Current Biology, Vol. 15, 1899–1911, November 8, 2005, ©2005 Elsevier Ltd All rights reserved. DOI 10.1016/j.cub.2005.09.052
Patterns of Auxin Transport and Gene Expression
during Primordium Development Revealed by Live
Imaging of the Arabidopsis Inflorescence Meristem
Marcus G. Heisler, Carolyn Ohno, Pradeep Das,
Patrick Sieber, Gonehal V. Reddy, Jeff A. Long,
and Elliot M. Meyerowitz*
Division of Biology
California Institute of Technology
Pasadena, California 91125
Summary
Background: Plants produce leaf and flower primordia
from a specialized tissue called the shoot apical meristem
(SAM). Genetic studies have identified a large
number of genes that affect various aspects of primordium
development including positioning, growth, and
differentiation. So far, however, a detailed understanding
of the spatio-temporal sequence of events leading
to primordium development has not been established.
Results: We use confocal imaging of green fluorescent
protein (GFP) reporter genes in living plants to
monitor the expression patterns of multiple proteins
and genes involved in flower primordial developmental
processes. By monitoring the expression and polarity of
PINFORMED1 (PIN1), the auxin efflux facilitator, and the
expression of the auxin-responsive reporter DR5, we
reveal stereotypical PIN1 polarity changes which, together
with auxin induction experiments, suggest that
cycles of auxin build-up and depletion accompany, and
may direct, different stages of primordium development.
Imaging of multiple GFP-protein fusions shows
that these dynamics also correlate with the specification
of primordial boundary domains, organ polarity
axes, and the sites of floral meristem initiation.
Conclusions: These results provide new insight into
auxin transport dynamics during primordial positioning
and suggest a role for auxin transport in influencing primordial
cell type.
Introduction
Primordium development is a fundamental process in
plants and animals. It involves a positioning mechanism
that defines where the tissue or organ will arise as well
as changes to growth and differentiation that result in
outgrowth and the correct patterning of cell types. Lateral
primordia in plants, unlike animals, initiate continuously
and at regular positions from the growing tip or
SAM, implying that the positioning mechanism is a dynamic
and potentially self-regulating system. Recently
it has been shown that the plant hormone auxin acts
centrally in this process. Plants genetically or chemically
impaired in their ability to transport auxin fail to
form floral primordia [1]. However, externally applied
auxin can rescue this defect in a position-specific manner,
suggesting that the distribution of auxin in the meristem
flank may determine where primordia arise [2].
Consistent with this proposal is the finding that, in the
*Correspondence: meyerow@caltech.edu
epidermis below initiating primordia, the subcellular localization
of the putative auxin efflux carrier PIN1 is polarized
apically, suggesting auxin is transported toward
primordia from the stem epidermis. Within primordia,
PIN1 is localized basally in the presumptive provascular
cells forming a connection from the primordium to the
vascular tissue of the stem [3, 4]. These and other observations
have led to the proposal that auxin is transported
apically in the epidermis toward developing primordia
that then act as auxin sinks by transporting
auxin into the provascular tissues, thereby depleting
auxin from surrounding cells. Since auxin acts to trigger
primordium development, new primordia are initiated
where auxin levels are highest, that is, at a location furthest
from the previous zone of depletion [3, 5].
Once primordia are positioned, boundaries are established,
and cell types that result in proper organ development
are specified. Genetic analysis has uncovered
a large number of genes involved in these processes.
The SHOOTMERISTEMLESS (STM) gene is required for
SAM formation and maintenance, and its activity is
necessary to keep cells in an undifferentiated state [6].
During primordium specification, STM is locally downregulated,
and this downregulation seems to be required
for differentiation to take place [7, 8]. The CUPSHAPED
COTYLEDON (CUC) genes CUC1, CUC2, and
CUC3 are required to demarcate primordium boundaries
as well as promote meristem formation [9–11]. In plants
mutant for all three genes, cotyledons are frequently
fused and the SAM can fail to develop [10]. Genetic
studies have also identified a number of genes involved
in adaxial/abaxial patterning including adaxially expressed
members of the class III HD-ZIP gene family
such as REVOLUTA (REV) [12–14] as well as the abaxially
expressed YABBY transcription factors such as
FILAMENTOUS FLOWER (FIL) [15–17], KANADI genes
[14, 18, 19], and miRNAs 165 and 166 [20]. In general,
adaxial and abaxial factors are thought to act antagonistically
to set up mutually exclusive expression domains
on either side of the adaxial/abaxial boundary.
However, the mechanism by which the boundary is positioned
is unknown, as is the mechanism specifying
the stereotypical asymmetry across this boundary,
although it is proposed to correspond to a prepattern
of signal emanating from the center of the meristem
[12, 21].
Although we know of many genes involved in various
aspects of primordial development, as discussed above,
we have very little knowledge of either their temporal
order of expression or how their dynamic expression
domains are spatially related to one another. Furthermore,
when gene function is perturbed, previously we
have not been able to examine, at high temporal and
spatial resolution, whether a given phenotype is a direct
or indirect consequence. These types of data are now
becoming available through the development of confocal
imaging techniques and the use of GFP. Live imaging
of Arabidopsis meristem cells recently enabled
an analysis of cell fate and division patterns [22] as well
as an analysis of the effects of various drugs that affect
Current Biology
1900
cell division parameters or meristem differentiation patterns
[23]. With the development of GFP-spectral variants
and spectral analysis, our abilities are enhanced
further by enabling us to document multiple events at
subcellular resolution simultaneously [24]. In this study,
we take advantage of these techniques to investigate
patterns of PIN1 activity and other gene and protein
expression patterns. We find that PIN1 undergoes cycles
of activity that not only give new insight into auxin
transport dynamics during primordium establishment
but also enable us to relate these dynamics to cell-type
patterning during early primordial development in the
inflorescence meristem.
Results
PIN1 Undergoes Cyclic Changes in Polarity
and Expression
Previous work has indicated that primordium positioning
is mediated by localized auxin concentrations that
are in turn patterned by the activity of the auxin efflux
mediator PIN1 [1, 3, 4, 25, 26]. In the meristem, the PIN1
protein is expressed predominantly in the epidermis
and provasculature. In cells located in the vicinity of
incipient primordia, PIN1 protein is localized laterally in
the plasma membranes closest to sites of subsequent
primordial emergence [3]. To investigate PIN1 dynamics,
we used confocal-based live imaging of a functional
pPIN1::PIN1-GFP fusion protein. We assessed PIN1 polarity
by identifying arcs of signal in low-resolution
images that we found, by analysis at high resolution, to
correspond to signal within cell corners (Figure S1) and
therefore to indicate the direction of auxin efflux.
Overall, our observed pattern of pPIN1::PIN1-GFP
expression and polarity resembles previous immunological
results [3], except that we find striking patterns
of PIN1 polarity and expression throughout the meristem
epidermis that change according to the stage of
primordial development occurring at different positions.
For convenience, we use “P” to designate primordia
that have clearly grown out from the meristem,
while “I” designates incipient primordia (assessed by
position) before the onset of rapid growth. Both primordia
and incipient primordia are numbered in order of
appearance, with P1 designating the youngest primordium
and I1 designating the oldest incipient primordium.
