Niche-independent high-purity cultures of Lgr5 intestinal ......Niche-independent high-purity cultures of Lgr5+ intestinal stem cells and their progeny Xiaolei Yin, Henner F Farin,

Niche-independent high-purity cultures of Lgr5+ intestinal stem cells and their progeny Xiaolei Yin, Henner F Farin, Johan H van Es, Hans Clevers, Robert Langer & Jeffrey M Karp

NATURE METHODS

Supplementary Figure 1 Cell growth and GFP expression as a function of the culture conditionsSupplementary Figure 2 Single Lgr5-GFP cell cultureSupplementary Figure 3 Colon, stomach and human small intestinal cell cultureSupplementary Figure 4 Differentiation of intestinal stem cellsSupplementary Figure 5 Gene Set Enrichment Analysis (GSEA) of organoids Supplementary Figure 6 Exploring the mechanism of action for CHIR and VPASupplementary Figure 7 Mechanism of VPASupplementary Table 1 Summary of growth factors and small molecules used in this studySupplementary Results Towards elucidating the mechanism of CHIR and VPA

Nature Methods: doi:10.1038/nmeth.2737

1

Supplementary Figure 1

Supplementary Figure 1 | Cell growth and GFP expression as a function of the culture conditions. (a) Mosaic expression of Lgr5-GFP in vivo. Small intestine was harvested from Lgr5-GFP mice and directly imaged under fluorescence microscopy. While all areas of the small intestine were covered by crypts containing Lgr5+ stem cells, approximately half of these crypts contained GFP+ cells. (b) Colony numbers and (c) live single cell numbers from triplicate wells were counted at each time point. Error bars indicate S.D. of triplicate wells. (d) FACS sorting of freshly isolated single Lgr5-GFP+ cells. GFPhigh single cell population was collected. Representative FACS analysis showing the gating strategy to define GFP+ cell population. Freshly isolated single cells from crypts showed two distinct GFPhigh and GFPlow populations while cultured cells did not show discriminated GFPhigh and GFPlow populations. ENR-CV cultured cells showed a single GFPhigh population, here all GFP+ cells were gated for analysis. Note that the GFP– population includes stem cell that lack GFP reporter expression (i.e. GFP silenced Lgr5+ stem cells). GFPneg cells were largely absent following culture of sorted single Lgr5-GFP (See Fig. 1d,e). A total of 10,000 live cells were analyzed for each sample. (e) Growth factor requirement for self-renewal of Lgr5+ stem cells in the CV culture condition. Crypts were cultured for 6 days in the presence of CHIR and VPA, with EGF, Noggin, R-spondin 1 and their combinations, as indicated. E: EGF (50 ng/ml); N: Noggin (100 ng/ml); R: R-spondin 1 (500 ng/ml); C: CHIR (3 µM); V: VPA (1 mM). Error bars indicate S.D. or triplicate wells. Experiments were performed three times using different animals (n = 3) with similar results. (f)Crypts cultured for 6 days in multiple conditions as indicated. GFP and brightfield images are shown.

b

ENRENR-V

ENR-C

ENR-CV

0

50

100

150

200

250

300

350Day0Day2Day4Day6Day8Day10

No.

of c

olon

ies

0 2 4 6 8 103.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5ENRENR-VENR-CENR-CV

Days

No.

of c

ells

(Log

10)

ca

f

E-CVN-C

VR-C

VEN-C

VER-C

V

NR-CV

ENR-CV

0

10

20

30

40

50

GFP

+ pe

rcen

tage

E-CVN-C

VR-C

VEN-C

VER-C

V

NR-CV

ENR-CV

0

100

200

300

400

500

600

700

No.

of c

ells

(X1,

000)

E-CVN-C

VR-C

VEN-C

VER-C

V

NR-CV

ENR-CV

0123456789

Relativ

e G

FP in

tens

ity (X

1,00

0)e

ENR-CV (C: 3 µM) EN-CV (C: 3 µM) EN-CV (C: 6 µM)

104d

ENR

GFP+GFP+

Single Cells from Crypts

GFPhigh

GFPlow

ENR-CV

GFP+

Wild-Type Lgr5-GFP

GFPhigh

FSC

GFP

101

102

103

100

0 64 128 192 2560 64 128 192 2560 64 128 192 2560 64 128 192 256 0 64 128 192 256


