LETTERS
PLETHORA proteins as dose-dependent master
regulators of Arabidopsis root development
Carla Galinha1
*{, Hugo Hofhuis1
*, Marijn Luijten1
, Viola Willemsen1
, Ikram Blilou1
, Renze Heidstra1
& Ben Scheres1
Factors with a graded distribution can program fields of cells in a
dose-dependent manner1,2
, but no evidence has hitherto surfaced
for such mechanisms in plants. In the Arabidopsis thaliana root,
two PLETHORA (PLT) genes encoding AP2-domain transcription
factors have been shown to maintain the activity of stem cells3
.
Here we show that a clade of four PLT homologues is necessary for
root formation. Promoter activity and protein fusions of PLT
homologues display gradient distributions with maxima in the
stem cell area. PLT activities are largely additive and dosage
dependent. High levels of PLT activity promote stem cell identity
and maintenance; lower levels promote mitotic activity of stem cell
daughters; and further reduction in levels is required for cell differentiation.
Our findings indicate that PLT protein dosage is
translated into distinct cellular responses.
During animal development, instructive molecules acquire a
graded distribution and induce distinct cellular responses in a concentration-dependent
manner. Whether similar mechanisms occur
in plants has been controversial; dosage-sensitive action of plant
hormones has been inferred only after external application4
. Plant
stem cell regions, which supply cells for the growing root and shoot
systems5
, are potential sites of action for instructive gradients. Stem
cells are maintained in local micro-environments, which are similar
to animal stem cell niches6
. Stem cell daughters undergo additional
divisions in transit-amplifying cell compartments called meristems;
when cells leave the meristem they rapidly expand and differentiate.
The PLETHORA1 (PLT1, At3g20840) and PLT2 (At1g51190) genes
encode AP2-domain transcription factor family members essential
for defining the root stem cell niche3
. plt1;plt2 mutants display stem
cell loss, loss of transit-amplifying cells and reduced cell expansion.
PLT1 and PLT2 expression strongly correlates with a transcriptional
response maximum to the plant hormone auxin in the root tip3,7
and this maximum has been shown to have profound organizing
activity8
—a property often associated with sources of instructive
gradients. Here, we reveal that the PLT gene family controls distinct
aspects of root development in a dose-dependent manner through
PLT expression gradients that culminate in the stem cell niche.
The proteins encoded by At5g10510/AINTEGUMENTA-LIKE6
(AIL6)/PLT3 and At5g17430/BABY BOOM(BBM) group with
PLT1 and PLT2 in the AP2/ERF transcription factor family (Supplementary
Fig. 1)9
, and these candidate redundant factors are predicted
to be expressed in the root10
.
From the heart-stage of embryogenesis onward, PLT3 is expressed
in provascular cells, the quiescent centre and columella progenitor
cells (Fig. 1a). Post-embryonically, PLT3 messenger RNA accumulates
in the root stem cell niche with the strongest signal in the
columella stem cell layer (Fig. 1b), in contrast to the predominant
quiescent-centre-localization of PLT1 and PLT2 transcript3
. At the
heart-stage of embryo development, BBM is expressed in provascular
cells and in the lens-shaped quiescent centre progenitor cell (Fig. 1c).
Post-embryonically, BBM transcript accumulates in the quiescent
centre and columella stem cells—in a similar manner to the PLT
mRNAs—andinprovascular cellsoftheproximal meristem (Fig.1d).
The plt3-1 mutant allele carries a T-DNA insertion interrupting
the first AP2 domain (Supplementary Fig. 2). No transcript was
detected by PCR with reverse transcription (RT–PCR) or by in situ
hybridization on plt3-1 seedlings (data not shown), suggesting that
plt3-1 is a null allele. Homozygous plt3 single mutants have slightly
shorter roots and meristems compared to wild type, but plt12/2
plt22/2
plt32/2
triple homozygotes are rootless (Fig. 1e, upper
inset). Progeny from plt12/2
plt22/2
plt31/2
, plt12/2
plt21/2
plt32/2
and plt11/2
plt22/2
plt32/2
plants segregate ,25% rootless triple
mutants (Supplementary Table 2), demonstrating linkage between
the rootless phenotype and the three PLT genes. The embryonic root
pole of triple homozygous seedlings is fully differentiated at 3 days
post germination (d.p.g.) and adventitious root primordia arrest at
6 d.p.g. (Supplementary Fig. 3). Mature plt12/2
plt22/2
embryos
have only subtle defects in the cellular organization of the distal-most
region3
(Supplementary Fig. 4), but plt11/2
plt22/2
plt32/2
parents
yield ,25% embryos with aberrant root poles that lack a lateral root
cap cell layer (Supplementary Fig. 4).
