Skip to main content
Download PDF

FTO-mediated DSP m6A demethylation promotes an aggressive subtype of growth hormone-secreting pituitary neuroendocrine tumors

Molecular Cancer volume 23, Article number: 205 (2024) Cite this article

Abstract

Background

Growth hormone-secreting pituitary neuroendocrine tumors can be pathologically classified into densely granulated (DGGH) and sparsely granulated types (SGGH). SGGH is more aggressive and associated with a poorer prognosis. While epigenetic regulation is vital in tumorigenesis and progression, the role of N6-methyladenosine (m6A) in aggressive behavior has yet to be elucidated.

Methods

We performed m6A-sequencing on tumor samples from 8 DGGH and 8 SGGH patients, complemented by a suite of assays including ELISA, immuno-histochemistry, -blotting and -fluorescence, qPCR, MeRIP, RIP, and RNA stability experiments, aiming to delineate the influence of m6A on tumor behavior. We further assessed the therapeutic potential of targeted drugs using cell cultures, organoid models, and animal studies.

Results

We discovered a significant reduction of m6A levels in SGGH compared to DGGH, with an elevated expression of fat mass and obesity-associated protein (FTO), an m6A demethylase, in SGGH subtype. Series of in vivo and in vitro experiments demonstrated that FTO inhibition in tumor cells robustly diminishes hypoxia resistance, attenuates growth hormone secretion, and augments responsiveness to octreotide. Mechanically, FTO-mediated m6A demethylation destabilizes desmoplakin (DSP) mRNA, mediated by the m6A reader FMR1, leading to prohibited desmosome integrity and enhanced tumor hypoxia tolerance. Targeting the FTO-DSP-SSTR2 axis curtailed growth hormone secretion, therefor sensitizing tumors to octreotide therapy.

Conclusion

Our study reveals the critical role of FTO in the aggressive growth hormone-secreting pituitary neuroendocrine tumors subtype and suggests FTO may represent a new therapeutic target for refractory/persistent SGGH.

Background

Growth hormone-secreting pituitary neuroendocrine tumors comprise approximately 10% of pituitary neuroendocrine tumors (PitNETs), which rank among the most prevalent intracranial tumors in adults [1]. Growth hormone-secreting pituitary neuroendocrine tumors are linked to significant comorbidities and increased mortality risk [2]. Prolonged exposure to excessive GH causes detrimental effects on various systems and organs, including the cardiovascular, respiratory, and musculoskeletal systems [3]. Moreover, patients with growth hormone-secreting pituitary neuroendocrine tumors face a significantly higher risk of colorectal, breast, and thyroid cancers. As a result, their long-term quality of life is generally poorer, with a lifespan reduced by approximately 30% compared to the general population [4, 5]. According to the 2022 WHO classification of PitNETs, growth hormone-secreting pituitary neuroendocrine tumors belong to the PIT1 lineage, further divided into densely granulated (DGGH) and sparsely granulated (SGGH) subtypes [6]. DGGH typically exhibits non-invasive growth patterns and has a favorable prognosis. SGGHs are characterized by invasive growth, are resistant to conventional therapies, have poorer prognosis, and are classified as refractory/persistent growth hormone-secreting pituitary neuroendocrine tumors [3, 6]. Despite recent advances in endoscopic techniques, about 15% of growth hormone-secreting pituitary neuroendocrine tumors that invade the cavernous sinus and bone do not achieve a biochemical cure, even with optimal postoperative radiation therapy and update medication [2, 3]. Thus, it is important to explore novel drugs for the treatment of refractory/persistent growth hormone-secreting pituitary neuroendocrine tumors. The distinct biological patterns of DGGH and SGGH serve as a suitable model for investigating the pathogenesis of aggressive behavior.

Desmosomes, comprising various junction proteins, act as vital intercellular junctions that facilitates cellular communication and maintain junctional integrity [7]. Studies have shown that desmosomes play an important role in the development and progression of various malignancies, including gastric, colorectal, and breast cancers. In contrast, the loss or weakening of desmosome structures promotes tumor invasion and metastasis [7]. Within the desmosome complex, desmoplakin (DSP) is a core protein in cell junctions, interacting with other proteins such as plakoglobin, desmoglein, and plakophilin [7]. In cardiac pathologies, mutations in the DSP gene impair intercellular desmosome function, contributing to the development of arrhythmogenic cardiomyopathy [8]. In addition, the loss of DSP disrupts intercellular junctions between tumor cells, promoting tumor cell invasion and metastasis [7]. A previous study found that transcripts of genes associated with desmosomal structure and function were consistently downregulated in SGGH, suggesting a critical role of desmosomes in the formation of invasive phenotypes [9].

N6-methyladenosine (m6A) modification is the most abundant endogenous RNA modification in eukaryotes. It has been reported to participate in various physiological processes, including embryonic development, immune system maturation, and neural development [10]. Through the action of RNA methyltransferases and demethylases, m6A recruits specific reader proteins to target RNAs, thereby influencing RNA stability, translation, alternative splicing, and other functions [11]. Recent studies have demonstrated that the dysregulation of m6A is associated with tumor initiation, invasion, and the formation of cell adhesion [12, 13]. For instance, the m6A demethylase FTO affects metastasis and invasion of Epstein-Barr virus-associated gastric cancer via an m6A-FOS-IGF2BP1/2-dependent manner, providing biomarkers for metastatic prediction and therapy of gastric cancer [14]. FMR1, a novel m6A reader, is upregulated in colorectal cancer (CRC) and plays a critical role in promoting CRC cell proliferation and metastasis by recognizing the m6A-modification site in EGFR mRNA [15]. Additionally, METTL14 influences neuronal activity and pain sensitivity through the GluN2A subunit of NMDAR in chemotherapy-induced neuropathic pain, highlighting a potential therapeutic target for pain management in cancer treatment [16]. However, minimal studies have shown that METTL3 is upregulated in growth hormone-secreting pituitary neuroendocrine tumors and promotes the proliferation and invasiveness of tumor cells [17]. The role of m6A in the pathogenesis of growth hormone-secreting pituitary neuroendocrine tumors remains to be elucidated.

Our previous studies uncovered the potential function of epigenetic regulation in the progression of invasive PitNETs [18, 19]. Based on these findings, we further analyzed the m6A modifications in DGGH and SGGH samples to explore the role of m6A modification in regulating the aggressive behavior of growth hormone-secreting pituitary neuroendocrine tumors. We identified a distinct m6A profile between DGGH and SGGH, primarily regulated by FTO. By diminishing the mRNA stability of DSP, a crucial component of the desmosome, FTO disrupts the desmosome structure and promotes hypoxic tolerance, octreotide resistance and growth hormone secretion of growth hormone-secreting pituitary neuroendocrine tumors. In summary, we propose a novel pathway for distinguishing growth hormone-secreting pituitary neuroendocrine tumors subtypes and reveal the potential pathogenesis of their aggressive growth pattern. The components in this pathway represent potential therapeutic targets for refractory/persistent growth hormone-secreting pituitary neuroendocrine tumors.

Methods

Samples preparation

Our study was conducted in accordance with the guidelines of the Declaration of Helsinki. All data were anonymously analyzed. A total of 39 cases of DGGH and 30 cases of SGGH were included in this study. Among them, 8 DGGH and 8 SGGH samples were recruited for high-throughput m6A-sequencing. The information for the 69 cohorts and high-throughput m6A-sequencing samples is in Table S1-S2. The patients were enrolled at Sun Yat-sen University Cancer Center (Guangzhou, China) from 2018 to 2021. Each patient signed an informed consent form. Ethical approval was obtained from the Medical Ethics Committee of Sun Yat-sen University Cancer Center (G2023-271). Before undergoing endoscopic sinus surgery, no treatments were given. The diagnosis of growth hormone-secreting pituitary neuroendocrine tumors was confirmed through histopathological and biochemical testing. Subtypes were classified based on fibrous bodies according to the latest consensus guidelines (Fig. S1a-b) [6]. DGGH has perinuclear cytokeratin expression, and SGGH shows a predominant (> 70%) fibrous body pattern.

High-throughput m6A-sequencing

Total RNA was isolated using TRIzol reagent (Invitrogen, USA) following the manufacturer’s procedure. The RNA amount and purity of each sample were quantified using NanoDrop ND-1000 (NanoDrop, USA). The RNA integrity was assessed by Bioanalyzer 2100 (Agilent, CA, USA) with RIN number > 7.0, and confirmed by electrophoresis with denaturing agarose gel. Poly (A) RNA was purified from 30 µg total RNA using Dynabeads Oligo (dT)25-61005 (Thermo Fisher, USA) using two rounds of purification. The poly(A) RNA was fragmented into small pieces using the Magnesium RNA Fragmentation Module (NEB, cat.e6150, USA) under 86℃ for 7 min. Then the cleaved RNA fragments were incubated for 2 h at 4℃ with m6A-specific antibody (Synaptic Systems, cat.202003, Germany) in IP buffer (50 mM Tris-HCl, 750 mM NaCl and 0.5% Igepal CA-630). The IP RNA was reverse-transcribed by SuperScript™ II Reverse Transcriptase (Invitrogen, cat.1896649, USA) to generate the cDNA which was then used to synthesize U-labeled second-stranded DNAs with E. coli DNA polymerase I (NEB, cat.m0209, USA), RNase H (NEB, cat.m0297, USA) and dUTP Solution (Thermo Fisher, cat.R0133, USA). An A-base was then added to the blunt ends of each strand, preparing them for ligation to the indexed adapters. Each adapter contains a T-base overhang for ligating the adapter to the A-tailed fragmented DNA. Single- or dual-index adapters were ligated to the fragments, and size selection was performed with AMPureXP beads. After the heat-labile UDG enzyme (NEB, cat.m0280, USA) treatment of the U-labeled second-stranded DNAs, the ligated products were amplified by PCR with the following conditions: initial denaturation at 95℃ for 3 min; 8 cycles of denaturation at 98℃ for 15 s, annealing at 60℃ for 15 s, and extension at 72℃ for 30 s; and then final extension at 72℃ for 5 min. The average insert size for the final cDNA library was 300 ± 50 bp. At last, we performed the 2 × 150 bp paired-end sequencing (PE150) on an Illumina Novaseq™ 6000 (LC-Bio Technology Co., Ltd., Hangzhou, China).

