38  

SUPPLEMENTAL FIGURE LEGENDS 966  

Supplemental Figure 1 Siglec-7 and -9 ligand expression on tumor cell lines and 967  

primary tumors assessed by flow cytometry (A,B) or confocal microscopy (C,D). (A) 968  

Histograms depict differential staining for ligands of Siglec-3, -7 and -9 on melanoma 969  

line A375, head and neck squamous cell carcinoma (HNSCC) cell line LAU2106, and 970  

colon adenocarcinoma line HCT116. Grey-shaded histograms represent control 971  

staining with secondary antibody. (B) Geometric mean fluorescence intensity (GMFI) 972  

ratios of Siglec-7 (open symbols) and Siglec-9 (filled symbols) ligand expression on 973  

CLL (n=3) and AML (n=3) leukemia cells. (C,D) Paraffin-embedded tissue biopsy 974  

sections of malignant melanoma lesions in dermal skin layers (C) or healthy dermal 975  

skin (D), co-stained for the melanoma marker Melan-A and Siglec-7 or Siglec-9 976  

ligands, respectively. Scale bar, 50 µm (C,D). Data are representative of at least 977  

three (A) independent experiments, and of six individual subjects (C,D). 978  

979  

Supplemental Figure 2. Neuraminidase treatment has no effect on 721.221 cells. 980  

Cytotoxicity of isolated peripheral blood NK cells from healthy donors (n=5) against 981  

721.221 as assessed in a 51Cr-release assay, without (open circles) or with 982  

neuraminidase-treatment (filled circles) of target cells. Cytotoxicity was evaluated at 983  

the indicated effector to target (E/T) ratios. 984  

 985  

Supplemental Figure 3. Flow cytometric intracellular cytokine measurement in NK 986  

cells exposed to neuraminidase treated or untreated K562 cells (10/1 E/T ratio, n=7). 987  

*P<0.05 and **P<0.005 (Student’s t-test). 988  

       

  39  

Supplemental Figure 4. Circulating NK cells of (hu)-NSG mice express high levels 989  

of Siglec-7. Flow cytometry of NK cells from peripheral blood of hu-NSG mice (n=13) 990  

for expression of human Siglec-7 and Siglec-9. Frequency (left of each pair) and 991  

representative histograms (right of each pair) are shown. Specific staining (black 992  

line), and isotype-matched control (shaded). 993  

994  

Supplemental Figure 5. Siglec-7 and Siglec-9 ligands on K562 cells are not 995  

reexpressed within 24 hours following neuraminidase treatment. Expression of 996  

Siglec-7 and Siglec-9 ligands on K562 cells before (t=0) or after neuraminidase-997  

treatment, as assessed by flow cytometry over a period of 24 hours. Values are 998  

expressed as GMFI ratio compared to control. 999  

1000  

Supplemental Figure 6. The combination of Siglec-7 and -9 Fab fragments has no 1001  

enhanced effect on NK cell cytotoxicity against K562 target cells, as assessed by a 1002  

51Cr assay at an E/T ratio of 10/1 (n=7). **P<0.005 (Student’s t-test). 1003  

1004  

Supplemental Figure 7. Effects of targeting Siglec-7 and -9 on NK cell cytotoxicity, 1005  

survival and proliferation. (A,B) Cytotoxicity of isolated peripheral blood NK cells from 1006  

healthy donors against the K562 cell line assessed in a 51Cr-release assay in the 1007  

presence of indicated mAbs (n=2 for Clone 191240, n=10 for Clone E10-286). E/T 1008  

ratio 20/1. (B) Cytotoxicity of sorted human CD56dim CD16+ Siglec-9- NK cells against 1009  

the K562 cell line assessed in a 51Cr-release assay in the presence of E10-286 mAb 1010  

or an isotype control (n=3). (C,D) Flow cytometric analysis of NK cell survival (C) and 1011  

       