2


Supplementary Figure 2 | Single Lgr5-GFP cell culture. (a) Single isolated Lgr5-GFP+ cells cultured for 9 days in CV con-dition. (b) 1500 FACS sorted single Lgr5+ cells were cultured in Matrigel under conditions as indicated. Representative images from day 7 cultures are shown and (c) quantification of colony numbers. Error bars indicate S.D. of triplicate wells. (d) Sorted Single Lgr5+ stem cells were seeded in 48-well plates. Viable cell numbers were quantified at 12 h after plat-ing. Colony numbers were counted at day 7 and colony-forming efficiency was quantified based on viable cell numbers. V: VPA; C: CHIR; W: Wnt3a at 100 ng/ml. Error bars indicate S.D. of triplicate wells. Experiments were performed three times using different animals (n = 3) and showed similar results. (e) Morphology and GFP images of Lgr5+ cells cultured in the CV condition at passage 24 (over 4 months) (top panel), and in spontaneous differentiation condition by transfer CV-p24 cells to ENR condition (bottom panel). (f) Representative image showing how colony-forming efficiency of single Lgr5+ cells was assessed. The colonies and dead single cells within Matrigel were fixed, air dried (to pellet Matrigel into a single layer) and stained with DAPI. The number of colonies (arrows) and dead cells (arrow heads) were numerated and colony-forming efficiency (CFE) was calculated using: CFE = Number of colonies / (Number of colonies + Number of dead cells). (g) Colony-forming efficiency of single Lgr5+ cells cultured in the CV condition at passage 6 and passage 24. (h) qPCR analysis of Lgr5+ cells cultured in the CV condition at passage 0, 6 and 24. CV-p6-Diff and CV-p24-Diff represent sponta-neous differentiation conditions by transferring cells from the CV condition to ENR condition. (i) FACS sorting of Lgr5-GFPhigh and GFPlow cells. (j) FACS analysis of GFP expression of sorted single GFPhigh and GFPlow cells after 9 days in culture, and (k, l) quantification of GFP expression. (m) Quantification of colony numbers formed by seeding 2,000 single Lgr5high or Lgr5low cells in ENR-C and ENR-CV conditions. In k–m, Error bars indicate S.D. of duplicate wells. Scale bars, (a) 200 µm, (e, f) 100 µm and (b) 1 mm. ***P<0.001; **P<0.01; *P<0.05; NS P>0.05.

ENRENR-V

ENR-C

ENR-W

ENR-CV

0

20

40

60

80

100

Colo

ny-fo

rmin

g e�

cien

cy (%

of l

ive

cells

)

***

NS

ENRENR-V

ENR-C

ENR-CV

0

100

200

300

400

500

600

Num

ber o

f col

onie

s (p

er 1

500

sort

ed c

ells

)

NS

***

***

a b c dENR ENR-V

ENR-CVENR-C

j

k l m

iLgr5-GFP Sorting

GFPhigh

GFPlow

FSC-H

Lgr5

-GFP

101

102

103

104

100

0 64 128 192 256

ENR-CV

Sort

ed s

ingl

e Lg

r5-G

FP+ c

ells

ENR-C ENR-CV0

20

40

60

80

100

GFP

+ p

erce

ntag

e GFPhigh

GFPlow

*

NS

ENR-C ENR-CV0

2000

4000

6000

Rela

tive

GFP

inte

nsity GFPhigh

GFPlow*

NS

0

100

200

300

400

500

ENR-C ENR-CV

GFPhigh

GFPlow

Num

ber o

f col

onie

s

**

***

ENR-CGFP+

ENR-CGFP+

FSC

GFP

GFP+

ENR-CVGFP+

ENR-CVGFP+

GFPhigh GFPlow

101

102

103

104

100

0 64 128 192 2560 64 128 192 2560 64 128 192 2560 64 128 192 256

0

10

20

30

40

50

Col

ony-

form

ing

effic

ienc

y (%

of t

otal

cel

ls)

