Supplementary information - media.nature.com · 1 Supplementary information Small molecule promotes...

1

Supplementary information

Small molecule promotes β-catenin citrullination and

inhibits Wnt signaling in cancer

Yi Qu1,2,3

, Jan Roger Olsen1, Xing Yuan

4, Phil F Cheng

5, Mitchell P. Levesque

5, Karl A.

Brokstad6, Paul S. Hoffman

7, Anne Margrete Oyan

1, Weidong Zhang

3,4*, Karl-Henning

Kalland1, 2, 8

*, Xisong Ke1,3

*

1 Department of Clinical Science, University of Bergen, Bergen, Norway

2 Department of Microbiology, Haukeland University Hospital, Bergen, Norway

3 Institute of Interdisciplinary Integrative Medicine Research, Shanghai University of

Traditional Chinese Medicine, Shanghai, P.R. China

4 Department of Phytochemistry, College of Pharmacy, Second Military Medical University,

Shanghai, P.R. China

5 Department of Dermatology, University of Zürich, University of Zürich Hospital, Zürich,

Switzerland

6 Broegelmann Research Laboratory, Department of Clinical Science, University of Bergen,

Bergen, Norway

7 Department of Medicine, Division of Infectious Diseases and International Health,

University of Virginia School of Medicine, Charlottesville, VA, USA

8 Centre for Cancer Biomarkers, University of Bergen, Bergen, Norway

*Correspondence to: [email protected], [email protected] or [email protected]

Nature Chemical Biology: doi:10.1038/nchembio.2510

mailto:[email protected]



2

Supplementary Results

Supplementary Table 1. Small molecule screening data.

Category Parameter Description

Assay Type of assay Cell based luciferase assay Target Wnt pathway Primary measurement Detection of TOPFlash activity Key reagents Dual-Glo® Luciferase Assay System

(Promega) Assay protocol See online methods Additional comments no

Library Library size 460 drugs Library composition FDA approved drugs Source Selleck Chemicals Additional comments no

Screen Format 96 well plates Concentration(s) tested 10μM, 1%DMSO Plate controls Each plate contained 6BIO only/DMSO

only as positive/negative control Reagent/ compound

dispensing system Manual

Detection instrument and software

biotech synergy H1 and Gene 5 software

Assay validation/QC 6BIO increases TOPFlash activation Correction factors no Normalization Luciferase activity was normalized by

Renilla activity Additional comments no

Post-HTS analysis

Hit criteria TOPFlash activity < 0.5 standard deviations from the mean

Hit rate Approximately 1% Additional assay(s) Fluorescence reporter and FACS

analysis Confirmation of hit purity and

structure Drugs were repurchased (selleckchem) and confirmed by LC-Q-TOF-MS analysis

Additional comments no


3

Supplementary Table 2. Peptide information of PAD2-treated β-catenin.

Calc. Mass Obsrv. Mass1 ± da ± ppm

Start Seq.

End Seq.2 Sequence

Ion Score3 C. I. %4 Modification

8,684,635 8,684,177 -0,0458 -53 643 649 LLHSRNE 49 100

8,705,043 8,704,656 -0,0387 -44 148 155 LATRAIPE 60 100

8,783,573 8,783,162 -0,0411 -47 739 747 MGGHHPGAD 8 99,904

8,783,573 8,783,636 0,0063 7 739 747 MGGHHPGAD 5 99,834

9,455,251 9,454,771 -0,048 -51 156 163 LTKLLNDE 41 100

9,944,042 9,943,515 -0,0527 -53 728 736 ALGMDPMME 6 99,868

11,235,089 11,234,879 -0,021 -19 106 115 GMQIPSTQFD 9 99,932

11,585,062 11,584,481 -0,0581 -50 68 77 QGFSQSFTQE 66 100

12,605,057 12,605,568 0,0511 41 728 738 ALGMDPMMEHE 2 99,643

12,765,007 12,765,471 0,0464 36 728 738 ALGMDPMMEHE 1 99,498 Oxidation (M)[7]

12,836,525 12,836,038 -0,0487 -38 572 583 GCTGALHILARD 8 99,908 Carbamidomethyl (C)[2]

13,296,943 13,296,472 -0,0471 -35 609 620 NIQRVAAGVLCE 35 100 Carbamidomethyl (C)[11]

