Clinical samples and information
Eight GHPA clinical samples were obtained during neurosurgery and sent for hPASC culture following RNA-sequencing. Schematic experimental procedure is illustrated in Fig. 1A. The detailed clinical information of each patient is shown in Supplemental Table 1. For verifying the correlation of interactive genes, eighteen GHPA samples were subsequently collected following the same procedures and used for Real-time PCR. Their detailed clinical information is available in Supplemental Table 2. All the samples were pathologically diagnosed as GHPA by the pathologists. These studies were approved by the Ethics Committee of Beijing Tiantan Hospital, and written informed consent was available for all patients.
Human PASC culture and verification
Briefly, immediately after the GHPA sample was resected during an operation, it was quickly divided into two portions. One portion was “snap-frozen” and stored in liquid nitrogen for pathological immunostaining and sequencing, the other portion was processed in the laboratory for hPASC culture. The protocol for culturing hPASCs was same as our previous report [8]. The hPASCs were cultured in the stem cell medium [DMEM/F-12 with 2 mM L-glutamine, 1% penicillin-streptomycin, 1X B27 (50X, Life Technologies, USA), 20ng/ml basic fibroblast growth factor (bFGF, Peprotech) and 20ng/ml epidermal growth factor (EGF, Peprotech)] at 37°C with 5% CO2 in an incubator. The stem cell medium was refreshed every 3 days. For differentiation, 14-day hPASCs were seeded in a culture plate and supplemented in the differentiation medium [DMEM/F-12 with 15% fetal bovine serum (FBS), 2 mM L-glutamine, and 1% penicillin-streptomycin] at 37°C with 5% CO2.
Immunohistochemistry
The hPASCs were cultured in a poly-D-lysine-coated plate and washed with 1X PBS three times. The cells were fixed with 4% paraformaldehyde for 20 min at room temperature and incubated in 0.3% Triton™ X-100 (Sigma-Aldrich, USA) for 10 min. After washing with 1X PBS three times, 5% bovine serum albumin (BSA) was added to the cells for blocking at room temperature for 1 h. The processed cells were incubated with primary antibody at 4°C overnight. The next day, the cells were washed with 1X PBS and incubated with secondary antibody conjugated to Alexa-Fluor 488 or Alexa-Fluor 647 (abcam, USA) at 1:1000 dilution for 1 h at room temperature. For tissue staining, the samples were sectioned at a thickness of 6 µm per slice and subjected to the same fixation procedure. The section was blocked with 5% serum in 1X PBS with 0.3% Triton™ X-100 and incubated overnight with primary antibodies. The next day, after washing with 1X PBS, the sections were incubated with secondary antibodies in 1X PBS for 1 h at room temperature. After washing, 4',6-diamidino-2-phenylindole (DAPI, Sigma-Aldrich, USA) was incubated for 10 min at room temperature.
The images were captured using a Zeiss microscopic imaging system with fluorescence emission system at different magnifications. All antibodies were purchased from Abcam unless otherwise stated. The antibodies used in the experiments were as follows: mouse anti-Sox2 (ab79351, 1:800), rabbit anti-Oct4 (ab19857, 1:500), rabbit anti-Nestin (ab105389,1:200), mouse anti-CD133 (ab264538, 1:500), rabbit anti-CXCR4 (ab181020, 1:500), rabbit anti-CXCL12(17402-1-AP, Proteintech).
Enzyme-linked immunosorbent assay (ELISA)
ELISA was performed to quantify the level of growth hormone in the culture medium or mouse blood according to manufacturer’s instructions (Elabscience, USA). The protocol was in consistence with previous report [8]. The final mixture was evaluated quantitatively at 450 nm using a spectrophotometer (Bio-Tek, USA).
RNA extraction and RNA-sequencing
Total RNA from the hPASCs or tumor samples was extracted by TRIzol reagent according to the manufacturer’s instructions (Invitrogen, USA). After purification, the quality and quantity of RNA were assessed using the Bioanalyzer 2100 system (Agilent Technologies, USA) and stored at − 80°C before use. RNA sequencing was performed by the commercial provider Novogene Co. Ltd. The double-stranded DNA was synthesized from the cDNA generated from the extracted RNA. The index-coded samples were performed on a cBot Cluster Generation System using TruSeq PE Cluster Kit v3-cBot-HS (Illumina) according to the manufacturer’s instructions. After cluster generation, the library preparations were sequenced on an Illumina Novaseq platform. After read mapping, the counts for differentially expressed genes from each sequenced library were adjusted and analyzed using the “edgeR” package from R software (3.22.5). The P-values were adjusted by the Benjamini & Hochberg method. An adjusted P-value < 0.05 and fold change of -0.5 > Log2FC > 0.5 were set as the threshold for selecting differentially expressed genes.