CV-p24

CV-p6

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Rela

tive

mRN

A ex

pres

sion

ENR-p0

CV-p0

CV-p6

CV-p6-D

i�

CV-p24

CV-p24-D

i�

e f g hENR-CV-p24 DAPI

CV-p24-Di�

AlpiMuc2ChgALyz1Lgr5


3


Supplementary Figure 3 | Colon, stomach and human small intestinal cell culture. (a) Lgr5-GFP and brightfield images of stomach organoids cultured for 6 days in conditions as indicated. F: FGF10, W: Wnt3a, g: gastrin. (b, c) FACS analysis of GFP expression of stomach cultures. (d) Lgr5-GFP and brightfield images of colonic organoids cultured for 6 days in conditions as indicated. (e, f) FACS analysis of GFP expression of the colonic culture. (g) qPCR analysis of marker expression of cultured colonic organoids in multiple conditions. Error bars indicate S.D., n = 3. (h) Representative image of freshly isolated human small intestinal crypts cultured in multiple conditions. W: Wnt3a, Ni: Nicotinamide, A: A83-01, S: SB202190, P: PGE2. Titles in bold indicate published culture conditions1, 2 with gastrin in basal media. (i) Quantification of colony numbers. ENR was added in all conditions. Error bars indicate S.D. of triplicate wells. (j) Representative image of human small intestinal organ-oids at passage 2. (k) qPCR analysis of marker gene expression of human intestinal crypts cultured in multiple conditions. Error bars indicate S.D., n = 3. All scale bars, 200 µm.

a

d

k

h

0

100

200

300

400

500N

umbe

r of C

olon

ies

WW-N

i

W-Ni-A-S

W-Ni-A-S-P C CV

C-V-N

iC-Tu

C-Tu-Ni

i

050

100

150

Cou

nt

102 103 104 105

ENR-CV

GFP

050

100

150

Cou

nt

ENR-C

050

100

150

Cou

nt

ENR-W

Lgr5

GFP

-IRES

-cre

ERT

colo

n

ENR-W ENR-CVENR-C

Lgr5

GFP

-IRES

-cre

ERT

stom

ach

ENR-FWg ENR-FWg+VPA b

e

Hum

an c

rypt

s

ENR-W-Ni-A-S-P ENR-CV ENR-CV-Ni

ENR-CV-NiENR-W-Ni-A-S-P

Hum

an c

rypt

s

j

g

Lgr5

-GFP

104103102101100

FL3-H

0

20

40

60

80

Control

+VPA

GFP

+ p

erce

ntag

e

c

W C CV0

10

20

30

40

50

GFP

+ p

erce

ntag

e

f

0

0.5

1.0

1.5

0

2

4

6

8ENR-WENR-CENR-CV

Rela

tive

mRN

A e

xpre

ssio

n

Alpi Muc2 ChgA Lyz Lgr5

LGR5

0

5

10

15

CC-

VC-

V-N

iC-

TuC-

Tu-N

i

W-N

i-AS

W-N

i-AS-P

Cryp

t

Rela

tive

mRN

A e

xpre

ssio

n

ALPI

1

2

3

0

CC-

VC-

V-N

iC-

TuC-

Tu-N

i

W-N

i-AS

W-N

i-AS-P

Cryp

t

Rela

tive

mRN

A e

xpre

ssio

n

MUC2

0

2

4

6

CC-

VC-

V-N

iC-

TuC-

Tu-N

i

W-N

i-AS

W-N

i-AS-P

Cryp

t

Rela

tive

mRN

A e

xpre

ssio

n

CHGA

0

2

4

6

8

10

12

CC-

VC-

V-N

iC-

TuC-

Tu-N

i

W-N

i-AS

W-N

i-AS-P

Cryp

t

Rela

tive

mRN

A e

xpre

ssio

n

LYZ

0.0

0.2

0.4

0.6

0.8

1.0

1.2

CC-

VC-

V-N

iC-

TuC-

Tu-N

i

W-N

i-AS

W-N

i-AS-P

Cryp

t

Rela

tive

mRN

A e

xpre

ssio

n

103

102

101

100

104

103

102

101

100


4


CD-Muc2/DAPIID-Muc2/DAPI

ENR-Alp

IV-Alp

ENR-ChgA/DAPI

ChgA

ENR-Muc2/GFP/DAPI

cENR

ENR-CV

ENR-CD

ENR-IV

ENR-ID

6 days

Transfer

ENR-CVENR

......