  40  

proliferation (D). (C) AnnexinV-GFP/PI staining of 24-hour cultures in the presence of 1012  

indicated mAbs at 10 µg/ml. Similar results were achieved at lower mAb 1013  

concentrations. (D) CFSE dilution upon stimulation with IL-2 and IL-15 for 9 days and 1014  

culture in the presence of anti-Siglec-7 mAb at 4 ug/ml, anti-Siglec-9 mAb at 3 ug/ml, 1015  

or isotype-matched control mAbs. Data are representative of four (C) or three (D) 1016  

independent experiments. 1017  

1018  

Supplemental Figure 8. Expression of siglecs on human NK cells. (A) Flow 1019  

cytometry of surface siglec receptor expression on peripheral blood NK cells of adult 1020  

healthy donors with and without neuraminidase treatment. Specific staining (black 1021  

line), neuraminidase treated cells (dashed line) and isotype-matched control 1022  

(shaded) (n=3). (B) Representative examples of Siglec-7 and Siglec-9 expression on 1023  

CD56bright CD16dim/- and CD56dim CD16+ NK cell subsets with or without unmasking 1024  

by neuraminidase treatment (removal of sialic acid residues). (C,D) Siglec-9 1025  

expression on NK cells  in presence of NK cell-relevant cytokines as assessed by flow 1026  

cytometry. C) Geometric mean fluorescence intensity (GMFI) ratios of gated Siglec-1027  

9+ NK cells from total NK cell cultures. D) Histograms are representative of freshly 1028  

fluorescence-activated cell sorted NK cell Siglec-9+ or Siglec-9- CD56dim NK cell 1029  

subsets (dotted line), isotype control (shaded), 24 (dashed line), 48 (solid line) or 120 1030  

(long dashed line) hour-cultures. Data are representative of at least three (A,D), six 1031  

(C), or 14 (B), experiments. 1032  

1033  

       

  41  

Supplemental Figure 9. (A,B) Flow cytometry of Siglec-9- (open circles) and Siglec-1034  

9+ (filled circles) gated CD56dim CD16+ NK cells from healthy donors (n=10-19) for 1035  

surface expression of NKG2D and NCRs. 1036  

1037  

Supplemental Figure 10. Frequency and Siglec-7 expression on peripheral blood 1038  

NK cells in cancer. (A) NK cell numbers in the peripheral blood of melanoma and 1039  

colon adenocarcinoma (CoACa) patients. (B) Cytotoxicity of NK cells from healthy 1040  

donors (HD) and melanoma patients, as assessed in a 51Cr release assay of K562 1041  

target cells at an E/T ratio of 20/1 (n=7). (C) Expression of Siglec-7 on peripheral 1042  

blood NK cells of healthy donors (HDs) and cancer patients. Shown are frequency 1043  

and GMFI ratio of specific staining to isotype-matched control, without (white bars) or 1044  

with (grey bars) neuraminidase treatment. *P<0.05, and **P<0.005 (Student’s t-test). 1045  

Supplemental Figure 1 Siglec-7 and -9 ligand expression on tumor cell lines and primary tumors assessed by flow cytometry (A,B) or confocal microscopy (C,D). (A) Histograms depict differential staining for ligands of Siglec-3, -7 and -9 on melanoma line A375, head and neck squamous cell carcinoma (HNSCC) cell line LAU2106, and colon adenocarcinoma line HCT116. Grey-shaded histograms represent control staining with secondary antibody. (B) Geometric mean fluorescence intensity (GMFI) ratios of Siglec-7 (open symbols) and Siglec-9 (filled symbols) ligand expression on CLL (n=3) and AML (n=3) leukemia cells. (C,D) Paraffin-embedded tissue biopsy sections of malignant melanoma lesions in dermal skin layers (C) or healthy dermal skin (D), co-stained for the melanoma marker Melan-A and Siglec-7 or Siglec-9 ligands, respectively. Scale bar, 50 mm (C,D). Data are representative of at least three (A) independent experiments, and of six individual subjects (C,D).

