Perforin-2 is a pore-forming effector of endocytic escape in cross-presenting dendritic cells

During initiation of antiviral and antitumour T cell-mediated immune responses, dendritic cells (DCs) cross-present exogenous antigens on MHC class I. Cross-presentation relies on the unique “leakiness” of endocytic compartments in DCs, whereby internalised proteins escape into the cytosol for proteasome-mediated generation of MHC-I-binding peptides. Given that type 1 conventional DCs excel at cross-presentation, we searched for cell-type specific effectors of endocytic escape. We devised an assay suitable for genetic screening and identified a pore-forming protein, perforin-2 (Mpeg1), as a dedicated effector exclusive to cross-presenting cells. Perforin-2 was recruited to antigen-containing compartments, where it underwent maturation, releasing its pore-forming domain. Mpeg1-/- mice failed to efficiently prime CD8+ T cells to cell-associated antigens, revealing an important role for perforin-2 in cytosolic entry of antigens during cross-presentation.


Reagents
All reagents used in this study are listed in table S5. All antibodies used in this study are listed in table S3.
For BMDC cultures, BM cells from 8-13-week-old male or female C57BL/6J mice were resuspended in media supplemented with 5 ng/mL GM-CSF and 20 ng/mL Flt3-L. 15 x 10 6 cells were seeded in a non-tissue culture treated 10 cm plate. Cells were differentiated for 9/10 days with the addition of 10 mL of fresh media with 5 ng/mL GM-CSF and 20 ng/mL Flt3-L on the fifth or sixth day and 5 mL of media containing 20 ng/mL GM-CSF added on the seventh day.

BFP only construct
For the generation of the BFP-only construct, mTagBFP2 (table S4) was cloned in replacing mScarlet in the mScarlet only construct using MluI and NotI sites. humanMpeg1 IRES -BFP construct This plasmid was derived from the Mpeg1 IRES -mScarlet by replacing murine Mpeg1 for a human Mpeg1 IDT gene block (table S4) using KpnI and NotI sites. mScarlet was then replaced with IRES mTagBFP2 using Mlul and NotI sites (table S4).
All oligos and Cas9 nuclease were initially resuspended in RNAse free 10 mM TrisHCL, 0.1 mM RNAse-free EDTA at 500 ng/mL and 200 ng/mL, respectively. Oligos and Cas9 were mixed in RNAse free 10 mM TrisHCL, 0.1 mM EDTA and incubated at room temperature for 15 min (final concentrations 20 ng/mL for both oligos and Cas9). The Cas9-gRNA complexes were injected into a C57/Ola zygote, which was then allowed to recover for 3 h prior to its transfer into a CD1 surrogate. The immunisation experiments were performed after two backcrosses, the remaining experiments after six backcrosses or where indicated, using littermate controls.
We assessed editing efficiency by Sanger sequencing of a 560 bp amplified DNA sequence from the target region (primers can be found in Supplementary Table 1). In doing so we identified three recurrent mutations in the different founder mice, which consisted of a 40-base deletion affecting CDS bases 526 to 566, a 17-base deletion spanning CDS bases 525 to 541, and lastly a 16-base deletion of CDS bases 524 to 540. Mice with the 17-base deletion were used for experiments. Lentiviral Production 3x10 6 HEK293T cells were seeded per 10 cm dish. After 24 hours, cells were transfected with the sgRNA plasmid and lentiviral packing plasmids (psPAX2 and VSVG-Pmd2) using TransIT-LT1 transfection reagent. Media was changed 18 hours post-transfection to DMEM, 10% FCS, 1% BSA. Virus-containing supernatant was first collected 48 h post-transfection and media was replenished. This first virus batch was kept overnight at 4 °C. A second virus batch was collected the following day (72 h post-transfection), and both aliquots were mixed. Cellular debris was removed by centrifugation at 380 g for 5 min, before concentrating the virus by centrifugation at 3000 g for 45 min in Amicon Ultra-15 centrifugal filters.

Flow cytometry
All flow cytometry staining was done on ice. For cell surface staining, nonspecific binding was blocked using αCD16/CD32. Cells were then stained with the corresponding antibodies for 30-60 min (table S3). Following antibody staining, cells were washed three times with PBS, 1mM EDTA, 1% FCS. For live/dead staining cells were either resuspended in a (0.1 µg/mL) DAPI PBS, 1mM EDTA, 1% FCS solution or stained for 10 min with a fixable viability dye (table S3). Intracellular staining was done using BD Cytofix/Cytoperm Fixation/Permeabilization Kit. Data was acquired on a LSRFortessa (BD) or a CytoFLEX (Beckman Coulter) and analysed on FlowJo 10 (BD). For cell sorting, cells were sorted into serum coated polypropylene tubes using a SY3200 Cell Sorter (Sony).