We previously showed that plt12/2
plt22/2
mutants have strongly
reduced transcription of the PIN4 gene, which encodes an auxin
efflux facilitator11
. In triple mutant embryos from plt11/2
plt22/2
plt32/2
parents, PIN1 and PIN3 mRNAs are strongly reduced
(Fig. 1g–j and Supplementary Table 1). Post-embryonic PIN2
mRNA is strongly reduced in triple mutant roots before differentiation
(Fig. 1k, l). Therefore, PLT1, PLT2 and PLT3 redundantly
control expression of multiple PIN genes in the embryonic and postembryonic
root.
bbm-1 and bbm-2 mutant alleles carry T-DNA insertions before
and in the beginning of the first AP2 domain, respectively (Supplementary
Fig. 2). Truncated transcripts are detected by RT–PCR and
may be translated, but genetic interactions (described below) suggest
that the insertions cause loss-of-function effects. plt32/2
bbm2/2
double mutants have a shorter root and root meristem than either
single mutant (Fig. 1f, and Supplementary Fig. 3).
Intriguingly, the progeny of plants segregating different plt and
bbm allele combinations lack root and hypocotyl (Fig. 1e, lower inset)
at significant frequencies (Supplementary Table 2), reaching ,10%
of the progeny of selfed plt12/2
plt21/2
plt32/2
bbm-22/2
. These
defects initiate in the early basal embryo (Supplementary Fig. 5)
and resemble those in mutants of the auxin response factor
MONOPTEROS12
and the auxin perception machinery13,14
. PLT
genes do not seem to strongly perturb early global auxin-dependent
patterning processes, as suggested by essentially normal cotyledon
vasculature in the triple mutant (Supplementary Fig. 4). Segregation
*These authors contributed equally to this work.
1
Department of Biology, Faculty of Science, Utrecht University, Padualaan 8, 3584 CH Utrecht, The Netherlands. {Present address: Department of Plant Sciences, University of Oxford,
South Parks Road, Oxford OX1 3RB, UK.
Vol 449|25 October 2007|doi:10.1038/nature06206
1053
Nature©2007 Publishing Group
of plt2 in a homozygous bbm background and vice versa yields ,25%
early arrested embryos, and homozygous double mutants could not
be recovered, indicating a redundant function in early embryogenesis
(data not shown).
Ectopic root structures are initiated by constitutive embryonic
expression of PLT genes3
and after induction of BBM expression15
.
To test whether PLT induction induces a developmental switch to
root development, we expressed a PLT2–GR fusion protein that
complements plt12/2
plt22/2
after dexamethasone (dex) induction,
when driven by its own promoter (Supplementary Fig. 6). When
35S-PLT2–GR is activated by application of dex, roots are produced
from the shoot apex (Fig. 1m, n). Our gain- and loss-of-function
experiments indicate that PLT genes are master switches for root
development.
plt12/2
plt21/2
plt32/2
mutants have intermediate root and meristem
size between plt12/2
plt32/2
and plt12/2
plt22/2
plt32/2
(Fig.
1e, f, and Supplementary Fig. 3) and 50% of plt12/2
plt21/2
plt32/2
bbm2/2
seedlings have shorter roots than plt12/2
plt32/2
bbm2/2
, whereas 50% have no primary root (Supplementary Table
2). plt3 alleles are also semi-dominant, because growth and meristem
maintenance defects in plt12/2
plt22/2
plt31/2
seedlings are intermediate
between plt12/2
plt22/2
and plt12/2
plt22/2
plt32/2
(Fig. 1f,
and Supplementary Fig. 3). The semi-dominance of plt2 and plt3
loss-of-function alleles indicates dose-dependent activity.
