Cancer-associated mesothelial cell–derived ANGPTL4 and STC1 promote the early steps of ovarian cancer metastasis

Ovarian cancer (OvCa) preferentially metastasizes in association with mesothelial cell–lined surfaces. We sought to determine if mesothelial cells are required for OvCa metastasis and detect alterations in mesothelial cell gene expression and cytokine secretion upon interaction with OvCa cells. Using omental samples from patients with high-grade serous OvCa and mouse models with Wt1-driven GFP-expressing mesothelial cells, we validated the intratumoral localization of mesothelial cells during human and mouse OvCa omental metastasis. Removing mesothelial cells ex vivo from human and mouse omenta or in vivo using diphtheria toxin-mediated ablation in Msln-Cre mice significantly inhibited OvCa cell adhesion and colonization. Human ascites induced angiopoietin-like 4 (ANGPTL4) and stanniocalcin 1 (STC1) expression and secretion by mesothelial cells. Inhibition of STC1 or ANGPTL4 via RNAi obstructed OvCa cell-induced mesothelial cell to mesenchymal transition while inhibition of ANGPTL4 alone obstructed OvCa cell-induced mesothelial cell migration and glycolysis. Inhibition of mesothelial cell ANGPTL4 secretion via RNAi prevented mesothelial cell–induced monocyte migration, endothelial cell vessel formation, and OvCa cell adhesion, migration, and proliferation. In contrast, inhibition of mesothelial cell STC1 secretion via RNAi prevented mesothelial cell–induced endothelial cell vessel formation and OvCa cell adhesion, migration, proliferation, and invasion. Additionally, blocking ANPTL4 function with Abs reduced the ex vivo colonization of 3 different OvCa cell lines on human omental tissue explants and in vivo colonization of ID8p53–/–Brca2–/– cells on mouse omenta. These findings indicate that mesothelial cells are important to the initial stages of OvCa metastasis and that the crosstalk between mesothelial cells and the tumor microenvironment promotes OvCa metastasis through the secretion of ANGPTL4.


Mouse tissue immunohistochemistry and immunofluorescence.
Harvested mouse omental tumors were formalin fixed prior to paraffin embedding. The fixed tissues were dehydrated using increasing dilutions of ethanol, cleared in xylene, embedded in paraffin wax, and 5 µm thick sections were mounted on Superfrost Plus charged slides (Thermo Fisher Scientific). The slides were deparaffinized in xylene and rehydrated in decreasing dilutions of ethanol. Antigen retrieval was performed in 10 mmol/L sodium citrate buffer (0.05% Tween-20, pH 6) for 30 minutes at 100°C followed by incubation with endogenous peroxide block using 3% (v/v) hydrogen peroxide in absolute methanol for 20 minutes at room temperature. The slides were blocked in 10% normal goat serum in PBS, 0.1% Triton-X 100, and incubated overnight at 4°C with primary antibodies against GFP (1:100). The staining was visualized using a ready to use VECTASTAIN Elite ABC-HRP kit and a DAB Substrate kit and counterstained with hematoxylin. Images were photographed using a Nikon Eclipse Ti2 (Nikon) microscope.
For immunofluorescence, all slides were blocked in 10% normal goat serum in PBS, 0.1% Triton-X 100 after the antigen retrieval step and incubated overnight at 4°C with primary antibodies against PDPN (1:100) and CK-19 (1:100) followed by incubation with Alexafluor 488 or Alexafluor 568-labeled secondary antibodies (1:250). Nuclei were stained with Hoechst 33342 nucleic acid stain (1:1000 in 1% BSA in PBST) and slides were washed and mounted with ProLong Gold Antifade. Appropriate negative controls for the immunostaining were prepared by omitting the primary antibody. Images were photographed using a live cell DSU Spinning Disk Confocal microscope (Olympus).

RNA isolation and Quantitative Reverse Transcription-Polymerase Chain Reaction.
Total RNA was isolated from the primary human mesothelial cells using Trizol (Invitrogen) as per the manufacturer's protocol. The RNA quality and concentration was determined using the NanoDrop 8000 Spectrophotometer (Thermo Fisher). Reverse transcription of 2 µg total RNA was carried out using the High-Capacity cDNA Reverse Transcription Kit. qPCR was performed with predesigned TaqMan probes (Supplemental Table 4) using TaqMan Fast Advanced Master Mix on an Applied Biosystems StepOnePlus Real-Time PCR System (Applied Biosystems). GAPDH was used as a housekeeping gene for normalization. The reactions were run in triplicate with at least 2 biological replicates for each individual experiment. Relative levels of mRNA expression were calculated using the 2 -ΔΔCt method. Differences between treatments were evaluated using an unpaired two-tailed Student's t-test.