Western Blotting
Cell pellets were lysed in RIPA buffer with protease inhibitor for 20 min at 4°C while shaking at 800 rpm. Lysates were centrifuged at 20,000g for 10 min at 4°C, the supernatant was then transferred to a clean 1.5 mL Eppendorf. Lysates were mixed with LDS NuPAGE + Bolt reducing agent and incubated at 70°C for 10 min. Samples were loaded onto an SDS-Page gel and transferred onto a nitrocellulose membrane using an iBlot1 (Invitrogen). The membrane was blocked for 1 h with 5% milk solution in PBS, 0.01% Tween (PBST), and blotted with the primary antibody overnight at 4°C. Secondary staining was performed for 1h at RT. Chemiluminescence was detected with ECL Prime Western Blotting Detection Reagent.
Saporin assay 3 x 10 5 cells were seeded per well in a tissue culture treated 96-well U-bottom plate. Cells were pulsed for 2 h at 37°C with saporin (Merck, S9896). Cells were spun at 380g for 3 min and resuspended in DC media containing 0.01 mg/mL puromycin and incubated for 30 min at 37°C.
To monitor saporin-mediated activation, 3 x 10 5 cells were seeded per well in a tissue culture treated 96-well U-bottom plate. Cells were pulsed with saporin for 3 h at 37°C degree. Cells were washed once in PBS, stained and analysed by flow cytometry.
For the CRISPR/Cas9 screen, 3 x 10 6 cells were seeded per well in a tissue culture treated 6well plate. Cells were pulsed with 0.5 mg/mL saporin for 2 h at 37°C. Cells were then washed twice with media and incubated with puromycin at 0.01 mg/mL in DC media for 30 min at 37°C.

Bead preparation
Saporin or BSA were conjugated to Ova-beads through disulfide bonds. For efficient conjugation, free sulfhydryl groups were introduced by reacting 0.6 mL of 2.5 mg/mL saporin or BSA with 12 μL of 2 mg/mL Traut's reagent in PBS + 2 mM EDTA for 60 min at room temperature. Excess reagent was removed using a Zeba spin desalting column equilibrated with PBS + 2 mM EDTA.
Ova-beads were resuspended in PBS + 2 mM EDTA and reacted with 1 mM SPDP for 30 min at room temperature. The beads were then washed twice in PBS + 2 mM EDTA and incubated with 1 mg/mL cysteine-modified saporin or BSA overnight at room temperature. After incubation, beads were washed twice in PBS and immediately used for phagocytosis.

Bead-based saporin assay
The assay was performed in cell culture medium lacking β-mercaptoethanol. MutuDCs were collected, washed once in PBS and seeded in a 96-well U-bottom plate with 5 x 10 5 cells per well in 100 µL cell culture medium. Saporin/Ova-or BSA/Ova-beads were diluted in cell culture medium such that adding 50 µL of each dilution to cells gave a 10:1 ratio of beads:cells. At the end of each incubation, puromycin was added at 0.01 mg/mL for 30 min and cells were washed in ice-cold PBS. Non-internalised beads were labelled in PBS containing 1% (vol/vol) BSA with a rabbit αOvalbumin antibody for 30 min on ice followed by donkey αRabbit-AF555 for 30 min on ice. After labelling dead cells with a fixable viability stain for 10 min on ice, cells were fixed and permeabilised using the BD Fix/Perm and Perm/Wash buffers. Puromycin incorporation was determined by staining with an αPuromycin-AF647 antibody in Perm/Wash buffer for 45 min on ice. Cells were analysed by flow cytometry. β-lactamase assay 4 x 10 6 HeLa cells were pulsed with 800 uL of the CCF4 solution (prepared as in (19) for 45 min, and washed by adding 5 ml PBS and spinning at 450 g, 15°C for 5 min. The cells were there resuspended in warm HeLa media with Probenecid at a 1:100 dilution (ThermoFished, P36400). To control for the background conversion of CCF4 observed in the absence of β-lactamase, each sample was split into two, media-only or media with β-lactamase (final concentration 2 mg/mL; P0389, Sigma). At each time point, 100 uL of the cell suspension was transferred into 100 uL of ice-cold PBS to stop trafficking. Finally, the cells were centrifuged for 2 min at 800 g, 4°C, stained with eFluor 780 live/dead solution containing Probenecid for 10 min on ice, centrifuged again, and resuspended in FACS buffer with Probenecid for flow cytometry.

MutuDC survival assays
For saporin, gelonin and cycloheximide treatments, 5 x 10 4 cells were seeded per well in a treated 96-well U-bottom plate. Cells were incubated with the different treatments for 24 hours. Cells were washed twice in PBS and then stained with a fixable viability stain. Cell viability was assessed by flow cytometry.
For treatment with Poly (I:C) 2.5 x 10 4 cells were seeded per well in a treated 96-well U-bottom plate. Poly (I:C) was added and cells were incubated for 48 hours. Cells were washed in twice in PBS and then stained with a fixable viability stain. Cell viability was assessed by flow cytometry.
For treatment with Bleomycin A1, 2 x 10 5 cells were seeded per well in a treated 96-well Ubottom plate. Bleomycin A1 was added and cells were incubated for 48 hours in the IncucyteÒ. The increase in cell area covered by the cells was monitored for each well and normalised to untreated cells.