To test whether PLT genes equally contribute to PLT ‘dosage’, we
transformed plt12/2
plt22/2
double mutants with PLT1, PLT2, PLT3
and BBM genes fused to the yellow fluorescent protein gene YFP and
driven by the full PLT2 promoter. In independent lines with similar
overall YFP levels, PLT1 and PLT2 fully complemented and PLT3 and
BBM partially complemented root growth in the double mutant. All
PLT proteins rescued columella stem cell activity (Supplementary
Figs 6 and 7). Thus, total PLT levels and to some extent intrinsic
differences in PLT protein activity contribute to root growth and
stem cell maintenance.
Transgenic lines carrying complete promoters of the PLT genes
fused to the cyan fluorescent protein gene CFP reveal highest promoter
activity in the stem cell niche, consistent with mRNA levels,
but they also show graded activity in the proximal meristem (Fig.
2a–d). Gradients can be observed in epidermal surface views, excluding
quenching effects, and they are specific to PLT promoters (Fig.
2e). To analyse whether this promoter activity drives a PLT protein
gradient, we combined the PLT–YFP fusions with their corresponding
full promoters. PLT1 and PLT2 gene fusions complemented
plt12/2
plt22/2
mutants (Supplementary Figs 6 and 7, and data not
shown).
j
a db c e
f g h i
Figure 2 | PLT promoter activity and PLT protein fusions display gradients.
a–d, CFP reporter driven by full-size promoters of PLT1 (a) PLT2
(b) PLT3 (c) and BBM (d). e, Epidermal gradient of PLT2 (left) but not
RCH2 (right) promoter. f–i, YFP reporter fused in-frame to genomic
fragments of PLT1 (f), PLT2 (g), PLT3 (h) and BBM (i). j, Co-localization in
one plant of PLT2 transcriptional (CFP, left magnification) and translational
(YFP, right magnification) fusion viewed in different regions using separate
channels.
PIN1 PIN2PIN3
300
60
120
360
180
240
Meristemlength(mm)
f
PLT3 BBM
col
pv
a b c d
e
g h i j k l
m n
WS
plt1
plt2
plt3
plt1;plt3;plt2/+
plt2;plt3
plt2;plt3;plt1/+
plt1;plt2
plt1;plt3
plt1;plt2;plt3/+
plt1;plt2;plt3
bbm
plt1;bbm
plt3;bbm
Col0
plt1;plt3;bbm
Figure 1 | Four PLT genes promote root formation. a–d, In situ
hybridization with PLT3- (a, b) and BBM- (c, d) specific probes in wild-type
embryos at heart-stage (a, c), and in roots of 3 d.p.g. wild-type plants
(b, d). Asterisk, quiescent centre; pv, provasculature; col, columella.
e, Seedlings 10 d.p.g., from left to right: wild type, plt32/2
, bbm-12/2
,
plt12/2
plt22/2
, plt12/2
plt22/2
plt32/2
and a plt12/2
plt21/2
plt32/2
bbm-
12/2
segregant. Insets show magnification of plt12/2
plt22/2
plt32/2
mutant
(upper) and plt12/2
plt21/2
bbm-12/2
segregant (lower). f, Meristem size in
wild type (Col0 and WS) and plt mutants at the indicated d.p.g. For each data
point, n 5 10 to 50; error bars, s.e.m. g–l, In situ hybridization using PIN
probes on wild-type (g, i) and plt12/2
plt22/2
plt32/2
(h, j) torpedo-stage
embryos and wild-type (k) and plt12/2
plt22/2
plt32/2
mutant (l) 2 d.p.g.
seedlings. m, Shoot of 9 d.p.g. 35S-PLT2–GR plant 6 days after
dexamethasone application. n, Magnification reveals cellular organization of
ectopic root including columella starch granules.
