Science Immunology - Supplementary Materials for · 2017. 4. 25. · Fig. S1. Characterization of soluble recombinant NCR-Fc fusion molecules. (A) NCR-Fc profiles on Coomassie blue-stained

Supplementary Materials for

Complement factor P is a ligand for the natural killer

cell–activating receptor NKp46

Emilie Narni-Mancinelli,* Laurent Gauthier, Myriam Baratin, Sophie Guia,

Aurore Fenis, Ala-Eddine Deghmane, Benjamin Rossi, Patrick Fourquet,

Bertrand Escalière, Yann M. Kerdiles, Sophie Ugolini, Muhamed-Kheir Taha,

Eric Vivier*

*Corresponding author. Email: [email protected] (E.N.-M.); [email protected] (E.V.)

Published 28 April 2017, Sci. Immunol. 2, eaam9628 (2017)

DOI: 10.1126/sciimmunol.aam9628

The PDF file includes:

Fig. S1. Characterization of soluble recombinant NCR-Fc fusion molecules.

Fig. S2. B12 cells express a ligand for NKp46.

Fig. S3. 27A1.7 mAb recognizes JAM1, but JAM1 is not a direct ligand for

NKp46.

Fig. S4. Characterization of soluble recombinant CFP-HIS fusion molecules.

Fig. S5. Native CFP oligomers bind to NKp46-Fc.

Fig. S6. B12 cell–derived CFP activates NKp46 reporter cells.

Fig. S7. CFP does not bind to T cells.

Fig. S8. CFP stimulation does not induce classical NK cell effector functions.

Fig. S9. Impact of anti-NK1.1 and anti-aGM1 treatments on NKp46+ ILC subsets.

Fig. S10. Impact of anti-aGM1 treatment on Nm infection.

Fig. S11. Therapeutic benefit of CFP delivery for Nm infection control in the

absence of NKp46+ ILCs or NKp46.

Table S1. Affinity/avidity of CFP for the NKp46 receptor.

Table S2. List of genes modulated upon NKp46 stimulation.

Table S3. List of genes modulated upon CFP stimulation.

Other Supplementary Material for this manuscript includes the following:

(available at immunology.sciencemag.org/cgi/content/full/2/10/eaam9628/DC1)

immunology.sciencemag.org/cgi/content/full/2/10/eaam9628/DC1

Raw data table (Microsoft Excel format)

Supplementary Materials:

Fig. S1. Characterization of soluble recombinant NCR-Fc fusion molecules. (A) NCR-Fc

profiles on Coomassie blue-stained gels. (B) NCR-Fcs were coupled to protein A-A488 bound to

latex beads. Each NCR-Fc bound to the beads was analyzed by FACS for NCR-specific mAb

staining. The antibodies used were the anti-HuNKp46 9E2, anti-MoNKp46 29A1.4 and anti-

NKp30 p30-15 clones.

Fig. S2. B12 cells express a ligand for NKp46. Reporter cells were cocultured with B12 cells in

the presence of the indicated anti-HuNKp46 (BAB281) or anti-NKp30 (AZ20) mAbs (A) or with

soluble recombinant Fc fusion proteins (B). Cell activation was assessed by evaluating IL-2

levels in the coculture supernatant in a standard CTLL-2 survival assay. Data are the means of

triplicates ± SD and are representative of two independent experiments.

Fig. S3. 27A1.7 mAb recognizes JAM1, but JAM1 is not a direct ligand for NKp46. (A)

Cytometric profiles of B12 and NIH3T3 cells stained with 27A1.7 or isotype control rIgG2b

mAbs. (B) Flow sensograms of B12 cell lysate in the presence of the indicated mAb flowing

over a flow cell immobilized with 27A1.7 mAb. (C) Flow sensograms of JAM1-Fc flowing over

a flow cell immobilized with 27A1.7 mAb. (D) B12 cells were silenced for Jam1, and JAM1

expression was assessed by flow cytometry. (E) Flow sensograms of mouse JAM1-Fc flowing

over a flow cell immobilized with mouse NKp46-Fc or JAM1-mAb. Data are representative of

two independent experiments.

Fig. S4. Characterization of soluble recombinant CFP-HIS fusion molecules. (A) Protein

profiles on Coomassie blue-stained gels. (B) Anti-HIS mAb-APC coupled to anti-mouse Ig

beads was incubated with HIS-tagged CFP molecules. CFP binding was then detected with an

anti-human CFP-FITC polyclonal Ab that also cross-reacts with mouse CFP.