ENR-CV

ENR-ID

a

b

d

f

Ce

IV-Alp

Supplementary Figure 4 | Differentiation of intestinal stem cells. (a) Staining of organoids cultured in ENR condition. (left) Alp staining of enterocytes, the organoid was cut open under a dissecting microscope by using a sharp blade and the luminal content was removed before staining. (middle) Muc2 staining of goblet cells (arrows) as well as mucin secreted by goblet cells, and (right) ChgA staining of enteroendocrine cells. GFP+ cells indicate Lgr5+ stem cells. (b) Scheme of dif-ferentiation protocol. Single Lgr5+ stem cells were cultured in the CV condition for 4–6 days to form colonies. Cell colonies were then harvested, washed, embedded within fresh Matrigel and cultured under multiple conditions. (c) Morphology of cell colonies transferred from the CV condition to the ENR condition and cultured for 4 days (upper panels). Colonies continuously cultured in the CV condition are shown as a control (lower panels). (d) Morphology of differentiated cells with low and high magnification images for each condition. Note the clear change in morphology for most cells in the CD and ID conditions, which reflects formation of Paneth cells and goblet cells, respectively. (e) Alp staining of colonies cultured in IV condition. Apical (left) and homogeneous (right) staining of Alp are shown. (f) Muc2 staining of colonies cultured in ID and CD conditions. All scale bars, 50 µm. For images from c–f, cells cultured in the CV condition at passage number 5 were used.


5


Supplementary Figure 5 | Gene Set Enrichment Analysis (GSEA) of organoids. Ranked gene expression changes (average of duplicate microarray experiments) in (left) organoids (ENR) treated for 6 days with CHIR, (middle) VPA, or (right) CHIR + VPA are shown on the x-axes. Up- and down-regulated genes are indicated with red and blue colors, respectively. Input gene sets were generated from public data sets: ‘Stem cell genes’ were defined as the 200 most differentially expressed probes in Lgr5-GFP high stem cells (vs. Lgr5-GFP low daughter cells; GEO data set GSE23672) and ‘Paneth cells genes’ (from sorted Paneth cells vs. Lgr5-GFP stem cells, GSE25109). The 200 most down-regulated probes in organoids 1 day following R-spondin withdrawal (using data set GSE28265) were defined as ‘Wnt-target genes’. For analysis of ‘proliferation genes’ the 44 experimentally verified myc target genes (Yu et al., 20053) were used as input. Position of marker gene sets is indi-cated with black stripes, the enrichment plot (green line) graphically represents the running enrichment score (ES) for a given gene set. Below the graphs the GSEA output is listed: ES and normalized ES (NES) can be used to compare gene sets of different sizes. The nominal p-value represents the statistical significance of the enrichment score, but is not adjusted for gene set size or multiple hypothesis testing. The FDR q-value (False discovery rate) represents the estimated probability that the normalized enrichment score represents a false positive finding.

Stem cell genes

Lgr5-GFP high vs low(top 200 probes)Munoz et al., 2012

Wnt target genes

Downregulated in organoids 1 day after R-spondin withdrawal(bottom 200 probes)de Lau et al., 2011

Proliferation genes

Myc induced genes(top 44 genes) Yu et al., 2005

Paneth cell genes

Genes enriched in sorted Paneth cells vs. sorted Lgr5 stem cells(top 200 probes)Sato et al., 2011

0 2,500 5,000 7,500 10,000 12,500 15,000 17,500Rank in Ordered Dataset

Upregulated in class: positiveEnrichment score (ES): 0.77Normalized Enrichment Score (NES): 3.13Nominal p-value: <0.001FDR q-value: <0.001

Rank in Ordered Dataset





Upregulated in class: negativeEnrichment score (ES): -0.43Normalized Enrichment Score (NES): -1.51Nominal p-value: 0.013FDR q-value: 0.013

Control vs. CHIR


Upregulated in class: positiveEnrichment score (ES): 0.47Normalized Enrichment Score (NES): 1.64Nominal p-value: 0.004FDR q-value: 0.004




Upregulated in class: negativeEnrichment score (ES): -0.40Normalized Enrichment Score (NES): -1.46Nominal p-value: 0.02FDR q-value: 0.03


Upregulated in class: negativeEnrichment score (ES): -0.63Normalized Enrichment Score (NES): -2.58Nominal p-value: <0.001FDR q-value: <0.001

Control vs. VPA

Rank in Ordered DatasetUpregulated in class: positiveEnrichment score (ES): 0.52Normalized Enrichment Score (NES): 1.88Nominal p-value: <0.001FDR q-value: <0.001






Upregulated in class: negativeEnrichment score (ES): -0.76Normalized Enrichment Score (NES): -2.80Nominal p-value: <0.001FDR q-value: <0.001