LETTERS NATURE|Vol 449|25 October 2007
1054
Nature©2007 Publishing Group
All PLT protein fusions revealed conspicuous gradients that
extend into the transit-amplifying cells and, for the PLT2 and
PLT3 fusions, into the elongation zone (Fig. 2f–i). The promoter
and protein gradients fully match when combined in one plant
(Fig. 2j). We previously reported accumulation of PLT transcripts
in the stem cell area3
, but, after extended staining, PLT1 in situ hybridizations
also reveal a broader expression domain (Supplementary
Fig. 8). We concluded that PLT promoter activity leads to protein
gradients with maximum expression in the stem cell niche. PLT1 and
PLT2 expression maxima broadly encompass the niche, whereas
PLT3 and BBM are more restricted.
We asked whether differences in PLT expression domains affect
the ability of PLT proteins to compensate for redundant partners.
Indeed, PLT1 and PLT2 only partially complement a plt12/2
plt22/2
mutant when driven by the BBM promoter (Supplementary Figs 6
and 7).
Our experiments suggested that the PLT protein concentration
gradient instructs different outputs in different regions, even though
each gene slightly differs in activity and expression profile. We therefore
tested whether altering the level or shape of the PLT2 gradient
affects the position of developmental boundaries. We expressed the
PLT2–YFP fusion in plt12/2
plt22/2
mutants under the RCH2 promoter,
which has low activity in the stem cell area but is active in
meristematic and elongating cells at a level comparable to that of
the PLT2 promoter (Fig. 3d–g). RCH2-PLT2–YFP prolongs transitamplifying
cell divisions but fails to maintain stem cells at 7 d.p.g.
(Fig. 3b, d). The transit-amplifying cell pool is lost at 12 d.p.g.
(Fig. 3c). We concluded that intermediate PLT levels in the meristem
promote transient cell cycling.
To validate that meristem size is controlled by a PLT gradient, we
analysed plt12/2
plt22/2
mutants complemented with the PLT2–YFP
construct driven by a truncated 1.3-kb PLT2 promoter fragment
(pPLT2s). This truncated promoter drives significant expression in
the stem cell area but the gradient declines more rapidly (Fig. 3e, f).
Accordingly, stem cells are rescued but root and meristem sizes are
,50% smaller (Supplementary Fig. 7). The amount of YFP signal per
mid-nuclear section in the stem cell zone, halfway the meristem, and
in the first expanding cells, provides three clearly separated intensity
ranges that match with zonation in the full- and truncated-promoter
driven gradients (Fig. 3g), suggesting that the PLT2 gradient defines
meristem zonation.
A dose-dependent gradient model predicts that PLT overexpression
shifts the meristem boundary. Indeed, dex induction of
35S-PLT2–GR plants promotes continuous growth of the transitamplifying
cell pool and meristem size increases (Fig. 4a–c). Ink
toner marks marking the elongation zone boundaries at the time of
induction reveal that PLT overexpression sustains cell division only
in cells that are still cycling and inhibits cell expansion in the elongation
zone. These data reinforce the idea that distinct PLT levels dictate
cell proliferation and mitotic exit.
The auxin response marker DR5-GUS8
and PIN3 transcription do
not change in 35S-PLT2–GR plants just before the onset of meristem
size expansion, but only at later stages, indicating that PLT-induced
expansion of the division zone is not caused by rapid changes in PIN
expression (Supplementary Fig. 9).
Notably, the stem cell area in PLT2–GR plants is not altered after
induction (Fig. 4c). The RETINOBLASTOMA (RBR) pathway was
recently identified as an independent stem cell input16
, so we reasoned
that this pathway might still limit stem cell pool size in the
presence of higher PLT levels. Therefore, we combined a root-specific
RNA interference (RNAi)-mediated silencing construct (RCH1-
RBRi)16
with 35S-PLT2–GR. After induction with dex in the double
transgenic, root meristem size increases as in 35S-PLT2–GR, but
clusters of dividing cells in the root cap area expand beyond that seen
in RCH1-RBRi alone (Fig. 4d, e). Moreover, periclinal divisions
normally associated with stem cells occur throughout the proximal
area (Fig. 4f). These data suggested that the high expression region of
the PLT gradient can be instructive for stem cell fate. Dramatic support
for this notion is provided by duplications of the distal stem cell
area in ,10% of RCH1-RBRi;35S-PLT2–GR root meristem zones
(Fig. 4g–i). We concluded that high PLT levels define the stem cell
domain, confirming PLT dosage-dependent stem cell specification.