From examining pPIN1::PIN1-GFP expression
and polarity in primordia of different ages in multiple
plants, a stereotypical pattern of activity can be discerned
(Figure 1A; rotated and enlarged views of Figure
1A are shown in Figures 1D–1K, 1M, and 1O). Before
an incipient primordium is first marked by pPIN1::PIN1GFP
expression, polarity at this location (I4 in Figures
1D and 1G) is directed from one adjacent site (P2 in
Figure 1D and magnified view in Figure 1G) toward the
other adjacent site (I1 in Figures 1D and 1G). A site is
first marked by relatively high levels of pPIN1::PIN1GFP
(I3 in Figures 1E and 1H) when PIN1-GFP polarity
is directed toward this site from both adjacent sites (P1
and P3 in Figures 1E and 1H). At this stage, pPIN1::
PIN1-GFP is limited to the epidermis (compare expression
at I3 position in Figures 1A and 1C). At slightly later
stages (I2 and I1), pPIN1::PIN1-GFP expression levels
increase (compare positions I2 and I1 to I3 in Figure 1A),
and PIN1-GFP polarity becomes directed toward these
sites from three adjacent primordia (P2, P3, and P5 for
I1 in Figures 1F and 1I). pPIN1::PIN1-GFP expression in
subepidermal layers also becomes detectable at these
stages (positions corresponding to I2 and I1 in Figure
1C). As primordial growth starts to accelerate (P1),
PIN1-GFP in epidermal cells located adaxial to the primordium
reverses polarity from being directed toward
the primordium to being directed away from the primordium
and back toward the meristem center and adjacent
regions (P1 in Figure 1J, compare to previous
stage I1 in Figure 1I). Subsequently, expression becomes
reduced within the primordial epidermis (arrows
in Figures 1K and 1M) but starts to reappear again by
stage two of flower development [27] (arrow in Figure
1O). In subepidermal cells below the primordium, expression
becomes restricted to the presumed provascular
tissues (arrowheads in Figures 1L, 1N, and 1P).
To better understand changes in PIN1 expression
and polarity, we conducted time-lapse confocal imaging
of growing, pPIN1::PIN1-GFP-expressing meristems.
Time-lapse images confirm the stereotypical
pattern of expression build-up followed by decrease at
primordial positions as well as the temporal correlation
between reversals at older primordia (asterisks in Figures
S2A–S2D, in the Supplemental Data available with
this article online) and newly localized pPIN1::PIN1GFP
expression at incipient primordia (arrows in Figure
S2 and Movie S1). However, we also find that pPIN1::
PIN1-GFP expression shifts relative to the underlying
cells in directions that are correlated with PIN1 polarity.
After PIN1-GFP reversal, GFP signal moves from the
site of reversal toward adjacent regions as well as toward
the primordium tip, leaving a zone of low expression
where the reversal originally took place (asterisk in
Figures S2E–S2H). The adjacent regions that subsequently
show increased expression correspond to sites
of incipient primordia (Figures S2A–S2D and Movies S1
and S2).
These observations show that PIN1 behavior associated
with flower primordium development can be divided
into two phases. Initially, the establishment of
PIN1 polarity toward cells destined to form an incipient
primordium is correlated with a reversal in PIN1 polarity
in cells surrounding an older adjacent primordium. PIN1
expression levels then increase in cells located at the
polarity focus, and this is correlated with further PIN1
reversals in adjacent primordia. The second phase is
characterized by a subsequent, rapid reversal in polarity
that occurs predominantly in adaxial cells surrounding
the primordium. As this occurs, pPIN1::PIN1-GFP
expression also shifts abaxially, to form a new focus
of expression at the primordium tip, and adaxially, to
adjacent positions destined to form new primordia.
PIN1 Polarity Changes Are Correlated with Auxin
Distribution Changes
Microarray data indicating that PIN1 may be auxin regulated
in pin1-1 meristems (within 30 min of treatment,
data not shown) as well as the consistent correlation
between PIN1-GFP polarity and pPIN1::PIN1-GFP expression
level prompted us to investigate PIN1 tran-
Auxin Transport and Gene Expression in Arabidopsis
1901
Figure 1. PIN1-GFP Expression and Polarity
Patterns in the Inflorescence Meristem
(A–P) pPIN1::PIN1-GFP expression and polarity
in a single representative meristem. In
color panels, signal intensities are coded
blue to yellow corresponding to increasing
intensity levels, respectively. (A) View of meristem
from above showing a three-dimensional
volume rendering of pPIN1::PIN1-GFP
expression and localization. (B) Identical
view as in (A) indicating the numbering of primordia
(P) and incipient primordia (I) with the
earliest primordium marked as P1 and later
primordia as P2–Pn. Incipient positions are
labeled from the oldest I1 to younger sites In.
(C) View of expression below the epidermis
with same orientation as in (A) and (B). Note
expression in presumptive provascular tissue
(pv arrows) below most positions except
I3 and I4. (D–F) Closer views of regions surrounding
I4, I3, and I1, respectively. (G–J)
Magnified views of cells surrounding I4, I3,
and I1, respectively (same orientation as [D]–
[F]) showing the apparent polarity of PIN1GFP
(arrows) in cells where it could be
clearly discerned. (G) There is no obvious
pattern of polarity associated with position
I4. Instead, polarity in this region is predominantly
directed toward I1 and away from P2.
(H) PIN1-GFP polarity at site I3 is directed toward
I3 from two adjacent positions, P1 and
P3. (I) pPIN1::PIN1-GFP expression at I1 is
stronger compared to I3 (compare I1 in [F]
with I3 in [E]), and this higher expression
level is coincident with PIN1-GFP polarity directed
toward I1 from three adjacent primordia,
P2, P3, and P5. (J) In primordium P1, cells
adaxial to the primordium show a reversed
direction of polarity compared to I1. Instead
of PIN1-GFP being localized toward the primordium
(as in Figure 1I), cells surrounding
the primordium exhibit PIN1-GFP polarity in
the direction away from the primordium, back toward the meristem and adjacent regions (arrows). (K, M, and O) Close-up views of P2, P3,
and P4 from (A). (L, N, and P) Reconstructed medial longitudinal sections (representative section orientation shown by dotted line in [K])
corresponding to the primordia shown in (K), (M), and (O), respectively. (K) At stage P2, pPIN1::PIN1-GFP expression (arrow in [K]) appears
reduced compared to P1 and I1. This pattern remains largely unchanged during stage P3 (arrow in [M]) until stage P4, when signal starts to
reappear in this region (arrow in [O]). Below the epidermis of developing primordial, there is a gradual restriction of pPIN1::PIN1-GFP expression
to the presumptive provasculature (arrowheads in [L], [N], and [P]). Scale bars in (A)–(C), 30 ␮m; in (D)–(J), (K), (M), and (O), 20 ␮m; and
in (L), (N), and (P), 10 ␮m.
scription in response to auxin using quantitative realtime
polymerase chain reaction (PCR). We found that
treatment of dissected wild-type inflorescences with
100 ␮M indole-3-acetic acid (IAA) resulted in PIN1 expression
levels increasing by a factor of 2.3 (2.2–2.3)
after 60 min (Figure 2A). A stronger response occurred
after treatment of pin1-1 apices with 5 mM IAA when,
after 30 min, we detected an upregulation of PIN1 by a
factor of 3.2 (2.9–3.5) (Figure 2A). Although these exogenous
IAA concentrations are much higher than those
thought to occur physiologically, similar exogenous
concentrations of IAA are required to induce a consistent
growth response in both wild-type and mutant apical
meristems [2, 3, 5], suggesting that high exogenous
IAA concentrations applied to shoot tissues are diluted
to physiologically relevant concentrations in vivo. To
examine the effects of exogenous auxin on the PIN1
expression pattern in more detail, we conducted timelapse
confocal imaging of pPIN1::PIN1-GFP-expressing
meristems treated with 5 mM 1-naphthaleneacetic
acid (NAA), a form of auxin that can diffuse into cells.