Control vs. CHIR+VPA0.80.70.60.50.40.30.20.10.0

0.70.60.50.40.30.20.10.0

0.60.50.40.30.20.10.0

0.30.20.10.0

-0.1-0.2-0.3-0.4

Enric

hmen

t sco

re (E

S)

Enric

hmen

t sco

re (E

S)En

richm

ent s

core

(ES)

Enric

hmen

t sco

re (E

S)En

richm

ent s

core

(ES)

Enric

hmen

t sco

re (E

S)En

richm

ent s

core

(ES)

Enric

hmen

t sco

re (E

S)En

richm

ent s

core

(ES)

Enric

hmen

t sco

re (E

S)En

richm

ent s

core

(ES)

Enric

hmen

t sco

re (E

S)

0.50.40.30.20.10.0

-0.1-0.2

0.200.25

0.15

0.050.10

0.00-0.05

-0.15-0.10

-0.20

0.10.0

-0.1-0.2-0.3

-0.5-0.4

-0.6

0.100.050.00

-0.10-0.05

-0.15

-0.25

-0.35

-0.20

-0.30

-0.40

0.50.40.30.20.10.0

-0.1-0.2

0.40

0.30

0.20

0.100.15

0.25

0.35

0.00

-0.10-0.05

0.05

0.1250.1000.0750.0500.0250.000

-0.025-0.050-0.075

0.10.0

-0.1-0.2-0.2-0.2-0.2-0.2-0.2-0.8

0 2,500 5,000 7,500 10,000 12,500

12,500

15,000 17,500

0 2,500 5,000 7,500 10,000 15,000 17,500

0 2,500 5,000 7,500 10,000 12,500 15,000 17,500 0 2,500 5,000 7,500 10,000 12,500 15,000 17,500

0 2,500 5,000 7,500 10,000 12,500 15,000 17,500

0 2,500 5,000 7,500 10,000 12,500 15,000 17,500

0 2,500 5,000 7,500 10,000 12,500 15,000 17,500 0 2,500 5,000 7,500 10,000 12,500 15,000 17,500

0 2,500 5,000 7,500 10,000 12,500 15,000 17,500

0 2,500 5,000 7,500 10,000 12,500 15,000 17,500

0 2,500 5,000 7,500 10,000 12,500 15,000 17,500


6


Supplementary Figure 6 | Exploring the mechanism of action for CHIR and VPA. (a) 6 day cultures of crypts in ENR-C (Control) condition or together with HDAC inhibitors. (b) Quantification of GFP percentage, total live cell number and rela-tive GFP intensity of cells in (a). (c) The effects of nicotinamide (Ni) in combination with Wnt3a (W, 100 ng/ml) or CHIR (C). Shown are cell numbers and percentage of GFP+ cells of crypts cultured for 6 days in multiple conditions. In all panels, error bars indicate S.D. or triplicate wells. ***P<0.001; **P<0.01; *P<0.05; NS P>0.05.

Compound 7-10uMTubastatin A-10uM

Control

TSA-20nMVPA-1.5mM

MC1568-10uMCI994-1uMMS275-0.5uM

b

a

No. of cells (X1,000)0 200 600

MC1568-10uMMC1568-5uM

Butyrate-0.5mMControl

MS275-1uMCompound 7-5uM

CI994-1uMCI994-2uM

Butyrate-1mMCompound 7-10uM

MS275-0.5uMSBHA-15uM

Tubastatin A-5uMTubastatin A-10uM

VPA-1.5mMTSA-20nM

Compound 7-20uM

400

Relative GFP intensity0 1,000 2,000 3,000 4,000 5,000 6,000 7,000

MC1568-10uMMC1568-5uM



CI994-1uMCI994-2uM




VPA-1.5mMTSA-20nM

Compound 7-20uM

GFP+ % cells0 20 40 50

MC1568-10uMMC1568-5uM



CI994-1uMCI994-2uM




VPA-1.5mMTSA-20nM

Compound 7-20uM

3010

Butyrate-0.5mMSBHA-15uM

Compound 7-5uMTubastatin A-5uM

c

0

100

200

300

400

500

No.

of c

ells

(X1,

000)

ENRENR-W+Ni(mM)

ENR-WV0.1 1010ENR-C+Ni(mM)

ENR-CV0.1 1010

0

10

20

30

40

GFP

+ p

erce

ntag

e

ENRENR-W+Ni(mM)

