The PIN auxin efflux facilitator network
controls growth and patterning in
Arabidopsis roots
Ikram Blilou
1
, Jian Xu
1
*, Marjolein Wildwater
1
*, Viola Willemsen
1
*, Ivan Paponov
2
, Jir
ˇ
ı
´
Friml
3
, Renze Heidstra
1
, Mitsuhiro Aida
1
,
Klaus Palme
2
& Ben Scheres
1
1
Department of Molecular Genetics, Utrecht University, Padualaan 8, 3584CH Utrecht, The Netherlands
2
Institut fu
¨
r Biologie II , Universita
¨
t Freiburg, Scha
¨
nzlestrasse1, 79104, Freiburg, Germany
3
Zentrum fu
¨
r Molekularbiologie der Pflanzen, Universita
¨
tTu
¨
bingen, Auf der Morgenstelle 3, 72076 Tu
¨
bingen, Germany
* These authors contributed equally to this work
...........................................................................................................................................................................................................................
Local accumulation of the p lant growth regulator auxin mediat es patter n formation in Arabidopsis roots and influences outgrowth
and development of lateral root- and shoot-derived primordia. However, it has remained unclear how auxin can simultaneously
regulate patterning and organ outgrowth and how its distribution is stabilized in a primordium-specific manner. Here we show that
five PIN genes collectively control auxin distribution to regulate cell division and cell expansion in the primary root. Furthermore,
the joint action of these genes has an important role in pattern formation by focusing the auxin maximum and restricting the
expression domain of PLETHORA (PLT) genes, major determinants for root stem cell specification. In turn, PLT genes are required
for PIN gene transcription to stabilize the auxin maximum at the distal root tip. Our data reveal an interaction network of auxin
transport facilitators and root fate determinants that control patterning and growth of the root primordium.
In Arabidopsis root development, a distal auxin maximum corre-
lates with pattern formation and the orientation and extent of cell
division
1
. Inhibition of polar auxin transport strongly affects these
processes
1
. The initiation of lateral roots and leaf primordia is also
associated with changes in auxin transport
2–4
. These observations
point to polar auxin transport as a major factor in organ formation.
Recent analyses of the Arabidopsis PIN proteins revealed their
association with auxin maxima in distal domains of organ primor-
dia
2–4
. The PIN genes are thought to encode components of the
auxin efflux machinery mediating polar auxin transport, as con-
cluded from the polar localization of PIN membrane proteins and
auxin uptake experiments
5–11
. PIN proteins might participate
directly in auxin transport or help in the assembly of other proteins
with efflux activity such as the AtMDR/PGP proteins
12,13
. Mutations
in the ARF-GEF GNOM, required for vesicle transport of PIN1,
also lead to developmental defects that resemble those caused by
interfering with auxin transport
14–16
.
All single pin mutants described so far display different weak
phenotypes in primary roots
4–8
, in contrast with gnom mutants
and polar auxin transport inhibitor treatments. Here, we investigate
the collective contribution of five PIN genes in the control of
cell division and cell expansion during root outgrowth. Further-
more, we assess how the PLETHORA (PLT) genes, which encode
auxin-inducible AP2-domain putative transcription factors necess-
ary and sufficient for stem cell specification in the root primor-
dium
17
, respond to and regulate PIN gene activity during pattern
formation.
PIN protein localization is changed in pin mutants
The PIN proteins described so far are expressed in specific but
overlapping regions of the root meristem
6–8
. PIN1 mainly resides at
the basal end of the vascular cells
8
but weak PIN1 signals can be
detected in the epidermis and the cortex (Fig. 1b). PIN2 localizes
apically in epidermal and lateral root cap cells and predominantly
basally in cortical cells
6
(Fig. 1e, f). PIN3 is expressed without
pronounced polarity in tiers two and three of the columella cells, at
the basal side of vascular cells and to the lateral side of pericycle cells
of the elongation zone
7
(Fig. 1h, i). PIN4 is detected around the
quiescent centre and cells surrounding it, and localizes basally in
provascular cells
8
(Fig. 1k). PIN7 resides at lateral and basal
membranes of provascular cells in the meristem and elongation
zone, whereas in the columella cells it coincides with the PIN3
domain (Fig. 1j).