Enrichment analysis
Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) functional enrichment analyses of differentially expressed genes were conducted with R package ‘clusterProfiler’. Biological processes (BP), molecular function (MF), and cellular components (CC) were covered in the GO enrichment analysis. A p value < 0.05 was set as the significance threshold. The KEGG and GO terms were visualized using the package ggplot2 of R software.
Protein-protein interaction (PPI) and MCODE analysis
Interaction of differentially expressed genes was predicted by the Search Tool for the Retrieval of Interacting Genes Database (STRING) (https://cn.string-db.org/) database, at a confidence score > 0.4. The PPI network was visualized using Cytoscape software. Molecular complex detection (MCODE) is a plugin in Cytoscape used to filter the significant modules of core genes from the PPI network complex. The criteria were set as follows: Degree Cutoff = 2, Node Score Cutoff = 0.2, K-Core = 2, and Max. Depth = 100 [9]. The enrichment analysis of the hub genes in each cluster were performed following the above methods.
Knockdown of CXCR4 in GH3 cells
The small interfering RNAs (siRNA) for knocking down CXCR4 and the negative control siRNA (siCtrl) (Supplemental Table 4) were synthesized by DIA-UP Biotech (Beijing, China). Transfection was performed according to the manufacturer’s instructions using lipofectamine 3000 (Invitrogen).
Real-time Polymerase chain reaction
Total RNA (2µg) was used as template for synthesizing the first-strand cDNA mixed with a reverse transcription mixture (RevertAid™ First Strand cDNA Synthesis Kit, Thermofisher, USA). Real-time PCR quantification was performed using SYBR Green PCR Master Mix (TaKaRa, Japan) in a TaKaRa 7500 Real-time system (TaKaRa, Japan). GAPDH served as the internal control. The relative expression level was quantified using the 2−∆∆ct method. The primers used for qPCR are mentioned in Supplemental Table 5.
Western blot
CXCR4 protein level was measured using Western blot. Briefly, total protein (50ug) from GH3 cells was extracted and separated on 12% SDS-PAGE gels, then it was transferred to polyvinyl difluoride membranes and immunoblotted with anti-CXCR4 antibody at 4°C overnight. Next day, the membrane was incubated with a bovine anti-rabbit IgG coupled to peroxidase as the secondary antibody at room temperature for 1h. The band was detected in ECL system (Thermo Scientific, Inc) using chemiluminescent horseradish peroxidase substrate.
Cell proliferation assay (CCK8)
Cell proliferation assay was performed using a Cell Counting Kit-8 assay according to the manufacturer’s instructions (CCK-8, Solarbio, China). The absorbance of cell supernatant aliquots at 450nm was measured using a spectrophotometer (Bio-Tek, USA).
Flow Cytometry Assay
GH3 cells were plated in 6-well plates at a concentration of 2×105 cells/well and treated with different concentrations of AMD3465 for 72 hours. For apoptosis assay, cells were rinsed with 1X PBS, resuspended in the staining buffer, and incubated with Annexin V-FITC and propidium iodide (BD Biosciences) for 15 min. The stained cells were washed with PBS and analyzed using the CytoFlex cytometer (Beckman). The data were analyzed with FlowJo software (V10, BD Biosciences).
Xenograft experiment
GH3 xenograft in vivo model was generated using BALB/c nude mice (Charles River, Beijing, China). Six- to eight-week-old female mice were subcutaneously injected with GH3 cells (1.0×105). Three weeks after injection, the mice were randomized allocated into three groups. AMD3465 was subcutaneously administered once daily at a dose of 2mg/kg (low dose group), 4mg/kg (high dose group) or PBS control group. Tumor size, weight and GH level were assessed every week. All mice were euthanized at week 6 since injection and the tumors were excised, weighed, and frozen at -80°C. A piece of individual tumor sample was used for immunostaining. All animal experiments were approved by the Institutional Animal Care and Use Committee, Beijing Neurosurgical Institute.
Molecular imaging
Seven GHPA patients were enrolled (clinical information is provided in Supplemental Table 3). All of the patients received pre-operative 68Ga-pentixafor PET/CT and 18F-FDG PET/CT and received surgical treatment later. The pathological diagnosis verified GHPA for each patient. Semi-quantitative analysis was used to evaluate PET imaging. For each patient, the volume of interest (VOI) was drawn, and the maximum standardized uptake values (SUVmax), mean standardized uptake values (SUVmean), and maximum tumor-to-background ratios (TBRmax) were recorded. The TBRmax was recorded by dividing the SUVmax of the VOIs by that of the brainstem. This pilot study was approved by the ethic committee of UPMC.
Statistical analysis
Statistical analyses were performed using GraphPad Prism 6.0 software (GraphPad Software, Inc., USA). Data are shown as the mean ± standard error of the mean (SEM) or the mean ± standard deviation (SD). A t-test was used to compare the statistical differences. Spearman correlations were used to test for correlated gene expression of CXCR4. P-values < 0.05 was considered to indicate a significant difference.