Bead preparation
To generate Ova-beads, amino-modified microspheres with a diameter of 3-μm were washed twice in PBS and preactivated with 8% (vol/vol) glutaraldehyde for 4 h at room temperature. Preactivated beads were washed once in PBS and then incubated overnight at 4 °C with ovalbumin at a concentration of 0.5 mg/mL in PBS. After incubation, beads were quenched in 0.4 M glycine in PBS, washed twice in PBS and used immediately for phagocytosis.

PhagoFACS assay
MutuDCs were collected, washed once in PBS and resuspended in ice-cold internalisation medium (CO2-independent medium containing 1X GlutaMAX) to a density of 20 x 10 6 cells/mL. Ova-beads were added at a 10:1 ratio of beads:cells and incubated for 25 min at 16 °C followed by a 5 min incubation at 37 °C to allow phagocytic binding and internalisation of beads. To remove non-internalised beads, cells were first washed twice with 10 mL ice-cold PBS at 100 g for 4 min at 4 °C and then resuspended in 1 mL PBS, applied to a 5 mL FCS cushion and centrifuged at 150 g for 4 min at 4 °C. The cell pellet was then resuspended to 20 x 10 6 cells/mL in cell culture medium (containing 5 µM BFA or 0.1 µM BafA1 for drug treated cells) and divided into different time points comprising 5 x 10 6 cells each. The chase was performed at 37 °C for different periods of time and stopped by adding ice-cold PBS. Non-internalised beads were labelled by staining with a goat αOvalbumin antibody for 30 min on ice followed by an αGoat-AF488 antibody for 30 min on ice. Cells from each time point were resuspended in 0.5 mL homogenization buffer (250 mM sucrose, 3 mM imidazole, 2 mM DTT, 2 mM PMSF and 1X protease inhibitor cocktail, ph 7.4) and passed 25 times through a 22-G needle. Intact cells and debris were pelleted by centrifugation at 150 g for 4 min and the phagosome-containing post-nuclear supernatants transferred to a V-bottom 96-well plate. The enriched phagosomes were washed with PBS containing 1% (vol/vol) BSA and stained with different primary antibodies overnight at 4 °C. The next day, the samples were incubated with appropriate secondary antibodies for 45 min on ice. Phagosomes were analysed by flow cytometry.

pHrodo assay
To generate pHrodo beads, amino-modified microspheres with a diameter of 3-μm were washed twice in PBS, resuspended in 100 mM sodium bicarbonate (pH 8.5) and reacted with 0.2 mM pHrodo iFL Red Ester dye for 1 h at room temperature. The beads were then washed once in PBS and any reactive ester moieties were quenched by incubating the beads in 1X TBS for 10 min. After an additional two washes in PBS, the beads were coated with 1 mg/mL ovalbumin by passive absorption for 1 h at room temperature. The beads were then washed twice in PBS and immediately used for phagocytosis as described for the phagoFACS assay. Non-internalised beads and dead cells were labelled by staining with a rabbit αOvalbumin antibody in PBS containing 1% (vol/vol) BSA for 30 min on ice followed by a staining with a donkey αRabbit-AF647 antibody and ViaKrome 808 in PBS for 30 min on ice. Stained cells were immediately analysed by flow cytometry using a chilled sample stage.

Microscopy
Immunofluorescence, galectin 3 recruitment 1 x 10 5 MutuDCs were plated on a μ-slide 8 well dish and allowed to adhere at 37°C overnight before treating them with 33 μM GPN for 10 min at 37°C. Cells were washed three times in PBS and fixed in 4% paraformaldehyde for 10 min at RT. Paraformaldehyde was washed away before permeating cells with 0.1% Triton-X100 for 10 min at RT. Cells were washed 3 times in PBS, and incubated for 30 min at RT with blocking buffer (1% BSA, 0.3M glycine, 0.1% Tween 20 in PBS). After three PBS washes, cells were stained with αGalectin3 for 40 min and then washed 3 times in PBS. Cells were then stained with donkey αMouse-Af647 and then washed three times in PBS. Images were acquired on a Zeiss 780 inverted confocal microscope. The images were processed and analysed in Fiji.
Immunofluorescence, co-localisation with perforin-2 1.5 x 10 5 MutuDCs were plated on poly-L-lysine coated μ-slide 8 well dish and allowed to adhere at 37°C overnight. To label acidic compartments, the cell culture medium was replaced with fresh medium containing 1 µM LysoTracker Red and cells were incubated for 30 min at 37 °C. Cells were then washed twice with PBS and fixed in 4% formaldehyde for 10 min at RT. After two washes with PBS, the samples were permeabilised with 0.15% Triton X-100 in PBS for 10 min at RT followed by three washes with PBS. To block unspecific binding and quench excess formaldehyde, the cells were incubated in blocking buffer (1% BSA, 0.3M glycine, 0.1% Tween 20 in PBS) for 30 min. Primary and secondary antibody incubations were performed in PBS containing 1% BSA for 1 hr at room temperature with three washes in PBS after each incubation. Nuclei were labelled by incubating the cells for 10 min with DAPI followed by two washes in PBS. Images were acquired on a VisiTech iSIM swept field confocal super resolution system coupled to a Nikon Ti2 inverted microscope stand equipped with a 100x/1.49 NA SR Apo TIRF objective lens. The images were processed and analysed in Fiji.