Auxin distribution appears to be altered in pin mutants
1,8
and
differences in auxin homeostasis affect PIN2 expression
18
.We
therefore determined whether pin mutants have altered expression
of remaining PIN proteins. In pin3pin4pin7, enhanced PIN1 protein
was detected in lateral-basal membranes of the endodermis (Fig. 1c,
d). Moreover, ectopic PIN2 protein was detected at the basal end of
provascular cells that normally express PIN3 and PIN7 (Fig. 1g).
PIN4 expression expands to tier three of the columella cells in pin3
single mutants and pin2pin3 double mutants with a membrane
localization similar to PIN3 in the wild type (Fig. 1l and data not
shown). In pin3pin7, PIN4 expands to the lateral root cap (Fig. 1m).
Our data show that defects in pin mutants can be masked by
ectopic activity of the remaining PIN genes. A comprehensive
mutant analysis is therefore necessary to uncover full gene function.
To this end, we generated all mutant combinations for PIN1, PIN2,
PIN3, PIN4 and PIN7, which group together within the eight-
member PIN gene family. We verified phenotypes in independent
allelic combinations to exclude influences of background modifiers
(Supplementary Table 1).
PIN genes control cell division zone size in the root meristem
Classical experiments revealed that externally added auxins can
stimulate cell division
19
but it is unknown whether internal auxin
distribution regulates cell division in primordia. PIN genes are
required for outgrowth of all organs
2,3,20
, and we assessed whether
this reflected their contribution to the in vivo regulation of cell
division. In roots, oriented cell divisions accompanied by a low rate
of cell expansion occur in the distal meristem zone (Fig. 2A).
pin1 and pin2 single mutants display a slight reduction of root
articles
NATURE | VOL 433 | 6 JANUARY 2005 | www.nature.com/nature 39
© 2005 Nature Publishing Group
length and root meristem size (Supplementary Fig. 2) whereas pin3,
pin4 and pin7 single mutants only display subtle division defects in
the quiescent centre and columella root cap
8
(Supplementary
Fig. 1d–h). Most double-mutant combinations show additive
defects in orientation of cell division, root length and root meristem
size (Supplementary Figs 1 and 2). However, pin1pin2 and all triple
and quadruple mutants containing pin2 show more-than-additive
reduction in root size and root meristem size (Fig. 2B, C and
Supplementary Fig. 2a, b), suggesting that PIN2 plays a pivotal role
in cell division control. The pin1pin2 double mutant suggests that
the role of PIN2 in meristem size control is masked in the pin2 single
mutant mainly by the activity of PIN1 in the PIN2 domain (Fig. 1b).
PIN2 is a main component for mediating proximal (basipetal)
auxin transport (Fig. 1e, f, n), which implies that basipetal transport
to meristematic cells has a critical role in meristem length regu-
lation. Consistent with the notion of auxin shortage in the meristem
zone, treatment with auxins restored meristem size of pin1pin2 and
pin2pin3pin7 to that of wild type (data not shown). Our data
substantiate physiological evidence indicating a role for basipetal
auxin transport in root growth
21,22
and identify control of cell
division as a major factor in this process.
The extreme reduction in root meristem size in pin2 mutant
combinations is reminiscent of the phenotype obtained upon
diphtheria toxin-mediated genetic ablation of root cap cells
23
.In
such plants we found high DR5::GUS accumulation in the pro-
vascular tissue (Fig. 2E, F), supporting the notion that lateral auxin
redistribution does not occur in the absence of columella cells with
laterally oriented PIN proteins.
