A paracrine circuit of IL-1β/IL-1R1 between myeloid and tumor cells drives genotype-dependent glioblastoma progression

Monocytes and monocyte-derived macrophages (MDMs) from blood circulation infiltrate glioblastoma (GBM) and promote growth. Here, we show that PDGFB-driven GBM cells induce the expression of the potent proinflammatory cytokine IL-1β in MDM, which engages IL-1R1 in tumor cells, activates the NF-κB pathway, and subsequently leads to induction of monocyte chemoattractant proteins (MCPs). Thus, a feedforward paracrine circuit of IL-1β/IL-1R1 between tumors and MDM creates an interdependence driving PDGFB-driven GBM progression. Genetic loss or locally antagonizing IL-1β/IL-1R1 leads to reduced MDM infiltration, diminished tumor growth, and reduced exhausted CD8+ T cells and thereby extends the survival of tumor-bearing mice. In contrast to IL-1β, IL-1α exhibits antitumor effects. Genetic deletion of Il1a/b is associated with decreased recruitment of lymphoid cells and loss-of-interferon signaling in various immune populations and subsets of malignant cells and is associated with decreased survival time of PDGFB-driven tumor-bearing mice. In contrast to PDGFB-driven GBM, Nf1-silenced tumors have a constitutively active NF-κB pathway, which drives the expression of MCPs to recruit monocytes into tumors. These results indicate local antagonism of IL-1β could be considered as an effective therapy specifically for proneural GBM.


Mice
Mice of both sexes (equal distribution) in the age range of 8-16 weeks were used for experiments (1,2).Previously-described Il1b -/-, and Il1a/b -/-mice were crossed with Ntv-a mice to generate Il1b -/-;Ntv-a, Il1a -/-;Ntv-a and Il1a -/-;Il1/b -/-;Ntv-a mice (3)(4)(5).Il1b mutant mice referred as Il1b -/- express RNA for Il1b but both Il1b -/-, and Il1a/b -/-mice lack protein expression (6).Mice used in this study are all in a C57BL/6J background, except for Ntv-a;Cdkn2a -/-;Pten fl/fl , which is in a mixed genetic background.C57BL/6J mice (000664) at 8-12 weeks old were purchased from the Jackson labs and used for BMDM isolation.All animals were housed in a climate-controlled, pathogenfree facility with access to food and water ad libitum under a 12-hour light/dark cycle.Tumorbearing mice were euthanized at humane endpoint to ensure comparable tumor burden across all animals.Humane endpoint was defined as the terminal stage of life where the mice exhibit symptoms including head tilt, lethargy, seizures, and excessive weight loss (10-15% total loss).All experimental procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of Emory University (Protocol #201700633) and the Icahn School of Medicine at Mount Sinai (Protocol #201900619).

Virus generation and tumor induction
We delivered RCAS-PDGFB in Cdkn2a -/-; Ntv-a mice, or the combination RCAS-PDGFB with RCAS-shRNA-p53-Rfp in Ntv-a mice, both of which have been shown to closely resemble human PN GBM in terms of histological and molecular features (1, 2).To generate a murine MES model, we chose to silence the tumor suppressor genes NF1, TP53, and PTEN by co-injection of RCAS virus caring short hairpin RNA of these molecular targets, along with introducing PDGFA, as previously described (2).To propagate RCAS viral vectors, DF-1 cells (ATCC, CRL-12203) were purchased and grown at 39°C according to the supplier's instructions.The cells were routinely tested for mycoplasma contamination to assure they are negative for mycoplasma.Cells were transfected with RCAS-hPDGFB-HA, RCAS shRNANf1, RCAS shRNAPten, and RCAS-shRNA-p53-Rfp using a Fugene 6 transfection kit (Roche, 11814443001) according to the manufacturer's instructions.DF-1 cells (4x10 4 ) in 1 µl neurobasal medium were stereotactically delivered with a Hamilton syringe equipped with a 30-gauge needle for tumor generation (7).The target coordinates were in the right-frontal striatum at AP 1.0 mm and right 1.5 mm from bregma; depth 1.5 mm from the dura surface (1, 2, 7).Mice were continually monitored for signs of tumor burden and were sacrificed upon observation of endpoint symptoms including head tilt, lethargy, seizures, and excessive weight loss.

