Cell line
A modified version of the ID8 mouse ovarian surface epithelial cell line, ID8-gTRP53-gBRCA1 (ITB1), was obtained from Dr. Iain A. McNeish [17] under an MTA agreement. This cell line was maintained at 37°C in a humidified atmosphere with 5% CO2 in DMEM medium supplemented with 10% fetal bovine serum (FBS), 1% l-glutamine, 200 mM, and 100 units each of penicillin and streptomycin. Mycoplasma infection was monitored regularly, and antibiotics (De-Plasma, TOKU-E, D022) were added to the culture medium as needed.
gRNA pool library production
The murine ID8-gTRP53-gBRCA1-Cas9 cell line was generated by co-transfecting a lentiviral Cas9-Blast vector (Addgene #52962) with the packaging plasmids pCMV-dR8.91 and pCMV-VSV-G (Addgene #8454) into HEK293T cells. Following 48 h of transfection, the virus was harvested and its titer was determined. The Cas9-Blast lentivirus was then stored at -80°C until it was used for overnight infection of ID8-gTRP53-gBRCA1 cells. Selection with 5 μg/mL of blasticidin was performed two days later. To acquire clones with high Cas9 activity, single-cell sorting into 96-well plates was performed on the ID8-gTRP53-gBRCA1-Cas9 cells. Subsequently, infection with a lentivirus driving expression of a gRNA specific for CD274 and mCherry was carried out on multiple clones. Ten days post-infection, each clone was stimulated with 10 ng/mL of interferon (IFN)g for 24 h, and the expression of PD-L1 was determined by FACS using an anti-CD274 antibody. The efficiency of Cas9 editing was determined by measuring the percentage of PD-L1 negative cells in the transduced population.
The mouse CRISPR Brie lentiviral pooled libraries consisting of 79,637 gRNAs were co-transfected with the packaging plasmids psPAX2 #12260 and pCMVVSV-G #8454 into HEK293T cells. Six hours after transfection, the medium was replaced with a medium supporting virus production (DMEM supplemented with 20% of FBS). After 48 h, the lentiviral medium was harvested and stored at -80℃. Before use in the in-vivo experiments, the ID8-gTRP53-gBRCA1-Cas9 cells were infected with the Brie lentivirus at a multiplicity of infection (MOI) of 0.06 and selected with puromycin for at least 10 days before injection, as described below in the section ‘In-vivo experiments.’
Generating knockout cell lines
To create knockout lines, the genome-scale CRISPR-Cas9 knockout (GeCKO) protocol was followed [19]. Oligos were designed using the same sequences as those provided in the gRNA library pool (described in ' gRNA pool library production') and were phosphorylated and annealed with T4 PNK ligase. The lentiviral CRISPR plasmid was digested and dephosphorylated using the BsmBI restriction enzyme, and the annealed oligos were ligated into the lenti CRISPRv2 vector. The plasmid was transformed into stbl2 bacteria, and positive colonies were selected using ampicillin selection (100 µg/mL). The positive colonies were expanded, and plasmid DNA was isolated using the QIAGEN® Plasmid Plus Midi Kit. The plasmid DNA was used to create lentiviruses and to infect ITB1 cell line to generate new lines expressing the desired genetic modifications. To this end, two TRAF3 guides, 5-CAGGTTCACGTGCTGTACCG-3 and 5‑CGGTACAGCACGTGAACCTG-3, were used to created two stable ITB1 TRAF3 knockout (KO) clones, designated TRAF3KO1 and TRAF3KO2, respectively.