Our data imply that basipetal transport and lateral redistribution
of auxin are both critical for maintenance of the meristem zone. A
potential mechanism to transport auxin to every cell in the
meristem is indicated by PIN2 localization at basal membranes of
cortex cells (Fig. 1e, f) and the expression domains of vascular PIN3
and PIN7, which suggest an ‘auxin reflux’ loop (Fig. 1h, j, n). We
investigated auxin transport in the root by expressing the bacterial
auxin biosynthetic enzyme IAAH under the WOX5 (ref. 24)
promoter, which allows induction of auxin biosynthesis in the
quiescent centre (Fig. 2G) by external addition of the precursor
IAM
25
. After different time points of induction, we monitored auxin
accumulation using the DR5-GFP reporter. In line with the
observed polar orientation of the PIN proteins, enhanced auxin
responses appear first in the columella region, subsequently in the
lateral root cap and then in the epidermis; enhanced responses in
provascular strands only emerge at later stages (Fig. 2H, a–d).
Epidermal and provascular auxin responses could be reduced by
the polar transport inhibitor N-naphthylphthalamic acid (NPA),
suggesting that accumulation of newly synthesized auxin in these
regions is due to an NPA-sensitive auxin transport loop (Fig. 2H, e).
In pin2 mutant background the induced responses are restricted to
the lateral root cap and epidermis, confirming a role for basipetal
transport of auxin towards provascular cells in the root meristem
(Supplementary Fig. 3).
Our results suggest that the capacity to circulate auxin through
loop-oriented PIN efflux facilitators regulates meristem size. Such
a loop system can redeploy auxin and hence operate at least
partially independently from shoot auxin supply, consistent with
the ability of isolated root systems to maintain growth without
external auxin application. In a wider perspective, these data
support that PIN protein localization predicts auxin transport
routes.
PIN genes regulate cell expansion and root elongation zone size
The capacity to stimulate cell expansion upon external application
has been a defining property for auxins
19
. Recent analysis of tropic
responses revealed that PIN genes play a part in differential auxin
distribution, which is accompanied by differential cell expan-
sion
7,9,11,26
. It is, however, unknown whether PIN genes also regulate
general cell expansion during organ growth. In roots, rapid cell
expansion without cell division occurs in the elongation zone
located proximal to the meristem zone (Fig. 2A).
Final cell size is affected in several pin mutants, but no additive
effects occur in mutant combinations, suggesting that PIN gene
action on cell expansion is complex. Nevertheless, in single allelic
combinations of pin1pin3 and pin3pin7, the change in final cell
length is the major factor accounting for the reduction in root
length (Supplementary Fig 2, red fonts). In contrast to mature cell
size, the size of cells immediately after departure from the meristem
is not affected in any pin mutant combination (Supplementary
Fig. 2).
We observed a reduction in the size of the elongation zone in pin
mutant combinations, which mostly correlated well with reduction
in meristem size. However, in pin3pin4pin7 only the elongation
zone size is reduced (Supplementary Fig. 2, blue fonts). These data
Figure 1 PIN expression and protein localization in roots of wild-type (WT) and pin mutant
Arabidopsis. a, Arabidopsis root meristem with columella (col), quiescent centre (qc),
lateral root cap (lrc), epidermis (e), cortex (c), endodermis (en) and vascular bundle (v).
b–d, PIN1 immunolocalization: b, wild type; inset shows boxed area enlarged.
Arrowheads depict polar localization. c, d, pin3pin4pin7
(Allelic Combination 1(AC1))
; boxed
area in c is enlarged in d. Arrowheads in d, PIN1 upregulation in the endodermis.
e–g, PIN2 protein: e, f, Wild type; boxed area in e is enlarged in f. Arrowheads in f, apical
PIN2 in the epidermis and basal in the cortex. g, pin3pin4pin7
(AC1)
; arrowhead shows
ectopic expression. h, i, PIN3 protein in wild-type root meristem. Arrowheads in h, PIN3 in
the pericycle. j, PIN7–green fluorescent protein (GFP) fusion. k–m, PIN4 protein: k, wild
type; white arrow, wild-type differentiated columella cell without PIN4. l, pin3 (white
arrow); m, pin3pin7
(AC1)
: black arrow, PIN4 expansion to the lateral root cap; white arrow
as in k. Insets: enlargements of boxed areas showing details of PIN4 localization.
n, Localization of PIN proteins suggests auxin transport routes. PIN1 (green), PIN2 (red),
PIN3 (yellow), PIN4 (violet) and PIN7 (blue). Immunolocalization signals are green in b–g
and k–m and red in h, i.
articles
NATURE | VOL 433 | 6 JANUARY 2005 | www.nature.com/nature40
© 2005 Nature Publishing Group
indicate that the region where cell elongation occurs can be
independently controlled by PIN gene activity.
