: IL1R1 is central to adhesion of EOC cells on the inflamed peritoneal walls in mice.

(A-B): Analysis of microarray data (TCGA-381-tpm-gencode36) of ovarian cancer patients using R2 Genomics Analysis and Visualization Platform tools showed that IL1β and IL1R1 mRNA overexpression predicts poor survival. (C): Schema of testing the role of Il1β, Il1r1 and Il1rn (Il1r1 antagonist) in ovarian cancer burden both at the needle injury and omental sites. (D-E): HGS-3 murine high-grade serous EOC cells (4.5 million/per mice) were implanted intraperitoneally using 21 sterile gauge needle in C57BL/6 WT and C57BL/6 Il1r1KO mice. Mice were observed for 45-50 days and euthanized. Tumors formed on needle injury site both protruding at the skin, and, in the peritoneum, and on the omentum, shown by red arrows, were isolated, weighed and frozen in liquid nitrogen. Lavages via washing with sterile PBS(5mL) were also collected. The studies were repeated thrice. A representative experiment is shown. (F): Weights of the omental did not differ between C57BL/6 wt and Il1r1KO mice. (G): Weights of the peritoneal/skin tumors differed significantly between C57BL/6 wt and Il1r1KO mice groups. * indicates <0.05. (H-K): Tumors formed at the needle injury site in C57BL/6 wt and Il1rnKO mice are pointed with the red arrows. (I-upper vs I-lower): Images of the omental and peritoneal /skin tumors harvested from the euthanized Il1r1KO mice are shown. (J): Omental tumor weights in C57BL/6 wt did not differ between Il1rnk° mice groups. (L-upper vs L-lower): Images of the omental and peritoneal /skin tumors harvested from the euthanized Il1rnKO mice are shown. (M): Weights of the tumors formed on the needle injury site in C57BL/6 WT mice did not differ for Il1rnKO mice groups. This experiment was repeated twice. Tumor sizes were analyzed via non-parametric T-test using Graph-Prism version -7 or higher. * indicates <0.05. (N): Il1r1 KO does not significantly impact myeloid percentages in the peritoneal lavage and tumor. A) Peritoneal lavage and tumor percentages and cell numbers of CD11b+Ly6G+ neutrophils from flow cytometry data of wild type and IL1R1KO mice injected with HGS3 tumors. B) Peritoneal lavage percentages and cell numbers of F4/80loMHCIIhi and F4/80hiMHCIILow macrophages from the peritoneal lavage and percentage of MHCIIhi and MHCIILow macrophages from tumors from flow cytometry data of wild type and IL1R1 KO mice injected with HGS3 tumors. N = 15-20 mice per group from 3 independent experiments. Statistical significance was determined using Mann-Whitney test, *p < 0.033, **p < 0.002, ***p < 0.001. (O): Table shows that none of the IL1β, IL1R1 and myd88/TLR targeted agents has been approved for treatment of a malignancy yet. (P): Schema of IL1β/IRAK4 signaling pathway. Scheme shows that inhibiting IRAK4 can centrally block IL1/TLR driven signaling in ovarian malignancies. (Q): Table summarizes the status of 3 IRAK4 inhibitors undergoing clinical trials. (R): Using GENT2 database we observed that compared to normal ovaries, malignant ovaries overexpress IRAK4 mRNA. (S): GENT2 database also showed that higher stages of EOC disease significantly increased IRAK4 mRNA expression. Analysis of the EOC patient’s microarray data using GENT2 tools showed that IRAK4 mRNA expression was altered in various stages of disease. Two-sample T-test showed statistical differentiations: IA vs I (p=0.004); IA vs II (p=0.007); IA vs III (p<0.001); IC vs IIA (p=0.002); IC vs III (p<0.001); IIA vs I (p<0.001); IIA vs II (p<0.001); IIA vs III (p<0.001); IIA vs IIC (p=0.006); IIA vs IIIC (p=0.001); IIC vs III (p<0.001); III vs IIB (p=0.004); IIIA vs I (p=0.001); IIIA vs II (p=0.004); IIIA vs III (p<0.001); IIB vs I (p=0.008); IIIB vs III (p=<0.001); IV vs I (p=0.004); IV vs IIA (p=0.005); IV vs III (p=<0.001). (http://gent2.appex.kr/gent2/, date accessed 10/2/2023). (T-U): Analysis of microarray data (TCGA-381-tpm-gencode36) of ovarian cancer patients using R2 Genomics Analysis and Visualization Platform tools showed that IRAK4 and IRAK1 mRNA overexpression predicts poor survival in EOC patients.