17,818,527 17,817,732 -0,0795 -45 678 692 LTSSLFRTEPMAWNE 32 100 17,978,477 1,797,764 -0,0837 -47 678 692 LTSSLFRTEPMAWNE 12 99,967 Oxidation (M)[11]

17,988,317 17,987,579 -0,0738 -41 678 692 LTSSLFRTEPMAWNE 32 100 Citrullination (R)[7]

18,178,929 18,178,179 -0,075 -41 18 32 RKAAVSHWQQQSYLD 57 100 Citrullination (R)[1]

21,231,641 21,230,774 -0,0867 -41 164 182 DQVVVNKAAVMVHQLSKKE 38 100

1. Mass value (Da) of the peptide sequence

2. Position of peptide sequence in the complete amino acid sequence

3. Derived for MS/MS (MALDI-TOF/TOF) data

4. Confidence of the peptide sequence ID


4

Supplementary Table 3. Peptide information of NTZ-treated endogenous β-catenin.

Calc. Mass Obsrv. Mass ± da ± ppm Start.Seq. End Seq. Sequence Ion Score

C. I. % Modification

8,043,774 8,043,608 -0,0166 -21 330 335 TYTYEK 35 100

8,405,665 8,405,665 0 0 536 542 LVQLLVR 37 100

8,554,067 855,398 -0,0087 -10 543 549 AHQDTQR 60 100

8,905,345 8,905,282 -0,0063 -7 226 233 EGLLAIFK 44 100

10,835,979 10,835,757 -0,0222 -20 171 180 AAVMVHQLSK 68 100

10,875,929 10,875,928 -0,0001 0 191 200 SPQMVSAIVR 80 100

12,405,514 1,240,515 -0,0364 -29 662 671 MSEDKPQDYK 45 100

13,737,092 13,737,002 -0,009 -7 613 625 VAAGVLCELAQDK 80 100 Carbamidomethyl

17,818,599 17,818,577 -0,0022 -1 550 565 RTSMGGTQQQFVEGVR 105 100 Citrullination (R)[1]

17,868,752 17,868,668 -0,0084 -5 454 469 AGDREDITEPAICALR 64 100 Carbamidomethyl


5

Supplementary Fig. 1. NTZ inhibits Wnt/β-catenin signaling. (a) Representative images of

293FT-7TGC cells treated with 6BIO (1μM) and NTZ (10μM) as indicated for 24 hours.

Scale bar: 20μm. (b) Flow cytometry analysis of 293FT-7TGC cells treated with 6BIO (1μM)

and NTZ at indicated concentrations. The percentages of TCF-GFP positive cells are

indicated. (c) Topflash assay of 293FT cells overexpressing β-catenin lacking the first 47

N-terminal residues (ΔN47) or with mutations of residues Ser45/Thr41/Ser37/Ser33 (4A).

Cells were treated with NTZ at indicated concentrations for 24 hours. Data represent Topflash

activities normalized to Renilla luciferase activities in three independent experiments (mean

±s.d). P-values were determined by a Student t-test, *p<0.05. **p<0.01.


6

Supplementary Fig. 2. CRISPR/Cas9 deletion of CTNNB1. (a) Schematic depiction of

GFP-tagged CRISPR/Cas9 lentiviral vector pL-CRISPR.EFS.GFP (Addgene #57818). (b)

Schematic program of CRISPR/Cas9 based genomic DNA deletion and PCR-based colony

screening. Yellow indicates the DNA sequence encoding the ARM repeats of CTNNB1.Two

sgRNAs (PAM sequences in red) were designed to target upstream and downstream of the

fragment for deletion. To screen deletion colonies, one primer pair matched the internal

region of the deletion (blue arrows) and one primer pair localized external to the deletion (red

arrows). (c) DNA sequencing confirmed the expected deletion at the sgRNA recognition sites

in HCT116 cells overexpressing GFP tagged Cas9 and sgRNAs. (d) PCR confirmed CTNNB1

deletion clones (M11, M32) in HCT116 cells. PCR products with primers detecting the

internal (Int) and external (Ext) regions are indicated with blue and red arrows, respectively.

(e) Immunoblot confirmed β-catenin deletion in clones (M11,M32) of HCT116 cells. Full blot

image for e is shown in Supplementary Figure 21.