Together, our data reveal that modulation of PIN activities can
separately affect meristem size, elongation zone size and final cell
size. These effects are not additive but probably result from
interactions between changes in auxin distribution and transcrip-
tional or translational responses influencing carrier components.
We conclude that PIN-mediated modulation of auxin distribution
controls both cell division and cell elongation and thereby contrib-
utes to the ‘organizing’ role of auxin in organ growth.
PIN genes regulate PLT expression and pattern the distal root
meristem
Polar auxin transport is a major contributor to root meristem
patterning in Arabidopsis, and the specification of distal cell types
correlates well with the auxin response maximum
1
. Close to the
auxin maximum, the quiescent centre maintains surrounding cells
as stem cells (Fig. 1a). Quiescent centre and stem cell specification
require SHR and SCR, putative GRAS family transcription fac-
tors
27,28
and the redundantly acting PLT1 and PLT2 AP2-domain
putative transcription factors
17
. PLT transcript accumulation is
Figure 2 PIN genes control meristem size and patterning in Arabidopsis roots.
A–F, Meristem size control. A, Wild-type meristem zone (MZ), elongation zone (EZ) and
differentiation zone (DZ). B, Columella cell staining. C, D, pin2pin3pin7
(AC1)
: arrow in C,
border of MZ and EZ; black arrowhead in D, columella stem cells. E, F, DR5::GUS
expression in wild type (E) and pRCP1::DT-A
tsM
(root cap expressed diphtheria toxin,
‘diftox’) (F). G, WOX5 promoter specificity for quiescent centre. H, a–e, DR5::GFP in
WOX5::IAAH plants. a, Control. b–d, DR5::GFP upregulation (arrowheads) after IAM
application. e, No provascular upregulation on 50
m
M NPA. I–L, Patterning in pin2 mutant
combinations. I, QC25::CFP (cyan fluorescent protein) in pin2pin3pin4
(AC2)
. J, SCR::YFP
(yellow fluorescent protein) in pin1pin2pin3pin7
(AC1)
. K, L, In wild type (K) and in
pin2pin3pin4
(AC2)
(L), PLT1 transcript is restricted to the quiescent centre and stem cells.
M, N, DR5::GFP in wild type (M) and pin2pin3pin7
(AC1)
(N). O –X, pin mutants with
patterning defects. Wild type treated with NPA (R). In 33% of pin1pin4pin7
(AC1)
DR5::GUS
(P), starch-granule-containing columella cells (S) and PLT1 expression (W) expand
proximally (white arrowhead). In pin1pin4pin7
(AC2)
, columella cells also expand laterally
(T). In pin3pin4pin7
(AC1)
, mature columella cells (U), their stem cells (arrowheads in V),
DR5::GFP expression (Q) and PLT1 mRNA (X) expand laterally. PLT1 transcript: whole-
mount in situ hybridization (blue/purple signals). Differentiated columella cells: starch
granule staining (purple). DR5::GUS, blue. GFP fluorescence, green. qc, quiescent centre.
A–F, K, L, O, P, R–X, Nomarski optics; G, H, I, J, M, N, Q, CLSM after propidium iodide
staining.
articles
NATURE | VOL 433 | 6 JANUARY 2005 | www.nature.com/nature 41
© 2005 Nature Publishing Group
correlated with auxin accumulation in embryos and primary and
lateral roots, and depends on auxin response factors
17
. The identi-
fication of these critical factors for specification of distal cell types as
well as the highly organized PIN gene expression domains in the
root primordium set the stage for an analysis of the role of PIN genes
in pattern formation.