Processing of MeRIP-seq and RNA-seq data

FastQC (https://www.bioinformatics.babraham.ac.uk/projects/fastqc/) and fastp [20] were employed to perform quality control and preprocessing of raw sequencing reads.

For MeRIP-seq data, analysis was performed using MeRIPseqPipe [21]. Clean reads were aligned to the hg38 genome using STAR [22]. MACS2 [23] (-p 1e-6 --keep-dup 5) and MetPeak [24] were used to identify m6A enriched peaks. The intersect peaks were retained for subsequent analysis. The read coverage of IP and Input data for each peak was calculated using Multicov [25] and normalized by the RPKM method. The radio of IP RPKM (with adding 1) and INPUT RPKM (with adding 1) was used to represent the methylation level of each m6A peak. FeatureCounts [26] was used to generate gene counts. For differential methylation analysis, the Wilcox test was used to examine the significant differences, while DESeq2 [27] was used to identify differentially expressed genes. The m6A peak annotation was performed with the human annotation file (GENCODE, version 39) downloaded from the GENCODE database (https://www.gencodegenes.org/) using our custom Perl scripts HOMER [28] was performed to find the m6A motifs.

For scRNA-seq data, raw sequencing data were processed with CellRanger (10X Genomics, v3.1.0) using default settings and aligned to the human genome (GRCh38). Feature-barcode matrices were then treated with CellBender (default parameters) to eliminate ambient RNA. The resulting clean matrices were analyzed using Seurat (v4.1.0). The specific methods involved in scRNA-seq analysis include the preparation of tumor samples and other analyses such as identifying differentially expressed genes (DEGs) and cell classification, and pathway enrichment analysis as previously described [29].

For RNA-seq data, clean reads were aligned to the Rn6 genome using STAR [22]. FeatureCounts was performed to quantify the expression of genes. DESeq2 was used to identify differentially expressed genes and clusterProfiler was used to do the pathway enrichment. Pathway activity was predicted through GSVA and ssGSEA analysis.

Cell lines and cell culture

The human growth hormone-secreting pituitary neuroendocrine tumor cells were isolated from primary growth hormone-secreting pituitary neuroendocrine tumors as described previously and cultured in DMEM/F12 medium (Gibco, NY, USA) supplemented with 20% FBS [19]. The rat growth hormone-secreting pituitary neuroendocrine tumors cell line GH3, which are more likely to be an in vitro model of sparsely granulated subtypes [9], was obtained from the National Infrastructure of Cell Line Resource. GH3 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM)/high glucose supplemented with 10% fetal bovine serum (FBS) (Gibco, USA), 100 U/mL penicillin, and 100 µg/mL streptomycin. All cultured cells were maintained at 37 °C in a humidified atmosphere of 5% CO2. All cell lines used in this study were tested and confirmed to be free of mycoplasma contamination.

RNA isolation and quantitative reverse transcription polymerase chain reaction (qRT‒PCR)

Total RNA was extracted from tissues and cells using TRIzol (Invitrogen, USA) following the provided protocol from the manufacturer. Subsequently, the total RNA was reverse-transcribed into complementary DNA (cDNA) utilizing a Reverse Transcription System Kit (Takara BIO INC, Kusatsu, Shiga, Japan). For quantitative real-time PCR, the cDNA served as the template and was amplified using specific primers and a SYBR Premix Ex Taq RNAse H kit (Takara Bio, Tokyo, Japan) in conjunction with the Roche LightCycler 480II detection system. The amplification protocol consisted of an initial 5-minute incubation at 95 °C, followed by 40 cycles of 10 s at 95 °C and 30 s at 60 °C. Each experiment was replicated at least three times. The resulting data were analyzed using the delta-delta CT method (formula: 2−(Ct target−Ct reference)) to calculate relative expression levels, which were then compared to control samples. The primers used are listed in Table S3.

m6A immunoprecipitation (MeRIP)

The total RNA was extracted using the method mentioned above. Specifically, 100 µg of total RNA was subjected to the MERIP experiment using the riboMeRIPTM m6A Transcriptome Profiling Kit (Ribobio, C11051-1), following the manufacturer’s protocol. After immunoprecipitation, the enrichment of RNA was examined using qPCR analysis.

m6A ELISA

We used the enzyme-linked EpiQuik m6A RNA methylation quantification kit (Epigentek, NY, USA) to detect the level of m6A RNA methylation in RNA according to the manufacturer’s instructions. Briefly, the control sample and test sample were incubated at 37℃ for 90 min to complete RNA binding, and the capture antibody solution was added and incubated at room temperature for 1 h. Next, the detection antibody solution was added and incubated at room temperature for 30 min to complete RNA capture. Finally, the developer was added and incubated at room temperature for 10 min. When the color of the positive control well changed to moderate blue, the stop solution was added to each well and the absorbance value at 450 nm was measured using an enzyme-linked immunosorbent assay (ELISA) reader. m6A %= (Sample OD value - NC OD value) ÷ RNA input/ (Positive control OD value - Negative control OD value) ÷ Positive control input.

RNA-Binding protein immunoprecipitation (RIP)

Wash freshly resected whole tissue three times with ice-cold PBS. Tumor cells were collected using the primary tumor cell isolation method described above [19]. Collect cells by centrifugation at 1500 rpm for 5 min at 4 °C and discard the supernatant. Subsequently, RIP experiments were performed using the Magna RIP™ RNA-Binding Protein Immunoprecipitation Kit (Millipore, Catalog No. 17–700). After immunoprecipitation, the enrichment of RNA was examined using qPCR analysis.

Immunohistochemistry

The tissues were fixed in 4% paraformaldehyde (PFA) for 24 h and then processed for paraffin embedding. Sectioning was performed at a thickness of 3 mm. Subsequently, the slides underwent deparaffinization and rehydration. To retrieve the heat-induced epitopes, the slides were submerged in an antigen-unmasking solution (Solarbio). To eliminate endogenous peroxidase and nonspecific binding sites, a sequential treatment with 0.3% H2O2 and 5% normal goat serum was carried out. Then antibodies were applied overnight at 4 °C (Table S4). Afterward, the slides were incubated with Dako REAL EnVision HRP rabbit/mouse (belonging to K5007, DAKO, Glostrup, Denmark) for 20 min at room temperature. To visualize the staining signals under light microscopy, Dako REAL DAB + CHROMOGEN and Dako REAL substrate buffer (belonging to K5007, DAKO, Glostrup, Denmark) were applied. Finally, the slides were counterstained using a hematoxylin solution. The stained slides were scanned using KFBIO Digital Pathology Slide Scanners (KFBIO, Ningbo, China) and analyzed with the Halo platform.

Western blot assays

Protein extracts were obtained from tumor tissues or cells by using RIPA buffer (Epizyme Biotech, PC101). The protein concentration was determined using a BCA protein assay kit from Thermo Fisher Scientific. After electrophoresis, proteins were transferred onto PVDF membranes and incubated overnight at 4 °C with the corresponding antibody (Table S4). Subsequently, the membranes were exposed to HRP-conjugated secondary antibodies (1:10000, Abcam, Cambridge, MA, USA) for 1 h at room temperature. The signals were detected using an ECL detection system from Bio-Rad Laboratories (Richmond, CA, USA).

RNA interference

The cells were seeded in six-well plates to reach a density of approximately 50–60% before transfection. 125 µL Opti MEM serum reducing medium (Gibco, Grand Island, NY, USA) containing the required amount of siRNA and Lipofectamine 3000 (Thermo Fisher Scientific, Waltham, MA, USA) was added respectively to two centrifuge tubes and incubated at room temperature for 10 min to form a transfection complex. Evenly add the transfection complex to the six-well plates and perform functional experiments after 48 h of transfection. The sequences of siRNA are listed in Table S5.

Stable cell line generation

Culture 293T cells to achieve a cell density of 70% for transfection. Mix the lentiviral vector with packaging plasmids PSPAX2 and pMD2. G in a 4:3:1 ratio and add them into the medium of 293T cells. 48 h and 72 h later, the virus was collected. Target cells at a density of 70% were incubated with virus and 0.001% polybrene and 48 h later 2 ug/ml puromycin was used to select transduced cells. The sequences of shRNA are listed in Table S5.

RNA stability assays

Cells with or without knockdown of FTO or FMR1 were subjected to treatment with actinomycin D at a final concentration of 2 µM for various time points. Subsequently, total RNA was extracted using the TRIzol reagent. The expression levels were assessed using RT-qPCR, and the mRNA half-life was calculated.

