Deficiency of factor-inhibiting HIF creates a tumor-promoting immune microenvironment

Significance Under pathological conditions, hypoxia and the hypoxic tissue microenvironment influence cell behavior and resistance to therapies. Increasing numbers of modulators, including those targeting factor-inhibiting hypoxia-inducible factor [(HIF) (FIH)], an enzymatic inhibitor of HIF, have been developed to alter HIF activity and to treat diseases including cancer. Understanding how FIH functions under physiological conditions is therefore vital. Here, we identify FIH as a key regulator of immune homeostasis and suppressor of B cell lymphomagenesis throughout the physiological aging process. FIH deficiency in the host or in myeloid cells alone creates a tumor-supportive tumor microenvironment to promote cancer cell growth. Although the current work is limited to transgenic mouse studies, this information may inform future therapeutic application of HIF and FIH modulators.

histologically diagnosed low-grade, low-grade with high-grade components, and high-grade B cell lymphomas in indicated tissues.Frequencies were obtained by dividing the number of mice with indicated type of lymphoma by the total number of mice of this genotype.A higher frequency of mice with low-grade B cell lymphoma in the lung was observed in the FIH Δ1-2/Δ1-2 mice (FIH +/+ vs FIH Δ1-2/Δ1-2 mice: 0 vs 50%, p=0.0020 by  2 test).

Materials and Methods
Antibodies and cell lines are listed in Supplementary Tables 9-11.

Mouse colonies
Mice were housed in the Functional Genetics Facility of the Wellcome Trust Centre for Human Genetics (University of Oxford) in individually ventilated cages with food and water provided ad libitum and on a 12-hour light/12-hour dark cycle.All animal procedures were approved by the University of Oxford's ethical review committee and licensed by the UK Home Office (license number PPL 30/2862).

Spontaneous tumor analysis
Mice that showed signs of illness or visible tumors were culled and subjected to necropsy.Heart, lung, thymus, liver, pancreas, stomach, intestine, and lymph nodes were collected, and tissue histopathology was analysed in collaboration with pathologists Profs.Robert Goldin (Imperial College London) and Elizabeth J. Soilleux (University of Cambridge).

Preparation of FIH protein and antibody generation
FIH constructs d25 and d36 (the d25 construct means the protein starts from the 26 th amino acid whereas the d36 starts from the 37 th amino acid) were subcloned in a pET28a between a NheI and BamHI restriction enzyme cleavage site.Plasmids were transformed in E. coli BL21 (DE3) cells and a single colony was picked and grown in LB medium (100 mL) supplemented with kanamycin (30 µg/mL).The stationary starter culture was diluted 1:100 in fresh medium supplemented with the appropriate antibiotics and grown to an OD600~0.8 at 37°C prior to induction with isopropyl β-D-thiogalactopyranoside (IPTG, final concentration 0.5 mM) for 4 h at 37°C.Cells were harvested via centrifugation (10,000 g, 8 min, JA-10 rotor, Beckman Coulter).The cell pellet was stored at -80 °C.FIH containing pellets were re-suspended in 50 mM HEPES buffer (pH = 7.5, 500 mM NaCl, 5 mM imidazole) with added DNaseI from bovine pancreas (Roche) and lysed by sonication.The lysate was centrifuged (4°C, 34,000 g, 30 min, JLA1625 rotor, Beckman Coulter) and the supernatant filtered to remove cell debris.His-tagged proteins were immobilized on a His-Trap (5 mL) column, washed with low imidazole buffer (50 mM HEPES pH = 7.5 buffer supplemented with mM HEPES pH = 7.5, 500 mM NaCl, 500 mM imidazole).The N-terminal His6-tag was cleaved by the addition Thrombin (4 units per mg) overnight at 4°C.Proteases and cleaved tag were removed via SEC using a Superdex S200 column equilibrated with 50 mM Tris-HCl pH = 7.5.Completeness of the cleavage was verified by Mass Spectrometry.
Splenocytes from BALB/c mice hyperimmunised with the recombinant FIH protein d25 were fused with a non-producing mouse myeloma cell line SP2 using polyethylene glycol as a fusogen.
Supernatants of selected hybrids were screened primarily using dot-blot on nitrocellulose membrane coated with FIH protein d25 and d36 followed by validation in various murine primary tissues with Western blotting.

Syngeneic tumor model
Mice with matched sex and age (8-14 weeks) were selected as experimental animals.4×10 5 LLC cells or 1×10 5 B16 cells suspended in 100 µL of PBS were injected subcutaneously.The length and width of the tumor mass was measured with a calliper during tumor growth.Tumor volume (V) was calculated using the formula V=(length × width 2 ) ÷ 2.