7

Supplementary Fig. 3. Clonogenic growth assay of SW480 and HCT116 cells. (a)

Representative images of colony formation of cancer cells overexpressing CTNNB1sgRNAs.

HCT116 and SW480 cells were transfected with GFP-tagged Cas9 sgRNAs for deletion of

CTNNB1, two days later the top 3% GFP positive cells were sorted and plated in low density

for 12 days. Colonies were imaged and quantified by Trypan blue staining and OD590nm in

three independent experiments. (b) Representative images of colony formation of HCT116

clones (M11 and M32) with CTNNB1 deletion. Cells were treated with NTZ at indicated

concentrations for 12 days. The colony growth was quantified by Trypan blue staining and

OD590 nm in three independent experiments. In a and b, P-values were determined by a

Student t-test, *p<0.05. **p<0.01.


8

Supplementary Fig. 4. Evaluation of NTZ in Apc

min/+ mice. (a) Representative

haematoxylin eosin (H&E) and immunohistochemistry staining of β-catenin in intestinal

adenomas of Apcmin/+

mice at 12 weeks. (b) Body weight of mice treated with vehicle or NTZ

at 200mg/kg daily at the indicated weeks, Significant difference was determined by a Student

t-test, n.s = not significant. (c) Representative H&E staining of normal-looking intestine in

mice treated with vehicle or NTZ for 5 weeks. Scale bars: 100μm.


9

Supplementary Fig. 5. NTZ promotes β-catenin degradation. (a) Immunoblotting analysis

of β-catenin in DLD1 and Colo320 cells treated with NTZ at indicated concentrations for 24

hours. (b) Immunoblotting analysis of SW480 cells treated with NTZ at indicated

concentrations for 24 hours. SW480 cells overexpressing FLAG-tagged STAT3 was used as a

control. Full blot images for a-b are shown in Supplementary Figure 21.


10

Supplementary Fig. 6. 2D-DIGE images of the DARTS samples. SW480 protein lysates were

incubated with NTZ (150μM) or equal volumes of DMSO for 1 hour and digested with pronase for 30

minutes before 2D-DIGE analysis. The region of the gel with differentially protected spots is indicated

and enlarged in Figure 3a.


11

Supplementary Fig. 7. PAD2 is the functional target of NTZ. (a) Immunoblots of SW480

control(Ctrl) and cells overexpressing ENOA, HA-tagged RNF115, FLAG-tagged IKZF1 and

PAD2, GFP-tagged CEP89 and mEmerald-tagged CALR. (b) Immunoblots of

active-β-catenin (ABC, non-phospho-Ser33/37/Thr41) in 293FT and SW480 cells with

overexpression of indicated candidate proteins. (c) Representative immunofluorescence

images of SW480 cells transfected with GFP-tagged PAD2 construct. An empty GFP

construct was used as negative control. Scale bars: 20μm. (d) Real-time quantitative RT-PCR

analysis of mRNA expression of indicated genes in SW480 control and PAD2 overexpressing

cells in three independent experiments. (e) Immunoblot of 293FT cells transfected with

FLAG-tagged PAD2, β-catenin-ΔN47 and β-catenin-4A as indicated for 24 hours. (f)

Topflash assay of SW480 cells treated with NTZ (10μM) and transfected with shPAD2

knockdown vector as indicated for 24 hours. Data represent triplicate experiments. (g)

Immunoblot of SW480 cells transfected with shPAD2 knockdown vector and treated with

NTZ (10μM) for 24 hours. Data represent triplicate experiments and the fold change ratios of

β-catenin and PAD2 were indicated. In d and f, P-values were determined by a Student t-test,

*p<0.05. Full blot images for a,b e and g are shown in Supplementary Figure 21.


12

Supplementary Fig. 8. CRISPR/Cas9 deletion of PAD2. (a) Schematic of CRISPR/Cas9

based genomic DNA deletion and PCR-based colony screening. Two sgRNAs (PAM

sequences in red) were designed to target upstream and downstream of the fragment for

deletion. To screen deletion colonies, one primer pair matched the internal region of the

deletion (blue arrows) and one primer pair localized external to the deletion (red arrows). (b)

DNA sequencing confirmed the expected deletion at the sgRNA recognition sites in SW480

cells overexpressing Cas9 and sgRNAs. (c) PCR confirmed deletion PAD2 deletion

clones(M10, M11)of SW480 cells. PCR products with primers detecting the internal (Int) and

external (Ext) regions are indicated with blue and red arrows, respectively. (d)

Immunoblotting confirmed PAD2 deletion in clones of SW480 cells. Full blot image for d is

shown in Supplementary Figure 21.