Flow cytometry

Cellular apoptosis of GH3 and primary cells before and after knockdown of FTO and DSP under normal cell culture oxygen concentration or hypoxic environment (1% O2) were analyzed by flow cytometry using an Annexin V Alexa Fluor 647/7AAD assay kit (Cat: FXP147, 4 A Biotech) according to the manufacturer’s instructions and analyzed with a Beckman flow cytometer (Beckman Coulter, Miami, FL, USA).

Cell viability assay

Cell viability was assessed by using CCK-8 assays (DOJINDO, CK04). Following transfection, GH3, and primary cells were seeded in 96-well plates at a density of 5 × 103 cells/well. On days 0, 1, 2, 3, and 4, 10 µl of CCK-8 solution was added to each well and incubated for 2 h. The absorbance at 450 nm was then measured using a Multiskan plate reader. To investigate the influence of FB23-2 (MCE, HY-127103) and OCT (MCE, HY-P0036) on growth hormone-secreting pituitary neuroendocrine tumor cell viability, GH3 and primary tumor cells were seeded into 96-well plates at a density of 5 × 103 cells/well and 2 × 104 cells/well, respectively. After 24 h, the cell medium was changed to a medium containing FB23-2 (4 µM) or OCT (100 nM) or their combination. Three days later, the absorbance was measured as mentioned above.

Colony formation assays

To conduct the cell colony formation assays, 5 × 102 GH3 with FTO knockdown were seeded into 6-well plates and incubated at 37 °C with 5% CO2. Approximately 10 days later, the cells were washed twice with PBS and fixed with methanol for 30 min. Subsequently, the cells were stained with 0.2% crystal violet for 30 min, and gently washed. The number of colonies was counted.

Growth hormone detection

Growth hormone concentrations in the culture medium of GH3 cells, as well as in serum, were assessed with ELISA kits from Millipore (EZRMGH), following the protocols provided by the manufacturer. Similarly, growth hormone concentrations in the culture medium of primary tumor cells and organoids were determined using a different set of ELISA kits from Elabscience (E-EL-H0177). The growth hormone levels were then normalized against the results from cell and organoid viability tests, as well as the weight of the tumors.

Organoid culture and drug response assay

Tumor samples were washed and minced, then dissociated and cultured into organoids using specific kits and media. After centrifugation, cells were mixed with Matrigel, deposited into plates, and incubated. Successful organoid formation was monitored by morphology and viability. Organoids larger than 100 μm were used for drug screening and passage. Recurring organoids were dissociated, collected, and cultured again. Drug screening used a modified medium without Y-27,632 on first-generation PDOs. Organoids were dissociated, mixed with Matrigel/modified medium, and seeded into Laminin-coated 384-well plates. After 48 h, a medium containing FB23-2 (4 µM) or OCT (100 nM) or their combination was added. After 4 days, viability was assessed with CellTiter-Glo 2.0, normalizing relative luminescence units to DMSO controls (100% viability) as previously reported [30]. All organoids used in this study were tested and confirmed to be free of mycoplasma contamination.

Animal experiments

BALB/c nude mice aged 4–5 weeks were obtained from Beijing Vital River Laboratory Animal Technology. Each group of mice (five mice per group) received subcutaneous injections of 3 × 106 GH3 cells suspended in 100 µl of PBS into the left axilla area. Tumor volume was measured and calculated using the formula: volume = length × width2 × 0.5. Mice were monitored regularly for signs of the defined end-point criteria. To observe the effect of OCT on tumors, once the tumors are visible to the naked eye, daily intraperitoneal injections of the drug (OCT, 50 µg/kg) were initiated. If the weight loss of any mouse exceeded 20% of the initial weight, breathing difficulties, or tumors approaching 15 mm in diameter, it was euthanized immediately and no tumors exceeded this size limit. The Institutional Animal Care and Use Committee of Sun Yat-sen University Cancer Center approved all animal experiments (Ethics Approval no: L102022020004Y), and the handling of the animals adhered to institutional guidelines.

Electron microscopy

Cells were collected by centrifugation and mixed with TEM fixative at 4℃ for 2–4 h, and stored at 4℃. Next, the agarose pre-embedding step was performed. After centrifugation and removal of the supernatant, the samples are washed with 0.1 M phosphate buffer (pH 7.4) and embedded in a 1% agarose solution. Subsequently, post-fixation was done with a 1% osmium tetroxide in 0.1 M phosphate buffer (pH 7.4), fixed at room temperature in the dark for 2 h. Dehydration involved sequential dehydration with different concentrations of alcohol, followed by two dehydration steps with 100% acetone. Permeation and embedding were performed with acetone and EMBed 812, followed by overnight baking at 37℃, and polymerization in an oven at 60℃ for 48 h. The resin blocks were cut into ultrathin sections of 60–80 nm using an ultramicrotome. Finally, staining was conducted with uranyl acetate and lead citrate, and the samples were observed and imaged under a transmission electron microscope for analysis.

Statistical analysis

Statistical analyses were conducted using R 4.1.3 and GraphPad Prism 8.4.2 (San Diego, CA, USA). To ensure biological accuracy, each experiment was replicated independently at least three times. For data following a normal distribution, results were analyzed using Student’s t-test. Non-normally distributed data were assessed using the Wilcoxon rank-sum test. Statistical significance was established at a p-value of less than 0.05. All tests were two-sided.

Results

FTO is the key factor leading to distinct m6A levels between different growth hormone-secreting pituitary neuroendocrine tumor subtypes

Firstly, we conducted MeRIP-seq and RNA-seq on 16 tumor samples from 16 individuals with growth hormone-secreting pituitary neuroendocrine tumors, including 8 SGGHs and 8 DGGH (Fig. 1a). Upon merging the m6A peaks from different subtypes, we obtained 48,528 m6A peaks for further analysis. The data showed that most genes exhibited a single m6A peak, and the identified m6A peaks were predominantly enriched in the classical GGACH motif (Fig. 1b). In line with previous studies, these m6A peaks were primarily localized within the coding sequence (CDS) regions and the regions near stop codons, with mRNA being the most enriched molecular form (Fig. 1c-e) [31]. Differential methylation analysis revealed that there were 8,660 (60.94%) hypo-methylated m6A peaks and 5,551 (39.06%) hyper-methylated m6A peaks in SGGHs compared to DGGH, indicating that SGGHs experienced a global decrease in m6A modification (Fig. 1f). Subsequent quantification of global m6A levels using ELISA assays confirmed these lower m6A modification levels in SGGH compared to DGGH (Fig. 1g). Through integrated analysis of m6A modification levels and gene expression profiles, we observed a positive correlation between m6A modification levels and gene expression in growth hormone-secreting pituitary neuroendocrine tumors (Fig. 1h). Our findings revealed distinct m6A modification landscapes between SGGH and DGGH, characterized by a pronounced downregulation of m6A levels in SGGH. This alteration in m6A modification may influence gene expression regulation, contributing to the pathogenesis and phenotypic characteristics of growth hormone-secreting pituitary neuroendocrine tumor subtypes.

Fig. 1
figure 1

The m6A modification landscape between subtypes of growth hormone-secreting pituitary neuroendocrine tumors. (a) Schematic representation of m6A -seq workflow comparing DGGH and SGGH samples. Eight samples in each group. (b) Bar plot of the distribution of m6A peaks across the genome with the predominant m6A motif. (c) The distribution of m6A sites across mRNA regions. (d) the genomic distribution of m6A modifications by region. (e) the genomic distribution of m6A modifications by RNA subtypes. (f) The distribution of the different methylation levels of dysregulated m6A peaks within subtypes. (g) The difference of global m6A methylation levels in DGGH and SGGH through ELISA. Experiment was replicated independently at least three times. (h) Scatter plot of m6A modification peak distribution with subtype-specific gene expression changes in 16 samples

Full size image

Next, we investigated which methylase or demethylase might account for the differential m6A levels among the two subtypes. Based on the sequencing data, both FTO and RBM15 were significantly upregulated in the SGGH (Fig. 2a). However, only FTO was confirmed to be consistently upregulated in another cohort of growth hormone-secreting pituitary neuroendocrine tumor patients (DGGH = 39, SGGH = 30) treated in our center, a public array dataset (DGGH = 10, SGGH = 10) and a scRNA-seq dataset (DGGH = 2, SGGH = 2) (Fig. 2b-g). Additionally, we define Knosp 1–2 as non-invasive according to the literature, and define Knosp 3–4 as invasive [32]. FTO was significantly overexpressed in invasive growth hormone-secreting pituitary neuroendocrine tumors compared to non-invasive ones (Fig. 2h). We then performed FTO perturbation in cell lines and primary tumor cells (Fig. 2i-l). Overexpression of FTO led to a decrease in overall m6A levels, while knockdown of FTO resulted in an upregulation of m6A levels (Fig. 1m-n). These findings demonstrate that FTO is the primary demethylase regulating m6A modifications in different growth hormone-secreting pituitary neuroendocrine tumor subtypes.