Haematoxylin and Eosin (H&E) staining
Tissues were collected and fixed in 10% neutral buffered formalin solution at 4°C for 24 h.Fixed tissues were processed in an Excelsior AS tissue processor (Thermo Fisher) using the routine overnight programme.Processed tissues were embedded into paraffin blocks and sectioned at 4 µm-thick.FFPE tissue sections were deparaffinised in two changes of histoclear (National diagnostics) and rehydrated in a gradient of ethanol solution (100, 90, and 70%).Slides were sequentially incubated in Harris haematoxylin for 3 min, acid alcohol (1% v/v) for 2 s, Scott's water for 30 s, and eosin for 3 min, with washes in tap water between each step.Sections were then dehydrated in a gradient of ethanol (70, 90, 100%) and two changes of histoclear before being mounted with mounting medium (Vectamount, Vector Labs).

Tissue immunohistochemistry (IHC) and immunofluorescence (IF) staining
For IHC staining on paraffin sections, slides were deparaffinised and hydrated by serial incubation with histoclear, 100% ethanol, 90% ethanol, and 70% ethanol.Sections were incubated in 3% (v/v) H2O2 in methanol for 10 min at RT to inactivate endogenous peroxidase.Antigenic epitopes were retrieved by incubating the slides in boiling sodium citrate buffer (pH 6.0) for 10 min.Slides were left in the buffer to cool for 20 min at RT.Samples were blocked with 10% (v/v) normal goat serum (NGS)/PBS for 2 h at RT.Sections were then incubated with primary antibodies diluted in 5% (v/v) NGS/PBS overnight at 4 °C in a humidified chamber, followed by incubation with biotinylated secondary antibodies diluted in 5% (v/v) NGS/PBS for 1 h at RT. Slides were washed with PBS for 15 min after each incubation with antibodies.Sections were then incubated in Avidin-Biotin Complexes (ABC Reagent, Vector Labs) for 20 min at RT, washed in PBS, and applied with HRP substrate solution (DAB substrate kit, Vector Labs) till a dark brown colour was visualized.Slides were counterstained with haematoxylin, dehydrated by immersion in increasing gradients of ethanol solution, and mounted with mounting medium (Vectamount, Vector Labs).
For immunofluorescence staining on paraffin sections, slides were deparaffinised and hydrated by serial incubation with histoclear, 100% ethanol, 90% ethanol, and 70% ethanol.Sections were rinsed in water.Antigenic epitopes were retrieved by incubating the slides in boiling sodium citrate buffer (pH 6.0) for 10 min.Slides were left in the buffer to cool for 20 min at RT. Slides were washed in PBS and then permeabilized with 0.2% triton for 10 min.Samples were blocked with 10% (v/v) normal goat serum (NGS)/PBS for 1h at RT.Sections were then incubated with primary antibodies diluted in 10% (v/v) NGS/PBS overnight at 4°C in a humidified chamber.Slides were washed with PBS for 30 min followed by incubation with Alexa-Fluor secondary antibodies (1/500) diluted in 10% (v/v) NGS/PBS for 1h at RT. Slides were washed with PBS for 30 min after each incubation with antibodies then incubated with Dapi 1/500 for 10 min.Slides then washed with PBS for 5 min before mounting.

In vitro stimulation of BMDMs
BMDMs were harvested, counted, and re-plated in 24-well plates with 5×10 5 cells per well in cDMEM without LCM.Cells were left quiescent overnight and stimulated the next day.BMDMs were polarised to M1 macrophages by stimulation with 5 ng/mL LPS plus 1 ng/mL IFNγ (PeproTech) for 24 hours, and to M2 macrophages by stimulation with 10 ng/mL IL-4 (PeproTech) for 24 hours.
For cells stimulated with B16-or LLC-derived tumor-conditioned medium (TCM), 1 volume of TCM was mixed with 1 volume of cDMEM before application on to cells.
Mycobacterium bovis BCG was cultured until an OD600 (optical density value at 600 nm) within the range of 0.6-1.0. 10 mL of bacterial culture was achieved.The desired volume of bacterial culture was washed with and resuspended in antibiotic-free cDMEM of an appropriate volume.
BMDMs were seeded in 24-well plates the day before infection.Before infection, BMDMs were washed twice with antibiotic-free cDMEM.Next, 500 µL of bacteria-containing medium was added to each well of BMDMs.After 4 h of incubation, BMDMs were washed with PBS to remove extracellular bacteria, and 1 mL of antibiotic-containing cDMEM was added to each well for prolonged incubation until samples were harvested at the indicated time points.