We were able to detect an increase in expression beginning
after 2 hr and increasing up to 6 hr post treatment
in primordial cells away from which PIN1 polarity
was directed, presumably after reversal (arrows, Figures
2B and 2C). Expression also increased in subepidermal
cells located at primordial boundaries (arrows,
Figures 2D and 2E). We also tested whether the pPIN1::
PIN1-GFP expression pattern was sensitive to the
auxin transport inhibitor N-1-naphthylphthalamic acid
(NPA). After treating pPIN1::PIN1-GFP-expressing meristems
for 6 hr with 100 ␮M NPA, we observed increasing
delocalization of pPIN1::PIN1-GFP expression that
by 14 hr had clearly spread to subepidermal cells surrounding
the presumed vascular tissue that initially
lacked expression (Figures 2F–2I). These experiments
indicate that the PIN1 expression pattern is dependent
on the distribution of auxin in the meristem, which in
Current Biology
1902
Figure 2. PIN1 Expression Responds to Auxin
(A) PIN1 mRNA levels as measured by real-time PCR analysis of dissected wild-type inflorescences immersed in 100 ␮M IAA show greater
than 2-fold upregulation after 60 min relative to mock-treated controls. Identical treatments carried out on pin1-1 mutant apices using 5 mM
IAA in lanolin paste resulted in approximately 3-fold induction after 30 min. (B), (C), (F), and (G) show maximum intensity projections of the
meristem viewed from above, while (D), (E), (H), and (I) show corresponding transverse optical sections below the epidermis, respectively. (B–E)
Response of pPIN1::PIN1-GFP (green) to exogenous auxin. (B and D) pPIN1::PIN1-GFP-expressing meristem before NAA treatment. (C) and
(E) show the same meristem as in (B) and (D) after treatment with 5 mM NAA for 6 hr. Expression becomes delocalized and increases in cells that
previously expressed pPIN1::PIN1-GFP at low levels (arrows in [B]–[E]). This occurs both in the epidermis (compare [B] and [C]) and layers below
(compare [D] and [E]). (F–I) Response of pPIN1::PIN1-GFP to treatment with 100 ␮M NPA for 14 hr. In both the epidermis (F and G) and subepidermal
layers (H and I), there is a delocalization of expression after 14 hr. (J–M) Time lapse of pDR5rev::3XVENUS-N7 (red) and pPIN1::PIN1-GFP
(green) expression together (J and K) and pDR5rev::3XVENUS-N7 (red) alone (L and M). At both the initial time point (J and L) and 12 hr later
(K and M), pDR5rev::3XVENUS-N7 expression initiates when pPIN1::PIN1-GFP expression first marks a new site (arrowheads in [J] and [K]).
After PIN1-GFP reverses polarity in cells adaxial to primordia, primordial expression of pDR5rev::3XVENUS-N7 persists and subsequently
appears in daughter cells of earlier-expressing cells (encircled by broken line in [M]). Expression in nondaughter cells occurs at a later stage
of floral bud development (arrowheads in [M]). (N and O) Recovery of fluorescence after bleaching. Cells located within incipient primordia (I2
in [N]) and more mature primordia (P2 in [N]) that expressed pDR5rev::3XVENUS-N7 were selectively irradiated with 514 nm laser light until
expression became undetectable (circled regions in [O]). Seven hours after bleaching, fluorescence could again be detected in the same cells
at I2 (arrow in [P]) but not in P2. Scale bars in (B), (F), and (J)–(N), 30 ␮m.
Auxin Transport and Gene Expression in Arabidopsis
1903
turn depends on auxin transport. Since pPIN1::PIN1GFP
expression levels are also correlated with the
observed changes in PIN1-GFP polarity, our data suggests
that changes in PIN1 polarity associated with different
stages of primordial development drive dynamic
changes in auxin concentration, which then influence
PIN1 transcription.
To gain further evidence of a correlation between auxin
concentration dynamics and PIN1 expression and polarity,
we visualized the activity of the DR5rev auxin-responsive
promoter driving three tandem copies of VENUS, a
rapidly folding YFP variant [28], fused to a nuclear localization
sequence, here termed pDR5rev::3XVENUS-N7,
in combination with pPIN1::PIN1-GFP. As predicted, we
found that local upregulation of both markers occurred
simultaneously at young sites predicted to form primordia
(arrowheads in Figures 2J and 2K). In contrast to
pPIN1::PIN1-GFP expression, which decreases in primordia
after PIN1-GFP polarity reverses toward the
meristem in adaxial cells, pDR5rev:3XVENUS-N7 expression
persists after reversal (compare circled area in
Figure 2M to corresponding region in Figure 2L). However,
by time-lapse imaging, we found that these
pDR5rev::3XVENUS-N7-expressing cells either corresponded
to or were daughters of cells expressing
pDR5rev::3XVENUS-N7 before reversal (Figure 2L, circled
cells in Figure 2M), suggesting the possibility that
persistent expression might be due to VENUS-N7 perdurance.
To test this hypothesis, we selectively radiated
cells expressing pDR5rev::3XVENUS-N7 at primordial
positions with 514 nm laser light until VENUS
fluorescence became undetectable. We then assessed
the recovery of fluorescence in these cells over time.
Of five meristems tested, three showed fluorescence
recovery specifically in cells within incipient primordia
but not in cells located in adjacent older primordia (Figures
2N–2P), while two plants showed no recovery at
either type of position. Overall, these results support
the proposal that transcription from the DR5 promoter
is active during the early stages of primordial specification
and that, like pPIN1::PIN1-GFP, DR5 expression
decreases around the time that PIN1-GFP polarity reverses.
De novo expression in cells unrelated to those
expressing pDR5rev::3XVENUS-N7 before PIN1-GFP reversal
was detected at a later stage (arrowheads in Figure
2M), again correlated with the reappearance of pPIN1::
PIN1-GFP expression (see also Figures 1A and 1O).
To summarize, both the pDR5rev::3XVENUS-N7 and
pPIN1::PIN1-GFP expression data support the proposal
that reversals in PIN1 polarity correlate with and are
likely to cause changes in auxin levels within developing
primordia.
The Meristem Marker SHOOTMERISTEMLESS (STM)
Is Downregulated as PIN1 Is Upregulated
at Incipient Primordia
The STM gene is required both for meristem formation
and maintenance [6]. In situ hybridization analysis
shows that STM is expressed throughout most of the
inflorescence meristem but is specifically downregulated
at positions where specification of the cryptic floral
bract is thought to occur [29]. To determine when
and where STM is downregulated relative to PIN1, we
visualized an STM-VENUS fusion protein under the
control of the STM promoter, designated pSTM::STMVENUS,
in combination with pPIN1::PIN1-GFP. The
overall expression pattern of the STM marker is similar
to that found by in situ hybridization analysis with
expression encompassing most of the meristem. Examination
of meristems expressing both pSTM::STMVENUS
and pPIN1::PIN1-GFP reveals that pSTM::STMVENUS
expression is downregulated at locations where
pPIN1::PIN1-GFP is upregulated (Figures 3A and 3B).
This complementary pattern starts to become apparent
at the earliest position marked by localized pPIN1::
PIN1-GFP expression and polarity, showing that STM
marks primordial positions with similar timing to PIN1
(Figures 3A and 3B). The reduction in pSTM::STMVENUS
expression in the vicinity of high pPIN1::PIN1GFP
expression is also evident several cell layers below
the meristem epidermis in regions corresponding to the
presumptive provascular cells (Figure 3C). We also note
that pSTM::STM-VENUS expression strengthens in a
boundary-like domain abaxial to those cells in which
PIN1 is directed toward the meristem and away from
the primordium (after reversal). This is most noticeable
in transgenic lines in which pSTM::STM-VENUS expression
levels are low (Figure 3F marked “b”). Lastly, we
also note a reduction in pSTM::STM-VENUS levels toward
the meristem center relative to the periphery in
cells below the L2 (Figure 3C).