Tau entry assay
Tau entry assays were performed as previously described (36). Briefly, HEK 239T cells expressing NLS-eGFP-LgBiT (NGL) were transduced with lentivirus harbouring humanMpeg1 IRES BFP or BFP only under the control of an SFFV promoter. Approximately 16 h prior to assay, 2x10 4 cells were seeded into a white 96-well plate coated with poly-L-lysine. The media was replaced 16 h later with recombinant 0N4R-P301S-tau-HiBiT (tau-HiBiT) protein at desired concentration in Assay Medium composed of CO2 independent medium supplemented with 1% penicillin-streptomycin, 1 mM GlutaMAX and 1 mM sodium pyruvate. Cells were washed once with PBS, and incubated in substrate solution (Assay Medium, LCS buffer and Live Cell Substrate) at RT for 5 min before loading onto a pre-warmed clarioSTAR Microplate Reader at 37ºC. The plate was mixed by 200 RPM double orbital shaking for 10 seconds prior to signal acquisition by spiral average (NanoLuc setting, 470 nM). Post-signal acquisition, cell viability per well was acquired by incubation with PrestoBlue Viability Reagent according to manufacturer instructions. The plate was then loaded onto the ClarioSTAR Microplate Reader and fluorescence read by excitation wavelength of 560 nm and emission at 590 nm.

Processing of mouse tissues
Spleens were perfused with a solution of RPMI-1640, 0.1 mg/mL Liberase-TL and 0.1 mg/mL DNAse I, minced and digested for 25 min at 37 ˚C. HI-FCS was added (10% v/v) to stop the digestion, before filtering tissues through a 70 μM filter. Red blood cells were then lysed with red blood cell lysis buffer hybrid-max for 3 min at RT. Lungs were minced in a solution of RPMI-1640, 750 U/mL collagenase type I and 0.3 mg/mL DNAse I. Samples were digested for 30 min at 37 ˚C while shaking at 800 rpm. Digest was then filtered through a 70 μM filter before washing twice with PBS 2% HI-FCS. Red blood cells were then lysed in 140 mM NH4Cl, 17mM Tris, pH 7.2 for 5 min at RT. To obtain the bone marrow, femurs and tibias from mice were cut at both ends, and the bone marrow was flushed into BMDC media by brief centrifugation at 10,000g. Red blood cells were lysed with red blood cell lysis buffer hybrid-max for 1 min at RT. Tissue CD11c+ cells were enriched using a Pan-Dendritic Cell Isolation Kit. OT-I T cells were obtained from OT-I spleens and lymph nodes. Both tissues were mashed and filtered through a 70 μM filter. OT-I T cells were enriched using either EasySep Mouse Naïve CD8+ T Cell Isolation Kit or Naïve CD8a+ T cell isolation kit.
Cross-priming assays 3T3 UVC-irradiation and antigen coating 1.5 x 10 6 3T3 cells were plated in 10 cm plate 16 hours prior to irradiation. For irradiation, media was replaced with 5 mL PBS. Cells were UVC irradiated (240 mJ/cm 2 ) with a UVP Crosslinker (AnalytikJena). Media was then replenished, and cells were incubated for 16 h at 37°C.
For antigen coating, the supernatant, a PBS wash and the trypsinised UVC-irradiated 3T3s were collected and resuspended at 10 x 10 6 cells/mL in 10mg/mL ovalbumin and 0.25 mg/mL HMW Poly(I:C). Cells were incubated for 1 h at 37°C. Cells were washed three/four times in ice-cold PBS and resuspended for injections.