Averagepixelintensity
0
50
100
150
200
250a
b
c
d e f
g
pPLT2–PLT2–YFPa2
pPLT2–PLT2–YFPb1
pPLT2–PLT2–YFPb2
pPLT2s–PLT2–YFPa1
pPLT2s–PLT2–YFPa2
pPLT2–PLT2–YFPa1
pRCH2–PLT2–YFP
Figure 3 | PLT expression regulates stem cell maintenance and meristem
boundary. a–d, Meristem prolongation but not stem cell rescue in RCH2PLT2–YFP
plants. Nomarski optics image of root tip of 7 d.p.g. plt1;plt2
(a), and of plt1;plt2 RCH2-PLT2–YFP at 7 d.p.g. (b) and 12 d.p.g. (c). Starch
granule staining (brown) shows no rescue of columella stem cells below the
quiescent centre. Confocal view of 7 d.p.g. plt1;plt2 RCH2-PLT2–YFP root
(d) shows that the meristem is rescued and reveals no expression of
PLT2–YFPin thestem cell area.Asterisk in b, quiescentcentre.e, f, Promoter
truncation shifts the meristem boundary. CLSM views at identical pinhole
and laser settings for RCH2-PLT2–YFP (d), pPLT2-PLT2–YFP (e) and
pPLT2s-PLT2–YFP (f). g, Quantification of fluorescence per nucleus in
pRCH2-PLT2–YFP transient meristem (red circles in d, and red graph
areas), and in stem cells (yellow in e, f and graph area), mid-meristem (green
in e, f and graph area) and first elongating cells (blue in e, f and graph area) of
pPLT2-PLT2–YFP and pPLT2s-PLT2–YFP (a and b indicate independent
transformants, 1 and 2 indicate different roots).
NATURE|Vol 449|25 October 2007 LETTERS
1055
Nature©2007 Publishing Group
This effect is normally limited by RBR. Low RBR levels in the RCH1RBRi
transgenic display limited expansion of the stem cell domain16
because the PLT levels dictated by the gradient are limiting.
Our data indicate that PLT protein gradients define three outputs
in the growing root primordium: stem cell programming, mitotic
activity and exit to differentiation. Analysis of PLT target genes will
be required to assess how much of the response to graded activity is
due to additive concentration effects on the same targets and to
differences in target specificity.
Although the molecular link between auxin action and PLT gene
activation may not be direct3
, auxin distribution and response
systems are essential for correct PLT gene transcription. This raises
the possibility that PLT proteins promote stem cells and transitamplifying
cells as a graded read-out of auxin distribution. In an
accompanying paper, we provide evidence that PIN-mediated polar
auxin transport establishes a dynamic gradient spanning the root
meristem17
. Hence it is tempting to speculate that an auxin gradient
underlies the observed PLT gradients. Classical morphogen systems
were conceptualized as independent from the response system.
However, several gradients in animal development involve complicated
dynamics (for example, ref. 18) and the static concept of positional
information is being challenged19
. We show that PIN polar
auxin transport facilitator expression that is essential for correct
auxin distribution is regulated by PLT activity, which is a clear
example of entanglement between positional information and its
response system.
METHODS SUMMARY
Plant work. plt1-4 and plt2-2 alleles were described in ref. 3, plt3-1, and bbm-1
and bbm-2 are salk T-DNA insertion lines 127417, 097021 and 067917, respectively,
provided by the Signal Insertion Mutant Library (http://signal.salk.edu/).