Western blot analysis
All cells were harvested using a cell scraper and washed with ice-cold PBS. A lysis buffer supplemented with phosphatase inhibitor cocktails (BioTools, B15001A/B) and a protease inhibitor (Millipore Sigma, P2714-1BTL) was used to lyse the cells, which were then placed on ice for 30 min, followed by 3 min of ultrasonic cell disruption. The lysed cells were centrifuged at 14,000 rpm for 10 min at 4°C, and the supernatant was collected. The protein concentration was determined using the Bradford assay (Bio-Rad, Protein Assay, cat# 500-0006), and 1 μg/μL of total cellular protein from each sample was used to run a PAGE and transferred onto a PVDF membrane (Bio-Rad, 1704157). The membranes were soaked in a blocking solution of 5% BSA (AMRESCO, 0332-TAM) in TBS-T [Tris-buffered saline (TBS)-Tween 20 (0.1%)] for 1 h and then hybridized overnight with primary antibodies diluted in 5% BSA and 0.1% Tween. The following day, the membranes were incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies (1:20,000, Jackson), diluted in the blocking solution. A chemiluminescent reaction was developed using enhanced chemiluminescence (ECL) (Westar Nova 2.0 Cyanagen XLS071.0250), and images were captured using an Azure Biosystems camera system. For nuclear and cytoplasmic separation, the Rockland Nuclear & Cytoplasmic Extract Protocol [18] was followed. In brief, the cells were lysed with a low-salt lysis buffer, which allowed the nuclear proteins to remain intact while the cytoplasmic proteins were lysed. Then, using a high-salt lysis buffer, the nuclear membrane and the nuclear proteins were lysed. The separation quality was verified by using an antibody against histone H3, which should be present only in the nuclear fraction.
RNA isolation and real-time quantitative PCR
The isolation of total RNA and its conversion to cDNA were performed using the ISOLATE II RNA Mini Kit (Bioline, BIO-52073) and the qScript cDNA synthesis kit (Quanta Bioscience, 95047-100), respectively, as per the manufacturers' protocols. A mix of cDNA and Fast SYBR qPCR Master mix (BioGate, EZ60) along with custom primer sets from IDT were used for qPCR analysis on a Roche light cycler 480 II machine.
Flow cytometry
Isolated cells were blocked with anti-CD16/32 (anti-FcγIII/II receptor, clone 2.4G2) for 10 min at 4°C. Zombie Aqua™ fixable viability dye was then used to identify dead cells by incubating the cells with the dye for 15 min at room temperature. Surface markers were then stained by incubating the cells with the desired antibodies for 20–30 min at 4°C. The stained cells were analyzed using a CytoFLEX Flow Cytometer, and the results were analyzed using the CytExpert software.
For intracellular cytokine staining, cells were first cultured for 5–6 h with PMA (25 ng/mL), ionomycin (1 μM) and brefeldin A (5 µg/mL) to increase the intracellular cytokine staining signals. Surface markers were then stained, and the cells were fixed using 4% paraformaldehyde for 20 min at room temperature. The cells were permeabilized using a permeabilization buffer (1´) (eBioscience, 00-8333-56), and intracellular staining was performed by incubating the cells for 20 min. The cells were then analyzed on a CytoFLEX flow cytometer.
In-vivo experiments
The in-vivo experiments were conducted using 6- to 8-week-old NSG mice (NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ, Jackson labs) and wild-type (WT) C57/BL/6 mice (Envigo, Huntingdon, UK). The mice were housed in air-filtered laminar flow cabinets and provided with food and water ad libitum. Animal experiments were conducted in compliance with protocols established by the Institutional Animal Care and Use Committee (IACUC) of Ben-Gurion University of the Negev for ensuring animal welfare and minimizing discomfort. The animal ethical clearance protocol number used for the study was IL-23-05-2020(E).
Two different experiments were performed with the WT and NSG mice, as described in the Results section. In an experiment designed to investigate the role of anti-tumor immunity in the progression of ITB1-generated disease, WT and NSG mice were each injected intraperitoneally (i.p.) with 4-5 million ITB1 cells. In a separate experiment, which was designed to identify potential immunomodulators in ITB1 cells, WT and NSG mice were injected with 4-5 million ITB1 cells that expressed a gRNA library
Two experiments were performed only with WT mice. In the first, designed to examine the involvement of the microbiome in OC, one group of WT mice was pretreated with an antibiotic cocktail (ABX) for 2 weeks to deplete the total microbiome before TRAF3KO cell injection. This was achieved by adding ampicillin (1 g/mL), vancomycin (0.5 g/mL), neomycin sulfate (1 g/mL), and metronidazole (1 g/mL) to the drinking water and changing the water every two days [20]. The supplementation of the drinking water ABX was continued throughout the entire experiment. Two other groups of WT mice were injected with ITB1 or TRAF3KO cells and received normal drinking water (vehicle) throughout the entire experiment. For the FACS experiments, WT mice were sacrificed 2 weeks after the injection, and the peritoneal cavity was washed out with phosphate-buffered saline (PBS) to collect the contents.