In the previously discussed pin2 mutant combinations, meristem
size is affected but distal patterning is normal, as judged by the
presence of columella stem cells (Fig. 2D), the QC25 marker for the
quiescent centre (Fig. 2I), SCR promoter activity (Fig. 2J), PLT1
(ref. 17) transcript distribution (Fig. 2K, L) and the DR5 auxin
response marker (Fig. 2M, N).
In pin1pin4pin7, however, the auxin response maximum and the
starch granules that mark the differentiated columella cells shift
proximally in 40% of roots (Fig. 2P, S). In pin3pin4pin7, the auxin
response maximum shifts laterally (Fig. 2Q), associated with
inappropriate lateral up-regulation of PIN1 (Fig. 1c, d). Correlated
with this lateral expansion, a broadening of the columella domain
including its stem cells occurs (Fig. 2U, V). Consistent with changes
in distal patterning, PLT1 messenger RNA shifts proximally in
pin1pin4pin7 (Fig. 2W) and expands laterally in pin3pin4pin7
(Fig. 2X).
The phenotypes of both triple mutants resemble effects of
treatment with inhibitors of polar auxin transport (Fig. 2R) and
support a role for PIN proteins in focusing and stabilizing an auxin
maximum in both proximo-distal and lateral dimensions. Our data
strongly suggest that the PIN proteins in this way focus expression
of the auxin-inducible PLT genes in the distal root region, which
specifies the position of the quiescent centre and stem cells.
PIN genes restrict PLT mRNA and root identity to the basal
embryo pole
We addressed whether PIN genes control PLT expression and root
stem cell patterning during embryogenesis. Distal quiescent centre
and columella cells originate from daughter cells of the hypophysis
at the early globular stage of embryogenesis (Fig. 3a–d). At that
stage an auxin perception maximum is detected in the hypophysis
4
.
Figure 3 PIN genes and embryonic patterning. a–d, Wild-type embryo at 2-cell (a)
globular (b) early heart (c) and torpedo stage (d). hyp, hypophysis; qc, quiescent centre;
col, columella; sc, stem cells; psc, provascular stem cells. e–h, pin2pin3pin4pin7
(AC1)
embryos: e, aberrant divisions in basal cells at octant stage; f, basal cell duplication at
globular stage; g, h, quiescent centre and stem cell division defects at heart (g) and
torpedo stages (h). i–l, pin1pin3pin4pin7
(AC1)
embryos with abnormal basal (i) and apical
cell divisions (j–l). m–t, Gene expression in pin quadruples: m, PLT1 mRNA in wild-type
preglobular embryos is restricted to basal cells; n, ubiquitous PLT1 in
pin2pin3pin4pin7
(AC1)
proembryos; p, o, STM in pin2pin3pin4pin7
(AC1)
(p) and wild type
(o); q, PLT1 in pin1pin3pin4pin7
(AC1)
; r–t, In comparison to wild type (r), WUS is slightly
reduced (s) or expands (t). u–x, Explanted embryos: u, wild type develops normal
seedlings; v, pin2pin3pin4pin7
(AC1)
develops reduced cotyledons (arrowheads) and
ectopic root hairs (arrow); w, x, pin1pin3pin4pin7
(AC1)
explants have reduced root
development and ectopic shoot-like structures at the apex (green tissues). PLT1, STM and
WUS transcripts: in situ hybridization (purple). Images: Nomarski optics.
Figure 4 PLT genes regulate PIN transcript levels. a–d, PIN4 mRNA localization. PIN4
transcript in the quiescent centre, stem cells and provascular initials of wild-type embryo
(a) and seedlings (c). No detectable transcripts in plt1plt2 double-mutant embryos (b) and
seedlings (d). e–l, PIN3 and PIN7 mRNA in columella cells and provascular region of wild-
type embryo (e, i) and seedlings (g, k, arrows). plt1plt2 with normal levels in embryonic
columella (f, j) and seedlings (h, l). Reduction of provascular expression in seedlings (h, l).