CUC2 Boundary Expression and PIN1 Dynamics
Previous work has shown that the CUC1, CUC2, and
CUC3 genes are required for proper organ separation
[9, 10] and that, in mutants in which auxin patterning is
disrupted, the expression patterns of CUC1 and CUC2
are also disrupted, suggesting a link between auxin
patterning and CUC gene function [25, 30, 31].
To start to investigate the temporal and spatial relationship
between the CUC genes and PIN1-mediated
auxin transport patterns, we fused the CUC2 coding
region to VENUS and expressed it under the CUC2 promoter,
designated pCUC2::CUC2-VENUS, together with
pPIN1::PIN1-GFP. pCUC2::CUC2-VENUS is expressed in
boundary-like domains surrounding primordia in a pattern
similar to that previously reported from in situ hybridization
experiments [32] (Figure 3D). However, weak
expression is also detected in the epidermal cells of the
central zone (CZ) (arrow in Figure 3D). Like pSTM::STMVENUS,
we found pCUC2::CUC2-VENUS to be downregulated
at pPIN1::PIN1-GFP expression and polarity
foci (arrowheads in Figure 3D). This downregulation
seems to occur at an earlier stage than pSTM::STMVENUS
and is more extensive. Like pSTM::STM-VENUS,
pCUC2::CUC2-VENUS is specifically upregulated in
those cells near primordia from which pPIN1::PIN1-GFP
polarity was directed after its reversal (Figure 3E).
These data suggest that both CUC2 and STM are
regulated in a similar manner, although with different
timing or sensitivity. Furthermore, they show strikingly
complementary expression patterns compared to PIN1
both at pPIN1::PIN1-GFP foci and in boundary regions
from which pPIN1::PIN1-GFP is directed.
Current Biology
1904
Figure 3. STM and CUC2 Expression
Both pSTM::STM-VENUS (red in [A]–[C] and
[F]) and pCUC2::CUC2-VENUS (red in [D] and
[E]) are expressed in a complementary domain
to pPIN1::PIN1-GFP (green in [B], [C], [E], and
[F]), with expression being absent or reduced
at early positions (arrowheads in [A] and [D])
marked by pPIN1::PIN1-GFP polarity foci (compare
[A] to [B] and [D] to [E]). However, while
pCUC2::CUC2-VENUS expression is low in
subepidermal layers (data not shown), pSTM::
STM-VENUS is expressed at high levels in a
similar pattern to its expression in the epidermis
(C). After PIN1-GFP polarity reverses,
both markers are upregulated in boundary
domains (marked “b” in [B], [E], and [F]) corresponding
to cells abaxial to pPIN1::PIN1GFP-expressing
cells exhibiting polarity away
from the primordium toward the meristem.
When pSTM::STM-VENUS expression is
weak (F), its expression pattern closely resembles
that of pCUC2::CUC2-VENUS (compare
[F] to [E]). Arrow in (D) denotes weak
expression. Scale bars in (A)–(C), 30 ␮m; in
(D)–(F), 20 ␮m.
The Adaxial Marker REVOLUTA Is Localized after
PIN1 and Remains Adaxial from Inception
The REV gene is involved in regulating axillary meristem
formation, organ polarity, and meristem size and is expressed
in a similar pattern to other related members
of the HD-ZIP family of transcription factors such as
PHABULOSA and PHAVOLUTA [12–14, 33–35]. It is also
thought to mark primordia at a particularly young stage
[13]. To monitor REV expression dynamics, we constructed
a functional REV-VENUS fusion and expressed
it under the REV promoter. We found the pREV::REVVENUS
transgene to be expressed in an identical manner
to the reported endogenous transcription pattern
[13]. pREV::REV-VENUS expression is present in the
upper layers of the CZ of the meristem as well as in
finger-like domains extending out to developing primordia
(Figure 4A). pREV::REV-VENUS signal is first evidently
associated with primordial positions between
zero and one plastochron after pPIN1::PIN1-GFP first
localizes, and the expression domain is contiguous
with expression in the CZ (I2 in Figures 4A and 4B).
However, before this, newly localized pPIN1::PIN1-GFP
expression always appears adjacent to pREV::REVVENUS-expressing
cells in the CZ (I3 in Figure 4A, corresponding
position in Figures 4B and 4C). As pPIN1::
PIN1-GFP signal increases at primordial initiation sites,
the pREV::REV-VENUS expression domain extends outward
to overlap with the adaxial half of the pPIN1::PIN1GFP
expression domain (I1 in Figure 4A, corresponding
position in Figures 4B and 4D). This pattern of overlap
also extends below the epidermis along the presumptive
provascular cells marked by newly localized pPIN1::
PIN1-GFP expression (Figure 4D). At approximately the
same time that pPIN1::PIN1-GFP polarity reverses and
expression levels decrease in subepidermal cells, pREV::
REV-VENUS expression is lost from the cells located
between the primordium and the CZ, thus isolating REV
expression from the CZ (P1 in Figure 4A, corresponding
position in Figures 4B and 4E).
To summarize, pREV::REV-VENUS expression extends
halfway into the pPIN1::PIN1-GFP expression domain
from nearby adjacent CZ cells that already express pREV::
REV-VENUS. Adaxial, localized pREV::REV-VENUS expression
associated with pPIN1::PIN1-GFP expression
is then maintained during primordial outgrowth and
separation from the CZ.