RNA-seq and bioinformatic analysis.
Total RNA was isolated from primary human mesothelial cells 36 hours after treatment with Tyk-nu spheroid conditioned media, ascites, or control media using the RNeasy Mini kit (Qiagen) following manufacturer's instructions. RNA purity was checked using the NanoPhotometer ® spectrophotometer (IMPLEN). RNA integrity and quantitation were assessed using the RNA Nano 6000 assay kit of the Bianalyzer 2100 system (Agilent Technologies, CA, USA). Library preparation and next generation RNA sequencing was carried out by Novogene. Sequencing libraries were generated using the NEBNext ® Ultra ™ RNA library Prep kit for Illumina ® (NEB, USA) following manufacturer's recommendations and index codes were added to attribute sequences to each sample. The library quality was assessed on the Agilent Bioanalyzer 2100 system. The clustering of the index-coded samples was performed on a cBot Cluster Generation System using the PE Cluster Kit cBot-HS (Illumina) according to the manufacturer's instructions. After cluster generation, the library preparations were sequenced on an Illumina platform and paired-end reads were generated. Raw data (raw reads) of the FASTQ format were first processed through fastp. In this step, clean reads were obtained by removing reads containing adapter and poly-N sequences and reads with low quality from raw data. At the same time, Q20, Q30 and GC content of the clean data were calculated. All the downstream analyses were based on the clean data with high quality. Paired-end clean reads were aligned to the reference genome using the Spliced Transcripts Alignment to a Reference (STAR) software. FeatureCounts was used to count the read numbers mapped of each gene. RPKM (Reads Per Kilobase of exon model per Million mapped reads) of each gene was calculated based on the length of the gene and reads count mapped to this gene. Differential expression analysis between two conditions/groups (three biological replicates per condition) was performed using DESeq2 R package. The resulting P values were adjusted using the Benjamini and Hochberg's approach for controlling the False Discovery Rate (FDR). Genes with an adjusted P value < 0.05 found by DESeq2 were identified as differentially expressed. Differential expression analysis of two conditions without biological replicates was performed using the EdgeR R package, A corrected p-value of 0.005 and log 2 (Fold CHANGE) of 1 were set as the threshold for significantly differential expression. Gene set enrichment analysis was performed for 50 Hallmark gene sets using the Molecular Signatures Database (MSigDB v7.4).
Hypoxia. Ten thousand primary human mesothelial cells were plated in black-walled 96-well plates and incubated at 37°C for 24 hours. The mesothelial cells were transfected with control, ANGPTL4, or STC1-targeted siRNAs (as described above). The mesothelial cells were then treated with malignant ascites or control growth media and cultured at 5%CO 2 , 37ºC. Forty-eight hours later, the medium was replaced with fresh growth medium containing Image-iT Green Hypoxia Reagent at a final concentration of 5 µM, and the cells were incubated at 5%CO 2 /37ºC. The cells were washed with PBS and fixed in 4% paraformaldehyde containing Hoechst 33342 (2 µM) for 10 minutes. The paraformaldehyde was exchanged with PBS. The fluorescence was quantified using the SpectraMax iD5 Microplate Reader (Molecular Devices).
Bioenergetic Glycolysis assay. Glycolysis was measured as previously described (60) with the seahorse Extracellular Flux XF-96 analyzer (Agilent). Briefly, human primary mesothelial cells were seeded in Seahorse XF-96 plates at a density of 10,000 cells per well and allowed to adhere for 24 hours. The next day, media was replaced with either control media or ascites from a patient with high grade serous OvCa, and control antibodies, ANGTPL4 antibody (2 µg/ml), STC1 antibody (200 ng/ml) or both antibodies were added to the assay. After 24 hours, cells were changed to unbuffered DMEM without glucose (D5030, Sigma-Aldrich supplemented with glutamine (2mmol/L), pH adjusted to 7.4, and incubated in a non-CO 2 incubator for 1 hour. The extracellular acidification rates (ECAR) were determined following sequential injections with D-glucose (10mM), Oligomycin (2µM), and 2-deoxyglucose (100mM). The extracellular acidification rates (ECAR) after the injection of D-glucose were a measure of glycolysis and after the injection of oligomycin represented glycolytic capacity. The glycolytic reserve is quantified by the measure of ECAR after the injection of 2-deoxyglucose. Samples were analyzed with 8 technical replicates. Data is representative of 3 independent experiments after normalization using the CyQuant cell proliferation assay kit.
Statistics. Data were analyzed by GraphPad Prism 8 (Version 9.3.1) and presented as mean± standard error mean (SEM) of the indicated number of samples. Student's t -test two-tailed and one-way ANOVA with pairwise comparisons were used to determine significance in two-group and multiple-group experiments, respectively. P values of less than 0.05 were statistically significant.

