Protoderm

MMC

Meristemoid

SLGC

GMC

Stoma secondarymeristemoid

secondarystoma

1

2

3a3b

4

5

4b

5b

SLGCs

SLGCs

Supplemental Figure 1. Stomatal development in Arabidopsis.

Developmental steps that lead to stomata formation. A protodermal cell acquires MMC fate (orange). The MMC executes the first asymmetric division that generates a meristemoid (yellow) and an SLGC (blue). The meristemoid can undergo additional asymmetric divisions, amplifying the number of SLGCs, and ultimately it differentiates into a GMC (purple), of which the symmetric division produces the guard cell (GC) pair that constitutes the stoma (green). Secondary or satellite stomata can be formed when a SLGC asymmetrically divides away from the primary stoma.

Col-0 ldc

Supplemental Figure 2. Adaxial cotyledon phenotype exhibited by the ldc mutant.

DIC images showing complete 15-day-old adaxial cotyledons of Col-0 and ldc. Bar = 0.5 mm.

Supplemental Figure 3. Genetic mapping of the ldc mutation and SPCH protein sequence indicating known domains and amino acid changes of the point mutants.

(A) Genetic map of Arabidopsis presenting the 32 microsatellite markers (purple lines) used for the genetic mapping of the ldc mutation. Each black rectangle represents a chromosome (I to V). Inset shows the genome region in which the ldc mutation had reduced recombination rates (r) and the physical position of the SPCH locus (blue line). (B) SPCH amino acid sequence with colored domains (blue for bHLH, yellow for MAP KINASE TARGET DOMAIN [MPKTD], and green for the plant specific C-terminal domain (SPCH, MUTE and FAMA) [SMF]) and point amino acid changes for the different mutants (red).

I II III IV V

MNB8

r=0%

MUA2

r=4%

SPCH

SPCH (At5g53210) aminoacid sequence

MQEIIPDFLEECEFVDTSLAGDDLFAILESLEGAGEISPTAAST

PKDGTTSSKELVKDQDYENSSPKRKKQRLETRKEEDEEEED

RàW (spch-5)

GDGEAEEDNKQDGQQKMSHVTVERNRRKQMNEHLTVLRSL

bHLH

MPCFYVKRGDQASIIGGVVEYISELQQVLQSLEAKKQRKTYA

EVLSPRVVPSPRPSPPVLSPRKPPLSPRINHHQIHHHLLLPPIS

MPKTD

PRTPQPTSPYRAIPPQLPLIPQPPLRSYSSLASCSSLGDPPPY

VàM (spch-2)

SPASSSSSPSVSSNHESSVINELVANSKSALADVEVKFSGAN

SMF

VLLKTVSHKIPGQVMKIIAALEDLALEILQVNINTVDETMLNSFT

Qàstop (spch-1)

IKIGIECQLSAEELAQQIQQTFC

A

B

* * *

Supplemental Figure 4. Epidermal DIC images of several mature organs in Col-0 and spch-5.

Stomata are false-colored in purple for better identification. Brackets mark stomatal clusters. Bars = 20 µm.

SE

PAL

PE

DIC

EL

SIL

IQU

ES

TAM

EN

FILA

ME

NT

PE

TAL

STA

ME

N

Col-0

] ]

STE

MH

YP

OC

OTY

L

spch-5

0

10

20

30

40

50

60

70

80

S

D (

sto

ma

ta

/m

m )2

0

5

10

15

20

25

SI

(%)

Col

-0

spch-5

Col

-0

spch-5

*

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6O

A (

cm

)2

Col

-0

spch-5

n.d

0

20

40

60

80

100

Col-0 spch-5

0

50

100

150

200

250

PC

D (

ce

lls/m

m2)

*

<1,000

1,000-2,000

2,000-5,000

5,000-10,000

10,000-15,000

>15,000

Pavement cell size distribution

size(µm2)

%

DC

ADAXIAL

]

]

ABAXIAL

spch-5

Col

-0

Col

-0

spch-5

A

B

Supplemental Figure 5. spch-5 third-leaf epidermal phenotype.