OT-I CTV labelling
Staining was performed with a CellTrace Violet Cell Proliferation Kit. Isolated cells were resuspended at 5 x 10 5 /mL in a 2.5 μM CellTrace Violet solution, and incubated at 37°C for 20 min in the dark. 10% v/v FSC was added, and cells were incubated a further 5 min at 37°C. Cells were centrifuged at 300 g for 5 min, resuspended in T cell media and incubated at 37°C for 10 min.
In vivo immunisation 8-12 week old male and female mice were i.v. injected 0.5 x 10 6 OT-I cells. The following day mice were injected with either 100 μg of ovalbumin + 50 μg Poly(I:C), or with 1x10 6 UVCirradiated 3T3s coated as described with 10 mg/mL ovalbumin and 0.5 mg/mL Poly(I:C). Three days later spleens were isolated and OT-I T cell abundance was assessed by flow cytometry.
In vitro cross-presentation assay XCR1 + cDC1s were magnetically enriched from day 9 Fl3L-GM/CSF cultures using an EasySep PE Positive Selection Kit and XCR1-PE antibody (2 µg/mL). Isolated cDC1s were then plated at 1 x 10 5 cells per well in a U-bottom 96 well plate. UVC irradiated 3T3s coated were added to cDC1s and incubated for 20 h at 37 ˚C. Cells were then washed twice in PBS and 5 x 10 4 CTVlabelled OT-Is were added and co-cultured for 3 days at 37 ˚C. OT-I proliferation was then assessed by flow cytometry.

Recombinant perforin-2 cleavage Expression and purification of recombinant perforin-2
The murine perforin-2 ectodomain (amino acids 20-652), tagged with an N-terminal signal peptide and a C-terminal hexahistidine tag was introduced into vector pHL-sec. Expi293F cells at 2x10 6 /mL were transfected with 3.8 mg of plasmid in the presence of 10.2 mg of Polyethyleneimine Max. Three to four hours later, valporic acid was added at a final concentration of 3.5 mM. Cells were allowed to grow for a further 5 days. Culture media was collected, centrifuged for 2 h at 4000g and filtered through a 0.22 μM membrane. An ÄKTA flux TM (Cytiva Life Sciences) was used for sample concentration (with a 10kDa cut-off filter) and buffer exchange to Buffer A (25mM Tris pH 7.5, 500Mm NaCL, 10 mM imidazole). The sample was then loaded onto a Ni-NTA column at RT. The column was washed, sequentially, with 10 mM and 40 mM imidazole, before collecting the perforin-2 elute with a 200 mM imidazole wash. The perforin-2 fractions were then dialyzed in PBS and concentrated to 2 mg/mL.

Asparagine Endopeptidase cleavage reactions
AEP (specific activity 350 pmol/min/μg) was resuspended in activation buffer (0.1 M NaOAc, 0.1 M NaCl, pH 4.5) at 50 μg/mL. Prior to the cleavage reactions, ΑEP was incubated for 4 h at 37°C. For the in vitro cleavage reactions, AEP was diluted in assay buffer (50 mM MES, 250 mM NaCl, adjusted pH 5.5) and 35 or 175 μU were added to a 50 μL final reaction volume. To ensure AEP was active, cleavage of a fluorogenic substrate was confirmed using manufacturer's protocols. For perforin-2 cleavage, 4 μL of purified perforin-2 was cleaved in the absence or presence of AEP inhibitor peptide at 0.5 mg/mL. Reactions were allowed to proceed for 2 h at 37°C before being terminated by addition of denaturing agent.

sgRNA library generation
The oligos to generate the sgRNA library were ordered from Twist Bioscience as part of a larger pool containing several libraries. The sequences are listed in table S1. The oligo pool was resuspended to 53 nM and minilib-PCR1 primers (table S1) were used to amplify the library.
The LentiBFP vector was digested with BsmBI, and a gibson reaction was set up with the library inserts. The Gibson reaction mix was electroporated into EnduraTM competent cells and electroporated at 1.8 kV/600Ο/10μF. Bacteria were allowed to recover in recovery media for 1 h at 37°C, before plating them in a Luria broth (LB) lennox NuncTm Square BioAssay dish and growing them overnight at 30°C. Bacteria were then collected by washing the plate 5 times with 5 mL LB. Bacteria were centrifuged at 4000 g for 15 min, before removing the supernatant and freezing the pellet. For DNA purification, the bacteria pellet was thawed, and DNA purified using a QIAGEN Plasmid Maxi Kit.

Lentiviral Transduction for the CRISPR-Cas9 screen
Cas9 expressing MutuDCs were seeded at 2x10 6 per 10 cm dish. Virus was added at a 0.3 MOI. Two days after transduction BFP positive cells were sorted and allowed to recover. Library representation was maintained at 800x.