Images: Nomarski optics.
articles
NATURE | VOL 433 | 6 JANUARY 2005 | www.nature.com/nature42
© 2005 Nature Publishing Group
Clonal analysis and stereotyped cell division patterns show that the
proximal stem cells are recruited around mid-heart stage (Fig. 3c)
29
.
We found no penetrant embryonic defects in single, double and
triple pin mutants and in pin1pin2pin3pin7 and pin1pin2pin3pin4.
However, in lines segregating pin2pin3pin4pin7 and pin1pin3pin4-
pin7 mutants, marked changes were observed in embryonic division
patterns and gene expression (Fig. 3e–t and Supplementary Table 2).
pin2pin3pin4pin7 mutant embryos exhibit cell division defects
mainly in the basal region (Fig. 3e–h) and ,30% produce viable
seedlings. In contrast, pin1pin3pin4pin7 mutants are embryo lethal
and cell division defects occur in apical and basal embryo regions
4
(Fig. 3i–l). The significance of the altered cell divisions in both
quadruple-mutant embryos in terms of patterning was investigated
by locating the expression domain of marker genes for root and
shoot identity. PLT1 transcript marks the basal domain of the octant
stage embryo (Fig. 3m) and restricts to the quiescent centre and
stem cell domain before mid-heart stage
17
. STM and WUS are
required for shoot meristem function and are transcribed in shoot
meristem precursor cells from early embryogenesis onwards
30,31
.
pin2pin3pin4pin7 mutants contain both aberrant cell divisions
and high levels of PLT1 transcript throughout the embryo from the
16-cell stage onwards (Fig. 3n). In contrast, PLT1 mRNA is correctly
excluded from the apical region in pin1pin3pin4pin7 mutants
(Fig. 3q). Interestingly, the mRNA localization of WUS and STM
does not change in pin2pin3pin4pin7 mutants (Fig. 3o, p), whereas
in pin1pin3pin4pin7, WUS transcripts are either slightly reduced
(Fig. 3s) or expanded in apical embryonic cells (Fig. 3t).
In explanted wild-type embryos, the shoot and root apical
meristems develop normally (Fig. 3u). In ,40% of pin2pin3pin4-
pin7 mutants, the explants develop reduced cotyledons and root
hairs emerge at more apical positions (Fig. 3v and Supplementary
Tab le 2). In ,80% of pin1pin3pin4pin7 mutants, the explants
completely arrest root growth and expand the shoot domain
(Fig. 3w, x and Supplementary Table 2). The explant phenotypes
of both mutants match the observed expansion of PLT1 and WUS
domains, because ectopic expression of these genes promotes root
and shoot identity, respectively
17,32
. PLT gene expression is strongly
dependent on the joint action of PIN proteins, in line with its
dependence on auxin response factors and the correlation between
PLT expression and auxin accumulation
17
. Most probably the PLT
expression domain is regulated by PIN4 and PIN7, which are
appropriately positioned for basal auxin transport at the pre-
globular stage
4
. PIN3 and PIN2 are not expressed at this stage but
ectopic PIN2 mRNA can be detected in pin3pin4pin7 siliques (data
not shown), suggesting that ectopic expression of PIN members in
embryos provides a remarkably versatile compensatory mechanism
for the loss of PIN4 and PIN7.
PIN-regulated early PLT, WUS and STM transcription suggests
that auxin transport in the embryo regulates the proper expression
of critical root and shoot stem cell regulators. The notion that early
cellular asymmetries in membrane localization of PIN proteins are
translated into the patterning of embryonic stem cell domains via
regulation of auxin flux provides a conceptual framework for initial
events in plant embryogenesis.
PLT genes regulate PIN gene expression in the root meristem
The PLT genes are required for specification of the stem cell niche
and convey root identity when ectopically expressed
17
. Therefore we
asked whether the PLT proteins could regulate the root-specific
distribution of PIN transcripts, thereby fine-tuning the position of
the stem-cell-associated auxin maximum, and cell division and cell
expansion domains. Remarkably, PIN4 transcript, which overlaps
with the PLT transcripts in wild type, is undetectable in 94% and
aberrant in 6% of the embryos and seedlings of plt1plt2 (Fig. 4a–d).