Fig. 2
figure 2

FTO is the key factor leading to distinct m6A levels among different growth hormone-secreting pituitary neuroendocrine tumor subtypes. a. Boxplot of m6A regulatory genes expression between DGGH and SGGH. b-c. Boxplot of quantitative PCR analysis of FTO (b) and RBM15 (c) level in a cohort of 69 DGGH and SGGH samples. d. Violin plot of FTO expression in DGGH and SGGH cells in Huashan scRNA-seq cohorts. e. Boxplot showing FTO expression levels in DGGH and SGGH based on dataset GSE214226. f-g. Immunohistochemical staining for FTO in DGGH and SGGH tissue samples. h. Boxplot of FTO expression between non-invasive and invasive GH adenomas. i-k. Verification of FTO knockdown and overexpression in GH3 cells and primary tumor cells in mRNA level through qPCR assays. l. Western blot assays demonstrating the efficacy of FTO knockdown and overexpression in GH3 cells and primary tumor cells. m-n. Bar plots representing global m6A modification levels following FTO manipulation in GH3 cells (m) and primary tumor cells (n). Each experiment was replicated independently at least three times

Full size image

Desmosome organization is a key signaling pathway for the aggressive phenotype of SGGH, which is regulated by FTO

Gene expression profiling indicated that multiple pathways related to cell-cell junctions were down-regulated in SGGH (Fig. 3a), specifically desmosome and cell-cell junction organization. Notably, desmosome organization is a subclass of cell-cell junction organization (Fig. S2a). Additionally, scRNA-seq data also demonstrated the significant downregulation of desmosome organization in SGGH (Fig. 3b). DGGH and SGGH were distinctly clustered and separated through PCA analysis, which was based on genes associated with desmosome organization (Fig. 3c, Fig. S2b). ROC analysis revealed that most genes related to desmosome organization possess strong diagnostic capabilities for differentiating DGGH from SGGH (Fig. 3d, Fig. S2c). Genes involved in desmosome organization, including DSP, DSG2, PKP2, and PKP3, crucial for cell junctions, were significantly downregulated in SGGH when compared with DGGH (Fig. 3e-f). We tested whether FTO influences the regulation of desmosome organization. Through RNA sequencing and single-sample gene set enrichment analysis (ssGSEA), we found that desmosome organization pathway activity was reduced after FTO overexpression and increased following FTO knockdown (Fig. 3g). Consistently, desmosome-related genes, including DSP, DSG2, PKP2, and PKP3, were downregulated by FTO overexpression and upregulated by FTO knockdown (Fig. 3h-j). The significant up-regulation of functions related to extracellular matrix and angiogenesis is also a characteristic of SGGH. However, overexpression of FTO has no significant effect on the pathway activity of extracellular matrix and angiogenesis. In contrast, FTO knockdown moderately upregulates the pathway activity of extracellular matrix and angiogenesis, which is in conflict with the observed upregulation of extracellular matrix and angiogenesis functions in SGGH (Fig. 3g, S2d-g). To further investigate the functional changes in desmosome organization, we defined desmoglein (DSG) as the marker of desmosome. The DSG protein family is an essential component of desmosome organization, which works as cell surface transmembrane proteins and consists of four members: DSG1, DSG2, DSG3, and DSG4. They are connected to the DSP to form a desmosome or hemidesmosome [33, 34]. Among these members, DSG2 is the only DSG protein that participates in desmosome organization, and it is the crucial component of the desmosome (https://amigo.geneontology.org/amigo/term/GO:0002934). Cell immunofluorescence showed an increase in DSG2 expression on the cell surface after FTO knockdown (Fig. 3k). These data further confirm the role of FTO in regulating the dysfunction observed in desmosome organization among growth hormone-secreting pituitary neuroendocrine tumor subtypes.

Fig. 3
figure 3

FTO influences the desmosome organization of growth hormone-secreting pituitary neuroendocrine tumors. a. Top five upregulated and downregulated pathways of GO-BP enrichment analysis in SGGH compared with DGGH. b. Violin plot showing pathway activity of desmosome organization in DGGH and SGGH cells in Huashan scRNA-seq cohorts. c. PCA plot revealing desmosome organization profiling differences between DGGH and SGGH. d. ROC curves evaluating the diagnostic performance of desmosome organization related genes in differentiating DGGH from SGGH. e. Boxplot showing expression of DSP, DSG2, PKP3 and PKP2 between DGGH and SGGH. f. Schematic illustrating downregulated genes of desmosome organization in SGGH. g. Pathway activity change of desmosome organization, regulation of angiogenesis and extracellular matrix organization following overexpression (above) and knockdown (below) of FTO. h. mRNA level changes of Dsp, Dsg2, Pkp2 and Pkp3 following FTO overexpression in GH3 cells. i-j. mRNA level changes of Dsp, Dsg2, Pkp2 and Pkp3 following FTO knockdown in GH3 cells (i) and primary tumor cells (j). k. Immunofluorescence images verified changes in DSG2 protein levels and localization after FTO knockdown in primary tumor cells. Each experiment was replicated independently at least three times

Full size image

FTO influences the desmosome organization of growth hormone-secreting pituitary neuroendocrine tumors by regulating the m6A level of DSP

Next, we investigated the specific mechanisms by which FTO affects desmosome organization. Since the global m6A level in SGGH is lower than that in DGGH, we primarily focused on genes that showed downregulation of m6A levels. By taking a strict intersection of the differentially expressed genes after FTO perturbation, the differentially expressed genes between growth hormone-secreting pituitary neuroendocrine tumor subtypes, and the genes showing downregulation of m6A levels in SGGH, we found 19 genes that exhibited a negative correlation with FTO expression (Fig. 4a). Among these, DSP was selected for further analysis since it is a key component of desmosome organization. We found that the level of DSP mRNA and protein was significantly decreased in SGGH (Fig. 4b-c). In addition, scRNA-seq data showed significant downregulation of DSP in SGGH (Fig. S3a). Consistently, the levels of FTO and DSP were negatively correlated in two independent datasets (Fig. 4d, Fig. S3b). ROC analysis showed FTO and DSP possess good diagnostic performance in distinguishing DGGH from SGGH (Fig. 4e, Fig. S2c). In cell lines, FTO knockdown and overexpression resulted in upregulation and downregulation of the protein level of DSP, respectively (Fig. 4f-g). To elucidate the regulatory impact of FTO on DSP-mediated desmosome organization, qRT-PCR, western blot and immunofluorescence assays were performed. Our data showed a decrease in both mRNA and protein levels of DSG2 after knocking down DSP whereas knocking down FTO partially restored the mRNA and protein levels of DSG2 (Fig. 4h-i). Similar results were observed in PKP2 and PKP3 (Figs. S3d-f). Immunofluorescence assays demonstrated a reduction of DSG2 expression at the cell surface following the knockdown of DSP whereas knocking down FTO partially restored the DSG2 expression at the cell surface (Fig. 4j). MeRIP-qPCR further confirmed that the m6A level on DSP in DGGH is significantly higher than that in SGGH (Fig. 4k, S3g). Significantly dysregulated m6A upon FTO knockdown were further verified by m6A-sequencing and MeRIP-qPCR (Fig. 4l-n). Taken together, these results provide evidence that FTO is involved in regulating DSP expression through m6A modification and influencing desmosome function in growth hormone-secreting pituitary neuroendocrine tumors.

Fig. 4
figure 4

FTO influences the desmosome organization of growth hormone-secreting pituitary neuroendocrine tumors through regulating the m6A level of DSP. a. Venn plot illustrating genes negatively correlated with FTO and with downregulated m6A level in SGGH. b. Difference of DSP mRNA level between DGGH and SGGH sample through qPCR assays in 69 samples. c. Western Blotting confirmation of higher DSP protein levels in DGGH tissues. d. Regression lines indicating the correlation between FTO and DSP expression. p and R value were calculated by linear models. e. ROC curves evaluating the diagnostic performance of FTO and DSP in differentiating DGGH from SGGH. f. Western blots displaying DSP protein levels following Fto overexpression in GH3 cells. g. Western blots displaying DSP protein levels following Fto knockdown in GH3 cells (above) and primary tumor cells (below). h. Changes in mRNA levels of Dsg2 in GH3 cells and primary tumor cells following Dsp and Fto knockdown through qPCR assays. i. Western blot presenting DSG2 protein expression changes following Fto and Dsp knockdown in GH3 cells (left) and primary tumor cells (right). j. Immunofluorescence images verified changes in DSG2 protein levels and localization following FTO and DSP knockdown in primary tumor cells (j). k. m6A level of DSP in DGGH and SGGH through MeRIP-qPCR assays. l. Graphical representation of differential m6A modification peaks on the DSP after FTO knockdown through m6A-seq. m-n. Bar plot demonstrating the impact of FTO knockdown on m6A levels of DSP in primary tumor cells (m) and GH3 cells (n) through MeRIP-qPCR assays. Each experiment was replicated independently at least three times

Full size image

FTO regulates the mRNA stability of DSP by interacting with m6A reader FMR1

We sought to understand how FTO-mediated m6A modification affects the mRNA level of DSP. Since DSP mRNA expression positively correlated with its m6A levels (Fig. 4a), we hypothesized that m6A modification might affect the stability of DSP mRNA. Upon treatment with actinomycin D, which inhibits de novo RNA synthesis, the stability of DSP mRNA increased in FTO knockdown and decreased in FTO overexpression. (Figs. 5a-c). This suggests that FTO influences the stability of DSP mRNA via m6A. Generally, m6A modification regulates mRNA stability through reader proteins, including IGF2BP1, IGF2BP2, and IGF2BP3 [12]. However, DSP mRNA and protein levels were unchanged knocking down these readers (Figs. S4a-d). Subsequently, we utilized SRAMP (https://www.cuilab.cn/sramp/) to identify m6A modification sites based on differential peaks [34] and used the RMVar database (https://rmvar.renlab.org/) to find RNA-binding proteins at those sites [35, 36]. A total of five high-confidence sites and one moderate-confidence site were found. Among them, site 485 is associated with three RNA-binding proteins, FMR1, HNRNPC, and SND1. Site 1038 is associated with two RNA-binding proteins, NUDT21 and ACIN1 (Fig. 5d). In a previous study, FMR1 has been reported to be involved in the regulation of mRNA stability through m6A modifications [37]. Additionally, the expression level of FMR1 in SGGH tends to be lower than in DGGH (Fig. 5e, Fig. S4e). We subsequently validated the binding of FMR1 to the 485 site using RIP-qPCR. FTO knockdown significantly reduced the binding of FMR1 and m6A on DSP (Fig. 5f). Importantly, our studies show that following knocking down FMR1, the mRNA, protein levels, and mRNA stability of DSP were decreased (Fig. 5g-j, S3f). The above results suggest that FMR1 acts as a reader protein for m6A modification to regulate DSP mRNA stability.