The Abaxial Primordium Marker FILAMENTOUS
FLOWER Is Asymmetrically Expressed at Initiation
To determine whether PIN1 expression also marks the
adaxial boundary of abaxially expressed genes, we examined
pPIN1::PIN1-GFP expression in combination
with nuclear localized dsRED under the control of the
FIL promoter. FIL is a member of the YABBY gene family
of transcription factors and functions to specify abaxial
cell types and to promote flower development and
lateral organ expansion [15–17]. pFIL::dsRED-N7 first
marks primordia between one and two plastochrons
later than upregulation of pPIN1::PIN1-GFP (I1 and I2 in
Figure 4F, arrows in Figure 4G). At inception, pFIL::
dsRED-N7 expression is detected abaxial to pPIN1::
PIN1-GFP in the two layers of cells located between
the epidermis and pPIN1::PIN1-GFP-marked provasculature
(Figure 4H). Slightly later, pFIL::dsRED-N7 extends
to the epidermis and into the pPIN1::PIN1-GFP
expression domain (Figure 4I). At later stages, pFIL::
dsRED-N7 expression almost entirely overlaps pPIN1::
PIN1-GFP in the primordium (Figure 4J). To investigate
this more thoroughly, we covisualized pFIL::dsRED-N7,
pREV::REV-VENUS, and pPIN1::PIN1-GFP together (represented
in red, green, and blue, respectively, in Figures
4K–4O). We found that, after pFIL::dsRED-N7 expression
first expands (approximately the same stage as
shown in Figure 4I), both pREV::REV-VENUS and pFIL::
dsRED-N7 are coexpressed in a number of cells centered
within the pPIN1::PIN1-GFP expression domain
(P1 in Figures 4K–4O). These data confirm that both
pREV::REV-VENUS and pFIL::dsRED-N7 are initially expressed
asymmetrically relative to pPIN1::PIN1-GFP,
showing that pPIN1::PIN1-GFP marks a domain be-
Auxin Transport and Gene Expression in Arabidopsis
1905
Figure 4. Relative Expression Patterns of REV, FIL, and PIN1
(C)–(E), (H)–(J), and (L)–(O) represent longitudinal sections of primordia along the planes of sections as depicted by dotted lines in (A), (F), and
(K), respectively. pREV::REV-VENUS (red in [A]–[E]) is expressed in the CZ as well as in radial domains extending out to where pPIN1::PIN1GFP
(green in [A]–[J]) has established expression and polarity foci (compare [A] and [B]). pREV::REV-VENUS expression is localized adaxially
to the first site marked by pPIN1::PIN1-GFP and does not overlap (I3 in [A] and [C]). At later stages, pREV::REV-VENUS expression extends
halfway into the cells expressing pPIN1::PIN1-GFP at high levels (I1 in [A] and [D]) and later separates from the central domain of expression
as subepidermal pPIN1::PIN1-GFP expression also becomes restricted to provascular cells (P1 in [A] and [E]). pFIL::dsRED-N7 (red in [F]–[O])
is expressed in primordia approximately one to two plastochrons after they are first marked by pPIN1::PIN1-GFP (I1 and I2 in [F], arrows in
[G]). pFIL::dsRED-N7 signal first appears in cells located abaxial to pPIN1::PIN1-GFP (I1 in [F] and [H]) but is then found to overlap later during
development (P1 and P4 in [I] and [J], respectively). Plants expressing all three markers (K) show that, compared to pPINI::PIN1-GFP (blue in
[K]–[N]), both the expression patterns of pFIL::dsRED-N7 (red) (M) and pREV::REV-VENUS (green) (N) are ab- and adaxially polarized within
the developing primordium but also overlapping within the pPIN1::PIN1-GFP expression domain (O). [L]–[O] correspond to different combinations
of markers expressed in the primordium P1 in [K]. Scale bars in (A), (F), and (K), 30 ␮m; in (C)–(E) and (H)–(J), 20 ␮m; in (L), 10 ␮m.
tween abaxial and adaxial cell identities at primordial
initiation. Soon after, as the expression domains of REV
and FIL expand into the anlagen, their domains partially
overlap during subsequent development.
Gene Expression Patterns during Early Initiation
of Floral Development
The LEAFY (LFY) gene encodes a transcription factor
involved in conferring floral meristem identity on primordia
that develop from the inflorescence meristem.
In lfy mutants, secondary inflorescence meristems subtended
by floral bracts frequently develop in the place
of flowers [36]. To understand where and when floral
meristem identity is specified as marked by LFY expression,
we examined the expression of the LFY promoter
driving endoplasmic-reticulum (ER)-localized
GFP in combination with pFIL::dsRED-N7. We detect
the onset of pLFY::GFP-ER expression approximately
two plastochrons later than pFIL::dsRED-N7 in a few
adaxial cells abutting the pFIL::dsRED-N7 expression
domain (Figures 5A and 5B). By one plastochron later,
pLFY::GFP-ER expression is much stronger and broader,
and this trend continues during later stages (Figures 5A
and 5B–5D). pLFY::GFP-ER expression is also sharply
bounded on the adaxial side of the primordium by cells
predicted to express the boundary marker CUC2. If this
were so, we would expect a gap to form between the
FIL and CUC2 expression domains at the time LFY expression
initiates. Visualization of pCUC2::3XVENUSN7
and pFIL::dsRED-N7 revealed that, although the expression
of these markers initially abuts (Figures 5E and
5F), later during development, a discernable gap in expression
forms between their domains. This gap continues
to expand during floral meristem development in
a manner similar to the expansion of the pLFY::GFPER
expression pattern (Figure 5H), suggesting that LFY
expression initiates specifically in the region between
the FIL and CUC2 expression domains and that the FIL
Current Biology
1906
Figure 5. pLFY::GFP-ER Expression and the Initiation of Floral Meristem Development
(B)–(D) and (F)–(H) represent longitudinal sections along the planes of sections depicted by dotted line in (A) and (E), respectively. pLFY::GFPER
(green) is initiated two plastochrons after pFIL::dsRED-N7 (red) in a small number of cells located at the adaxial boundary of pFIL:dsREDN7
expression (arrow in [B]). During subsequent development, the pLFY::GFP-ER expression domain expands to encompass the floral meristem
(B–D). (E–H) The pCUC2::3XVENUS-N7 (green) and pFIL::dsRED-N7 (red) domains abut each other early in primordium development (F)
but become separated in a pattern complementary to the expansion of pLFY::GFP-ER expression (compare [A]–[D] to [E]–[H]). (I and J)
During stage two of flower development, pREV::REV-VENUS (red) expression expands (I) as pPIN1::PIN1-GFP (green) also expands (J), while
pFILdsRED-N7 (red) remains restricted to the region marked by the original boundary of pPIN1::PIN1-GFP-expressing cells (K). (L) Also during
stage two, pDR5rev:3XVENUS-N7 expression (red) starts to reappear together with the expansion of pPIN1::PIN1-GFP expression (green). (M
and N) pREV::REV-VENUS (green) and pLFY::GFP-ER (red) overlap during stage two. (O and P) Direction of meristem is toward the top of the
pictures. During late stage two, pSTM::STM-VENUS expression (red) also undergoes a rapid expansion into the flower primordium from a
boundary-like domain (compare [O] and [P]). Scale bars in (A), 30 ␮m; in (B) and (C), 20 ␮m; in (D) and (E), 30 ␮m; in (F)–(P), 20 ␮m.
and CUC2 expression domains move apart, possibly
through the proliferation of cells within the developing
floral meristem. Since pREV::REV-VENUS expression
also abuts pFIL::dsRED-N7 expression, we examined
pREV::REV-VENUS expression during stage two of
flower development [27] and found that its expression
also spreads adaxially in a similar manner to pLFY:GFPER
(compare Figures 5C and 5I, 5M and 5N). This is
further correlated with an expansion of pPIN1::PIN1-GFP
(Figure 5J) and pDR5rev::3XVENUS-N7 expression (Figure
5L). Although pSTM::STM-VENUS expression is still
restricted to an adaxial boundary domain during early
stage two of flower development, before stage three,
pSTM::STM-VENUS expression also expands rapidly
(compare Figures 5O and 5P).
To summarize, the pCUC2::CUC2-VENUS and pFIL::
dsRED-N7 expression domains are located adjacent to
each other, with the boundary between these domains
corresponding to the region where pLFY::GFP-ER expression
later initiates. pLFY::GFP-ER expression then
expands as cells proliferate to form the flower meristem,
and this is correlated with an expansion of
pREV::REV-VENUS and pSTM::STM-VENUS expression
into the primordium as well as an increase in primordial
pDR5rev::3XVENUS-N7 expression.
Discussion
PIN1 Reversals and Primordium Positioning
Our data suggest that, within floral anlagen, there is a
cycle of auxin build-up followed by decrease correlated
with and likely caused by a rapid reversal in PIN1 polar-
Auxin Transport and Gene Expression in Arabidopsis
1907
ity in cells located adaxial to the primordium (Figure
6A). Thus, our results support the proposal that auxin
is depleted from primordial regions during their development
[3, 5]. However, instead of depletion occurring
simply because of auxin transport into the vasculature,
as previously proposed [3, 5], we show that it may result
from specific reversals in PIN1 polarity away from
primordia. Given this result, our data provide new insights
into how PIN1 acts to position primordia. We
show that upregulation of PIN1 marks anlagen first in
the epidermis, suggesting that the patterning mechanism
may act primarily within this cell layer, as is also
suggested by laser ablation experiments [37]. We also
reveal correlations between PIN1 polarity reversals at
older primordia and the establishment of foci at new
sites, suggesting causal links between these events.