13

Supplementary Fig. 9. Evaluation of SW480 cells with PAD2 deletion. (a) MTS assay of

wildtype (WT) and PAD2 deleted SW480 cells in time course. (b) Representative images of

SW480 cells that continuously grew for 7 days. Scale bar: 100μm. (c) Representative images

of colony formation of SW480 wildtype (WT) and PAD2 deletion clones (M10, M11) treated

with 10µM NTZ for 12 days. The colony growth was quantified by Trypan blue staining and

OD590 nm in triplicate experiments. In a and c, P-values were determined by a Student t-test,

*p<0.05.


14

Supplementary Fig. 10. NTZ and PAD2 promote protein citrullination. (a) Sequence of

wildtype (WT) and mutant (W348A/C647A) PAD2 with change of residues shown in red.

(b) Immunoblotting analysis of SW480 cells transfected with FLAG-tagged wildtype (W) or

mutant (M) PAD2 for 24 hours. (c) Citrullination analysis of SW480 cells treated with NTZ

(10μM) for indicated times. (d) Citrullination analysis of 293FT and HCT116 cells treated

with NTZ (10μM) for 24 hours. Full blot images for b-d are shown in Supplementary

Figure 22.


15

Supplementary Fig. 11. Prediction of the interaction between NTZ and PAD2. (a) Sequence and counts of detected PAD2 peptides in the spot enriched in NTZ-treated SW480

lysates. (b) Mapping the detected peptides to the protein sequences and domains of PAD2.

The region (aa138-175) with higher frequencies of detected peptides is highlighted with red.

(c) Visualization of the potential NTZ binding region (aa138-175) of the PAD2 holoenzyme

(PDB: 4N2C)(http://www.pymol.org/). Calcium was marked by number 1-6. (d) PAD2

citrullination of β-catenin in the presence of calcium and NTZ at indicated concentrations.

Full blot image for d is shown in Supplementary Figure 22.


http://www.pymol.org/

16

Supplementary Fig. 12. NTZ increases PAD2 level and protein citrullination. (a) NTZ

does not regulate the enzyme activity of PAD2. Recombinant β-catenin (200ng) was

incubated with PAD2 (20ng) and NTZ at indicated concentrations for 1 hour before

immunoblotting with indicated antibodies. (b) Representative immunohistochemistry staining

of PAD2 in adenomas of Apcmin/+

mice treated with vehicle or NTZ. The adenomas and

adjoining normal tissue are indicated as N and T, respectively. The ratio of PAD2 staining

levels between normal tissue and tumor tissue was quantified. Scale bars: 100μm.

(c).Immunoblot analysis of protein citrullination in formalin-fixed, paraffin-embedded (FFPE)

small intestines in vehicle or NTZ-treated Apcmin/+

mice (5 mice per group). Data represent

triplicate experiments, protein citrullination was quantified and normalized to the GAPDH

level. Lysates were blotted with antibodies against citrulline (anti-Cit) and GAPDH separately.

Full blot images for a and c are shown in Supplementary Figure 22.


17

Supplementary Fig. 13. Tizoxanide (TIZ) regulates PAD2 stabilization. SW480 cells were

treated with protein synthesis inhibitor cycloheximide (CHX) together with DMSO (D),

10μM TIZ (T) or 10μM NTZ (N) for indicated times before immunoblotting. Full blot image

is shown in Supplementary Figure 23.


18

Supplementary Fig. 14. NTZ hydrolyzes and metabolizes to tizoxanide. (a) Total ion

chromatographies (TICs) of NTZ samples dissolved in DMSO (10μM, 0 minute) or water

(10μM) for indicated times. Graphs are representative of 3 independent experiments. (b) Mass

spectrum of the peaks in a with retention times of 6.851 and 7.133 minutes indicates NTZ

(molecular weight 306 Da) and its metabolite TIZ (molecular weight 264 Da), respectively. (c)

NTZ and TIZ percentages in NTZ samples dissolved in DMSO (10μM, 0 minute) or water

(10μM) for indicated times. Data are representative of 3 independent experiments. *p<0.05.