The T-DNA insertion in PLT3 was confirmed by genotyping. The plt1;plt2;plt3
triple mutant was generated by crossing plt3-1 to plt1-4;plt2-2. bbm-1 and bbm-2
were crossed to plt1-4;plt2-2 and plt3-1 and allelic combinations were selected
from F2 populations. The T-DNA insertion site on bbm-1 and bbm-2 lines was
verified by genotyping. Primers for genotyping are indicated in Supplementary
Table 3. Promoter and genomic sequences were amplified from Col-0 genomic
DNA using the primer combinations listed in Supplementary Table 3.. Promoter
fragments were fused to the endoplasmic reticulum targeted CFP coding
sequence in a pGreenII vector20
. For translational fusions, PLT genomic
sequences were fused at the 39 end to either the YFP coding sequence or the
carboxy-terminal-encoding region of the rat glucocorticoid (GR) receptor21
and
placed under the control of particular promoters (amplified regions are
described in Supplementary Table 3). Promoter swaps were performed by fusing
5.8 kb of PLT2 and 4.2 kb of BBM promoter fragments to the YFP-fused PLT
genomic sequences. Transgenic plants were generated by transforming Col-0
wild-type or plt1-4;plt2-2 plants, as described22
.
Phenotype analysis and microscopy. Light microscopy23
, confocal microscopy
and aniline blue staining24
of mature embryos was performed as described. Root
length was measured, as before3
. Meristem cell length was measured using
ImageJ (v.1.36) and mature cortical cell length as well as fluorescence levels were
determined using Zeiss LSM Pascal (3.2SP2) software.
In situ hybridization. Whole-mount RNA in situ hybridization was performed
as described11
. The PLT3 and BBM riboprobes, specific for non-conserved
sequences downstream of the AP2 repeats, were prepared from templates
amplified from complementary DNA (for primers, see Supplementary Table
3). The PLT1 probe is as in ref. 3; the PIN1, PIN2 and PIN3 probes are as in
ref. 25.
Received 5 July; accepted 30 August 2007.
1. Tabata, T. & Takei, Y. Morphogens, their identification and regulation.
Development 131, 703–712 (2004).
2. Gurdon, J. B. & Bourillot, P. Y. Morphogen gradient interpretation. Nature 413,
797–803 (2001).
3. Aida, M. et al. The PLETHORA genes mediate patterning of the Arabidopsis root
stem cell niche. Cell 119, 109–120 (2004).
4. Skoog, F. & Miller, C. O. Chemical regulation of growth and organ
formation in plant tissues cultured in vitro. Symp. Soc. Exp. Biol. 54, 118–130 (1957).
5. Weigel, D. & Jurgens, G. Stem cells that make stems. Nature 415, 751–754
(2002).
6. Spradling, A., Drummond-Barbosa, D. & Kai, T. Stem cells find their niche. Nature
414, 98–104 (2001).
7. Xu, J. et al. A molecular framework for plant regeneration. Science 311, 385–388
(2006).
8. Sabatini, S. et al. An auxin-dependent distal organizer of pattern and polarity in the
Arabidopsis root. Cell 99, 463–472 (1999).
9. Nole-Wilson, S., Tranby, T. L. & Krizek, B. A. AINTEGUMENTA-like (AIL) genes
are expressed in young tissues and may specify meristematic or divisioncompetent
states. Plant Mol. Biol. 57, 613–628 (2005).
10. Birnbaum, K. et al. A gene expression map of the Arabidopsis root. Science 302,
1956–1960 (2003).
11. Blilou, I. et al. The PIN auxin efflux facilitator network controls growth and
patterning in Arabidopsis roots. Nature 433, 39–44 (2005).
12. Hardtke, C. S. & Berleth, T. The Arabidopsis gene MONOPTEROS encodes a
transcription factor mediating embryo axis formation and vascular development.
EMBO J. 17, 1405–1411 (1998).
13. Hellmann, H. et al. Arabidopsis AXR6 encodes CUL1 implicating SCF
E3 ligases in auxin regulation of embryogenesis. EMBO J. 22, 3314–3325
(2003).
14. Dharmasiri, N. et al. Plant development is regulated by a family of auxin receptor F
box proteins. Dev. Cell 9, 109–119 (2005).
15. Srinivasan, C. et al. Heterologous expression of the BABY BOOM AP2/ERF
transcription factor enhances the regeneration capacity of tobacco (Nicotiana
tabacum L.). Planta 225, 341–351 (2007).