Orthotopic glioma generation
The same procedure was used as described above, except 3x10 4 of freshly-dissociated tumor cells were injected in the right-frontal striatum AP 1.0 mm and right 1.5 mm from bregma; depth 1.5 mm from the dura surface of recipient animals.Two or three donor tumors of either sex were used to obtain single cell suspension for orthotopic glioma generation in male and female recipient animals.

Microglia and monocyte-derived macrophage isolation and culture
Microglia were isolated from postnatal day 0 (p0) to p3 pups using a modification of a previouslydescribed protocol (8).Briefly, whole brains were extracted from p0-p3 pups using sterile instruments followed by consecutive washes in DPBS.Brains were then digested in 0.5% trypsin (ThermoFisher, 15400054) and DNase I (Sigma-Aldrich, 11284932001) followed by mechanical dissociation prior to filtration with a 70 µm cell strainer.The filtrate was plated in flasks precoated with poly-D-lysine (Sigma-Aldrich, P6407) in DMEM (ThermoFisher, 10569010) with 10% FBS (HyClone, SH30396.03) and M-CSF (Peprotech, 315-02) at a concentration of 40 ng/ml.The cells grew until the astrocytes reached confluence on the bottom of the flask and microglia were observed ballooning off of the surface.During this time, additional M-CSF was periodically added.Microglia were collected by gently rinsing the back of the flask with culturing media and were plated in 6-well plates in 2 ml of the culturing media with M-CSF at a concentration of 40 ng/ml for experimentation.C57BL6/J mice (8-12 weeks of age) were euthanized via CO2 asphyxiation.The whole, intact femur and tibia were stripped of muscle and collected in sterile Dulbecco's phosphate-buffered saline (DPBS, ThermoFisher, 14190-250).Both ends of the bone were cut and the marrow was flushed into a clean petri dish using DPBS supplemented with 4% bovine serum albumin (BSA, ThermoFisher, 15260037), heparin (StemCell, 07980), DNase I (Sigma-Aldrich, 11284932001), and penicillin/streptomycin (ThermoFisher, SV30010).The resulting mixture was briefly triturated and passed through a 70 µm cell strainer.The cells were plated in a 15 cm non-cell culture-treated plate for a six-day differentiation in 15 ml of DMEM (ThermoFisher, 10569010) with 10% fetal bovine serum (FBS; HyClone, SH30396.03) and 40 ng/ml macrophage colonystimulating factor (M-CSF, Peprotech, 315-02).An additional 15 ml of media with M-CSF was added after day three.The cells were harvested for experimentation via a 10-minute incubation in DPBS with 5mM EDTA on ice.Cells were plated in 6-well plates in 2 ml of culturing media with M-CSF at a concentration of 20 ng/ml for experimentation.

Organotypic tumor slice culture
Ntv-a mice harboring PDGFB-overexpressing and p53-silenced tumors were euthanized at endpoint via carbon dioxide asphyxiation.The brain was rapidly extracted and embedded in 4% low-melt agarose in PBS.The embedded brain was then mounted on a vibratome (Leica, 1220S) and submerged in ice cold Hank's balanced salt solution (HBSS, Gibco 14185052).The brain was cut into 300 µm-thick sections and the slices were transferred to inserts in a 6-well plate.Slices were cultured in Neurobasal media (StemCell, 05700) supplemented with B27 supplement (ThermoFisher, 17504044), sodium pyruvate (ThermoFisher, 11360070), and glutamine (ThermoFisher, 35050061).M-CSF (BioLegend, 576406) at a concentration of 40 ng/ml was included in the media during co-culture experiments with BMDM.

MES tumor cell cultures and in vitro stimulation
Murine MES GBM lines (1816 and 4622) were cultured as either neurospheres or FBS cultures as previously described (9)(10)(11).Mycoplasma tests (Lonza, LT07) were performed monthly to ensure no such contamination occurred in the cultures.For treatment in the presence of NF-kB inhibitors, the cells were sub-cultured at 10^5 cells/well in a 24-well plate pre-coated with Geltrex (Life Technologies, A14132-01) so that the cells grew in a monolayer fashion, evenly exposed to the compounds.To stimulate BMDM or microglia with spent medium of GBM cell lines, supernatant was collected from each well of the cultured GBM cells and transferred to BMDM or microglia cultures.