Preparation of libraries for next generation sequencing
Genomic DNA was isolated using a DNeasy Blood and Tissue Kit. Samples were digested for 4 hours with Proteinase K. sgRNAs were then amplified in a two-step PCR with Herculase II Fusion DNA Polymerase. For the first PCR a maximum of 5 ug of DNA per 50 μL reaction was amplified using all available genomic DNA and libgen-PCR1 primers (table S1). The reaction mixed consisted of 10 μL 5X Herculase II reaction buffer, 0.5 μL dNTP mix, 0.5 μL Herculase II fusion DNA polymerase and 1.25 mL of 10 μM both the forward and reverse primers. PCR cycling conditions were: 1 min at 95 °C; 18x (30 s at 95°C, 30 s at 55 °C, and 30 s 72 °C); 10 min at 72°C).
The number of cycles for the second PCR was determined by qPCR using a KAPA Library Quantification Kit. Cycling conditions were: 2 min at 95 °C, followed by 30 s at 95 °C, 30 s at 53 °C and 30 s at 72 °C for a total of 40 cycles. Samples with similar CT values were then pooled for the second PCR.
Following sgRNA amplification from gDNA, a second PCR was performed to barcode the sgRNAs using 10 μL from the first PCR. PCR cycling conditions were: 2 minutes at 95 ˚C, followed by 30 s 95 °C, 30 s at 53 °C and 30 s at 72 °C for the number of cycles determined by the previous qPCR step, and a final 10 min at 72 °C. The amplified products were separated using a 2% agarose gel and purified using a Gel Extraction Kit. Agarose was melted at 40 °C instead of the recommended 50 °C. Samples were then further purified using a Charge Switch PCR Clean-Up Kit. Finally, the concentration of DNA fragments containing P5 and P7 adaptors was determined via qPCR using KAPA Library Quantification Kit. This ensured adequate cluster density during NSG. This was done using a KAPA Library Quantification Kit. Samples were analysed by SE50 sequencing on a HiSeq4000 sequencer loaded with 15% spike-in PhiX Control Library (Illumina).

Analysis of the CRISPR-Cas9 screening data
Processing of the sequencing data The analysis pipeline was adapted from (56). Demultiplexing was performed using the demuxFQ package. Next, the fastq files were processed using cutadapt-1.4.1 to remove the flanking sequences: GACGAAACACCG and GTTTTAGA on the 5' and 3' end respectively (analysis parameters: -e 0.2 --minimum-length 20 --discard-untrimmed). The trimmed 20 bp long reads were matched to the library of reference sgRNA sequences and counted (analysis parameters:f -v 0 -m 1 --norc -a --best --strata --un). Raw counts for each sgRNA are provided in table S1.

Hit calling
Each screen repeat was analysed initially using a modified stat.wilcox function from the caRpools (v 0.83) package with the following parameters and modifications. Guides with less than 20 counts in either of the populations selected for comparison were excluded from the analysis. The counts were then normalized to median of the population. The function returns enrichment score for each population of four sgRNAs targeting one gene relative to 100 random guides. The p-values are calculated using a two-sided Mann-Whitney-U test and non-adjusted P-values were used in the next step of the analysis.
To combine the data from the biological replicates of the screen, for each gene we calculated the mean enrichment score and used Fisher's method to combine the p-values. P-values were then adjusted using the Benjamini-Hochberg method. Genes with enrichment scores greater than 0.5 and adj p-values < 0.01 were considered as hits.

Analysis of organellar mapping data
Peptide data from organellar mapping experiments (19) in control, prazosin-and tamoxifentreated cells were included in the analysis. As in the original paper, SILAC ratios were inverted, weighted with fraction yields, and divided by the sum of all five ratios across the map. This yielded for each peptide a 'per map' normalized profile (summing to 1). Mean profiles from each treatment group were used to prepare the map using the prcomp function in R.

Whole cell lysate proteomic analysis
Cell culture and treatments 10x10 6 MutuDCs were plated in a 90mm tissue culture treated plate and either CpG-(1 µM), BafA1-(1 µM) or mock treated for 16 h at 37 °C. Cells were then harvested with PBS 5mM EDTA, and washed three times in PBS.

Cell lysis and in-solution digestion of proteins
Cell pellets were thawed, lysed in 200 µl 2.5% (w/v) SDS/50 mM Tris pH 8 and incubated at 72 °C for 5 min. Lysates were then sonicated at 4 °C (three times 5 s bursts with an amplitude of 10 µm) to break-up DNA. Estimations of protein concentrations were made using a Pierce BCA Protein Assay Kit. For each sample, 100 µg protein was precipitated by the addition of 5 volumes of icecold acetone, incubated at −20 °C overnight and pelleted by centrifugation at 4 °C for 5 min at 10,000×g. All subsequent steps were performed at room temperature. Precipitated protein pellets were air-dried for 5 min, resuspended in 50 µl digestion buffer (50 mM Tris pH 8.1, 8 M Urea, 1 mM DTT) and incubated for 20 min. Protein was alkylated by addition of 5 mM iodoacetamide for 20 min (in the dark) and then enzymatically digested by addition of LysC (1 mg per 50 mg of protein) for an overnight incubation. Digests were then diluted four-fold with 50 mM Tris pH 8.1 before addition of Trypsin (1 mg per 50 mg of protein) for 4 hours. The peptide mixtures were then acidified to 1% (v/v) TFA in preparation for peptide purification and fractionation.