Since we find that, in the epidermis, polarity is mainly
directed from older to younger sites, one possible scenario
is that new sites are first specified by a reversal
event at an older primordium. Such an event may be
mediated by the PINOID protein kinase [38, 39], since
it is expressed in boundary domains and can act as a
PIN1 polarity switch [30, 40]. In accordance with this
proposal, it has been known for some time that older
primordia influence the position of younger primordia
[37, 41, 42]. However, this hypothesis does not explain
the simultaneous appearance of regularly positioned
primordia in whorled phyllotaxis. Thus, an important
goal for future investigation will be to determine the direction
of causality between focus establishment and
polarity reversal and to see how disruptions to these
processes affect primordial positioning.
Boundary Establishment, Organ Specification,
and PIN1 Expression
Both STM and CUC gene function are required for meristem
establishment and organ separation [7, 9–11, 43].
Also, ectopic STM and CUC2 expression is associated
with ectopic meristem formation and abnormal development
of lateral organs, indicating that tight regulation
of these genes is important [8, 44, 45]. Genetic studies
have also shown that PIN1 activity in the embryo is required
for CUC2 expression in cotyledon boundaries
[31]. In addition, both CUC2 and STM expression in the
pin-like inflorescence meristem of pin1 mutants is uniform
throughout the periphery [25]. These studies suggest
that both local downregulation and the specific
upregulation of these genes depend on auxin transport
(and therefore auxin levels) in some manner [31]. By
linking STM and CUC2 expression to auxin concentration
and transport dynamics, we reveal how auxin
transport may control these genes. We find that both
pSTM::STM-VENUS and pCUC2::CUC2-VENUS are downregulated
at a time when PIN1 first establishes polarity
foci, coincident with a presumed auxin build-up as evidenced
by pPIN1::PIN1-GFP and pDR5rev::VENUS-N7
expression. Both pCUC2::CUC2-VENUS and pSTM::STMVENUS
are also upregulated in primordial boundaries at
a similar time to when PIN1-GFP polarity reverses away
from the primordia and apparent auxin depletion occurs
in epidermal and subepidermal layers, as evidenced by
PIN1 expression. Thus, there appears to be a consistent
and complementary relationship between the expression
of auxin reporter genes, the direction of presumed
auxin flow, and the activation of STM and CUC2, suggesting
that both the STM and CUC2 expression domains
may be patterned by the cycles of PIN1 focus
establishment and reversal during primordium development
(illustrated in Figure 6A). However, given the limited
time resolution of this study, it is also possible that
CUC2 and/or STM may act upstream of PIN1 to influence
PIN1 behavior. This proposal is also suggested by
the observation that upregulation of CUC1 and CUC2
can lead to increases in organ number [45, 46]. Further
experiments testing the effect of auxin on CUC2 and
STM expression may help to distinguish these possi-
bilities.
Auxin Transport Routes and the Establishment
of Organ Polarity
We find that, in the inflorescence meristem, the polarity
markers pFIL::dsRED-N7 and pREV::REV-VENUS are
expressed predominantly abaxially and adaxially, respectively,
at initiation, although they overlap to some
degree later on. Although this finding contrasts with
some previous reports that polar expression patterns
arise gradually during primordial development [12, 15],
it is consistent with more recent observations showing
that a mutated FIL promoter that drives GFP expression
in both adaxial and abaxial leaf domains also drives
expression more widely than the wild-type promoter
during organ initiation [47]. pREV::REV-VENUS expression
starts in the CZ completely adaxial to pPIN1::PIN1GFP
and then expands into, but does not completely
overlap, the domain of pPIN1::PIN1-GFP-marked cells.
In contrast, pFIL::dsRED-N7 appears to be initiated in
cell layers lying in between pPIN1::PIN1-GFP-marked
cells and the epidermis, although its domain quickly expands
to almost completely encompass the pPIN1::
PIN1-GFP domain (summarized in Figure 6B). These results
show that, in flower primordia, polarity is established
at least as early as the onset of REV and FIL
expression. It is also intriguing to note that pPIN1::
PIN1-GFP marks the abaxial boundary of pREV::REVVENUS
expression. This suggests that, in addition to
specifying primordial positions and boundary domains
(as defined by CUC2 and STM expression), auxin transport
routes also play a role in positioning boundaries
between adaxial and abaxial cell types.
Early Development of the Arabidopsis Flower
Our results indicate that LFY expression arises in a
boundary region between the CUC2 (and presumably
STM) and FIL expression domains. From our analysis of
pFIL::dsRED-N7 and pPIN1::PIN1-GFP, this region also
corresponds to cells marked by PIN1, suggesting yet
another role for auxin transport routes in either directly
or indirectly positioning LFY expression and thus influencing
floral development. During subsequent stages,
this domain undergoes significant expansion. From our
observations, it appears that, during this expansion, the
FIL and CUC2 expression domains do not become reduced,
suggesting either that their expression domains
are moving relative to the LFY-expressing cells or that
the LFY domain is undergoing rapid growth such that
the CUC2 and FIL domains are physically pushed apart
(see Figure 6B). The latter proposal is supported by previous
work showing that, although LFY expression initi-
Current Biology
1908
Figure 6. Conceptual Summary
(A) PIN1 polarity (arrows), presumed auxin concentration dynamics, and CUC2/STM expression levels. High auxin, low STM/CUC2 (yellow);
low auxin, high STM/CUC2 (blue).
(B) Summary of observed expression patterns during early stages of primordial development. Arrows indicated direction of auxin transport
in the epidermis.
ates first by cell recruitment, this is later followed by
expansion through cell proliferation [23]. That LFY may
in fact be required for such growth is suggested by the
reduced size of floral meristems that arise in lfy mutants
[36]. Lastly, it is worth noting that, at the time when
pLFY::GFP-ER expression expands, we also see the reappearance
of pDR5rev::3XVENUS-N7 expression and
a widening and strengthening of pPIN1::PIN1-GFP and
pREV::REV-VENUS. Slightly later, pSTM::STM-VENUS
expression also expands. These events presumably indicate
the onset of floral meristem activity.
Conclusion
Besides determining primordial position, auxin transport
patterns play a crucial role in other developmental
processes including venation patterning [48, 49], establishing
the embryo apical-basal axis [50], promoting differential
growth in response to gravity [51], and formation
of the root meristem [40, 52–54]. Our study shows
that this patterning system also potentially acts to influence
later differentiation events that occur during flower
primordium formation, including boundary formation, organ
polarity, and floral meristem initiation. Thus, auxin
transport routes may provide positional information for
many developmental processes throughout the plant
life cycle. An important goal for future investigation will
be to test our causal hypotheses and understand the
mechanisms that determine the transport routes them-
selves.