Tumor and cultured cell RNA isolation and qPCR analysis
At humane endpoint, mice were sacrificed with an overdose of ketamine/xylazine and immediately perfused with ice cold Ringer's solution (Sigma-Aldrich, 96724-100TAB).The brain was extracted, and a piece of tumor was immediately snap-frozen in liquid nitrogen for storage at -80°C.Alternatively, cultured cells were harvested from plates using TRIzol (ThermoFisher, 15596026).RNA was isolated from the frozen tumor pieces or cells with the RNeasy Lipid Tissue Mini Kit (Qiagen, 74804) according to the manufacturer's instructions.RNA quantity was assessed with a NanoDrop 2000 spectrometer (ThermoFisher), while quality was confirmed via electrophoresis of samples in a 1% bleach gel as previously described (12).RNA was used to generate cDNA with a First Strand Superscript III cDNA synthesis kit (ThermoFisher, 18080051) according to the manufacturer's instructions and with equal amounts of starting RNA.Quantitative-PCR was performed with the validated BioRad PCR primers using SsoAdvanced Universal green Supermix (BioRad, 1725271, detailed in Table S1).Fold changes in gene expression were determined relative to a defined control group using the 2 -ΔΔCt method or by zscore, with β-Actin or HPRT used as housekeeping genes.

Human tissue samples and pathological appraisal
Archived formalin-fixed, paraffin-embedded (FFPE) human GBM samples and de-identified clinical information were provided by Emory University (IRB Study 18-177) and patient information is included in our published manuscript (13).Fresh tumor tissues used for ELISA quantification were collected at Mount Sinai Hospital through the biorepository, under IRBapproved protocols (18-00983).All patient samples were de-identified.Ten randomly selected tumor samples were included in this study, with de-identified age, sex, and molecular information summarized in Table S3."Normal appearing" brain tissues away from the tumor mass, obtained during surgical resection, from three patient samples were used as controls.Board-certified neuropathologists diagnosed and graded both the human tumor tissues and murine samples according to the 2016 World Health Organization Classification of Tumors of the Central Nervous System (14).Gene expression profiling to determine transcriptional subtypes was performed using targeted genome sequencing (Foundation One/Sema4).

TCGA analysis
U133 Microarray data for the GBM (TCGA, provisional) dataset were downloaded from cBioPortal (https://www.cbioportal.org) in August 2021 and sorted into subtypes based upon a proprietary key.G-CIMP-positive tumors were excluded from analysis.We included 372 patient samples for which covariate information (survival information, age, and gender) was available.All of the TCGA dataset samples analyzed in this manuscript were IDH-wildtype malignant gliomas, consistent with the diagnosis of (IDH-wildtype) Glioblastoma according to the WHO 2021 CNS tumor classification.We have now added this information to the methods section.Cox Proportional Hazard Models were fitted in R using age and gene expression as continuous covariates, and gender as a binary variable.Forest plots were done using the function ggforest.

Tissue processing and immunohistochemistry
Archived FFPE human GBM samples were sectioned at 5 μm thickness, slide-mounted, and stored at -80°C until use.To process mouse tumor tissues, animals at humane endpoint were anesthetized with an overdose of ketamine/xylazine mix and transcardially perfused with icecold sterile Ringer's solution.Brains were removed and processed according to the different applications.For H&E tumor validation and immunohistochemistry staining, brains were fixed in 10% neutral buffered formalin for 72 hours at room temperature (RT), processed in a tissue processor (Leica, TP1050), embedded in paraffin, sectioned (5 μm), and slide mounted.