BMDM migration assay
Each well of a cell invasion and migration (CIM) plate is composed of an upper chamber and a lower chamber, separated by a microporous membrane.Macrophages added to the upper chamber were allowed to migrate through the membrane to the chemokine-containing lower chamber.As cells passed through the membrane they adhered to the underside of the membrane, which was embedded with electrodes.This generated a signal of electrical impedance and was reflected as an arbitrary unit (cell index) as shown in the migration trace.It has been confirmed that rising cell impedance correlates with an increasing number of migrated cells (3).

BMDM phagocytosis assay
First, 1×10 5 BMDMs were plated in a 96-well optical plate with black walls (Thermo Fisher) one day before the experiment.The next day, one vial of pHrodoTM Green Zymosan Bioparticles (Thermo Fisher) was thawed in 4 mL of Live Cell Imaging Solution (Thermo Fisher), and sonicated for 5 min on ice.The culture medium of the BMDMs was removed and replaced with 100 µL of bioparticle suspension.BMDMs in the control wells were incubated with bioparticle-free Live Cell Imaging Solution.The plate was transferred to an incubator at 37 °C.At each time point, cells were washed with PBS twice to remove extracellular bioparticles before fluorescence intensity was read using a Clariostar Microplate Reader (BMG LABTECH).Phagocytic activity was indicated by relative fluorescence activity, which was calculated by subtracting the fluorescence intensity of the particlefree control well from value of the corresponding experimental well.

Flow cytometry
Murine tissues were cut, crushed and filtered with a 70 µm cell strainer (Falcon).The resultant single cell suspensions were washed and resuspended in FACS buffer.Immunostaining was performed in a 96-well V-bottom plate (Corning).Splenocytes were used for single-colour staining for fluorescence compensation during data analysis.Cells were washed with FACS buffer twice and blocked with anti-mouse CD16/32 antibody (BioLegend) for 10 min at RT.When staining cell surface antigens, samples were incubated with a master mix of fluorophore-conjugated antibodies for 30 min at 4 °C.After one wash with FACS buffer, samples were fixed by incubation with fixation buffer (BioLegend) for 10 min at RT followed by FACS wash.When intracellular staining (ICS) was also required, cells that were already stained with required cell surface antigens were fixed as described above, washed twice with 1× Permeabilisation Wash Buffer (BioLegend), and incubated with fluorophore-conjugated antibodies diluted in 1× Permeablisation Wash Buffer for 20 min in the dark at RT. Stained cells were washed and resuspended in FACS buffer for analysis.To detect nuclear FOXP3, samples that were already stained with required surface antigens were incubated with Fixation Permeabilisation Buffer (Thermo Fisher) for 30 min at RT.Samples were then washed with Permeabilisation Buffer (Thermo Fisher) twice followed by incubation with anti-FOXP3 antibody diluted in Permeabilisation Buffer for 30 min in the dark at RT. Next, cells were washed and resuspended in FACS buffer.A live cell fixable dye (L10119, Thermo Fisher) was used in all staining reactions to exclude dead cells.All of the centrifugation steps were performed at 350 g at 4 °C.Samples were analysed using an LSRFortessa X-20 cell analyser (BD Biosciences).FACS data were analysed using FlowJo v10.

Protein analysis
Adherent cells were lysed in tissue culture plates whereas mouse tissues were snap freezed, meshed, and lysed in Eppendorf tubes.Samples were lysed with urea buffer and supernatants were collected after centrifuging at 15,000 g for 15 min at 4°C .Protein concentration was determined using the Bio-Rad Bradford protein assay (Bio-Rad).Cell lysate of known volume was mixed with an appropriate volume of 6× Laemmli buffer.Samples were boiled for 10 min at 95 °C and were then ready for gel electrophoresis.SDS-PAGE gels were prepared with a Mini-PROTEAN® Tetra Cell system (Bio-Rad or pre-cast 26well Midi Protein Gels (Thermo Fisher).Prepared samples of equal amounts of protein and a prestained protein marker (New England Biolabs) were loaded into SDS-PAGE gels.Protein samples separated by SDS-PAGE were transferred onto a nitrocellulose membrane (Whatman) using a wet transfer system (Hoefer) loaded with 1× transfer buffer.The transfer was performed at a constant voltage of 80 V at 4 °C for 3 hours.The membrane was then blocked with 5% (w/v) milk (Marvel) in 1× TBST solution for 1 hour at room temperature before being incubated with primary antibody diluted in 5% (w/v) milk or 5% (w/v) BSA overnight at 4°C.The membrane was washed with 1× TBST for 20 min and incubated with HRP-conjugated secondary antibody at room temperature for 1 hour.Next, the membrane was washed with 1× TBST for 20 min followed by the application of enhanced chemiluminescence (ECL) Western blotting detection reagent (GE Healthcare).The result was visualised with X-ray film (Fujifilm) using a film developer in the dark.If another antibody was required for immunoblotting, the membrane was incubated with stripping buffer for 20 min at 55°C and blocked again before application of the new primary antibody.