(d) TICs of SW480 cells treated with DMSO or NTZ (10μM) for 30 minutes. (e) Mass

spectrum of the peak in d with retention time of 5.894 minutes indicates that only TIZ

(molecule weight 264 Da) was detected in SW480 cell lysate.


19

Supplementary Fig. 15. Examination of intracellular calcium and protein citrullination

in NTZ-treated SW480 cells. (a) SW480 cells loaded with Fluo-3 AM were incubated with

DMSO, ionomycin, CaCl2, NTZ or TIZ at indicated concentrations for 10 minutes before

flow cytometry. Fluo-3 emission was measured and normalized to the mean value of baseline

emission (DMSO sample) to determine intracellular Ca2+

levels following indicated treatment.

Data represent results in triplicate experiments (b) SW480 cells were treated with DMSO,

ionomycin, CaCl2, NTZ or TIZ at indicated concentrations for 24 hours before examination of

protein citrullination. (c) SW480 cells overexpressing PAD2 were treated with protein

synthesis inhibitor cycloheximide (CHX) and 1µM ionomycin plus 200µM CaCl2 at the

indicated times before immunoblotting. Full blot images for b and c are shown in

Supplementary Figure 23.


20

Supplementary Fig. 16. NTZ preferentially affects PAD2 in SW480 cells. (a) SW480 cells

were transfected with FLAG-tagged PAD1, PAD2, PAD3 and PAD4, respectively, for 24

hours followed by treatment with 10µM NTZ and 100ng/ml CHX at indicated times before

immunoblotting. (b) SW480 cells were transfected with FLAG-tagged PAD1, PAD2, PAD3

and PAD4, respectively, for 24 hours followed by treatment with 10µM NTZ for 4 hours

before immunoblotting. The ratio of protein citrullination in NTZ-treated cells was quantified

and normalized by the corresponding control cells. Full blot images for a and b are shown in

Supplementary Figure 23.


21


22

Supplementary Fig. 17. PAD2 inhibits the growth of Wnt activated cells. (a) Representative images of colony formation of SW480 cells with overexpression of wildtype

and mutant PAD2 for 2 weeks. The colony growth was quantified by crystal violet staining

and OD590 nm in 3 independent experiments. *p<0.05. (b) Representative images of colony

formation of HCT116 CTNNB1 deletion clones (M11, M32) overxpressing PAD2 or control

vector for 12 days. The colony growth was quantified by Trypan blue staining and OD590 nm

in 3 independent experiments. (c) Quantitation of sphere formation. Ctnnb1lox(ex3)/lox(ex3)

organoids with overexpression of PAD2 or control vector were dissociated to single cells and

seeded at equal densities, organoids with diameter more than 100 µm were counted 3 days

later. Data show percentage of spheres overexpressing PAD2compared to control in triplicate

experiments. (d) Representative images of the growth of Ctnnb1lox(ex3)/lox(ex3)

organoids with

overexpression of PAD2 or control vector. Scan bar: 200µm. (e) Representative images of

Ctnnb1lox(ex3)/lox(ex3)

organoids with a subpopulation of cells overexpressing GFP-tagged PAD2.

Organoids were grown in complete medium (left) and changed to medium without R-spondin

for one and two days (right). Scale bar: 200µm.


23

Supplementary Fig. 18. Full-size immunoblots of Figure 2a, 2b, 2c, 2d, 2f and Figure

3c.


24

Supplementary Fig. 19. Full-size immunoblots of Figure 4a, 4c, 4d, 4e, Figure 4f and 4h.


25

Supplementary Fig. 20. Full-size immunoblots of Figure 5a, 5b, 5c and 5d.


26

Supplementary Fig. 21. Full-size immunoblots of Supplementary Figure 2e, Figure 5a,

5b, Figure 7a, 7b, 7e, 7g and Figure 8d.


27

Supplementary Fig. 22. Full-size immunoblots of Supplementary Figure 10b, 10c, 10d,

Figure 11d, Figure 12a and 12c.


28

Supplementary Fig. 23. Full-size immunoblots of Supplementary Figure 13, Figure 15b,

15c, Figure 16a, and 16b.


Supplementary information - media.nature.com · 1 Supplementary information Small molecule promotes...

Documents

Transcript of Supplementary information - media.nature.com · 1 Supplementary information Small molecule promotes...