Experimental Procedures
Construction of GFP Reporters
The pPIN1::PIN1-GFP fusion construct was made by inserting
mGFP5 (a gift from J. Hasseloff) into a unique XhoI site located in
a 8763 bp genomic DNA fragment obtained by PCR containing the
PIN1 gene (3506 bp upstream of ATG to 2181 bp downstream of
the stop codon). This fragment was then transformed into plants
heterozygous for the pin1-4 mutation using the pART27 binary
transformation vector [55]. Functionality was confirmed by identifying
several T3 families in which the wild-type phenotype strictly
cosegregated with the transgene. 3XVENUS-N7 was constructed
by first PCR amplifying both VENUS [27] and N7, a DNA-fragmentencoding
nuclear-localization protein sequence [56], with restriction
sites and cloning them in frame to yield BamHI-VENUS-4XalaN7-XbaI
(pPD19). The BamHI-XbaI fragment from pPD19 was then
subcloned to pBJ36 [28] to yield pPD31. Separately, VENUS was
also PCR amplified and cloned to pCR2.1-TOPO, flanked by BamHI
and BglII sites (pPD30). The BamHI-BglII fragment from pPD30 was
then ligated to BamHI-linearized pPD31 and screened for clones
with multiple insertions to yield 3XVENUS-4XAlanine-N7 (pPD35).
pDR5rev::3XVENUS-N7 was constructed by cloning the DR5rev
promoter [50] upstream of the 3XVENUS-N7 coding region, and a
NotI fragment containing the reporter construct was then transfer-
Auxin Transport and Gene Expression in Arabidopsis
1909
red to pMLBART [57] and transformed into pPIN1::PIN1-GFPexpressing
plants. For constructing both the pSTM::STM-VENUSand
pREV::REV-VENUS-translational fusions, a version of BJ36
containing a 9X Alanine linker followed by VENUS was made. For
STM, a full-length STM cDNA was cloned and inserted in-frame
next to the Alanine linker. A 5.76 kb region of upstream regulatory
sequence was then inserted upstream of the cDNA. A NotI fragment
containing this reporter was cloned into pMLBART and transformed
into pPIN1::PIN1-GFP as described above. The ability of
this construct to rescue the stm mutant phenotype has not been
tested. pREV::REV-VENUS was constructed by amplifying a 7972
bp genomic DNA fragment containing 3779 bp of sequence upstream
of the ATG as well as the whole genomic coding region into
BJ36 containing 9X Alanine VENUS and transformed into Landsberg
erecta plants expressing pPIN1::PIN1-GFP and rev-1 homozygous
mutants. Although the phenotype of transformed rev-1 plants
was variable, many selected T1 plants exhibited a wild-type phenotype.
We used a CUC2 promoter reporter consisting of 3.2 kb of 5#
regulatory region fused to the 3XVENUS-N7 reporter as described
above. pCUC2::CUC2-VENUS was constructed by switching both
the CaMV 35S promoter for the 3.2 kb CUC2 upstream regulatory
sequence and VENUS for mGFP5 in a previously reported 35S::
CUC2-GFP construct [45]. Subsequently, a 9X Alanine linker was
introduced as a translational fusion into the PstI site between
CUC2 and VENUS and transformed as described above. Functionality
of this fusion protein has not been tested. The pFIL::dsREDN7
marker was constructed by fusing dsRED Express (Clontech)
to the nuclear localizing sequence N7 as described for VENUS-N7
above. DNA (5867 bp) upstream of the ATG start was then cloned
upstream of dsRED-N7 in BJ36 and transformed into pPIN1::PIN1GFP-expressing
plants as described above. pLFY::GFP-ER was
made by amplifying a 2288 bp fragment upstream of the LFY coding
region and inserting it upstream of the mGFP5-ER coding region
(a gift from J. Hasseloff) in the pZP222 vector [58].
Auxin Treatments
For quantitative real-time PCR analysis, inflorescences were dissected
and older flowers removed. The inflorescences were then
submerged in either 1% dimethylsulfoxide (DMSO) or 100 ␮M IAA
(Sigma) dissolved in 1% DMSO. For treatment of pin1-1 apices, an
auxin paste was applied made from 5 mM IAA, 1% DMSO in lanolin
(Sigma), or a mock treatment of 1% DMSO in lanolin. For live imaging,
we found that 1% DMSO alone altered PIN1::GFP expression,
and so both IAA and NAA (Sigma) were dissolved in 1 M KOH
to make 0.5 M IAA or NAA stock, which was then diluted to 5 mM
in H2O. A 1 M stock of NPA (Chem Service) dissolved in DMSO was
diluted in H2O to give a 100 ␮M treatment solution. Approximately
10 ␮l of solution was added to the meristem after each imaging
session every 2 hr. Each treatment experiment was repeated at
least three times with mock treatment controls.
Quantitative Real-Time PCR
Total RNA was extracted using RNAeasy columns (QIAGEN) and
subjected to on-column DNase treatment according to the manufacturer’s
instructions. RNA (1 ␮g) was then reverse transcribed
using oligo dT (20-mer) and SuperScript II Rnase H Reverse Transcriptase
(Invitrogen) according to the manufacturer’s instructions.
The reaction mixture was diluted to 30 ␮l, and 1 ␮l was used per
real-time PCR reaction. Each primer pair was designed to cross
introns in order to eliminate the possibility of genomic DNA amplification.
The primers for PIN1 were 5#-ACAAAACGACGCAGGC
TAAG-3# and 5#-AGCTGGCATTTCAATGTTCC-3#. We used both the
ACTIN2 and ACTIN8 genes as internal controls by using the following
primers that amplify both genes 5#-GGTAACATTGTGCTCAG
TGGTGG-3# and 5#-AACGACCTTAATCTTCATGCTGC-3#(3). Each
PCR reaction contained 12.5 ␮l of SYBR Green PCR Master kit
containing Hot Start Taq polymerase (Perkin-Elmer Applied Biosystems),
3 ␮l of 2.5 ␮M primers, 1 ␮l of cDNA, and 7.5 ␮l of H20 for a
reaction volume of 25 ␮l. Reactions were performed in triplicate on
a GeneAmp 7500 thermocycler (Perkin-Elmer Applied Biosystems).
Data was analyzed using the 2−⌬⌬CT
method [59]. The amplification
efficiencies for the ACTIN2/8 and PIN1 primers were found to be
approximately equal. The data were collected from three independent
treatments and are expressed as a mean fold change and
range in brackets corresponding to the mean −⌬⌬CT ± SEM converted
to a fold-change value.
Tissue Preparation for Confocal Analysis
Inflorescence meristems were prepared for live imaging as described
[22]. Ler plants expressing the GFP markers were generated
by transforming a chosen pPIN1::PIN1-GFP-transgenic line
with all other markers. Triple-marker combinations were generated
by crossing the double-marker plants and screening the F2 families.
For high-resolution images, fixed stained tissue was prepared
by applying 10 ␮g ml−1
FM4-64 (Molecular Probes) to intact inflorescences.
After 30 min, inflorescences were detached and fixed in
4% paraformaldehyde containing 0.1% Tween 20 and 0.1% Triton
X-100 at 4°C for 1 hr. For whole mounts, mature buds were dissected
away and the meristem immersed in 50% glycerol under a
coverslip ready for imaging using a Zeiss Plan-Apochromat 63×/
1.40 NA objective.
Confocal Settings for GFP Combinations
All imaging was done using a Zeiss 510LSM Meta using a waterdipping
63× objective with the plant temporarily under water. For
time-lapse imaging of GFP alone, we used a 488 nm laser line together
with a HFT KP 650 (short-pass) main dichroic and a 505 nm
long-pass filter. The laser was attenuated to 7%, and we used 0.7 s
scans of a 512 × 512 pixel frame. Stacks of 30 sections spaced
approximately 1.5 ␮m apart were collected every 2 hr. In between
imaging, the water was decanted and the boxes containing the
plants placed in a growth cabinet. Non-time-lapse images of GFP
alone were collected as above, except we used a 505–550 nm
band-pass filter and a higher laser setting and smaller pinhole. To
image GFP and dsRED together, we used multitracking in line-scan
mode and a 488/543 main dichroic. We used a 545 secondary dichroic
to split the emission in conjunction with a 543 nm laser line
and 560–610 nm band-pass filter for dsRED, and the 488 nm laser
line and a 505–550 nm band pass filter for GFP. The conditions
for imaging GFP/VENUS combinations depended on the relative
strengths of the markers. For strong VENUS with GFP, both fluorescent
proteins were excited using the 488 nm laser line, and the
emission was separated using the Meta spectral analyser. For
weaker VENUS signal, we used multitracking in frame mode.