IRAK4 participates in survival, cell-cycle progression and cell adhesion.

(A): IRAK4shRNA or null shRNA knockdown followed by clonal selection of HGS-3 murine EOC cells generated a C2-knockdown clone that showed nearly complete deletion of IRAK4. Null vector clones NV-1 or NV-2 were also generated following same procedures. (B): Proteomics analysis of the HGS-3 null vector NV vs partial IRAK4kd showed very limited off-target effects. Notably, downregulation of WNT-4, a poor prognostic factor of EOC was observed. (C) : C2-IRAK4kd HGS-2 clone cells (7 million/mice, IP) when implanted in C57BL/6 mice showed significant increase in survival than the NV1 null vector implanted mice (log rank p=0.0494). (D) : IRAK4kd reduced S-phase arrest. S-phase population upon partial IRAK4kd rose from 30.79 to 52.42% in HGS-3 EOC cells. (E) : IRAK4kd reduced mitotic index in HGS-3 murine EOC cells (p=0.054). (F) : IRAK4kd reduced adhesion to murine-Collagen. p=0.0071, p=0.0105, p=0.0009, error bars: one stdev in either direction. (G) : IRAK4kd HGS clones resulted in reduced expression of E-cadherin. (H) : E-cadherin and IRAK4 mRNA show strong correlation of expression. R=0.471, p=2.7e-10. mRNA expression data available at R2-genomics was analyzed using the system inbuilt tools (I) : Structure-activity relationship (SAR) guided optimization of JH-I-25 scaffold, a literature described IRAK4 inhibitor leading to 3 potent novel analogs, is shown. The chemical structures of UR241-2 (sulfone), PSP-099 (sulfide) and PSP-100 (sulfoxide) are shown. (J) : Calculated drug-likeness/physicochemical properties (LogP, cLogP and topological polar surface area-tPSA are shown. These properties were calculated using Chemdraw software. (K) : Comparison of IRAK1, IRAK2, and IRAK4-kinase inhibitory IC50s of UR241-2 versus CA4948, PF-06650833, PSP-099 and PSP-100 are shown. Compounds were screened using HotSpot Kinase assay available in Reaction Biology Laboratories using 1μM ATP concentration under a 10-dose singlet screening program. (L) : Dendrograms of global kinome activity of UR241-2 at 50- and 500nM doses. Red indicates kinase affected. At 50nM dose, 13 of 682 kinases were inhibited, whereas increasing dose to 500nM, 34 kinases, shown as red dots, from among 682 total kinases, were inhibited. (M) : A HEK293 cell-based nanoBret target Engagement screening assay was conducted to determine the selectivity of UR241-2 among 10 most affected kinases revealed by HotSpot kinase assay. The nanoBret assay showed that UR241-2 inhibits IRAK4 kinase activity selectively and other kinases including IRAK1 are affected at 10-100 folds higher doses, qualifying UR241-2 as one of the selective and specific IRAK4 kinase inhibitors.

In silico docking and molecular simulations exhibit UR241-2’s interactions with IRAk4 protein.

