RadD from Fusobacterium nucleatum Engages NKp46 to Promote Antitumor Cytotoxicity

Ahmed Rishiq
Johanna Galski
Reem Bsoul
Mingdong Liu
Rema Darawshe
Renate Lux
Gilad Bachrach
Ofer Mandelboim author has email address

The Concern Foundation Laboratories at the Lautenberg Center for Immunology and Cancer Research, Institute for Medical Research Israel Canada (IMRIC), Hebrew University-Hadassah Medical School, Jerusalem, Israel
The Institute of Dental Sciences, The Hebrew University-Hadassah School of Dental Medicine, Jerusalem, Israel
Section of Biosystems and Function, Division of Oral and Systemic Health Sciences, UCLA School of Dentistry, Los Angeles, United States

https://doi.org/10.7554/eLife.108439.1

Open access
Copyright information

Figures and data

NKp46 expression modifies the prognostic effect of F. nucleatum in a tumor-type-specific manner.
(A) Kaplan–Meier survival curves for head and neck squamous cell carcinoma (HNSC) patients stratified by F. nucleatum status and NKp46 (NCR1) expression. Patients with concurrent F. nucleatum positivity (n=87) and high NKp46 expression exhibited significantly improved overall survival compared to those who were F. nucleatum-negative (n=44) but NKp46-positive (log-rank P < 0.05).
(B) Kaplan–Meier survival curves for colorectal cancer (CRC) patients stratified by the same criteria showed no significant difference in survival between F. nucleatum–positive (n = 44) and F. nucleatum–negative (n = 31) groups.
(C) Table summarizing hazard ratios (HR) for F. nucleatum–negative cases among NKp46⁺ patients. In HNSC, absence of F. nucleatum was associated with a significantly poorer prognosis (HR = 2.08, 95% CI: 1.20–3.61), whereas in CRC, F. nucleatum absence showed no significant association with patient prognosis (HR = 0.71, 95% CI: 0.26–1.95).
(D) Comparison of NKp46 expression across HNSC and CRC tumors. Log₂ expression levels of NKp46 mRNA were compared across HNSC and CRC cohorts, stratified by F. nucleatum positive and negative. Results were analyzed by one-way ANOVA with Bonferroni post hoc correction.

Binding of Fusobacterium nucleatum to NKp46 and its D1 domain.
The figure shows histograms of FITC-labeled F. nucleatum subsp. nucleatum ATCC 23726 (FN23726) (upper histograms) and ATCC 10953 (lower histograms) incubated with 2 μg of NKp46 Ig, D1 domain of NKp46 (D1 Ig), Ncr-1 Ig, and CD16 Ig fusion proteins. Representative staining from one of two independent experiments is shown.

RadD is the bacterial ligand for NKp46.
A-B. Density plot of FITC labeled ATCC 10953 (A) and its ΔfadI mutant derivative ATCC 10953 ΔFad-I (B) stained with the various fusion proteins (listed in the X axis). C. Schematic representation of ATCC 10953 wild type (WT) strain and RadD surface expression (Left) compared to ATCC 10953 ΔFad-I (Right). D. Density plot of the FITC-labeled ΔRadD mutant strain of ATCC 10953 stained with various fusion proteins (listed in the X axis). The figure shows data from one representative experiment out of three to five independent experiments. E. Fold change quantification of FITC-labeled bacteria binding to the fusion proteins Ccm1-Ig, NKp46 Ig and Ncr-1 Ig in ATCC 10953 (left) and ATCC 23726 (right).
Summary of three to five independent experiments. The mean value ±SD of the experiments is presented. *P < 0.05, **P ≤ 0.01, ***P ≤ 0.001 and ****P ≤ 0.0001.

NKp46-02 antibody and arginine block ATCC 10953 binding to NKp46.
A. Quantification of Median Fluorescent Intensity (MFI) of FITC-labeled ATCC 10953 binding to Ccm1-Ig, NKp46-Ig, and Ncr-1 Ig, without or with 5 and 10 mM of L-Arginine. Data combined from three to four independent experiments are presented. B. NKp46-Ig (2 µg) was pre-incubated with 1 µg of a control anti-PVR antibody and NKp46 monoclonal antibodies (9E2, 461-G1, and 02) to evaluate the blocking of ATCC 10953 interaction with the NKp46 receptor. C. Shows the quantification results of histograms depicted in A. The mean value ±SD of the experiments is presented. *P < 0.05, **P ≤ 0.01, ***P ≤ 0.001 and ****P ≤ 0.0001.

Cytotoxicity and tumor growth is RadD and Ncr1-dependent.
A. Schematic diagram showing the design of the NK cells cytotoxicity assay against breast cancer cell lines T47D and MCF7. 1. Tumor cells were stained with Calcein-AM dye and then incubated either with tumor cells (T47D or MCF7) only, tumor + NK, tumor + bacteria (ATCC 10953WT and ATCC 10953 ΔRadD) + NK with/without preincubation with 02 antibody. 2. Killing assays were performed in a 37°C incubator for 4 hours. 3. The fluorescence intensity of Calcein was measured to determine cell viability using a spectrophotometer (Tecan Spark). Summary of NK cytotoxicity against T47D (B) and MCF7 (C) breast cancer cell lines. Combined results from five independent experiments. D. C57BL\6 or NCR1-KO mice were shaved and AT3 cells (1 x 10^{^6} cells in 100 μl PBS) were injected one day later into the mammary fat pad. When tumors reached a size of about 500 mm³, mice were inoculated intravenously with 5 × 10⁷ ATCC 10953WT and 7.5 × 10⁷ ATCC 10953 ΔRadD bacteria. Eight days later, mice were sacrificed and tumor weight was determined. E. The tumor weight of C57BL or NCR1-KO (F) mice. The figure shows the combination of 4-5 experiments performed. The mean value ±SD of the experiments is presented. *P < 0.05, **P ≤ 0.01, ***P ≤ 0.001 and ****P ≤ 0.0001.

Postulated model for the RadD-NKp46 interaction impact on NK cytotoxicity and tumor growth.
NKp46 interaction with RadD expressed by Fusobacterium nucleatum triggers NK cell cytotoxicity. This activation enhances tumor cell killing in vitro and in vivo. Conversely, the absence of RadD or the blocking of NKp46 impairs NK cell activity, leading to tumor progression. This figure was created using BioRender.

Sign up for email alerts