific antigen was blocked using 10% goat serum for 30 min at room temperature, followed by incubation of the sections with primary antibody at 4 °C overnight. After washing with PBS for 3 × 5 min, slices were treated with a Histostain-Streptavidin-Peroxidase kit (SP9001, ZSZB-Bio), followed by 3,3′-diaminobenzidine (DAB) staining and hematoxylin counterstaining. All tissue slices were photographed using a DP72 microscope (Olympus Corporation, Tokyo, Japan). A double-blind procedure was used to assess the sections. Two pathologists used a semi-quantitative scoring system to examine the sections. The number of positive cells was scored as 4 (> 75%), 3 (51–75%), 2 (25–50%), 1 (5–25%), or 0 (5%), while the staining intensity was assessed as 3 (brown), 2 (light brown), 1 (light yellow), or 0 (colorless). The immunoreactivity score (IRS) was calculated by adding the two grades. Negative (0), weak (1–3); moderate (4 and 5); and strong (6 and 7) were used to categorize the total score. For each group, the IRS median value was computed.
Kaplan-Meier survival analysis
Cervical cancer cases were divided into two groups based on the immunohistochemical score mentioned above: high expression (moderate and strong) and low expression (negative and weak). Progression-free survival was calculated for follow-up data. In the first 2 years, the follow-up period was 3 months. In the next 3 to 5 years, the follow-up period was 6 months, and in the subsequent years, the follow-up period was 12 months. The survival period was described using the Kaplan-Meier curve, and the log-rank test was used to compare the survival periods of each group. P value < 0.05 was considered statistically significant (P < 0.05).
RNA isolation and qRT-PCR
Total RNA Extraction Kit (R1200, Solarbio) was used to extract total RNAs, which was then reverse transcribed into cDNA using the TransScript
® One-Step gDNA Removal and cDNA Synthesis SuperMix (TransGen Biotech) according to the manufacturer’s instructions. FastStartTM SYBR Green Master (Roche Diagnostics) was used to perform RT-qPCR according to the manufacturer’s protocol. The sequence information for each primer used for human gene expression analysis was as follows:
-
FAT4-Forward: 5′- CAAATGCTGTGATTGCGTAT-3′,
-
FAT4-Reverse: 5′- AACAGTGGCAAAGCTACACCT-3′
-
STT3A-Forward: 5′- GAAGCAACAGGATTCCACCTACC-3′,
-
STT3A- Reverse: 5′- CAATGGACGGAGAAGAGTAGGC-3′
-
CD274-Forward: 5′- TGGCATTTGCTGAACGCATTT-3′,
-
CD274-Reverse: 5′- TGCAGCCAGGTCTAATTGTTTT-3′
-
ACTB-Forward: 5′- CGTGCGTGACATTAAGGAGAAG-3’,
-
ACTB-Reverse: 5′- GGAAGGAAGGCTGGAAGAGTG-3’.
The sequence information for each primer used for mouse gene expression analysis was as follows:
-
Fat4-Forward: 5′-GAGCCAATCCTTCAGAGGAGG-3′,
-
Fat4-Reverse: 5′-CCATGACCAGAAGTCCACACA-3′
-
Stt3a-Forward: 5′-ACCATCGTTACGTACCACCT-3′,
-
Stt3a-Reverse: 5′-AGCCAGCTACAGATCGAGAA-3′
-
Cd274-Forward: 5′-AATCGTGGTCCCCAAGCCTC-3′,
-
Cd274-Reverse: 5′-ACAGCAGGCTGTGAATATAATGC-3′
-
Actb-Forward: 5′-GATCCTGACCGAGCGTG-3′,
-
Actb-Reverse: 5′-GTTGGCATAGAGGTCTTTACGG-3′.
For quantification of gene expression, the 2-ΔΔCt method was used. β-actin expression was used for normalization.
Cell proliferation and colony formation
For CCK-8 cell proliferation assays, cells (1000 per well) were cultured in 96-well plates and cell proliferation was detected using a cell counting kit-8 (96992, Sigma-Aldrich) at 450 nm for 3 days. In the colony formation assay, 500 cells were inoculated in 6-well plates with 3 replicate wells per group. 7 d later, cell colonies were stained with Giemsa (G1015, Solarbio) and colonies of more than 50 cells were counted under the microscope.
Edu proliferation assay
pEGFP-N1/PD-L1 was a gift from Mien-Chie Hung (Addgene plasmid # 121478) [
24]. CTRL and sgRNA-
FAT4 ME180 cells were transfected with plasmid DNA using X-tremeGENE™ HP DNA Transfection Reagent (Roche, 06366236001). Click-iT
® c5-ethynyl-20-deoxyuridine (EdU) kit (C10337, Invitrogen) was used to detect the cell proliferation ability, and the cells were seeded in confocal plates at a density of 5 × 10
5 cells per well, and mixed with Incubate with 50 μM EdU buffer. After 2 h at 37 °C, fix with 4% formaldehyde for 0.5 h, permeabilize with 0.1% Triton X-100 for 20 min, add Click-iT
® reaction mixture, and stain Hoechst cell nuclei. Fluorescence microscopy was used to visualize the results.
Plasmid transfection and luciferase activity assay
pLV-β-catenin deltaN90 was a gift from Bob Weinberg (Addgene plasmid # 36985) [
28]. M50 Super 8 × TOPFlash (Addgene plasmid # 12456) and M51 Super 8 × FOPFlash (Addgene plasmid # 12457) were gifts from Randall Moon [
29]. ME180 cells were transfected with plasmid DNA using X-tremeGENE™ HP DNA Transfection Reagent (Roche, 06366236001) for 24 h. Cells were lysed and subjected to luciferase reporter assay by using a Double-Luciferase Reporter Assay Kit (TransGen Biotech, Beijing, China) and was normalized to pSV40-Renilla luciferase activity.