Mass spectrometry
For proteomic analysis of BafA1 and CpG induced changes in the abundance of perforin-2 peptides, 500 ng of peptides were loaded on a 50 cm by 75 µm inner diameter column, packed in-house with 1.8 µm C18 particles (Dr Maisch GmbH, Germany). Peptide separation by reverse phase chromatography was performed using an EASY-nLC 1000 (Thermo Fisher Scientific), running a linear gradient over 95 min at 300 nl/min flow rate and 55 °C. The gradient ran from buffer A (0.1% (v/v) formic acid) containing 5% buffer B (80% (v/v) acetonitrile, 0.1% (v/v) formic acid) to buffer A containing 30% buffer B. Runs were separated by 5 min wash-outs with 95% buffer B and re-equilibration. The LC was coupled to a Q Exactive HF-X Hybrid Quadrupole-Orbitrap mass spectrometer via a nanoelectrospray source (Thermo Fisher Scientific). MS data were acquired using a data-dependent top-15 method that dynamically excludes precursors picked during the last 30 seconds. MS1 survey scans were acquired at a resolution of 60,000 in a 300-1650 Th range. The maximum injection time was 20 ms for up to 3e6 target ions, as determined with predictive automatic gain control. Sequencing was performed via higher energy collisional dissociation fragmentation of ions isolated from a 1.4 Th window. The maximum injection time was 28 ms for 1e5 target ions. MS2 fragment scans were acquired at a resolution of 15,000 in a 200-200,000 Th range.
For proteomic analysis of perforin-2 peptides in AEP knockout cells, 300 ng of peptides were loaded on a 50 cm by 75 µm inner diameter column, as above. Peptide separation by reverse phase chromatography was performed using an EASY-nLC 1200 (Thermo Fisher Scientific), running a linear gradient over 100 min at 300 nl/min flow rate and 50 °C. The gradient ran from buffer A (0.1% (v/v) formic acid) containing 5% buffer B (80% (v/v) acetonitrile, 0.1% (v/v) formic acid), to buffer A containing 30% buffer B in 84 min, followed by an increase to 60% buffer B in 8 min, a further increase to 95% buffer B in 4 min, and a constant phase at 95% buffer B for 4 min. Runs were separated by 5 min wash-outs with 95% buffer B and re-equilibration. The LC was coupled to an Orbitrap Exploris 480 mass spectrometer via a nanoelectrospray source (Thermo Fisher Scientific). MS data were acquired using a data-dependent top-15 method as described above. The maximum injection time for MS1 survey scans was 25 ms for up to 3e6 target ions. MS2 fragment scans were acquired at a resolution of 15,000 with a scan range starting from 100 Th.

Processing of mass spectrometry data
Mass spectrometry raw files were processed in MaxQuant Version 1.6.10.43 (58), using the mouse SwissProt canonical and isoform protein database, retrieved from UniProt (2019_10_22; www.uniprot.org). Label-free quantification was performed using the MaxLFQ algorithm (59). No matching between runs was used. To detect peptide fragments resulting from other cleavage events than the in-solution digest with LysC and trypsin, the enzyme mode was set to semi-specific. LFQ minimum ratio count was set to 1 and default parameters were used for all other settings.
To assess abundance of peptides independent of protein abundance changes, peptide intensities were divided by their corresponding protein intensities. Data were filtered for 3 valid values in at least one experimental condition and then subjected to a two-sided student's t-test. Multiple hypothesis correction was done by permutation based FDR with s0=0.1 using Perseus (60).
The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium [http://proteomecentral.proteomexchange.org] via the PRIDE partner repository with the dataset identified PXD041861.

Fig S1. Saporin-puromycin escape assay in MutuDCs.
(A) Puromycylation in MutuDCs is sensitive to changes in the rate of translation. MutuDCs were incubated for 2 h at 37°C with 10 μg/mL cycloheximide followed by a 30 min 0.01 mg/mL puromycin chase in the presence or absence of 10 μg/mL cycloheximide. Puromycin incorporation was monitored with a αPuromycin antibody. Histograms are representative for three independent experiments.
(C, D) GPN facilitates endosomal escape of saporin. To demonstrate that endocytic escape of saporin is the rate-limiting step in the saporin-puromycin assay, we disrupted the integrity of endocytic compartments with glycyl-L-phenylalanine 2-naphthylamide (GPN). GPN is a cathepsin C substrate that induces osmotic lysis of lysosomes leading to release of lysosomal contents (61). (C) Confocal microscopy images of MutuDCs treated with 33 μM GPN for 10 min. Cells were stained with DAPI (white) and for galectin-3 (magenta). (D) MutuDCs were incubated with 0.5 mg/mL saporin for the indicated time in the presence or absence of 33 μM GPN. Following a 30 min puromycin chase, translation inhibition was monitored by flow cytometry. Data represent three independent experiments. Fig S2. Saporin-puromycin assay can capture physiological differences in endocytic escape.
(A, C) CD11c+ enriched splenic DCs from C57BL/6J were incubated with saporin for 2 h, followed by a 30 min puromycin chase. Translation inhibition was monitored by flow cytometry. (A) Gating strategy for C57BL/6J splenic cDC1s and cDC2s for the ex vivo saporin assay. (C) Data represent mean and SEM for five independent experiments, ns, not significant; **, P<0.01 using a multiple unpaired t-test (two-stage step-up, Benjamini, Krieger and Yekutieli).