Fig. 5
figure 5

FTO regulates the mRNA stability of DSP by interacting with m6A reader FMR1. a-c. mRNA stability changes of DSP after FTO perturbation in GH3 cells (a-b) and primary tumor cells (c) after actinomycin D treatment. d. Prediction score distributions for m6A modification site with related RNA binding proteins, as determined using the SRAMP prediction tool and RMVar database. e. qPCR result show FMR1 mRNA level in DGGH and SGGH. f. Bar plot demonstrating the impact of FTO knockdown on the binding of FMR1 and m6A on DSP through RIP-qPCR. g. DSP mRNA level change after FMR1 knockdown in GH3 cells (left) and primary tumor cells (right), quantified by qPCR. h. Western blots display DSP protein level alterations after FMR1 knockdown in GH3 cells (left) and primary tumor cells (right). i-j. mRNA stability changes of DSP after FMR1 knockdown in GH3 cells (i) and primary tumor cells (j) under actinomycin D treatment. Each experiment was replicated independently at least three times

Full size image

FTO knockdown inhibits hypoxia tolerance and formation of fibrous bodies of growth hormone-secreting pituitary neuroendocrine tumors

To further elucidate the clinical relevance of FTO in managing aggressive growth hormone-secreting pituitary neuroendocrine tumors, we conducted a series of functional experiments using cell lines, primary tumor cells and animal models. FTO does not affect the proliferation or clonal formation of cell lines and primary tumor cells (Fig. 6a-d). Besides, FTO does not affect the tumor volume and weight in subcutaneous tumor formation (Fig. 6e-h). Some research reports that hypoxic conditions may play an important role in pituitary tumorigenesis, and dysfunction in desmosomes may affect the hypoxia tolerance of growth hormone-secreting pituitary neuroendocrine tumor cells [9, 38,39,40,41]. We tested the importance of FTO on cell viability under hypoxic conditions (1% oxygen). By performing flow cytometry analysis, we determined that FTO knockdown does not influence cell apoptosis under normoxic conditions, but significantly increases the apoptosis rate under hypoxic conditions. Moreover, DSP knockdown elevates cell apoptosis under normoxic conditions but reduces the apoptosis rate under hypoxic conditions (Fig. 6i). Additionally, DGGH and SGGH exhibit pathological differences in fibrous bodies. In the study by Wierman et al., the authors speculated that disruption of desmosome organization may be associated with the formation of these fibrous bodies [9]. Indeed, our electron microscopy studies revealed a sparser distribution of filaments within the fibrous bodies of primary tumor cells following FTO knockdown (Fig. 6j).

Fig. 6
figure 6

FTO knockdown influences hypoxia tolerance and formation of fibrous bodies of growth hormone-secreting pituitary neuroendocrine tumors cells. a-c. Cell viability measured by CCK8 assays in GH3 (a-b) and primary tumor cells (c) cells under FTO perturbation. d. Colony formation was performed in GH3 cells following Fto knockdown. e. Line graph detailing changes in mouse body weight over time. f. Image of subcutaneous tumors from xenograft model by injecting GH3 cells following Fto knockdown. g. Detailed changes in mouse. h. Bar plot showing the effect of Fto knockdown on tumor wight. i. Flow cytometry analysis of percentage of apoptosis under normal and hypoxia after FTO and DSP knockdown in primary tumor cells and GH3 cells. j. Electron microscopy images show differences in fiber density within fibrous bodies after FTO knockdown in primary tumor cells. Each experiment was replicated independently at least three times

Full size image

Targeting FTO reduces the GH-secreting capability of tumor cells and enhances their sensitivity to somatostatin analogs

In clinical settings, patients diagnosed with growth hormone-secreting pituitary neuroendocrine tumors commonly suffer from a variety of systemic complications affecting different organs as a result of excess growth hormone secretion. The primary medications recommended for growth hormone-secreting pituitary neuroendocrine tumor patients are somatostatin analogs, such as octreotide [2]. However, individuals with SGGH often exhibit resistance to these drugs. Therefore, we further assessed the influence of FTO on growth hormone secretion capacity and sensitivity to somatostatin analogs in these patients.

Octreotide treatment for growth hormone-secreting pituitary neuroendocrine tumor primarily works by targeting somatostatin receptor, which downregulates the transcription level of growth hormone, thereby inhibiting its secretion and, to some extent, suppressing the growth of the growth hormone-secreting pituitary neuroendocrine tumors [42,43,44,45]. Reduced somatostatin receptor 2 (SSTR2) and elevated somatostatin receptor 5 (SSTR5) expression is a key characteristic of refractory/persistent growth hormone-secreting pituitary neuroendocrine tumors. Therefore, we first examined the impact of FTO on the transcription and protein levels of SSTR2, SSTR5 and growth hormone. RNA sequencing results showed that after FTO knockdown, SSTR2 levels were significantly upregulated, while growth hormone levels were significantly downregulated (Fig. 7a). Additionally, DSP has a significant positive correlation with SSTR2 (Fig. S5a).We subsequently validated these findings using qPCR, western blot, and growth hormone ELISA experiments (Fig. 7b-c, S5b). FTO knockdown can also downregulate the level of SSTR5 (Fig. S5c-d). Electron microscopy demonstrated a decrease in secretory granules after FTO knockdown in primary tumor cells derived from a patient with growth hormone-secreting pituitary neuroendocrine tumor (Fig. 7d). Concerning sensitivity to octreotide, knockdown of FTO enhances the sensitivity of GH3 and primary tumor cells to octreotide (Fig. 7e). Treatment with FB23-2, a methyltransferase inhibitor of FTO, enhanced the sensitivity of growth hormone-secreting pituitary neuroendocrine tumors organoids and cells to octreotide, and inhibited the secretion of growth hormone (Fig. 7f-h). In subcutaneous tumors, we showed that octreotide inhibited the growth of FTO-knockdown cells more effectively and further reduced the secretion of growth hormone (Fig. 7i-l, S5e).

Fig. 7
figure 7

Targeting FTO reduces GH secreting capability of tumor cells and enhances their sensitivity to somatostatin analogs. a. Bar plots showing the Sstr2 and Gh1 level change after Fto knockdown through RNA-seq analysis. b. Bar plots showing the Sstr2 and Gh1 level change after FTO and DSP knockdown in GH3 cells (left) and primary tumor cells (right) through qPCR analysis under the stimulation of octreotide (100nM). c. Bar plots showing the growth hormone level changes following FTO and DSP knockdown in GH3 cells (left) and primary tumor cells (right) under the stimulation of octreotide (100nM). d. Electron microscopy showing fewer secretory granules after FTO knockdown in primary tumor cells. e. Bar plots showing the octreotide sensitivity change after FTO knockdown in GH3 cells (left) and primary tumor cells (right). f. Bar plots showing the octreotide sensitivity change after combing with FB23-2 treatment in GH3 cells (left) and primary tumor cells (right). g. Bar plots showing the growth hormone level changes following octreotide and FB23-2 treatment in GH3 cells (left) and primary tumor cells (right). h. Bar plots showing the octreotide sensitivity (left) and growth hormone level (right) change after combining with FB23-2 in organoids. i-j. In vivo assessment of octreotide sensitivity in GH3 xenografts by injecting with or without Fto knockdown GH3 cells according to tumor volume (j), tumor weight (k) and growth hormone level (l). Each experiment was replicated independently at least three times

Full size image

Integrating our results, targeting FTO reduces tolerance in hypoxic microenvironments, inhibites growth hormone secretion function and enhances sensitivity to octreotide in growth hormone-secreting pituitary neuroendocrine tumors. Additionally, it also has some impact on the formation of fibrous bodies (Fig. 8).