Supplemental Figures
Supplemental Figure 1. Tracing mesothelial cells. GFP expression in mesothelial cells of mouse omentum from a transgenic mouse. Immunohistochemical localization of GFP-labeled mesothelial cells in mouse omentum collected from C57BL/6 (wild-type) or heterozygous Wt1 tm1(EGFP/Cre)Wtp /J mice. Black arrow, GFP-expressing mesothelial cell. Scale bar 500 μm.  2 or 3). B. Differentially expressed genes (DEGs) between different treatment groups shown in a hierarchical clustering heatmap. Rows are expression levels denoted as the z-score, displayed in a high-low (red-blue) color scale. The numeric scale indicates z-transformation. C. Gene set enrichment analysis plots demonstrating normalized enrichment score (NES) of the RNA-seq data. The common pathways that both CM (top) and ASC (bottom) regulate are shown. FDR, false discovery rate. D-E. Expression levels of HMGCR (D) or INSIG1 (E) mRNA in mesothelial cells treated with CM or ASC were measured using qRT-PCR. Data is represented mean±SEM (n=3). *, p<0.05 and **, p<0.01 calculated by one-way ANOVA.

Supplemental Figure 6. Confirmation of transient knockdown of ANGPTL4 AND STC1 in primary human mesothelial cells. A-B.
The expression and secretion levels of ANGPTL-4 (A) and STC1 (B) from mesothelial cells was measured using qRT-PCR (left) or enzyme linked immunoassays (ELISA, right). Mesothelial cells were transfected with control, STC-1, or ANGPTL-4 specific siRNA and treated with human ascites (ASC). After 48 hours qRT-PCR was performed, or mesothelial cell conditioned media was collected for ELISA analysis. Data shown as mean±SEM (n=5). *, p<0.05 and **, p<0.01 calculated by one-way ANOVA

Supplemental Figure 7. Combination treatment with ANGPTL4 and STC1 neutralizing antibodies has effect on ascites-induced glycolysis in mesothelial cells. A-B.
Seahorse glucose metabolism in mesothelial cells was measured. Primary human mesothelial cells were treated with control (IgG1 and/or IgG2) STC1 or ANGPTL4 specific neutralizing antibodies and stimulated with human ascites (ASC) or control media (48 hours). A. The extracellular acidification rate (ECAR) profile of mesothelial cells following glucose, oligomycin, and 2-DG treatments. B. Changes in basal ECAR, glycolytic capacity and glycolytic reserve in the mesothelial cells. Data represented as Mean±SEM (n=3). *, p<0.05, **, p<0.01, and ***, p<0.001 comparison by one-way ANOVA. C. Mesothelial cells were treated with control (IgG1 and/or IgG2) STC1 or ANGPTL4 specific neutralizing antibodies and stimulated with human ascites (ASC) or control media (36 hours). qRT-PCR for mesenchymal (snail and fibronectin) markers. Data shown as mean±SEM (n=5). *, p<0.05, **, p<0.01, ***, p<0.001, and ****, p<0.0001 calculated by one-way ANOVA. Figure 8. Ascites-induced hypoxia is not regulated by ANGPTL4 or STC1. Hypoxia was measured in mesothelial cells using the Image-iT Green Hypoxia Reagent. The nuclei were stained with Hoechst. Primary human mesothelial cells were transfected with control, ANGPTL4 or STC1 specific siRNA and 24 hours later treated with human ascites (ASC) or control media for 48 hours. Data represented as mean±SEM (n=5). *, p<0.05 calculated by one-way ANOVA. Figure 9. Mesothelial cell derived ANGTPLT4 and STC1 regulate ovarian cancer cell function. Mesothelial cells were treated with control siRNA or siRNA targeting STC-1 or ANGPTL4 followed by treatment with ascites (ASC) for twenty-four hours. Functional assays were performed using fluorescently labeled Kuramochi or Ovcar5 ovarian cancer (OvCa) cells. A. OvCa cell adhesion (1 hour) was tested on top of the transfected and pre-treated mesothelial cell monolayer. B. OvCa cell migration (12-24 hours) was tested using a Boyden chamber lined with transfected mesothelial cells. C. OvCa cell proliferation (72 hours) was tested on top of the transfected and pre-treated mesothelial cell monolayer. D. OvCa cell invasion (24-48 hours) was tested through transfected and pre-treated mesothelial cell monolayer and rat tail collagen type I using a Boyden chamber. All data shown as Mean ± SEM (n=5). *, p<0.05, **, p<0.01, ***, p<0.001, and ****, p<0.0001 calculated by one-way ANOVA. Figure 10. Treatment with an ANGPTL4 neutralizing antibody inhibits Tyk-nu cell induced fibronectin expression on the surface cells of the human omentum. Human ex-vivo colonization assay. GFP-labeled Tyk-nu, OVCAR5 or Kuramochi, OvCa cells were seeded on human omental tissue explants. The cultures were untreated or treated with an ANGPTL4 neutralizing antibody, a STC1 neutralizing antibody, the respective control IgGs, or both neutralizing antibodies. The cancer cells and mesothelial cells were digested off the omentum and the unlabeled omental surface cells were collected using fluorescently-activated cell sorting and qRT-PCR was performed. Data represented as mean ±SEM (n=8 patients/ group). *, p<0.05 and **, p<0.01 by one-way ANOVA.

Supplemental Tables
Supplemental Table 1. Genes regulated in mesothelial cells by ascites and OvCa-conditioned media (Top 100 genes compared, all positively regulated)