(A): JH-I-25 and its newer analogs UR241-2, PSP-099 and PSP-100 were docked individually to IRAK4 crystal structure using Gnina docking software, which is built on neural networks (CNNs) as a scoring function. (B): The interacting amino acid residues and their nature of interactions are shown. Green=aliphatic, magenta=aromatic, blue=basic, sky blue=polar, and yellow=sulfur. Red indicates unfavorable interactions. (C): JH-I-25, UR241-2, PSP-099 and PSP-100 analogs were docked together to IRAK4 protein using Gnina tools. (D): Root-mean square density (RMSD) simulations of JH-I-25, UR241-2, PSP-099 and PSP-100 ligands docked to IRAK4 falling within 2Å unit range are shown. RMSD-ligand is the root mean deviation of the bounded ligand compared to the ligand in the crystal structure. (E): Root-mean square density (RMSD) simulations of IRAK4 protein docked with JH-I-25, UR241-2, PSP-099 and PSP-100 ligands falling with 2Å unit range are shown. RMSD-Protein is the root mean deviation of the protein for each frame of the simulation compared to the crystal structure. Both D and E indicate that docking quality was acceptable. (F): RMSF (root mean square fluctuation) of JH-I-25, UR241-2, PSP-099 and PSP-100 ligands are shown. The ligands, when bound, induce the most notable structural changes in residues 172-181 of IRAK4 which are reflected in the spike in the RMSF charts. RMSF is the mean fluctuation of IRAK4 protein residues after binding of ligands. RMSF indicates the conformational flexibility of the complex. The time/frame element is removed from RMSF in order to show an average fluctuation of the residues. (H): The affinity score measures the affinity of a ligand (in that particular frame) binding to the protein. The frames represent different conformations that a ligand may adopt. The affinity is measured in kCal/mol and the lowest frame (lowest energy) is selected as the most stable binder. UR241-2 showed the best affinity score (-11.95), hence was considered the most stable binder ligand of IRAK4 crystal structure among the compounds tested.

UR241-2 blocks IL1β induced IRAK4 phosphorylation.

(A-B): UR241-2 (5-150nM) treatment for 4hrs blocks human and murine IL1β (5nM, 30 minutes) induced IRAK4-phosphorylation in both human (HCH-1) and murine high-grade serous EOC cells (HGS-1 and -3) cells. (C): UR241-2 (2.5μM) treatment reduced NF-κβ luciferase reporter activity induced by human TNF-α (3, 10 and 30nM) in stably transfected OVCAR-3-NF-κβ-Luc cell-lines. * indicates p=0.01 to 0.008 (two tailed unpaired T-test). (J): Similarly, UR241-2 (2.5μM) treatment, IL-1β (5ng), TNF-α (10ng), LPS (100nM) and R848 (TLR-7/8 agonist, 10μM). Ratio of positive/negative NF-κβ nuclei are shown. NF-κβ (Green) nuclear migration was inhibited significantly in IL-1β and LPS stimulated cells. The number of NF-κβ nuclear positive cells were counted using ImageJ software. * indicates <0.05 (Student T-test). (D): Summary of ADMET profiling of UR241-2. ADMET characteristics of UR241-2 were examined in the laboratories of Curia Inc, NY, USA. Complete ADMET along with positive controls is shown in the Supplementary Figure-3. (E): Pharmacokinetic PK profiling of UR241-2 in CD-1 mice. Mice were intraperitoneally injected with UR241-2 and blood concentration of UR241-2 was quantified by HPLC at the indicated hours. See also Supplementary Figure-3.

UR241-2 treatment colonies, proliferation, mitotic index in vitro and tumor growth in vivo.