Fig. 8
figure 8

The schematic model of targeting FTO in the aggressive subtype of growth hormone-secreting pituitary neuroendocrine tumor

Full size image

Discussion

Growth hormone-secreting pituitary neuroendocrine tumors have significant effects on the body and are associated with various diseases. However, the molecular mechanisms underlying the development and classification differences of these tumors are not yet fully understood. Initial findings suggested that an activating mutation in the G protein subunit A, leading to constant activation of cyclic adenosine monophosphate (cAMP), was associated with DGGH [46]. However, further research revealed that G protein subunit A mutations are present in 40–65% of growth hormone-secreting pituitary neuroendocrine tumors, including 23–38% of SGGH, indicating that these mutations do not align with specific tumor histological subtypes [46,47,48,49,50,51]. In Ezzat’s study, it was found that 43% of SGGH tumors exhibited mutations in the growth hormone receptor (GHR), while no mutations were found in DGGH tumors [52]. However, other research groups did not identify GHR mutations in their study cohorts [49, 53]. In 2017, Wierman’s study provided transcriptomic high-throughput data comparing DGGH and SGGH. The study found a consistent downregulation of E-cadherin, SSTR2, and p27 kip in SGGH, indicating enhanced epithelial-mesenchymal transition (EMT) functionality compared to DGGH, which aligned with previous findings [9]. Our dataset al.so confirmed the consistent differences observed in the Wierman study (Figs. S5a-c). Moreover, Wierman’s study highlighted that the major difference between DGGH and SGGH lies in desmosome organization. The expression of key desmosome components such as DSP, PKP2, plakophilin-like protein (PERP), and others were significantly downregulated in SGGH, suggesting that alterations in desmosomes may be a crucial factor and may be used for the classification differences between SGGH and DGGH [9]. Desmosomes typically act as tumor-suppressive complexes. The absence of desmosome proteins and desmosome-mediated adhesion is associated with the development and/or progression of cancer [7].

In this study, we aimed to investigate the molecular characteristics of different histological subtypes of growth hormone-secreting pituitary neuroendocrine tumors, specifically focusing on the role of m6A modification and its impact on desmosome organization. Initially, we conducted m6A sequencing and observed the downregulation of m6A levels in the SGGH subtype. We also validated the results using m6A ELISA, external datasets, and immunohistochemistry. Through these comprehensive analyses, we discovered that FTO was significantly upregulated in SGGH. FTO is known to play a role in regulating m6A modification levels in mRNA. In our study, we found that FTO regulates the m6A modification levels of the mRNA encoding the critical desmosomal component DSP. This dysregulation of m6A modification led to a reduction in desmosome organization in SGGH. Furthermore, we investigated the value of the clinical application of FTO in growth hormone-secreting pituitary neuroendocrine tumors. We found that the downregulation of FTO decreases hypoxia tolerance in pituitary tumor cells. Additionally, we found that FTO knockdown upregulated SSTR2 levels, which in turn led to a downregulation of growth hormone synthesis and secretion. Due to the potential relationship between fibrous bodies and desmosomes [9], we also found that FTO knockdown also makes the filaments sparser in fibrous bodies. These findings highlight the role of FTO in regulating the malignant phenotype of growth hormone-secreting pituitary neuroendocrine tumors and contribute to a better understanding of the molecular mechanisms underlying the differences between SGGH and DGGH. Our studies suggest that FTO may represent an effective therapeutic target and diagnostic marker for growth hormone-secreting pituitary neuroendocrine tumors. However, further research is needed to fully elucidate the underlying mechanisms and validate the therapeutic potential of targeting FTO.

Notably, desmosome organization is not only the primary difference between the SGGH and DGGH (Fig. 3a) but also plays a significant role in the development of growth hormone-secreting pituitary neuroendocrine tumors. Using GSVA analysis, we observed a downregulation of desmosome organization in growth hormone-secreting pituitary neuroendocrine tumors compared to normal pituitary (Fig. S6a). Additionally, based on AUC analysis, most of the desmosome genes exhibited good discriminatory power in distinguishing growth hormone-secreting pituitary neuroendocrine tumors from normal pituitary samples. Among them, DSP showed remarkable discriminatory power (Fig. S6b). Principal component analysis revealed that desmosome genes can effectively separate growth hormone-secreting pituitary neuroendocrine tumors from normal pituitary samples (Fig. S6c). Besides, through the analysis of gene expression in the brain using the Human Protein Atlas database (https://www.proteinatlas.org/), we discovered genes downregulated in SGGH have a significantly higher expression in the pituitary gland compared with other brain regions, except for the retina (Figs. S6d-g). These results indicate the importance of desmosome organization in pituitary gland, and the alterations in their function are crucial for the onset and development of PitNETs.

There are some areas that deserve attention. Compared to DGGH, the proliferative capacity of SGGH is significantly enhanced. In our study, we found that the disturbance of FTO has no significant effect on the cell proliferation of growth hormone-secreting pituitary neuroendocrine tumors. Since we only performed single-gene interference at the cellular level, our interpretation of the SGGH phenotype is relatively limited. The interactions between cells, extracellular matrix, and angiogenesis functions in the tumor microenvironment may play an important role in the high proliferation of SGGH. Additionally, we found that the number of secretory granules decreased after FTO knockdown. However, the number of secretory granules in SGGH is less than in DGGH, which is very interesting and important. This contradictory finding might be partly accounted for by the comprehensive regulatory mechanisms of growth hormone secretion. Overall, FTO has the potential to inhibit growth hormone secretion and thereby improve patient prognosis. Many studies have reported that targeting growth hormone secretion can significantly improve treatment outcomes and reduce risks for a range of tumors, including neuroblastomas, glioblastomas, breast cancer, prostate cancer, and non-small-cell lung cancer [54,55,56,57]. We will continue to explore these in the future.

Conclusions

Overall, our study demonstrates the significance of FTO-mediated pathogenesis in the aggressive growth hormone-secreting pituitary neuroendocrine tumors and reveals new therapeutic targets. Moreover, the desmosome-associated genes modulated by FTO could function as innovative factors for categorizing growth hormone-secreting pituitary neuroendocrine tumors.

Data availability

Public RNA-seq datasets for growth hormone-secreting pituitary neuroendocrine tumors are accessible under GEO accessions: GSE214226. Public scRNA-seq datasets for growth hormone-secreting pituitary neuroendocrine tumors are generously provided by Professor Zhao’s team at Huashan Hospital (https://pubmed.ncbi.nlm.nih.gov/36754052/). The raw sequence data reported in this paper have been deposited in the Genome Sequence Archive in BIG Data Center, Beijing Institute of Genomics (BIG), Chinese Academy of Sciences [http://bigd.big.ac.cn/] and GEO database under restricted access: HRA007337 (MeRIP-seq for growth hormone-secreting pituitary neuroendocrine tumors samples, https://ngdc.cncb.ac.cn/gsa-human/s/v52im56T for review only), CRA016311 or GSE269321 (MeRIP-seq for GH3 cell line, https://ngdc.cncb.ac.cn/gsa/s/8WVmDLmV or https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE269321 for review only), CRA016340 or GSE269044 (RNA-seq for GH3 cell line, https://ngdc.cncb.ac.cn/gsa/s/zpBrK8CM or https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE269044 for review only). The researchers can register and login to the GSA database website [https://ngdc.cncb.ac.cn/gsa-human/] and follow the guidance of “Request Data” to request the data step by step [https://ngdc.cncb.ac.cn/gsa-human/document/GSA-Human_Request_Guide_for_Users_us.pdf] and/or by contacting zuozhx@sysucc.org.cn or jiangxiaob1@sysucc.org.cn. All requests will be reviewed by corresponding authors and the SYSUCC institutional review board. The approximate response time for accession requests is about two weeks. The access authority can be obtained for scientific research and not-for-profit use only. Once access has been granted, the data will be available to download for two months. All data supporting the findings of this work are included in this article and the supplementary materials files.

Abbreviations

PitNETs:

Pituitary neuroendocrine tumors

DGGH:

Densely granulated growth hormone-secreting pituitary neuroendocrine tumors

SGGH:

Sparsely granulated growth hormone-secreting pituitary neuroendocrine tumors

m6A:

N6-methyladenosine

DSP:

Desmoplakin

pkp2:

Plakophilin 2

PERP:

Plakophilin-like protein

CRC:

Colorectal cancer

CDS:

Coding sequence

ssGSEA:

Single-sample gene set enrichment analysis

SSTR2:

Somatostatin receptor 2

SSTR5:

Somatostatin receptor 5

cAMP:

Cyclic adenosine monophosphate

GHR:

Growth hormone receptor

EMT:

Epithelial-mesenchymal transition

References

  1. Molitch ME. Diagnosis and treatment of Pituitary adenomas: a review. JAMA. 2017;317:516–24.

    Article  PubMed  Google Scholar 

  2. Fleseriu M, Biller BMK, Freda PU, Gadelha MR, Giustina A, Katznelson L, Molitch ME, Samson SL, Strasburger CJ, van der Lely AJ, Melmed S. A Pituitary Society update to acromegaly management guidelines. Pituitary. 2021;24:1–13.

    Article  PubMed  Google Scholar 

  3. Colao A, Grasso LFS, Giustina A, Melmed S, Chanson P, Pereira AM, Pivonello R. Acromegaly. Nat Rev Dis Primers. 2019;5:20.

    Article  PubMed  Google Scholar 

  4. Wu JC, Huang WC, Chang HK, Ko CC, Lirng JF, Chen YC. Natural history of Acromegaly: Incidences, re-operations, cancers, and Mortality Rates in a National Cohort. Neuroendocrinology. 2020;110:977–87.

    Article  CAS  PubMed  Google Scholar 

  5. Terzolo M, Reimondo G, Berchialla P, Ferrante E, Malchiodi E, De Marinis L, Pivonello R, Grottoli S, Losa M, Cannavo S, et al. Acromegaly is associated with increased cancer risk: a survey in Italy. Endocr Relat Cancer. 2017;24:495–504.

    Article  PubMed  Google Scholar 

  6. Asa SL, Mete O, Perry A, Osamura RY. Overview of the 2022 WHO classification of Pituitary tumors. Endocr Pathol. 2022;33:6–26.