RNAi
siRNA oligos against mouse FIH, HIF1α, and HIF2α were purchased from Dharmacon.Sequences are available from Dharmacon or on request.We used siGENOME RISC-Free siRNA (Dharmacon) as a negative control.J774 cells were transfected with the indicated siRNA oligos at a final concentration of 20 nM using Dharmafect 1 reagent (Dharmacon) according to the manufacturer's instructions.

RNA extraction, reverse transcription, and real time-quantitative PCR (RT-qPCR)
RNA was extracted from cells using an RNeasy Mini Kit (QIAGEN) following the manufacturer's instructions.First, 200 ng of total RNA was used to prepare cDNA with the SuperScript III First-Strand Synthesis System (Thermo Fisher), according to the manufacturer's protocol.Next, 1 unit volume of cDNA sample obtained from reverse transcription was diluted by 4 unit volumes of RNase-free water.Each qPCR reaction was performed in duplicate using 2 µL of cDNA template, 2 µL of RNase-free water, 1 µL of primer mix at a final concentration of 100 nM, and 5 µL of SYBR green (QIAGEN).Quantitative PCR was performed using the gene-specific primers listed in Supplementary Table 8.StepOnePlus Real-Time PCR System was used to conduct the following thermal cycles: 95°C (10 min); 40 cycles of 94°C (15 s), -55°C (30 s), and -72°C (1 min); followed by the melting curve run.Data were collected at the elongation phase of each cycle.All the primers used were tested for amplification efficiency.The expression level of each target gene was analysed based on the ΔΔCt method with Actg as a reference gene.For genes that were not expressed in the control sample, the expression level was expressed as relative expression normalised to Actg.expression analysis was performed using the exactTest function and a gene was considered to be differentially expressed between FIH +/+ and FIH Δ1-2/Δ1-2 if absolute log2 fold change was > 1 and FDR was < 0.05.The resulting differentially expressed genes (DEGs) were visualized as heatmaps using R package pheatmap (6) (version 1.0.12).The R script can be accessed via https://zenodo.org/record/7339349#.Y3oot-xBxBw.

Statistical analysis
Statistical analysis was formed in GraphPad.Differences were considered significant at p<0.05.

Fig. S1 .
Fig. S1.Reduced FIH expression led to increased lymphomagenesis (A) Schematic representation of the targeted FIH allele.(B) Western blots showing FIH expression in indicated tissue lysates prepared from FIH +/+ and FIH Δ1-2/Δ1-2 mice with β-tubulin being the loading control.(C) FIH expression in indicated tissues from FIH +/+ mice compared with a mouse fibroblast cell line NIH3T3 by Western blotting.Both GAPDH and β-tubulin were used as loading controls showing the variation in their expression levels.(D) Representative haematoxylin and eosin (H&E) and immunohistochemistry (IHC) staining showing a case of B cell lymphoma in the spleen found in an FIH Δ1-2/Δ1-2 mice.Scale bar, 25 µm.(E) Representative H&E staining with B cell lymphomas in non-lymphoid tissues: lung, liver, and kidney.Scale bar, 100 µm.(F) Representative H&E and IHC staining showing examples of splenic B cell lymphoma of low-grade, low-grade with high-grade components and high-grade lymphoma.Black arrows point to cells with low-grade features and white arrows indicate cells with high-grade features.Scale bar, 25 µm.(G) Frequencies of Representative FACS plots are shown below.(B and C) Proliferation of follicular B cells (CD21 Int CD23 + ) (B) and marginal zone B cells (CD21 + CD23 Int ) (C) by CFSE dilution assay.Horizontal lines and error bars represent mean ± SD. * indicates p<0.05 by two-tailed t-tests.Exemplar replicating profiles are shown on the right, with peaks corresponding to each round of cell division.XCGB50.1acorresponds to samples from a single WT mouse.N indicates the generation, i.e., N0 indicates a population which has not yet undergone mitosis.(D) Representative FACS plots from flow cytometry analysis of B cell subsets in the spleens of 18-week-old (young) and 104-week-old (aged) FIH +/+ mice (N=4) and FIH Δ1-2/Δ1-2 mice (N=3) showing gating of follicular B cells, marginal zone B cells, and transitional B cells on CD19+ CD220+ cells.(E) Representative FACS plots from flow cytometry profiling of immune cell subtypes in the indicated tissues of 104week-old (aged) FIH +/+ (N=4) and FIH Δ1-2/Δ1-2 (N=3) mice showing gating of CD11b+ CD11c-cells, neutrophils, inflammatory monocytes, and T cells.