PIN3 and PIN7 transcripts are normal in columella cells (Fig. 4e–l)
but are markedly reduced in the provascular domain of the post-
embryonic root elongation zone (Fig. 4g, h, k, l). Thus, the PLT
genes control PIN mRNA distribution.
Discussion
Our findings suggest an elegant mechanism for embryonic root
primordium formation and stabilization. In our model, PIN pro-
teins restrict PLT expression in the basal embryo region to initiate
root primordium formation (Fig. 5a). In turn, PLT genes maintain
PIN transcription, which stabilizes the position of the distal stem
cell niche (Fig. 5b). At a distance from the auxin maximum, PLT
genes maintain PIN3 and PIN7, which reinforce provascular acro-
petal auxin flux. In this way a ‘reflux’ loop is created that controls
auxin distribution in the growing primordium and meristem
(Fig. 5b, c). The loop stabilizes the auxin maximum and the PLT-
dependent stem cell domain in the distal root tip. Moreover, it
localizes meristem and cell expansion zones in the proximal
meristem and regulates final cell size (Fig. 5c).
Reporter genes and direct auxin measurements are consistent
with the presence of a transport-regulated auxin gradient in the root
meristem
1
, but local auxin biosynthesis and catabolism may also
contribute to the auxin concentration profile
33
. Furthermore, auxin
response may be regulated differently in the different root zones by
processes such as differential SCF
TIR1
-mediated proteolysis
34
. A
Methods
Materials
All pin mutants were in Columbia (Col-0) background, except for the Enkheim allele pin1-
1. pin1-1, pin1En134; eir1-1, pin2En701; and pin3-3, pin4-2 were described in refs 7–10.
pin3 salk_005544 and pin7 salk_048791 were provided by the Signal Insertion Mutant
Library (http://signal.salk.edu/cgi-bin/tdnaexpress/). plt1-4 and plt2-2 alleles were
described in ref. 17.
Triple mutants were generated by crossing double mutants sharing one allele, while
quadruple mutants were generated by crossing triple homozygotes with two alleles in
common. DR5::GUS (DR5-
b
-glucuronidase) was described in ref. 1 and was crossed with
pin1, pin2, pin4 and pin7. Homozygous lines were then used to generate double and triple
mutants homozygous for DR5::GUS. DR5-GFP, described in ref. 26, was transformed to
wild type, pin2pin3pin7 and pin3pin4pin7. Promoter constructs of QC25::CFP and
SCR::YFP were transformed to pin2pin3pin4 and to pin1pin2pin3pin7 respectively.
WOX5::GFP and WOX5::IAAH constructs were generated by fusing a 4.5-kb WOX5
(ref. 24) promotor fragment in front of GFPor IAAH
25
in the pGreenII0229 (ref. 35) vector
and transformed into the wild type or DR5::GFP lines.
Phenotype analysis and microscopy
Plant material for light microscopy was prepared as in ref. 36. Starch granules and
b
-glucoronidase activity were visualized as in ref. 36. For embryo phenotype analysis,
Figure 5 Model for primordium formation by PIN–PLT interactions. a, PIN-mediated root
primordium specification by restriction of PLT transcripts in octant/16-cell embryo stage.
b, At later stages of embryogenesis, PIN action further restricts PLT transcripts to define
the stem cell region and PLT genes start controlling root-specific PIN gene expression.
c, In post-embryonic roots, PIN-mediated auxin transport stabilizes the stem cell region
and regulates cell division (DIV) in the meristem zone and cell expansion in the elongation
zone (EL). PLT genes control several members of the PIN gene family to generate
primordium-specific auxin distribution.
articles
NATURE | VOL 433 | 6 JANUARY 2005 | www.nature.com/nature 43
© 2005 Nature Publishing Group