ENR-WV0.1 1010ENR-C+Ni(mM)

ENR-CV0.1 1010

NS

***

NS *

***


7


Supplementary Figure 7 | Mechanism of VPA. (a) VPA rescues Lgr5-GFP expression following Notch inhibition, titration experiment. Crypts were cultured in ENR-C condition with or without DAPT (D, 5 µM) and varying concentration of VPA (V, 0.25–2 mM) for 3 days. Scale bars, 200 µm. (b) Quantification of cell numbers and Lgr5-GFP expression of crypts cultured in ENR-C condition with or without DAPT and varying concentration of VPA (0–2 mM). Error bars indicate S.D. of duplicate wells. (c) Intestinal organoids from Dll1EGFP-IRES-CreERT2/+ mice were cultured in multiple conditions for 5 days. Dll1-GFP and brightfield images are shown. Overlay with the red channel shows autofluorescene (yellow signal). (d) FACS analysis of Dll1-GFP expression of cells in (c). (f and g) Crypts were cultured in the (f) ENR or (g) ENR+CHIR conditions for 4 days fol-lowed by addition of VPA at different concentrations for another 24 hours. The expression of Notch1, Hes1 and Atoh1 was analyzed by Real-time RT-PCR. (h) Lgr5-GFP expression in wild-type (WT) and Atoh1 knock-out (Atoh1-KO) organoids cul-tures. (i) qPCR analysis of marker gene expression in WT and Atoh1-KO organoids cultured in conditions as indicated. Error bars indicate S.D., n = 3. (j) Maintenance of Lgr5-GFP expression in intestinal adenomas lacking RBP-J. Immunfluorescence of paraffin sections 4 weeks following injection of Tamoxifen to induce Cre-activity in Lgr5-EGFP-ires-CreERT2 mice. (left) Control animal, (middle) homozygous deletion of APC alone and (right) APC/RPB-J deletion are shown. Top panels show GFP staining (Lgr5-reporter, green) and Lysozyme staining (red). Bottom panels show Immunostaining of RBP-J (white) in the same regions. Arrows show that adenomas lacking RBP-J are Lgr5 positive but Lysozyme negative. Scale bars, (a, h) 200 µm and (j) 50 µm.

Dll1

-GFP

/Aut

oBr

ight

�eld

ENR ENR-C ENR-V ENR-CVca

ENR-C ENR-C-D ENR-C-D-V(0.25mM)

ENR-C-D-V(0.5mM) ENR-C-D-V(1mM) ENR-C-D-V(2mM)

f

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Rela

tive

mRN

A ex

pres

sion

Notch1 Hes1Atoh1

ENR+VPA

0.0

0.2

0.4

0.6

0.8

1.0

1.2

00.5123

Rela

tive

mRN

A ex

pres

sion

Notch1 Hes1 Atoh1

ENR-C+VPAVPA (mM)

g

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Rela

tive

mRN

A ex

pres

sion

Notch1 Hes1 Atoh1

ENRENR-VENR-CENR-CV

eb

0

20

40

60

80

100

ENRENR-C

ENR-CVENR-CD+VPA(mM)

0 210.5

GFP

+ p

erce

ntag

e

0

2000

4000

6000

Rela

tive

GFP

inte

nsity

ENRENR-C

ENR-CVENR-CD+VPA(mM)

0 210.50

50

100

150

200

No.

of c

ells

(X1,

000)