Siglec-3L Siglec-7L Siglec-9L

A375

LAU2106

HTC116

A

B

AML CLL 10

100

1000

10000

GM

FI R

atio

Siglec-7L Melan A

Siglec-9L Melan A Overlay

Overlay

C

CLLAML

10

100

1000

10000

Data 1

Siglec-7 LigandSiglec-9 Ligand

Siglec-9L Melan A Overlay

Siglec-7L Melan A Overlay

D

Neuraminidase

No neuraminidase

Effector to target ratio

Spe

cific

lysi

s (%

)

Supplemental Figure 2. Neuraminidase treatment has no effect on 721.221 cells. Cytotoxicity of isolated peripheral blood NK cells from healthy donors (n=5) against 721.221 as assessed in a 51Cr-release assay, without (open circles) or with neuraminidase-treatment (filled circles) of target cells. Cytotoxicity was evaluated at the indicated effector to target (E/T) ratios.

Supplemental Figure 3. Flow cytometric intracellular cytokine measurement in NK cells exposed to neuraminidase treated or untreated K562 cells (10/1 E/T ratio, n=7). *P<0.05 and **P<0.005 (Student’s t-test).

0

10

20

30

40

50

IFN-γ MIP-1β TNF-α

Pos

itive

cel

ls (%

)

**

**

*

NK NK + K562 NK + sialidase treated K562

0

10

20

30

40

50

Supplemental Figure 4. Circulating NK cells of (hu)-NSG mice express high levels of Siglec-7. Flow cytometry of NK cells from peripheral blood of hu-NSG mice (n=13) for expression of human Siglec-7 and Siglec-9. Frequency (left of each pair) and representative histograms (right of each pair) are shown. Specific staining (black line), and isotype-matched control (shaded).

Time (h) Time (h)

Rat

io G

MFI

Rat

io G

MFI

Siglec-7 ligand Siglec-9 ligand

Supplemental Figure 5. Siglec-7 and Siglec-9 ligands on K562 cells are not reexpressed within 24 hours following neuraminidase treatment. Expression of Siglec-7 and Siglec-9 ligands on K562 cells before (t=0) or after neuraminidase-treatment, as assessed by flow cytometry over a period of 24 hours. Values are expressed as GMFI ratio compared to control.

Standar

t

Fab co

ntrol (

mix)

Fab co

ntrol

Siglec-7

Siglec-9

Siglec-7

/ -90

25

50

75

100

Data 12

Spe

cific

lysi

s (%

)

100

75

50

25

0

n.s.!

Supplemental Figure 6. The combination of Siglec-7 and -9 Fab fragments has no enhanced effect on NK cell cytotoxicity against K562 target cells, as assessed by a 51Cr assay at an E/T ratio of 10/1 (n=7). **P<0.005 (Student’s t-test).

n.s.!**!

Inhibition of specific lysis (%)

100 80 60 40 20 0

Anti-Siglec-9 (Clone E10-286)

Isotype control

Anti-Siglec-9 (Clone 191240)

Isotype control

Inhibition of specific lysis (%)

Anti-Siglec-9

Isotype control

100 80 60 40 20 0

A B

C D

AnnexinV

PI

Anti-Siglec-7 (IgG2b)

Isotype Control(IgG2b)

Anti-Siglec-9 (IgG1)

Isotype Control(IgG1)

CFSE

Cou

nts

Anti-Siglec-7 (IgG2b)

Isotype Control(IgG2b)

Anti-Siglec-7 (IgG1)

Isotype Control(IgG1)

Supplemental Figure 7. Effects of targeting Siglec-7 and -9 on NK cell cytotoxicity, survival and proliferation. (A,B) Cytotoxicity of isolated peripheral blood NK cells from healthy donors against the K562 cell line assessed in a 51Cr-release assay in the presence of indicated mAbs (n=2 for Clone 191240, n=10 for Clone E10-286). E/T ratio 20/1. (B) Cytotoxicity of sorted human CD56dim CD16+ Siglec-9- NK cells against the K562 cell line assessed in a 51Cr-release assay in the presence of E10-286 mAb or an isotype control (n=3). (C,D) Flow cytometric analysis of NK cell survival (C) and proliferation (D). (C) AnnexinV-GFP/PI staining of 24-hour cultures in the presence of indicated mAbs at 10 µg/ml. Similar results were achieved at lower mAb concentrations. (D) CFSE dilution upon stimulation with IL-2 and IL-15 for 9 days and culture in the presence of anti-Siglec-7 mAb at 4 ug/ml, anti-Siglec-9 mAb at 3 ug/ml, or isotype-matched control mAbs. Data are representative of four (C) or three (D) independent experiments.