(A) DIC images of the adaxial and abaxial epidermis of Col-0 and spch-5 in third leaves of 28-day-old plants. Stomata are highlighted in purple and clusters marked by brackets. Bars = 50 µm. (B) Graphs representing stomatal index, stomatal density, organ area and pavement cell density in abaxial epidermis of 28-day-old third leaves. Grey and orange denote Col-0 and spch-5, respectively. Asterisks indicate P < 0.05 (Student’s t test; n = 10) and n.d P > 0.05. Error bars represent SE. (C) Representative drawings of Col-0 and spch-5 with color code as in (D). Stomata are indicated in purple. (D) Size distribution of pavement cells in Col-0 and spch-5. Frequency of large pavement cells is much higher in the mutant than in the wild type. Bar = 100 µm.

Supplemental Figure 6. Additional data derived from cell tracking of leaf primordia shown in Figure 2.

Graphs elaborated from leaves (n = 5) per genotype and with 50 cells at initial field (T0). (A) Percentage of entry divisions. (B) Number of SLGCs per lineage. (C) Percentage of primary and secondary lineages. (D) Stomatal differentiation rate. Asterisk in (B) denotes P < 0.05 (Student’s t test). Error bars represent SE.

0

0.5

1

1.5

2

2.5

Col-0 spch-5nu

mbe

r

SLGCs/lineage

*

B

0

0.2

0.4

0.6

0.8

1

T0-T1 T1-T2 T2-T3 T3-T4

stom

ata

prec

urso

rs-1

day

-1

Stomatal differentiation rate

D

Col-0 spch-5

15%

85%

Do not enterEnter into lineage

70%30%

spch-5

Col-0

A

65%35%

Primary lineagesSecondary lineages

62%38%

spch-5

Col-0

C

Supplemental Figure 7. Orientation of cell division planes in

spch-5 and spch-5 mute-3 stomatal lineages.

Confocal images of 13-day-old adaxial cotyledon epidermis,

showing cell contours with PI (red) and young lineage cells with

GFP (green) expressed from a TMMpro:GFP reporter construct.

mute-3 plants form rosette-like lineages with several small cells

in a regular inward spiral arrangement indicated as a dotted white

line. Only the most recently formed meristemoid shows a strong

TMMpro:GFP expression. mute-3 spch-5 double mutant lineages

also display extra small cells, but arranged in irregular patterns

and of which several express TMMpro:GFP. Note the absence of

stomata due to the lack of the MUTE function. Bars = 50 µm.

mute-3 mute-3 spch-5

mute-3 mute-3 spch-5

TMMpro:GFP TMMpro:GFP

Supplemental Figure 8. Homology-based three-dimensional

structural prediction for the SPCH-5 protein.

Modeled protein structures of SPCH versions. (A) Wild type

SPCH. (B) SPCH-5 predicted from spch-5. The amino acid triad

critical for DNA binding in other bHLH proteins is highlighted

(pink). The wild-type Arg111 substituted by Trp111 in spch-5 is

marked in orange. The hydrogen bond involving this Arg111 is

indicated as a dashed line with the distance in Ångströms (A), but

is lost in the mutant (B).

2.38

His104Glu108

Arg112

Arg111

SPCH

Trp111

SPCH-5

His104Glu108

Arg112

A

B

Supplemental Figure 9. DNA-protein interactions involving

SPCH-5.

(A) Scheme of the domain structure of SPCH and the truncated

SPCHΔ273 version lacking the C-terminal part (the ACT-like or SMF domain). (B) Yeast one-hybrid experiment with the

tandem-repeated G-box sequence as bait for the wild type (WT) or a mutated version (mut). Colored circles indicate LacZ report-

er-normalized expression.

50 aa

bHLH MPKTD

bHLH SPCHMPKTD SMF

SPCHΔ273

A

Bait

TF

No vector

Empty AD vector

SPCH

SPCH-5

SPCHΔ273

PIF4

ICE1

SPCH-5 + ICE1

SPCH + ICE1

normalized value

0 1

B

G-boxWT mut

Supplemental Figure 10. Protein--protein interactions involving

SPCH-5.

(A) to (C) BiFC experiments with transient expression assays in

Nicotiana benthamiana leaves. Image compositions were made

with representative cells showing the recomposed GFP signal

diagnostic for physical interactions between proteins.

SP

CH

SP

CH

-5N

o p

artn

er

ICE1 SCRM2 No partner

GFP BF

Merge100 µm

A

100 µmSP

CH

SPCHB

SP

CHPPP

100 µm

ICE1 SCRM2 No partnerC

Supplemental Figure 11. Phenotypic and protein expression analyses of spch-3 lines complemented with different SPCH versions.