For all the survival experiments, the body weight of the mice was monitored once a week, and the mice were sacrificed when a 20% increase in body weight was observed, indicating the development of ascites. The ascites fluid was collected for the isolation of bacteria or tumor cells. The digestive and reproductive systems were also collected for histology analysis.
Bacterial DNA extraction
Bacterial DNA was isolated from ascites fluid or PBS wash using a modified phenol-chloroform protocol [21]. In brief, samples were lysed with a lysis buffer containing phenol, chloroform, and isoamyl alcohol (IAA) and then homogenized by bead-beating with bacterial disruption beads (RPI, 9833) for two cycles of 1 min each. The aqueous and phenol phases were separated by centrifugation, and the aqueous phase was re-suspended in IAA in a sterile tube. The bacterial DNA was then concentrated and isolated using isopropanol precipitation and ethanol washes. To minimize contamination, sterile conditions were used, and negative controls were taken by sampling the environment and collecting tools.
Sorting of IgA+ and IgA− bacteria
The above-described PBS washes were collected from co-housed mice and then centrifuged at 8,000 ´ g for 5 min at 4°C. The pellet was then resuspended in 100 μL of a blocking buffer containing 20% normal rat serum and incubated for 20 min on ice. Thereafter, the samples were stained with 100 μL of staining buffer containing phycoerythrin (PE)-conjugated anti-mouse IgA (1:12.5; eBioscience clone mA-6E1) for 30 min on ice. Following this step, the samples were washed 3 times with 1 mL of staining buffer. The anti-IgA-stained bacteria were then incubated in 1 mL of staining buffer containing 50 μL of anti-PE magnetic activated cell sorting (MACSÒ) beads (Miltenyi Biotec) for 15 min at 4°C. Finally, the samples were washed twice with 1 mL of staining buffer at 8,000 ´ g for 5 min at 4°C, and sorted using MACS. The IgA‑positive and the IgA-negative samples were then stored at -80°C for future use.
Sequencing and sequence analysis
The genome-scale CRISPR-Cas9 library was sequenced on an Illumina MiSeq platform using a 250-bp paired-end chip. The sequences were mapped using a modified version of the caRpools R-package [22]. Differentially expressed guides were identified by loading the normalized gRNA count table into Deseq2 and determining the top differential genes based on the mean log2 fold change and false discovery rate (FDR). To identify significant pathways that were enriched or depleted in the screen, gene set enrichment analysis, including genes with a threshold of log2 fold change <(-1), was conducted using the ReactomePA [23] R package.
To amplify the V3-V4 region of the bacterial DNA isolated from the ascites of mice, a 2-step PCR-based protocol published by Holm et al. [24] was followed. The amplicons were then sequenced using the Illumina MiSeq platform (MiSeq Reagent Kit v2, 250-bp paired-end). Quality-checking of the demultiplexed sequencing reads was carried out using FastQC (Version 0.11.8) [25] and MultiQC (Version v1.7) [26]. Adapters, primers, and low-quality reads were removed using Fastp (Version 0.2, defaults parameters) [27]. The sequences were denoised by using Qiime2 version 2021.11 [28], and an amplicon sequence variant (ASV) table was created with taxonomic assignment using the SILVA database version 138, 99% NR [29]. Contaminant sequences were identified and removed based on control samples and sequence prevalence using the decontam package (Version 1.12) [30] and squeegee tool (Version 0.1.3) [31]. Further analysis of the ASV table was done using the MicroEco package (Version 0.11) [32] and phyloseq package (Version 1.38) [33] in R (Version 4.1.1).
A similar proof-of-concept analysis was performed for ascites from women with OC (see ‘Results section’).
Statistical analysis
The in-vitro experiments were repeated a minimum of two to three times, with representative data/images being presented in the ‘Results’ section. For the in-vivo experiments, a minimum of 5 mice per group were used. Statistical analysis was conducted using GraphPad Prism 9 software and results are presented as means ± SEM. For experiments with two or more groups, a two-way ANOVA with Tukey's multiple comparison tests was used. Significance levels of p-values of 0.05, 0.01, 0.001, or 0.0001 were calculated and are indicated in the figures by *, **, ***, or ****, respectively.