(A): UR241-2 (5, 10 and 20 µM) treatment for 7 days blocks colonies formed from 2000 cells seeded in a 6 well plate compared to control. One representative image of multiple images of each well that were collected are shown. (B): Average size of colonies in vehicle versus 5, 10 and 20 μM groups were analyzed by ImageJ software. Non-parametric T-test was conducted to determine the statistical differences between the control and various treatment groups. * indicates <0.05 (Student T-test). (C): UR241-2 (20-60μM) treatment reduced cell viability in ES2, HCH1, SKOV-3, OVCAR-8 and OVCAR-3 EOC cells. The cell viability was measured by SRB assay which measures the total protein synthesis in drug treated cells compared to control vehicle. (D): UR241-2 (10-20μM) treatment reduced cell division in SKOV-3 and OVCAR-3 cells. The dividing cells were captured by pHistone-3 staining. SKOV-3 and OVCAR-3 cells (10,000/well) were seeded in an 8-well EasyCell chamber slides overnight. The cells were treated with vehicle and UR241-2 (10-20μM). Cells were fixed after 24-hours using neutral buffered formalin solution (35µL) for 20 minutes. The media was removed, and cells were washed repeatedly with TBST (500µL/5 min). Fixed cells were stained with p-Histone-H3 antibody (Cell Signaling Technology, cat#9701, 1:500 dilution, overnight) and then counterstained with Dylight-488(Vector laboratories, cat#DI-1488) secondary antibody (1:2000). The cells were washed repeatedly with TBST (500µL/5 min). Washed cells were stained again with the Vectashield mounting media containing DAPI (Vector Lab, cat#H-1200-10), protected with a cover glass slide, then imaged on an Olympus BX41 microscope. All images taken with a 10x ocular and 20x objective. All cells in each field were manually counted and assessed as either actively dividing (metaphase, anaphase, telophase) or not. OVCAR-3 vehicle group had significantly more actively dividing cells than the 20μM treatment group (p=0.0024). SKOV3 vehicle group had significantly more actively dividing cells than both 10μM and 20μM treatment groups (**p=0.0047, ***p=0.0003). (E): UR241-2 treatment (20mg, M-F, IP) reduced the growth of SKOV-3 cells derived xenografts growing in NSG mice. 10 NSG mice implanted subcutaneously with SKOV-3 cells (1 million/mice in DMEM+Matrigel (1:1, 100µL/mice) were randomized and treated for 28-days. Tumor sizes were measured on the days indicated in the X-axis. Mice were euthanized and tumors and peripheral blood were harvested. Averaged tumor volumes differed significantly between the control and treatment groups. *=p<0.05, **=p<0.005, ***=p<0.0005. (F): Images of the tumors formed in vehicle and UR241-2 treated mice are shown. (G): Animal weights did not differ. (H): Tumor weights between vehicle and treated groups differed significantly. The tumor weights of the individual mice in the control and treatment groups were plotted using GraphPrism Version 8.0. p=0.0001.

IRAK4 inhibition via UR241-2 decreases tumor burden at the needle injury site and increases the number of MHCII expressing myeloid cells in the peritoneal cavity and tumor.

(A): Tumor weights on the needle injury site were significantly lower in the UR241-2 treatment group than vehicle treated animals. Statistical significance was determined using Mann-Whitney test, *p < 0.033, **p < 0.002, ***p < 0.001. (B): Tumor weights on omentum did not differ between UR241-2 treatment group than vehicle treated animals (NS). Two of the three replicates are combined and shown. IRAK4 inhibition via UR241-2 increases the number of MHCII expressing myeloid cells and decreases MHCII low and CD206+ macrophages in the peritoneal cavity. Leukocytes were isolated from the peritoneal lavage or tumor of HGS-3-tumor bearing mice and stained for flow cytometry. (C) Representative flow cytometry plots of F4/80loMHCIIhi M1 macrophages and F4/80hiMHCIILow macrophages from the peritoneal lavage of HGS-3 tumor bearing mice treated with Vehicle or UR241-2. (D) Percentage and cell number of F4/80loMHCIIhi M1 macrophages and F4/80hiMHCIILow macrophages from C. (E) Representative flow cytometry plots of F4/80hiCD206+ macrophages from the peritoneal lavage of HGS-3 tumor bearing mice treated with Vehicle (Vx) or UR241-2. (E-right) Percentage and cell number of F4/80hiCD206+ macrophages. IRAK4 inhibition reduces neutrophil numbers in the peritoneal cavity and tumors of HGS3 bearing mice: (F): Flow cytometry plots showing CD11b+Ly6G+ neutrophils and MHCII+ neutrophils in the HGS-3 tumor. Percentage and cell number per gram are quantified to the right. (G): Flow cytometry plots showing CD11b+Ly6G+ and MHCII+ neutrophils in the peritoneal lavage of HGS-3 tumor bearing mice. Percentage and cell number are quantified to the right. N= 9 mice per group from two independent experiments. Gating strategies are shown in Supplementary Data-7. One mouse from the UR241-2 group died prior to analysis.