Table S1. Affinity/avidity of CFP for the NKp46 receptor. Binding studies were performed

with the classical kinetic wizard (as recommended by the manufacturer). Serial dilutions of

soluble analytes (recombinant CFP), ranging from 9 to 600 nM, were injected over the NKp46-

Fc immobilized proteins and allowed to dissociate for 10 minutes before regeneration. The entire

sensorgram sets were fitted with the 1:1 kinetic binding model. Calculated values relative to Fig.

2B are shown.

Fig. S5. Native CFP oligomers bind to NKp46-Fc. Binding expressed as relative units of

indicated CFP oligomers or control buffer over a flow cell immobilized with MoNKp46-Fc or

NKp30-Fc. Data are representative of two independent experiments.

Fig. S6. B12 cell–derived CFP activates NKp46 reporter cells. (A) Cfp transcripts were

quantified by RT-qPCR, in B12, 2sv and NIH-3T3 mouse fibroblasts. Results are normalized

with respect to Gapdh (glyceraldehyde phosphate dehydrogenase) and expressed as the mean

SD in arbitrary units. Data are representative of three independent experiments. (B) NKp46

reporter cells were cocultured with B12 cells in the presence or absence of polyclonal rabbit anti-

mouse CFP (provided by W. Song, University of Pennsylvania) or control serum (R&D

Systems), at a concentration of 30 µg/ml. Cell activation was assessed by evaluating IL-2

concentration in the coculture supernatant in a standard CTLL-2 survival assay. Results are

expressed as the % inhibition relative to control NKp46 reporter cells without antibody. Data are

the means of triplicates ± SD for three independent experiments. Unpaired T test was performed.

Fig. S7. CFP does not bind to T cells. Binding of soluble recombinant HIS-tagged HuCFP to

remaining CD3+ T cells from primary human IL-2-activated bulk NK cells. Data are

representative of two independent experiments. Relative to Fig. 4C.

Fig. S8. CFP stimulation does not induce classical NK cell effector functions. (A-B)

Ncr1WT/WT and Ncr1GFP/GFP NK cells, either resting (A, B) or stimulated for seven days with

5,000 units of IL-2 (A) were incubated with 100 µg/ml of HIS-tagged MoCFP alone or together

with the indicated stimuli, for 4 hours. The cytometric profiles show the frequencies of CD107a

detection and IFN-γ secretion. (C) NK cell preparations enriched in Ncr1WT/WT and Ncr1GFP/GFP

NK cells were left unstimulated or were incubated overnight with 100 µg/ml of HIS-tagged

MoCFP and then with the indicated stimuli for 4 hours. Numbers in quadrants indicate

frequencies. Data are representative of two independent experiments.

Table S2. List of genes modulated upon NKp46 stimulation. Relative to Fig. 5. List of genes

down and upregulated following anti-NKp46 mAb stimulation. Only genes modulated by a 2

fold change are shown.

Table S3. List of genes modulated upon CFP stimulation. Relative to Fig. 5. List of genes

down and upregulated following CFP stimulation. Only genes modulated by a 2 fold change are

shown.

Fig. S9. Impact of anti-NK1.1 and anti-aGM1 treatments on NKp46+ ILC subsets. Flow

cytometry analysis (A) and cell counts (B), to determine the ILCs content in various organs of

WT mice with and without injections of the depleting anti-NK1.1 mAb or anti-aGM1 pAb. (C)

Cytometric profiles of NK1.1 and aGM1 expression on ILCs. n=4 to 6. Data shown are the

pooled results of two independent experiments.

Fig. S10. Impact of anti-aGM1 treatment on Nm infection. WT control mice with and without

anti-aGM1 pAbs treatment were infected i.p. with 2x106 Nm bacteria. Bacterial load after

infection was monitored over time. The experiment was performed once with n=6-10, Two-way

ANOVA, Dunnett correction for multiple testing.

Fig. S11. Therapeutic benefit of CFP delivery for Nm infection control in the absence of

NKp46+ ILCs or NKp46. WT mice with and without injections of the depleting anti-NK1.1

mAb (A) and Ncr1WT/WT and Ncr1GFP/GFP littermates (B, C) were left untreated or treated with

100 µg CFP for 6 hours before i.p. infection with 1x107 (A) or 2x106 (B, C) Nm bacteria. Mouse

survival data are shown in (A, B) and bacterial load 48 hours post infection are shown in (C).

n=6 (A) n=5-8 (B, C). Related to Fig. 8.