Statistical analysis
GraphPad Prism 8.0 software was used for statistical analysis (GraphPad Software, Inc. La Jolla, CA). In all graphs, data points and bars represent the means of independent biological replicates, error bars represent standard deviations, and all data are reported as mean ± standard deviation (SD). All data represent similar results from at least three independent experiments. Data were analyzed for significance using Student’s t-test (two groups) or one-way ANOVA with Tukey’s post hoc test (multiple groups), correlations were analyzed using Pearson’s correlation coefficient, and data on clinicopathological characteristics of CC patients were analyzed using the χ2 test. P < 0.05 was considered a statistically significant difference.
Discussion
The FAT cadherin family displays similar extracellular domain structures, consisting of single-pass transmembrane receptors with 32–34 cadherin repeat sequences. The four FAT proteins invertebrates have different intracellular cytoplasmic domains, which may reflect the specific function of each FAT cadherin member [
1,
4]. FAT4, the vertebrate FAT protein with the most sequence similarity to Drosophila Fat; there have been a few studies looking into the relationship between FAT4 and human cancers, and it has been proposed that FAT4 inhibits tumorigenesis and progression in endometrial [
33], gastric [
32], and colorectal [
31] cancers. However, to our knowledge, the mechanism of FAT4-mediated β-catenin expression and whether FAT4 is associated with immune responses have not been discussed. Here, we report that FAT4 activates antitumor immunity through β-catenin by promoting aberrant glycosylation and degradation of PD-L1 and inhibiting the Wnt/β-catenin signaling pathway. Our findings provide a novel mechanism to regulate the β-catenin/STT3/PD-L1 axis.
Furthermore, the Wnt signaling is essential in maintaining stem cellularity, driving invasiveness, and immune escape [
12,
38]. Activation of the Wnt/β-catenin pathway in solid tumors prevents T cells spontaneous activation and infiltration into the tumor microenvironment, increasing resistance to immune checkpoint inhibitor (PD-1 and CTLA4) therapy. Blockade of β-catenin combined with immune checkpoint inhibitor has been demonstrated to promote complete tumor regression in homozygous mouse models of melanoma, breast cancer, neuroblastoma, and renal adenocarcinoma [
41,
42]. Therefore, exploring the molecular mechanisms by which Wnt/β-catenin signaling affects immune surveillance in cervical cancer will provide new directions for cancer immunotherapy [
43]. In this study, we found that FAT4 overexpression decreased β-catenin nuclear accumulation and increased FAT4 chelating β-catenin at the cell membrane, which is consistent with FAT4 ubiquitinating β-catenin and subsequently promoting its degradation, thereby preventing the Wnt signaling and tumor growth. Analysis of the TCGA database revealed that the FAT4 gene is mutated or deleted in multiple squamous cell carcinomas, suggesting that FAT4 mutations may be responsible for dysregulation of the Wnt signaling.
Previous studies have shown that PD-L1 levels can be regulated at both the transcriptional and post-translational modifications (PTMs) and that PD-L1 regulation is an important mechanism affecting the efficacy of PD-L1/PD-1 immunotherapy [
27]. In particular, several key transcription factors, including STAT3 [
44], c-Myc [
45], HIF1/2α [
46], and nuclear factor-κB (NF-κB) [
47], as well as mitogen-activated protein kinase (MAPK) [
48], epidermal growth factor receptor (EGFR) [
49], AKT/mTOR pathway [
50] and Wnt/β-catenin pathway [
13] can also boost PD-L1 mRNA expression when they are mutated or hyperactivated. Aberrant alterations in PTMs directly affect PD-L1-mediated immune resistance, including STT3-dependent N-linked glycosylation of PD-L1, GSK3β-dependent polyubiquitination, and degradation of PD-L1, and B3GNT3-dependent palmitoylation of PD-L1. It was reported that the β-catenin/TCF/LEF complex binds to the CD274 gene promoter region to induce PD-L1 expression, and the β-catenin/TCF7L2 complex binds to the STT3 gene promoter region to induce STT3A/STT3B expression [
13,
14]. Importantly, N-linked glycosylation of PD-L1 is required for its stabilization and translocation to the cell membrane, where it exerts immunological escape [
24]. In the present study, we demonstrate that FAT4 overexpression inhibits PD-L1 transcription and further glycosylation modifications in a β-catenin-dependent manner. The RNA-seq data indicate that FAT4 regulates N-Glycan biosynthesis. Specifically, FAT4 can cause aberrant N-linked glycosylation and ER retention of PD-L1 via the ER-associated N-glycosyltransferase STT3A, preventing PD-L1 from translocating to the membrane. It has been reported that GSK3β binds to the C-terminal structural domain of non-glycosylated PD-L1, causing PD-L1 phosphorylation followed by polyubiquitination [
24]. This is consistent with our study, where we confirmed that FAT4 was able to promote the interaction of GSK3β with PD-L1 and further ubiquitination-dependent degradation. Functionally, through T cell-mediated cancer cell killing assays and PD-L1/PD-1 binding assays, we demonstrated that FAT4 overexpression in tumor cells significantly activated CTL activity by downregulating PD-L1 levels. Thus, our study reveals the molecular mechanism of tumor PD-L1 regulation. Finally, we discovered a novel FAT4 protein expression profile in cervical cancer and showed that FAT4 regulates immune editing, suggesting that the FAT4/β-catenin/STT3/PD-L1 signaling axis could be a potential target for cervical cancer.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.