Supplemental Figure 8. Expression of siglecs on human NK cells. (A) Flow cytometry of surface siglec receptor expression on peripheral blood NK cells of adult healthy donors with and without neuraminidase treatment. Specific staining (black line), neuraminidase treated cells (dashed line) and isotype-matched control (shaded) (n=3). (B) Representative examples of Siglec-7 and Siglec-9 expression on CD56bright CD16dim/- and CD56dim CD16+ NK cell subsets with or without unmasking by neuraminidase treatment (removal of sialic acid residues). (C,D) Siglec-9 expression on NK cells in presence of NK cell-relevant cytokines as assessed by flow cytometry. C) Geometric mean fluorescence intensity (GMFI) ratios of gated Siglec-9+ NK cells from total NK cell cultures. D) Histograms are representative of freshly fluorescence-activated cell sorted NK cell Siglec-9+ or Siglec-9- CD56dim NK cell subsets (dotted line), isotype control (shaded), 24 (dashed line), 48 (solid line) or 120 (long dashed line) hour-cultures. Data are representative of at least three (A,D), six (C), or 14 (B), experiments.

A

Siglec-7 Siglec-8 Siglec-9 Siglec-10

Siglec-1 Siglec-2 Siglec-3 Siglec-5

Cou

nts

B CD56bright CD16dim / - CD56dim CD16+

Siglec-7

+ Neuraminidase

46.5% 2.8

21.7% 1.9

2.0% 1

0.0 % 1

99.9% 387

99.7% 258

98.8% 99

98.5% 54

Cou

nts

+ Neuraminidase

Siglec-9

C

Siglec-9 Siglec-9

Medium

or IL-15

Medium or

IL-12 + IL-18

Sig

lec-

9 (G

MFI

ratio

)

0 24 48 120

0

1

2

3

4

IL-15

1

2

3

4

24 48 120

Time (h)

unstimulatedIL-12 IL-18IL-15

unstimulatedIL-12 IL-18IL-15

unstimulatedIL-12 IL-18IL-15unstimulated IL-12 + IL-18 IL-15

D CD56dim Siglec-9- CD56dim Siglec-9+

Cou

nts

Pos

itive

cel

ls (%

) 100

80

60

40

20

0

CD56dim Siglec-9- CD56dim Siglec-9+

Rat

io G

MFI

100

80

60

40

20

0

CD56dim Siglec-9- CD56dim Siglec-9+

A B

Supplemental Figure 9. (A,B) Flow cytometry of Siglec-9- (open circles) and Siglec-9+ (filled circles) gated CD56dim CD16+ NK cells from healthy donors (n=10-19) for surface expression of NKG2D and NCRs.

No neuraminidase

Neuraminidase

No neuraminidase

Neuraminidase

A

C

Supplemental Figure 10. Frequency and Siglec-7 expression on peripheral blood NK cells in cancer. (A) NK cell numbers in the peripheral blood of melanoma and colon adenocarcinoma (CoACa) patients. (B) Cytotoxicity of NK cells from healthy donors (HD) and melanoma patients, as assessed in a 51Cr release assay of K562 target cells at an E/T ratio of 20/1 (n=7). (C) Expression of Siglec-7 on peripheral blood NK cells of healthy donors (HDs) and cancer patients. Shown are frequency and GMFI ratio of specific staining to isotype-matched control, without (white bars) or with (grey bars) neuraminidase treatment. *P<0.05, and **P<0.005 (Student’s t-test).

total NK

CD56 dim

CD56 bright

0

5

10

15

20100

200

300

400

CD56 bright per ul blood MelanomaCoACa

Total NK cell population

CD56dim NK cell population

CD56bright NK cell population

0

5

10

15

20 100

200

300

400

# C

ells

/ µl

blo

od

HD Melanoma CoACa

n.s. n.s. n.s. n.s.

n.s.

80

60

40

20

0

Spe

cific

lysi

s (%

)

HD Melanoma

B Chromium 20/1 HDs vs Mel Pat n=15 HDs, n=7 Pat

HDs Melanoma0

20

40

60

80

% S

pecif

ic lys

is

n.s.