(A) to (C) DIC images of the epidermis of spch-3 plants transformed with translational GFP fusions to SPCH, SPCH-5, or SPCHPPP under

the control of the SPCH promoter. Adaxial (A) and abaxial (B) 23-day-old cotyledons and 10-days-old hypocotyls (C) are shown. Stomata

are colored purple. Stomatal clusters are enclosed into brackets. (D) to (F) Confocal images of the same lines in (A) to (C). Red outlines

are PI-stained cell contours. Images were taken 3 days after sowing. GFP fluorescence marks the expression of the different protein

versions. Adaxial (D) and abaxial (E) cotyledons and hypocotyls (F) are shown. Bars = 100 µm.

Hyp

oco

tyl

Ad

axia

l C

ot.

Ab

axia

l C

ot.

]

SPCHpro:SPCH-GFP/spch-3 SPCH

pro:SPCHPPP-GFP/spch-3SPCH

pro:SPCH-5-GFP/spch-3

A

B

C

D

E

F

spch-2 (DMSO)

spch-2 (BL)

0

5

10

15

20

25

SI (%

)

spch-2

+ D

MSO

spch-2

+BL

n.dC

B

A

Supplemental Figure 12. BL effect on spch-2 stomatal develop-

ment.

(A) and (B) DIC images of the abaxial epidermis of spch-2 in

DMSO (A) and BL (B). (C) Stomatal index quantification of

spch-2 in medium with (purple) and without BL (grey). Error bars

represent SE. n.d denotes absence of statistically significant

differences between mean values (95% confident Student’s t test).

Supplemental Figure 13. Quantitative PCR for microarray validation.

Gene expression data in the microarrays (orange) validated by qRT-PCR (green) for 10 genes related to stomata development or BR

pathway, selected in pairwise comparisons between genotypes and BL treatments. (A) Nontreated spch-5 vs Col-0. (B) Nontreated

SPCHPPP/spch-3 vs Col-0. (C) BL-treated vs control spch-5. (D) BL-treated vs control SPCHPPP/spch-3. UBIQUITIN10 and ACTIN2 served

as reference genes. The Log2-fold change of the qRT-PCR expression data was calculated and represented. The qRT-PCR results were

averaged from three (or two in the case of SPCHPPP/spch-3) independent experiments, with the error bar representing the SE of the mean.

qPCR microarray

Aspch-5+DMSO vs Col-0+DMSO

Log

2(F

C)

TMM EPF2 CPD BASL EPF1 BIN2 ICE1 SCRM2 SPCH SDD10

-2

2

-4

-6

BSPCHPPP/spch-3+DMSO vs Col-0+DMSO

-6

-8

-4

-2

0

2

TMM EPF2 CPD BASL EPF1 BIN2 ICE1 SCRM2 SPCH SDD1

Log

2(F

C)

qPCR microarray

qPCR microarray

C spch-5+BL vs spch-5+DMSO

Log

2(F

C)

TMM EPF2 CPD BASL EPF1 BIN2 ICE1 SCRM2 SPCH SDD10

2

1

-1

-2

qPCR microarray

D SPCHPPP/spch-3+BL vs SPCHPPP/spch-3+DMSO Log

2(F

C)

0

1

-1

TMM EPF2 CPD BASL EPF1 BIN2 ICE1 SCRM2 SPCH SDD1

Supplemental Figure 14. Biological processes overrepresented in the spch-5 DEGs.

Genes differentially expressed in spch-5 as compared to Col-0 were analyzed with Gene Ontology (GO) term enrichment by means of ClueGO. Circles represent GO categories and lines depict relationships between nodes. Only the main categories enriched in DEGs in spch-5 and the involved main nodes are shown. Nodes containing down-regulated genes are in green; nodes containing both up- and down-regulated genes are in grey; no nodes were found with only up-regulated genes. Node size is relative to its P-value (inset). (A) Cell division and expansion. (B) Stomatal and epidermal development. (C) Hormonal responses. (D) Histone kinases

response to organic substance histone H3−S10

phosphorylation

microtubule−based movement

plant epidermis development

response to hormone

microtubule−based process cytokinesis

cell division

stomatal complex development

AB

CD

pV 0.005 - 0.01

pV 0.0005 - 0.005

pV 0.00005 - 0.0005

pV < 0.00005% up-regulated

unspecific terms

100%

50%

100%

50%

% down-regulated

<50%

Node Colour Node Size