16. Wildwater, M. et al. The RETINOBLASTOMA-RELATED gene regulates stem cell
maintenance in Arabidopsis roots. Cell 123, 1337–1349 (2005).
17. Grieneisen, V. A., Xu, J., Mare´e, A. F. M., Hogeweg, P. & Scheres, B. Auxin
transport is sufficient to generate a maximum and gradient guiding root growth.
Nature doi:10.1038/nature06215 (this issue).
18. O’Connor, M. B., Umulis, D., Othmer, H. G. & Blair, S. S. Shaping BMP morphogen
gradients in the Drosophila embryo and pupal wing. Development 133, 183–193
(2006).
19. Jaeger, J. & Reinitz, J. On the dynamic nature of positional information. Bioessays
28, 1102–1111 (2006).
20. Hellens, R. P., Edwards, E. A., Leyland, N. R., Bean, S. & Mullineaux, P. M. pGreen: a
versatile and flexible binary Ti vector for Agrobacterium-mediated plant
transformation. Plant Mol. Biol. 42, 819–832 (2000).
21. Aoyama, T. & Chua, N. H. A glucocorticoid-mediated transcriptional induction
system in transgenic plants. Plant J. 11, 605–612 (1997).
a b c d e f
g h i
Figure 4 | Inducible expansion of meristem and stem cell area with
PLT2–GR fusions. a–c, 35S-PLT2–GR 7 d.p.g. without dex (a) and 1 d after
5 mM dex application (b, c). Overview shows positioning of ink toner
particles that mark the meristem boundary (black arrowhead) and upper
elongation zone boundary at the onset of induction (b); the elongation zone
boundary is defined as the position where cortical cells rapidly expand.
Induced PLT2–GR roots reveal cell division below the meristem boundary
and incomplete cell elongation (c). d–f, 35S-PLT2–GR;pRCH1-RBR RNAi
plants: 10 d.p.g. without dex revealing the two RBRi-induced stem cell layers
below the quiescent centre (blue arrowhead, inset), asterisk indicates the
quiescent centre (d); with 3 d of dex application, revealing excessive root cap
stem cells (blue arrowhead) and periclinal divisions in the proximal
meristem (e); magnification with ectopic periclinal divisions (f , white
arrowhead). g–i, Duplication of the stem cell area (red arrowheads) and
distal cell types (brown starch granules) in ,10% of 8 d.p.g. 35S-PLT2–GR,
pRCH1-RBRi plants after dex application. Early (g), mid- (h) and late
(i) stages of ectopic stem cell centre; note the prolonged activity of both stem
cell centres (i, inset).
LETTERS NATURE|Vol 449|25 October 2007
1056
Nature©2007 Publishing Group
22. Clough, S. J. & Bent, A. F. Floral dip: a simplified method for Agrobacteriummediated
transformation of Arabidopsis thaliana. Plant J. 16, 735–743 (1998).
23. Willemsen, V., Wolkenfelt, H., deVries, G., Weisbeek, P. & Scheres, B. The HOBBIT
gene is required for formation of the root meristem in the Arabidopsis embryo.
Development 125, 521–531 (1998).
24. Bougourd, S., Marrison, J. & Haseloff, J. Technical advance: an aniline blue staining
procedure for confocal microscopy and 3D imaging of normal and perturbed
cellular phenotypes in mature Arabidopsis embryos. Plant J. 24, 543–550 (2000).
25. Friml, J. et al. AtPIN4 mediates sink-driven auxin gradients and root patterning in
Arabidopsis. Cell 108, 661–673 (2002).
Supplementary Information is linked to the online version of the paper at
www.nature.com/nature.
Acknowledgements We thank the Netherlands Genomics Initiative (M.L.) and the
Portuguese Foundation for Science and Technology (C.G.) for funding,
A. Shimotohno and J. M. Perez-Perez for sharing data and Frits Kindt for
photography.
Author Information Reprints and permissions information is available at
www.nature.com/reprints. Correspondence and requests for materials should be
addressed to B.S. (b.scheres@uu.nl).
NATURE|Vol 449|25 October 2007 LETTERS
1057
Nature©2007 Publishing Group