Bulk-seq of UR241-2 treated tumors show downregulation of ECM genes and upregulation of neutrophil activation genes.

(A): PCA plot of Bulk-Seq analysis of control and UR241-2 treated tumors. (B): Volcano plot of genes, overexpressed and downregulated between control and treatment groups are shown. Downregulated genes belonging to ECM family are shown. (C): Gene-families downregulated by UR241-2 are shown. (D): Gene-families upregulated by UR241-2 are shown. (E): Heatmap of the genes altered between control and treatment. (F): Heatmap of the ECM family of genes downregulated by UR241-2 are shown. (G): HGS-3 cells treated with UR241-2 demonstrated an inhibited ability to adhere to collagen compared to control, with cells treated at 5μM showing significantly less adhesion than the 1μM group at both the sixty (p<0.00005) and thirty (p=0.00068) minute timepoints. (H): Cell migration in HGS-3 cells treated with 500nM UR241-2 were significantly inhibited compared to control in just 12 hours (p<0.02), and both the 250nM and 500nM groups showed a significantly greater impact than the 100nM, 50nM, and control treatment groups at 24 hours (p<0.05).

Based on the data, it is proposed that IRAK4 kinase inhibition by UR241-2 led to blocked activity of NF-κβ resulting in the activation of MHCII macrophages and neutrophils which suppressed tumor growth at the site of injury.

Created with BioRender.com.

HGS-3 murine high-grade serous EOC cells (3-4.5 million/per mice) were implanted intraperitoneally using 21-gauge needle in C57BL/6 WT and C57BL/6 Nlrp3KO mice.

Mice were observed for 45-50 days and euthanized. Tumors formed on needle injury site, protruding at the skin as well as in the peritoneum and on the omentum, shown by red arrows, were isolated, weighed and frozen in liquid -nitrogen. Lavages via washing with sterile PBS(5mL) were also collected. The studies were repeated twice. A representative experiment is shown. Weights of the omental did not differ between C57BL/6 WT (Figure-1H) and Nlrp3KO mice. Similarly, the tumor sizes at the site of injury did not differ between C57BL/6 WT and Nlrp3KO mice.

Analysis of ovarian serous cystadenocarcinoma (2022-v32) microarray data (TCGA-381-tpm-gencode36) of ovarian cancer patients using R2 Genomics Analysis and Visualization Platform tools showed that FLT-3 and CLK-3 mRNA overexpression predicts poor survival.

(A): Kinetic solubility of UR241-1 is shown. (B): Quantified stability of UR241-2 in human and mouse liver microsomes is shown. (C): Inhibition of CYP450 isoforms by UR241-2 is quantified using HPLC. Controls used were furafyllin, Ticlopidine HCL, Montelukast sodium hydrate, sulfaphenazole, N-3-benznirvanol, Qunidine, ketoconazole. CYP isoforms affected by UR241-2 in terms of %-inhibition are shown. (D): Human and murine plasma protein binding of UR241-1 is shown in % units. Propranolol was used as a control. (E): CaCo-2 cell permeability of UR241-2 is shown. Efflux ratio of 4.14 indicates that UR241-2 faces significant efflux. (F): UR241-2 does not inhibit hERG. IC50 is 101.2µM. E-4031 was used as control. IC50 for E-4031 was 1.62e-08. (G): Pharmacokinetic (PK) of UR241-2 at 20mg/kg administered IP is shown. It is shown that ∼8ng/ml concentration of UR241-2 is maintained up until 8th hour of monitoring.

Analysis of the peripheral blood showed that hemoglobin (HB) levels did not differ between the vehicle and UR241-2 treated mice.

List and catalog details (reference no, manufacturer, name, other name, Clone, fluorochrome, max and min excitation, laser) of flow cytometry antibodies used in this study.

Gating strategy