ENR

ENR-C

ENR-CV

ENR-CD+VPA(mM)0 210.5

i

0

1

2

3AlpiMuc2ChgALyz1Lgr5

Rela

tive

mRN

A ex

pres

sion

ENRENR-C

ENR-V

ENR-CV

ENR-C

ENR-CV

WT Atoh1-KO

100

101

102

103

Dll1-GFP

0

20

40

60

80

100

% o

f Max

ControlENRENR-CENR-VENR-CV

d

hAtoh1-KOWT

ENR-CVENR-CENR-CV

Lgr5

-GFP

Brig

ht�e

ld

j

Lgr5

-GFP

/Lyz

RBP-

J

APC�/+ RBP-J�/+ APC�/� RBP-J�/+ APC�/� RBP-J�/�


8

Supplementary Table 1

Summary of growth factors and small molecules used in this study

Reagent Name Abbreviation Final Concentration CompanyEGF E 50 ng/ml InvitrogenNoggin N 100 ng/ml PeprotechR-spondin 1 R 500 ng/ml R&DWnt3a W 100 ng/ml PeprotechCHIR99021 C 3 µM StemgentValproic Acid V 1–2 mM SigmaIWP-2 I 2 µM StemgentDAPT D 10 µM StemgentLiCl Li 5 mM SigmaTrichostatin A TSA 25 nM SigmaSuberohydroxamic Acid SBHA 15 µM SigmaSodium Butyrate Butyrate 0.5 mM SigmaEGF, Noggin, R-spondin 1 ENR As aboveNicotinamide Ni 10 mM Sigmagastrin g 10 nM SigmaPGE2 P 0.02 µM SigmaA-83-01 A 0.5 µM TocrisSB202190 S 10 µM TocrisY-27632 Y 10 µM TocrisTubastatin A 5–10 µM ChemtekCompound 7 5–20 µM ChemtekMS275 0.5–1 µM CaymanCI994 1–2 µM CaymanMC1568 5–10 µM Sigma


9

Supplementary Results

Towards elucidating the mechanism of CHIR and VPA

Besides VPA, TSA, Tubastatin A and Compound 7, we also tested other pan-HDAC inhibitors including SBHA and Butyr-ate, as well as class I (CI-994, MS275, Supplementary Fig. 6a,b), class IIa (MC1568, Supplementary Fig. 6a,b) and class III (Nicotinamide, Supplementary Fig. 6c) HDAC inhibitors showed no or only moderate effects on GFP expression (Supple-mentary Fig. 6a-c).

Previous reports have shown that Notch pathway activation is required to inhibit secretory cell differentiation and to main-tain self-renewal of stem cells4-8, similar to the effects of VPA. To explore if VPA might target elements of the Notch pathway, we first tested whether Notch inhibition could be rescued by the addition of VPA. Notch/γ-secretase inhibition using DAPT led to impaired cell proliferation and GFP expression, which was rescued by VPA in a dose-dependent manner (Supple-mentary Fig. 7a,b). This suggests VPA acts downstream of Notch-receptor cleavage and could bypass the requirement of ligand-receptor mediated Notch activation. Furthermore, we found that although the addition of VPA alone induced weak global Dll1-GFP expression in Dll1EGFP-IRES-CreERT2 intestinal organoids, VPA+CHIR reduced Dll1-GFP expression (Supple-mentary Fig. 7c,d), which also suggests VPA+CHIR does not function through enhancing Delta-Notch interaction to acti-vate Notch pathway.

VPA has been previously shown to activate Notch1 and Hes1 in cancer cells lines9, 10 and therefore we tested the effect of VPA on the expression of Notch effectors. We found, however, that addition of VPA moderately decreased the expression of Notch1 and Hes1 in cells cultured for 6 days (Supplementary Fig. 7e) or 24 hours (Supplementary Fig. 7f,g). Additionally, we observed a pronounced decrease of the negative Notch target Atoh1 (Math1), a transcription factor that is essential for secretory cell differentiation11, 12. It has been reported that Atoh1 opposes Hes1 expression8, constituting a lateral inhibition signaling module that controls the specification of secretory and enterocyte lineages13. Intestinal stem cells remain functional both in vivo and in vitro after Paneth cell ablation induced by Atoh1 deficiency14, 15, which phenocopies the effects of VPA. We therefore tested whether Atoh1 deficiency mimics the effects of VPA in our system. Indeed, similarly to the effect of VPA, Atoh1 knockout intestinal organoids suppressed secretory marker expression when cultured in the ENR-C condition. Further addition of VPA, however, increased GFP and Lgr5 mRNA expression and downregulated the enterocyte marker Alpi expression compared with CHIR alone, which suggests VPA exhibits other roles in addition to Atoh1 modulation (Sup-plementary Fig. 7h,i).