(B)
The CRISPR/Cas9 library consisted of 4 sgRNAs per gene in a BFP-expressing lentiviral vector. Cas9-expressing MutuDCs were transduced with at an MOI of 0.3. Cells were sorted for BFP expression and expanded.
(C) Schematic representation of the CRISPR/Cas9 screen. For the screen, the saporin-puromycin assay was performed with a 2 h pulse of 0.5 mg/mL saporin (11:1 ratio of unlabelled to Atto550labelled saporin). Cells were first gated on Atto550 to control for uptake efficiency and split into two bins: puro high (saporin escape) or puro low (saporin retention). Puro high and puro low cells were also collected in the absence of saporin to identify guides that might have a global effect on translation. Genomic DNA was then isolated and prepared for next generation sequencing of the sgRNAs.

(D)
Volcano plot showing the sgRNAs enrichment analysis for the MutuDC library relative to the starting plasmid library. Each of the dots represents one targeted gene. Guides against Irf8 were depleted from the cell MutuDC cell library, in line with the role of Irf8 in survival of terminally differentiated cDC1s (62).

(E)
Volcano plots showing the sgRNAs enrichment analysis for the control screen in the absence of saporin. Each of the dots represents one targeted gene. Data represent the combined mean enrichment scores and the non-adjusted p values from three independent experiments (Fisher's method).
(F) Relative abundance of sgRNAs in purohigh and purolow populations from the saporin-puromycinbased genetic screen. The screen was performed in three biological repeats. The sgRNA counts from each population in each screen replicate were normalised, and the average of the counts for purohigh and purolow populations is plotted. Each dot corresponds to one sgRNA. The sgRNAs targeting Mpeg1 are highlighted in red.  . Cells were stained with a fixable live/dead stain, and viability was assessed by flow cytometry. For bleomycin, the cells were plated in the presence or absence of bleomycin and cultured in an IncucyteÒ for 48 h to monitor the growth rate. Data represent mean and SEM of three independent experiments, ns, not significant; *P<0.5; **P<0.01; ***P<0.001; ****P<0.0001 using a multiple t-test (Bonferroni-Dunn).    Table S4. BFP+ cells were sorted but not clonally selected. Knock out efficiency was assessed by reducing Western blot in untreated and CpGand BafA1-treated cells. b-Actin was used as a loading control. (B) Schematic representation of the saporin-bead puromycin assay. Cells are pulsed with saporinconjugated Ova-beads and allowed to internalise them for the indicated time. Translation is then monitored with a 30 min puromycin chase. Incorporated puromycin can then be detected with an αPuromycin antibody and flow cytometry.
(C) Mpeg1 KO and NT MutuDCs were pulsed for 3 or 5 h with saporin-conjugated Ova-beads and translation was monitored by a 30 min puromycin chase. Histograms are representative of three independent experiments each with two technical replicates. The quantification of translation inhibition represents three independent experiments each with at least two technical replicates, ns, not significant; *P<0.5; **P<0.01; ***P<0.001; ****P<0.0001 using an unpaired t test.  (B) Perforin-2 levels in Mpeg1 +/+ , Mpeg1 +/and Mpeg1 -/splenocytes were assessed by Western blot under reducing conditions. β-tubulin was used as a loading control.
(C) Flow cytometry gating strategy for identification of lung resident and migratory cDCs.
(C) Flt3-L/GM-CSF bone marrow cultures from wild-type and Mpeg1 -/mice were pulsed with saporin for 2 h, and translation was monitored by a 30 min puromycin chase. Flt3-L/GM-CSF cDC1s and cDC2s are defined as in (B). Histograms are representative for three independent experiments. Tables   Table S1. Genetic screen in MutuDCs. Data pertaining to the genetic screen in Fig. 2 including immgen gene expression data (cDC1s and cDC2s), a list of the sgRNAs in the custom-made library, sgRNA raw counts, CRISPR/Cas9 screen sample ID and CRISPR/Cas9 screen results. Table S2. Proteomics data.

List of Supplementary
(A) Peptide data used for Fig 3A. Normalised, averaged peptide profiles, and results of the PCA analysis.

(B) Mass spectrometry of CpG-and BafA1-treated MutuDCs and of AEP KO MutuDCs
Proteomics data including protein name, normalised LFQ intensities, p-values (Student's t-test) as well as N-term and C-term cleavage windows.