    Article  CAS  PubMed  Google Scholar 

  7. Dusek RL, Attardi LD. Desmosomes: new perpetrators in tumour suppression. Nat Rev Cancer. 2011;11:317–23.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Basso C, Bauce B, Corrado D, Thiene G. Pathophysiology of arrhythmogenic cardiomyopathy. Nat Rev Cardiol. 2011;9:223–33.

    Article  PubMed  Google Scholar 

  9. Kiseljak-Vassiliades K, Mills TS, Zhang Y, Xu M, Lillehei KO, Kleinschmidt-DeMasters BK, Wierman ME. Elucidating the role of the desmosome protein p53 apoptosis effector related to PMP-22 in growth hormone tumors. Endocrinology. 2017;158:1450–60.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Liu J, Dou X, Chen C, Chen C, Liu C, Xu MM, Zhao S, Shen B, Gao Y, Han D, He C. N (6)-methyladenosine of chromosome-associated regulatory RNA regulates chromatin state and transcription. Science. 2020;367:580–6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. Frye M, Harada BT, Behm M, He C. RNA modifications modulate gene expression during development. Science. 2018;361:1346–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. Zeng L, Huang X, Zhang J, Lin D, Zheng J. Roles and implications of mRNA N(6) -methyladenosine in cancer. Cancer Commun (Lond). 2023;43:729–48.

    Article  PubMed  Google Scholar 

  13. Zou Y, Zheng S, Xie X, Ye F, Hu X, Tian Z, Yan SM, Yang L, Kong Y, Tang Y, et al. N6-methyladenosine regulated FGFR4 attenuates ferroptotic cell death in recalcitrant HER2-positive breast cancer. Nat Commun. 2022;13:2672.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Xu YY, Li T, Shen A, Bao XQ, Lin JF, Guo LZ, Meng Q, Ruan DY, Zhang QH, Zuo ZX, Zeng ZL. FTO up-regulation induced by MYC suppresses tumour progression in Epstein-Barr virus-associated gastric cancer. Clin Transl Med. 2023;13:e1505.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Hu Y, Gao Q, Ma S, Yu P, Ding S, Yao X, Zhang Z, Lu S, Lu M, Zhang J, et al. FMR1 promotes the progression of colorectal cancer cell by stabilizing EGFR mRNA in an m(6)A-dependent manner. Cell Death Dis. 2022;13:941.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Lu W, Yang X, Zhong W, Chen G, Guo X, Ye Q, Xu Y, Qi Z, Ye Y, Zhang J et al. METTL14-mediated m6A epitranscriptomic modification contributes to chemotherapy-induced neuropathic pain by stabilizing GluN2A expression via IGF2BP2. J Clin Invest 2024, 134.

  17. Chang M, Wang Z, Gao J, Yang C, Feng M, Niu Y, Tong WM, Bao X, Wang R. METTL3-mediated RNA m6A hypermethylation promotes tumorigenesis and GH Secretion of Pituitary Somatotroph Adenomas. J Clin Endocrinol Metab. 2022;107:136–49.

    Article  PubMed  Google Scholar 

  18. Zhang W, Chen S, Du Q, Bian P, Chen Y, Liu Z, Zheng J, Sai K, Mou Y, Chen Z, et al. CircVPS13C promotes pituitary adenoma growth by decreasing the stability of IFITM1 mRNA via interacting with RRBP1. Oncogene. 2022;41:1550–62.

    Article  CAS  PubMed  Google Scholar 

  19. Du Q, Hu B, Feng Y, Wang Z, Wang X, Zhu D, Zhu Y, Jiang X, Wang H. circOMA1-Mediated mir-145-5p suppresses Tumor Growth of Nonfunctioning Pituitary Adenomas by Targeting TPT1. J Clin Endocrinol Metab. 2019;104:2419–34.

    Article  PubMed  Google Scholar 

  20. Chen S, Zhou Y, Chen Y, Gu J. Fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics. 2018;34:i884–90.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Bao X, Zhu K, Liu X, Chen Z, Luo Z, Zhao Q, Ren J, Zuo Z. MeRIPseqPipe: an integrated analysis pipeline for MeRIP-seq data based on Nextflow. Bioinformatics. 2022;38:2054–6.

    Article  CAS  PubMed  Google Scholar 

  22. Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M, Gingeras TR. STAR: ultrafast universal RNA-seq aligner. Bioinformatics. 2013;29:15–21.

    Article  CAS  PubMed  Google Scholar 

  23. Zhang Y, Liu T, Meyer CA, Eeckhoute J, Johnson DS, Bernstein BE, Nusbaum C, Myers RM, Brown M, Li W, Liu XS. Model-based analysis of ChIP-Seq (MACS). Genome Biol. 2008;9:R137.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Cui X, Meng J, Zhang S, Chen Y, Huang Y. A novel algorithm for calling mRNA m6A peaks by modeling biological variances in MeRIP-seq data. Bioinformatics. 2016;32:i378–85.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Quinlan AR, Hall IM. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics. 2010;26:841–2.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  26. Liao Y, Smyth GK, Shi W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics. 2014;30:923–30.

    Article  CAS  PubMed  Google Scholar 

  27. Love MI, Huber W, Anders S. Moderated estimation of Fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15:550.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Heinz S, Benner C, Spann N, Bertolino E, Lin YC, Laslo P, Cheng JX, Murre C, Singh H, Glass CK. Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol Cell. 2010;38:576–89.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Zhang Q, Yao B, Long X, Chen Z, He M, Wu Y, Qiao N, Ma Z, Ye Z, Zhang Y, et al. Single-cell sequencing identifies differentiation-related markers for molecular classification and recurrence prediction of PitNET. Cell Rep Med. 2023;4:100934.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Cui R, Duan H, Hu W, Li C, Zhong S, Liang L, Chen S, Hu H, He Z, Wang Z et al. Establishment of human pituitary neuroendocrine tumor derived Organoid and its pilot application for drug screening. J Clin Endocrinol Metab 2024.

  31. Meyer KD, Saletore Y, Zumbo P, Elemento O, Mason CE, Jaffrey SR. Comprehensive analysis of mRNA methylation reveals enrichment in 3’ UTRs and near stop codons. Cell. 2012;149:1635–46.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Asmaro K, Zhang M, Rodrigues AJ, Mohyeldin A, Vigo V, Nernekli K, Vogel H, Born DE, Katznelson L, Fernandez-Miranda JC. Cytodifferentiation of pituitary tumors influences pathogenesis and cavernous sinus invasion. J Neurosurg. 2023;139:1216–24.

    Article  CAS  PubMed  Google Scholar 

  33. Sokol E. Structure of Desmosomes. In Autoimmune Bullous Diseases: Text and Review. Edited by Horváth B. Cham: Springer International Publishing; 2022: 61–64.

  34. Zhou Y, Zeng P, Li YH, Zhang Z, Cui Q. SRAMP: prediction of mammalian N6-methyladenosine (m6A) sites based on sequence-derived features. Nucleic Acids Res. 2016;44:e91.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Luo X, Li H, Liang J, Zhao Q, Xie Y, Ren J, Zuo Z. RMVar: an updated database of functional variants involved in RNA modifications. Nucleic Acids Res. 2021;49:D1405–12.

    Article  CAS  PubMed  Google Scholar 

  36. Zheng Y, Nie P, Peng D, He Z, Liu M, Xie Y, Miao Y, Zuo Z, Ren J. m6AVar: a database of functional variants involved in m6A modification. Nucleic Acids Res. 2018;46:D139–45.

    Article  CAS  PubMed  Google Scholar 

  37. Richter JD, Zhao X. The molecular biology of FMRP: new insights into fragile X syndrome. Nat Rev Neurosci. 2021;22:209–22.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Xiong W, Knox AJ, Xu M, Kiseljak-Vassiliades K, Colgan SP, Brodsky KS, Kleinschmidt-Demasters BK, Lillehei KO, Wierman ME. Mammalian Ste20-like kinase 4 promotes pituitary cell proliferation and survival under hypoxia. Mol Endocrinol. 2015;29:460–72.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Kristof RA, Aliashkevich AF, Hans V, Haun D, Meyer B, Thees C, Schramm J. The regional oxygen saturation of pituitary adenomas is lower than that of the pituitary gland: microspectrophotometric study with potential clinical implications. Neurosurgery. 2003;53:880–5. discussion 885–886.

    Article  PubMed  Google Scholar 

  40. Vidal S, Scheithauer BW, Kovacs K. Vascularity in Nontumorous Human pituitaries and Incidental Microadenomas: a morphometric study. Endocr Pathol. 2000;11:215–27.

    Article  PubMed  Google Scholar 

  41. Vidal S, Horvath E, Kovacs K, Kuroki T, Lloyd RV, Scheithauer BW. Expression of hypoxia-inducible factor-1alpha (HIF-1alpha) in pituitary tumours. Histol Histopathol. 2003;18:679–86.

    CAS  PubMed  Google Scholar 

  42. Shimon I, Yan X, Taylor JE, Weiss MH, Culler MD, Melmed S. Somatostatin receptor (SSTR) subtype-selective analogues differentially suppress in vitro growth hormone and prolactin in human pituitary adenomas. Novel potential therapy for functional pituitary tumors. J Clin Invest. 1997;100:2386–92.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Tuggle CK, Trenkle A. Control of growth hormone synthesis. Domest Anim Endocrinol. 1996;13:1–33.