Flow Cytometry and spectral flow cytometry
Mice at humane endpoint were anesthetized with an overdose of ketamine/xylazine mix and transcardially perfused with ice-cold sterile Ringer's solution (catalog number).Initial steps of the enzymatic dissociation of the tumors are the same as described above, except 0.5% collagenase D (Sigma, 11088858001) and DNase I (Roche, 11284932001) were used in place of papain.Singlecell suspensions were passed through 70 μm cell strainers, centrifuged, and resuspended in 30% Percoll (GE Healthcare, 17-0891-01) solution containing 10% FBS (Hyclone SH30396.03).Cells were separated by centrifugation at 800g for 15 minutes at 4°C.The supernatant was carefully removed to discard debris and lipids.The cells were then washed in cold PBS and resuspended in RBC lysis buffer (BioLegend, 420301) for 1 min at 37°C.Cells were transferred to an Eppendorf tube and washed once with FACS buffer (DPBS with 0.5% BSA) and blocked with 100 μl of 2x blocking solution (2% FBS, 5% normal rat serum, 5% normal mouse serum, 5% normal rabbit serum, 10 μg/ml anti-FcR (BioLegend, 101319) and 0.2% NaN3 in DPBS) on ice for 30 minutes.Cells were then stained with primary antibodies (Table S2) on ice for 30 minutes and washed with PBS.The cells were subsequently incubated in 100 μl viability dye (Zombie UV, BioLegend, 1:800) at room temperature for 20 min.The cells were washed and fixed with fixation buffer (eBioscience, 00-5123-43, 00-5223-56) for 30 min.Cells stained with the cocktail of antibodies examined myeloid lineage are set aside in the fridge until loading to the cytometer.Cells stained for the lymphoid panel were then permeabilized with a permeabilization buffer (eBioscience, 00-8333-56) before the intracellular markers were stained.The cells were washed and stored in fridge till analysis.Antibodies used in this study include are listed in Table S2.All data were collected on a BD LSR II flow cytometer or Cytek Aurora spectral flow cytometer.Data were analyzed offline using FlowJo 10 software (Tree Star Inc.).
Fluorescence-activated cell sorting (FACS) was used to enrich tumor-associated microglia, BMDM or glioma cells allowing downstream quantitative PCR analysis.Mice at humane endpoint were anesthetized with an overdose of ketamine/xylazine mix and transcardially perfused with icecold sterile Ringer's solution.Tumors are dissected and dissociated with papain as described above.The cells were stained with a cocktail of antibodies as previously described (15) and were sorted on a Sony SH800 cell sorter.Normal microglia was similarly sorted, except naïve mouse brains were used.Microglia are defined as CD45 Lo CD11b + Ly6c Neg Ly6g Neg ; BMDM are CD45 Hi CD11b + Ly6c + Ly6g Neg ; Glioma cells are negative for all markers.

Single-cell RNA-seq and data analysis
Mice at humane endpoint were anesthetized with an overdose of ketamine/xylazine mix and transcardially perfused with ice-cold sterile Ringer's solution.Single cell suspensions of the tumors were obtained by papain dissociation as described above.Viability of single cells was assessed using Trypan Blue staining, and debris-free suspensions of >80% viability were deemed suitable for single cell RNA Seq.Single cell RNA Seq was performed on these samples using the Chromium platform (10X Genomics) with the 3' gene expression (3' GEX) V3 kit, using an input of ~10,000 cells.Briefly, Gel-Bead in Emulsions (GEMs) were generated on the sample chip in the Chromium controller.Barcoded cDNA was extracted from the GEMs by Post-GEM RT-cleanup and amplified for 12 cycles.Amplified cDNA was fragmented and subjected to end-repair, poly A-tailing, adapter ligation, and 10X-specific sample indexing following the manufacturer's protocol.Libraries were quantified using Bioanalyzer (Agilent) and QuBit (ThermoFisher) analyses and were sequenced in paired end mode on a NovaSeq instrument (Illumina) targeting a depth of 50,000-100,000 reads per cell.
Raw fastq files were aligned to mouse genome reference mm10 customized to include the Rfp sequence, using CellRanger v5.0.0 (10X Genomics).Count matrices filtered by CellRanges algorithm were further filtered by discarding cells with either < 400 genes, < 1000 UMI (unique molecular identifier), or > 25% mitochondrial genes expressed.Data was processed and analyzed using R package Seurat v4.0.5.Normalization was performed using NormalizeData function with normalization.method= 'LogNormalize'.Dimensionality reduction was computed on the top 2,000 variable features using FindVariableFeatures, ScaleData and RunPCA functions.UMAPs were generated using the top 15 PCs.For subclustering the immune compartment, we used R package Harmony v0.1 to mitigate batch effects driven by technical variation between replicates.De novo clustering using the Louvain algorithm was applied at different resolutions (0.2; 0.8; 2; 5) on the KNN graph space.For high-level annotation, cell types were identified in an iterative and semi-supervised fashion by assigning de novo discovered clusters to cell types based on expression of known marker genes that define each cluster.Annotation of cell subtypes at a lowerlevel was performed in a similar manner as for the high-level and further aided by de novo marker discovery using the Seurat FindMarkers function and Wilcoxon Rank Sum differential expression test.To identify doublet-enriched clusters we looked for clusters of cells displaying expression of canonical markers for two or more different cell types and higher number of genes/UMI; such clusters were removed from further analysis.Cell-level proliferation analysis was carried out with Seurat function CellCycleScoring using a list of murine cell cycle genes.Cells were then assigned as 'cycling' or 'non-cycling' if either their S-phase score or G2M-score was above a threshold of 0.05 for the tumor compartment and 0.1 for the immune compartment.
To generate the pie charts shown in Supplementary Figures 15 and 16, we tallied the total counts of the molecules of interest and traced their source to either malignant cells or non-malignant cells.The total counts obtained from the tumor compartment were then divided by the numbers of tumor cells expressing these molecules.Similarly, this ratio was also obtained from the nonmalignant compartment.These two ratios were subsequently used to create the pie chart.