In the nucleus Notch ON and OFF signals result in activation or repression of effector genes that depend on the transcription factor RBP-J16. We then asked whether genetic disruption of both transcriptional programs (i.e. the entire Notch lateral in-hibition module) could lead to similar effects as VPA under conditions of high Wnt activation. To test this, we conditionally deleted both RBP-J and the negative Wnt regulator APC in intestinal stem cells by Tamoxifen injection in Lgr5-EGFP-ires-CreERT2 mice. After 4 weeks the deletion of APC alone caused formation of multiple adenomas17 that were composed of Lgr5+ cells, which were intermingled with Paneth cells. Consistent with a previous report the additional deletion of RBP-J also resulted in adenoma formation18. Here, co-immunostaining showed maintenance of Lgr5 expression in the absence of Paneth cells in RBP-J-null adenomas (Supplementary Fig. 7j). This result suggests the stem cell marker Lgr5 expression in vivo does not rely on the Notch-ON signal when the Notch lateral inhibition module is disrupted, but can be imposed by high Wnt signaling alone.

References:

1. Jung, P. et al. Isolation and in vitro expansion of human colonic stem cells. Nat Med 17, 1225-1227 (2011).2. Sato, T. et al. Long-term expansion of epithelial organoids from human colon, adenoma, adenocarcinoma, and Bar-

rett’s epithelium. Gastroenterology 141, 1762-1772 (2011).3. Yu, D., Cozma, D., Park, A. & Thomas-Tikhonenko, A. Functional validation of genes implicated in lymphomagen-

esis: an in vivo selection assay using a Myc-induced B-cell tumor. Ann N Y Acad Sci 1059, 145-159 (2005).4. Pellegrinet, L. et al. Dll1- and dll4-mediated notch signaling are required for homeostasis of intestinal stem cells.

Gastroenterology 140, 1230-1240 e1231-1237 (2011).5. Milano, J. et al. Modulation of notch processing by gamma-secretase inhibitors causes intestinal goblet cell metapla-

sia and induction of genes known to specify gut secretory lineage differentiation. Toxicological sciences : an official journal of the Society of Toxicology 82, 341-358 (2004).


10

6. VanDussen, K.L. et al. Notch signaling modulates proliferation and differentiation of intestinal crypt base columnar stem cells. Development 139, 488-497 (2012).

7. van Es, J.H. et al. Notch/gamma-secretase inhibition turns proliferative cells in intestinal crypts and adenomas into goblet cells. Nature 435, 959-963 (2005).

8. Vooijs, M., Liu, Z. & Kopan, R. Notch: architect, landscaper, and guardian of the intestine. Gastroenterology 141, 448-459 (2011).

9. Greenblatt, D.Y. et al. Valproic acid activates notch-1 signaling and regulates the neuroendocrine phenotype in car-cinoid cancer cells. The oncologist 12, 942-951 (2007).

10. Stockhausen, M.T., Sjolund, J., Manetopoulos, C. & Axelson, H. Effects of the histone deacetylase inhibitor valproic acid on Notch signalling in human neuroblastoma cells. Br J Cancer 92, 751-759 (2005).

11. van Es, J.H., de Geest, N., van de Born, M., Clevers, H. & Hassan, B.A. Intestinal stem cells lacking the Math1 tumour suppressor are refractory to Notch inhibitors. Nat Commun 1, 1-5 (2010).

12. Yang, Q., Bermingham, N.A., Finegold, M.J. & Zoghbi, H.Y. Requirement of Math1 for secretory cell lineage com-mitment in the mouse intestine. Science 294, 2155-2158 (2001).

13. van Es, J.H. et al. Dll1+ secretory progenitor cells revert to stem cells upon crypt damage. Nat Cell Biol 14, 1099-1104 (2012).

14. Durand, A. et al. Functional intestinal stem cells after Paneth cell ablation induced by the loss of transcription factor Math1 (Atoh1). Proc Natl Acad Sci U S A 109, 8965-8970 (2012).

15. Kim, T.H., Escudero, S. & Shivdasani, R.A. Intact function of Lgr5 receptor-expressing intestinal stem cells in the absence of Paneth cells. Proc Natl Acad Sci U S A 109, 3932-3937 (2012).

16. Tanigaki, K. & Honjo, T. Two opposing roles of RBP-J in Notch signaling. Current topics in developmental biology 92, 231-252 (2010).

17. Barker, N. et al. Crypt stem cells as the cells-of-origin of intestinal cancer. Nature 457, 608-611 (2009).18. Peignon, G. et al. Complex interplay between beta-catenin signalling and Notch effectors in intestinal tumorigenesis.

Gut 60, 166-176 (2011).


Documents

Niche-independent high-purity cultures of Lgr5 intestinal ......Niche-independent high-purity cultures of Lgr5+ intestinal stem cells and their progeny Xiaolei Yin, Henner F Farin,