Fig. S4 .
Fig. S4.Myeloid FIH expression does not affect spontaneous tumorigenesis but suppresses B16 tumor growth in young mice.(A) Western blots showing FIH expression in bone marrow-derived macrophages (BMDMs) obtained from WT and FIH MKO animals.(B) Western blots showing FIH expression in bone marrow neutrophils in WT and FIH MKO animals.In all Western blots, β-tubulin is used as a loading control.(C and D) A cohort of WT (N=20) and FIH MKO (N=29) mice were monitored for 125 weeks.Tumor observed at various sites during the course of the study are presented as count (C) and percentage (D).(E and F) Flow cytometry profiling of immune cells (B cells, CD4 + T cells, CD8 + T cells, CD11b + myeloid cells, Treg (CD25 + CD4 + ), NKT cells (CD3 + CD49 + ), γδ T cells (γδ TCR + CD3 + ), NK cells (CD3 -CD49b + ), DC (dendritic cells, CD11b -CD11c + ), TAMs, M-MDSCs, and G-MDSCs ) in the spleen (E) and liver (F) of WT and FIH MKO mice.(G) Western blots showing FIH expression in TAMs isolated from LLC tumors grown in WT and FIH MKO mice.(H) WT (N=6) and FIH MKO (N=4) mice were subjected to subcutaneous injection of 1×10 5 B16 cells on both flanks.Average volumes of B16 tumors from day 1-13.Tumor volumes, number of tumors measured and number of mice with measurable tumors at each time point are shown in Supplementary Table5.

Fig. S5 .
Fig. S5.Mice with HIF2α deletion in myeloid cells show increased γδT cells in the spleen (A and B) Flow cytometry profiling of B cells, CD4 + T cells, CD8 + T cells, CD11b + myeloid cells, Treg, NKT cells, γδ T cells, NK cells, DCs, macrophages, M-MDSCs, and G-MDSCs in (A) the spleen and (B) liver of WT (N = 15) and HIF1α MKO (N = 6) mice.(C and D) Flow cytometry profiling of immune cells in (C) the spleen and (D) liver of WT (N = 14) and HIF2α MKO (N= 3) mice.Small horizontal lines indicate the mean ± SD. **** indicates p<0.0001 by two-tailed t-test.(E and F) Murine macrophage cell line J774 cells were knocked down with indicated oligos for 48 hours and stimulated accordingly for 48 hours (E) ARG1 and β-tubulin expression was detected by Western blotting.(F) Expression of Arg1 and other known targets of HIFs were examined by qRT-PCR.(NT: no treatment).
Generation of double stranded cDNA and library construction were performed using TruSeq® Stranded mRNA HT (RS-122-2103) with minor modifications to manufacturer specifications.The following custom primers (25 µM each) were used for the PCR enrichment step: multiplex PCR primer 1. Indices were according to the eight bases tags developed by the Wellcome Centre for Human Genetics, University of Oxford (4).Amplified libraries were analysed for size distribution using the Agilent Tapestation 2200 D1000.Libraries were quantified using Picogreen and relative volumes were pooled accordingly.Sequencing was performed paired end read on a HiSeq4000 according to Illumina specifications with a read length of 125 bp.RNA seq reads were aligned to the mouse reference genome (GRCm38.85)using STAR and quantified by FeatureCount.The edgeR package (5) (version 3.26.1)from Bioconductor (version 3.1) was used to analyse the RNA sequencing data.Genes with at least one count per million in at least 4 samples were kept for downstream analysis.Gene expression in terms of raw counts was normalized using the trimmed means of M values (TMM) method, implemented by the calNormFactors function.Differential

Table 2 . LLC tumour volumes in FIH +/+ , FIH +/Δ1-2 , and FIH Δ1-2/Δ1-2 mice Genotype/tumour volume (mm 3 )
The Mantel-Cox test was conducted to determine the statistical significance of difference in spontaneous tumorigenesis between indicated mouse strains.The  2 test was used to compare the tumor incidence of mice of different genotypes.Grubbs' test was employed to identify outliers in the datasets.A two-tailed, unpaired t-test was used to compare sizes of tumors at each time point, and data analysis of flow cytometry and RT-qPCR experiments, before which sample data were checked for normal distribution.The statistical tests used in each experiment are described within the relevant section of text.