Hematoxylin and eosin tumor volume reconstruction
Mice were sacrificed 25 days post-tumor cell transplantation with an overdose of ketamine and xylazine and perfused with 4% paraformaldehyde in PBS.The brain was carefully extracted and incubated in 4% paraformaldehyde in PBS for 24 hours following a 72-hour incubation in 30% sucrose in PBS.The brain was then embedded in O.C.T. compound (VWR, 25608-930) and frozen on dry ice.The entire brain was then serially sectioned on a cryostat (Leica) set to cut 30 µm sections.Every tenth section was collected and mounted on a slide for automated hematoxylin and eosin staining as described above.The slides were scanned at 20x magnification with a whole-slide scanner (Hamamatsu).Tumor area in each section was determined in a blinded fashion in NDP.view2 and multiplied by the thickness of ten slices.The resulting volumes for the slides of each tumor were then summed, producing an estimation of the total volume.

Cannula installation and drug administration in tumor-bearing mice
Mice were anaesthetized with a ketamine (100 mg/kg) and xylazine (10 mg/kg) cocktail, their heads shaved, and they were placed into a stereotactic device with a three-axis micromanipulator.The burr hole used to inject the RCAS virus was re-probed with a compact drill.A guide cannula (Plastics One) was implanted into the brain through the craniotomy and fixed in place with dental acrylic.A matching stylet or "dummy-cannula" was then screwed onto the guide cannula to prevent back flowing of the CSF or environmental debris from entering the guide cannula.Once the dental acrylic is cured, the skin around the pedestal of the guide cannula was sealed with Gluture liquid stitches (World Precision Instruments, 503763).
On the day of the infusion, the mice were anesthetized with inhalable isoflurane at 3% (volume/volume in pure oxygen) with a nebulizer.The dummy cannula was removed, and a Hamilton syringe mounted on a stereotactic frame was lowered into the guide cannula.Up to 2 µl of vehicle (artificial CSF, Harvard Instrument, containing isotype antibody, BE0091, BioXcell), IL-1RA (US Biological Corporation, I7663-62E), or purified anti-IL-1β IgG (BioXcell, BE0246) were injected into the lateral ventricle over 1 min.Once the injection was completed, the dummy cannula was screwed back, and mice were returned to their home cages.

In situ proteomics by NanoString GeoMx assay
Formalin-fixed, paraffin-embedded sections of 5 µm thickness were placed onto positivelycharged slides (ThermoFisher).The GeoMx Immune Cell Profiling Panel Mouse Protein Core with a 20-gene selection was used for nCounter readout along with morphology markers pancytokeratin (Alexa-488), leukocyte common antigen CD45 (Alexa-647), and DAPI (nuclei) staining for immunofluorescence labelling.Six circular ROIs 300 µm in diameter were utilized for targeted spatial transcriptomics using the GeoMx platform (NanoString, Seattle, WA).ROIs were generated in the tumor-rich region of the tissue sections, particularly adjacent to perivascular or perinecrotic spaces.Once digital spatial profiling data was completed, the samples were pooled for nCounter readout in RCC format.
RCC files generated from the nCounter readout were plugged into the digital spatial profiling software for initial QC and further analysis.ROIs with nuclei count less than 20 were excluded from further analysis.Data was subsequently normalized against GAPDH and Histone H3, two housekeeping genes.The normalized data were used for comparisons between treatment groups and vehicle controls, and to generate corresponding heatmaps.

STable 1 .
qPCR Primers Used in Study.The Bio-Rad qPCR primers used in the study are listed with their catalog numbers.