    Article  CAS  PubMed  Google Scholar 

  44. Bertherat J. Nuclear effects of the cAMP pathway activation in somatotrophs. Horm Res. 1997;47:245–50.

    Article  CAS  PubMed  Google Scholar 

  45. Mayo KE, Godfrey PA, Suhr ST, Kulik DJ, Rahal JO. Growth hormone-releasing hormone: synthesis and signaling. Recent Prog Horm Res. 1995;50:35–73.

    CAS  PubMed  Google Scholar 

  46. Spada A, Arosio M, Bochicchio D, Bazzoni N, Vallar L, Bassetti M, Faglia G. Clinical, biochemical, and morphological correlates in patients bearing growth hormone-secreting pituitary tumors with or without constitutively active adenylyl cyclase. J Clin Endocrinol Metab. 1990;71:1421–6.

    Article  CAS  PubMed  Google Scholar 

  47. Bakhtiar Y, Hirano H, Arita K, Yunoue S, Fujio S, Tominaga A, Sakoguchi T, Sugiyama K, Kurisu K, Yasufuku-Takano J, Takano K. Relationship between cytokeratin staining patterns and clinico-pathological features in somatotropinomae. Eur J Endocrinol. 2010;163:531–9.

    Article  CAS  PubMed  Google Scholar 

  48. Fougner SL, Casar-Borota O, Heck A, Berg JP, Bollerslev J. Adenoma granulation pattern correlates with clinical variables and effect of somatostatin analogue treatment in a large series of patients with acromegaly. Clin Endocrinol (Oxf). 2012;76:96–102.

    Article  CAS  PubMed  Google Scholar 

  49. Larkin S, Reddy R, Karavitaki N, Cudlip S, Wass J, Ansorge O. Granulation pattern, but not GSP or GHR mutation, is associated with clinical characteristics in somatostatin-naive patients with somatotroph adenomas. Eur J Endocrinol. 2013;168:491–9.

    Article  CAS  PubMed  Google Scholar 

  50. Landis CA, Masters SB, Spada A, Pace AM, Bourne HR, Vallar L. GTPase inhibiting mutations activate the alpha chain of gs and stimulate adenylyl cyclase in human pituitary tumours. Nature. 1989;340:692–6.

    Article  CAS  PubMed  Google Scholar 

  51. Adams EF, Brockmeier S, Friedmann E, Roth M, Buchfelder M, Fahlbusch R. Clinical and biochemical characteristics of acromegalic patients harboring gsp-positive and gsp-negative pituitary tumors. Neurosurgery. 1993;33:198–203. discussion 203.

    Article  CAS  PubMed  Google Scholar 

  52. Asa SL, Digiovanni R, Jiang J, Ward ML, Loesch K, Yamada S, Sano T, Yoshimoto K, Frank SJ, Ezzat S. A growth hormone receptor mutation impairs growth hormone autofeedback signaling in pituitary tumors. Cancer Res. 2007;67:7505–11.

    Article  CAS  PubMed  Google Scholar 

  53. Kola B, Korbonits M, Diaz-Cano S, Kaltsas G, Morris DG, Jordan S, Metherell L, Powell M, Czirjak S, Arnaldi G, et al. Reduced expression of the growth hormone and type 1 insulin-like growth factor receptors in human somatotroph tumours and an analysis of possible mutations of the growth hormone receptor. Clin Endocrinol (Oxf). 2003;59:328–38.

    Article  CAS  PubMed  Google Scholar 

  54. Di Bella G, Colori B, Scanferlato R. The over-expression of GH/GHR in tumour tissues with respect to healthy ones confirms its oncogenic role and the consequent oncosuppressor role of its physiological inhibitor, somatostatin: a review of the literature. Neuro Endocrinol Lett. 2018;39:179–88.

    PubMed  Google Scholar 

  55. Kopchick JJ, Basu R, Berryman DE, Jorgensen JOL, Johannsson G, Puri V. Covert actions of growth hormone: fibrosis, cardiovascular diseases and cancer. Nat Rev Endocrinol. 2022;18:558–73.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  56. Clayton PE, Banerjee I, Murray PG, Renehan AG. Growth hormone, the insulin-like growth factor axis, insulin and cancer risk. Nat Rev Endocrinol. 2011;7:11–24.

    Article  CAS  PubMed  Google Scholar 

  57. Boguszewski CL, Boguszewski M. Growth hormone’s links to Cancer. Endocr Rev. 2019;40:558–74.

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

We would like to express our gratitude to other members of Dongxin Lin and Jian Zheng’s research group at Sun Yat-sen University Cancer Center for their valuable suggestions on the research design.

Funding

This study was supported by the National Natural Science Foundation of China (82372624 to Xiaobing Jiang), Guangdong Basic and Applied Basic Research Foundation (2022A1515012430 and 2024A1515013102 to Xiaobing Jiang, 2021B1515020108 to Zhixiang Zuo), Guangdong Esophageal Cancer Institute Science and Technology Program (M202206 to Zhixiang Zuo) and Young Talents Program of Sun Yat-sen University Cancer Center (YTP-SYSUCC-0062 to Jialiang Zhang).

Author information

Author notes

  1. Yunzhi Zou, Xiaoqiong Bao, Depei Li and Zhen Ye contributed equally to this work.

Authors and Affiliations

  1. State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-sen University Cancer Center, Guangzhou, 510060, P. R. China

    Yunzhi Zou, Xiaoqiong Bao, Depei Li, Rong Xiang, Yuanzhong Yang, Ziming Chen, Lingxing Zeng, Chunling Xue, Hongzhe Zhao, Zeming Yan, Zekun Deng, Jintong Cheng, Guanghao Yue, Wanming Hu, Jixiang Zhao, Ruihong Bai, Jialiang Zhang, Zhixiang Zuo & Xiaobing Jiang

  2. Department of Neurosurgery, Huashan Hospital, Shanghai Medical College, Fudan University, Shanghai, 200040, China

    Zhen Ye, Boyuan Yao & Qilin Zhang

  3. Department of Pathology and Cell Biology, New York-Presbyterian Hospital, Columbia University Irving Medical Center, New York, NY, USA

    Zhe Zhu

  4. School of Pharmaceutical Sciences, Sun Yat-sen University, Guangzhou, Guangdong, 510006, China

    Zhenhua Zhang

  5. The First Affiliated Hospital of Guangdong Pharmaceutical University, Guangzhou, China

    Aiqun Liu

Authors
  1. Yunzhi Zou
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Xiaoqiong Bao
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Depei Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Zhen Ye
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Rong Xiang
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Yuanzhong Yang
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Zhe Zhu
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Ziming Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Lingxing Zeng
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Chunling Xue
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Hongzhe Zhao
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Boyuan Yao
    View author publications

    You can also search for this author in PubMed Google Scholar

  13. Qilin Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  14. Zeming Yan
    View author publications

    You can also search for this author in PubMed Google Scholar

  15. Zekun Deng
    View author publications

    You can also search for this author in PubMed Google Scholar

  16. Jintong Cheng
    View author publications

    You can also search for this author in PubMed Google Scholar

  17. Guanghao Yue
    View author publications

    You can also search for this author in PubMed Google Scholar

  18. Wanming Hu
    View author publications

    You can also search for this author in PubMed Google Scholar

  19. Jixiang Zhao
    View author publications

    You can also search for this author in PubMed Google Scholar

  20. Ruihong Bai
    View author publications

    You can also search for this author in PubMed Google Scholar

  21. Zhenhua Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  22. Aiqun Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  23. Jialiang Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  24. Zhixiang Zuo
    View author publications

    You can also search for this author in PubMed Google Scholar

  25. Xiaobing Jiang
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

Xiaobing Jiang, Zhixiang Zuo, Jialiang Zhang, Zhe Zhu and Aiqun Liu conceived and designed the study. Yunzhi Zou, Depei Li, and Yuanzhong Yang did most of the experiments. Ziming Chen, Lingxing Zeng, Chunling Xue, Hongzhe Zhao, and Ruihong Bai assisted in carrying out the experiments. Xiaoqiong Bao, Yunzhi Zou, Zhen Ye, Rong Xiang, Jixiang Zhao, and Zhenhua Zhang contributed to data processing and analysis. Zhixiang Zuo, Boyuan Yao, and Qilin Zhang supervised the data processing and analysis. Yunzhi Zou, Zhen Ye, Depei Li, Zeming Yan, Zekun Deng, Jintong Cheng, and Guanghao Yue contributed to the sample and information collection. Yuanzhong Yang, Zhe Zhu, and Wanming Hu conducted pathological analysis. Yunzhi Zou, Jialiang Zhang, Xiaoqiong Bao, and Rong Xiang prepared and revised the figures and drafted the manuscript. Xiaobing Jiang and Zhixiang Zuo supervised the research. All authors read and approved the final manuscript.

Corresponding authors

Correspondence to Aiqun Liu, Jialiang Zhang, Zhixiang Zuo or Xiaobing Jiang.

Ethics declarations

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary Material 2

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zou, Y., Bao, X., Li, D. et al. FTO-mediated DSP m6A demethylation promotes an aggressive subtype of growth hormone-secreting pituitary neuroendocrine tumors. Mol Cancer 23, 205 (2024). https://doi.org/10.1186/s12943-024-02117-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12943-024-02117-5

Keywords

Molecular Cancer

ISSN: 1476-4598