VENUS was excited using the 514 nm laser line in conjunction with
a 530–600 band-pass filter. GFP was excited with the 477 nm laser
line and collected using a 545 secondary dichroic in conjunction
with a 505–530 band-pass filter. For triple combinations of GFP,
VENUS, and dsRED, we used frame multitracking. The settings
were as for GFP/VENUS multitracking except with an additional
track for dsRED using the 543 nm laser line in conjunction with a
543 secondary dichroic to separate the excitation light and 560–
610 band-pass filter.
Bleaching Experiments
Images of five meristems were recorded prior to bleaching. Then,
using the Zeiss LSM software, two regions of interest corresponding
to pDR5rev::3XVENUS-N7-expressing cells located in incipient
and older primordia were selected in each of the meristems. These
regions were then continuously scanned using the 514 nm laser
line at 80% power until signal could not be detected using the
settings initially used for imaging the prebleached meristem. After
allowing 7 hr for recovery, we reimaged the meristems to assess
fluorescence recovery.
Image Analysis and Assessment of PIN1-GFP Polarity
Visualization of volume renderings was performed using Zeiss LSM
software and Amira (Mercury Computer Systems). At high resolution,
the subcellular localization of PIN1-GFP could be assessed
either by direct visualization or by analysis with image-processing
software specifically developed for this purpose (H. Jonsson,
M.G.H., B. Shapiro, E.M.M., and E. Mjolsness, unpublished data).
Analysis of meristems imaged in this manner confirms that, in general,
arcs of signal visible in approximately two thirds of meristem
epidermal cells at lower resolution in 3D projections corresponded
to PIN1-GFP protein localization within cell corners as shown in
Current Biology
1910
Figure S1. 3D projections give a clearer indication of lateral subcellular
PIN1-GFP polarity in the meristem compared to 2D optical
transverse sections because the viewing angle can be manipulated
to be normal to the curved epidermal surface. The pattern of PIN1GFP
polarity and expression associated with different stages of
primordium development shown in Figure 1 was representative of
data gathered from greater than 50 plants, as assessed visually.
Supplemental Data
Supplemental Data include two figures and two movies and can be
found with this article online at http://www.current-biology.com/
cgi/content/full/15/21/1899/DC1/.
Acknowledgments
We are grateful to members of the Meyerowitz laboratory for critical
comments and suggestions for the manuscript. We thank Jan Traas
for communicating unpublished results. This work was supported
by Department of Energy grant DOE FG02-88ER13873 and National
Science Foundation grant FIBR 0330786.
Received: August 11, 2005
Revised: September 27, 2005
Accepted: September 27, 2005
Published: November 7, 2005
References
1. Okada, K., Ueda, J., Komaki, M.K., Bell, C.J., and Shimura, Y.
(1991). Requirement of the auxin polar transport system in early
stages of Arabidopsis floral bud formation. Plant Cell 3, 677–
684.
2. Reinhardt, D., Mandel, T., and Kuhlemeier, C. (2000). Auxin regulates
the initiation and radial position of plant lateral organs.
Plant Cell 12, 507–518.
3. Reinhardt, D., Pesce, E.R., Stieger, P., Mandel, T., Baltensperger,
K., Bennett, M., Traas, J., Friml, J., and Kuhlemeier, C.
(2003). Regulation of phyllotaxis by polar auxin transport. Nature
426, 255–260.
4. Benkova, E., Michniewicz, M., Sauer, M., Teichmann, T., Seifertova,
D., Jurgens, G., and Friml, J. (2003). Local, efflux-dependent
auxin gradients as a common module for plant organ formation.
Cell 115, 591–602.
5. Stieger, P.A., Reinhardt, D., and Kuhlemeier, C. (2002). The
auxin influx carrier is essential for correct leaf positioning. Plant
J. 32, 509–517.
6. Endrizzi, K., Moussian, B., Haecker, A., Levin, J.Z., and Laux,
T. (1996). The SHOOT MERISTEMLESS gene is required for
maintenance of undifferentiated cells in Arabidopsis shoot and
floral meristems and acts at a different regulatory level than the
meristem genes WUSCHEL and ZWILLE. Plant J. 10, 967–979.
7. Long, J.A., Moan, E.I., Medford, J.I., and Barton, M.K. (1996).
A member of the KNOTTED class of homeodomain proteins
encoded by the STM gene of Arabidopsis. Nature 379, 66–69.
8. Lenhard, M., Jurgens, G., and Laux, T. (2002). The WUSCHEL
and SHOOTMERISTEMLESS genes fulfil complementary roles
in Arabidopsis shoot meristem regulation. Development 129,
3195–3206.
9. Aida, M., Ishida, T., Fukaki, H., Fujisawa, H., and Tasaka, M.
(1997). Genes involved in organ separation in Arabidopsis: an
analysis of the cup-shaped cotyledon mutant. Plant Cell 9,
841–857.
10. Vroemen, C.W., Mordhorst, A.P., Albrecht, C., Kwaaitaal, M.A.,
and de Vries, S.C. (2003). The CUP-SHAPED COTYLEDON3
gene is required for boundary and shoot meristem formation in
Arabidopsis. Plant Cell 15, 1563–1577.
11. Takada, S., Hibara, K., Ishida, T., and Tasaka, M. (2001). The
CUP-SHAPED COTYLEDON1 gene of Arabidopsis regulates
shoot apical meristem formation. Development 128, 1127–
1135.
12. McConnell, J.R., Emery, J., Eshed, Y., Bao, N., Bowman, J., and
Barton, M.K. (2001). Role of PHABULOSA and PHAVOLUTA in
determining radial patterning in shoots. Nature 411, 709–713.
13. Otsuga, D., DeGuzman, B., Prigge, M.J., Drews, G.N., and
Clark, S.E. (2001). REVOLUTA regulates meristem initiation at
lateral positions. Plant J. 25, 223–236.
14. Emery, J.F., Floyd, S.K., Alvarez, J., Eshed, Y., Hawker, N.P.,
Izhaki, A., Baum, S.F., and Bowman, J.L. (2003). Radial patterning
of Arabidopsis shoots by class IIIHD-ZIP and KANADI
genes. Curr. Biol. 13, 1768–1774.
15. Siegfried, K.R., Eshed, Y., Baum, S.F., Otsuga, D., Drews, G.N.,
and Bowman, J.L. (1999). Members of the YABBY gene family
specify abaxial cell fate in Arabidopsis. Development 126,
4117–4128.
16. Chen, Q., Atkinson, A., Otsuga, D., Christensen, T., Reynolds,
L., and Drews, G.N. (1999). The Arabidopsis FILAMENTOUS
FLOWER gene is required for flower formation. Development
126, 2715–2726.
17. Sawa, S., Watanabe, K., Goto, K., Liu, Y.G., Shibata, D., Kanaya,
E., Morita, E.H., and Okada, K. (1999). FILAMENTOUS
FLOWER, a meristem and organ identity gene of Arabidopsis,
encodes a protein with a zinc finger and HMG-related domains.
Genes Dev. 13, 1079–1088.
18. Eshed, Y., Izhaki, A., Baum, S.F., Floyd, S.K., and Bowman, J.L.
(2004). Asymmetric leaf development and blade expansion in
Arabidopsis are mediated by KANADI and YABBY activities.
Development 131, 2997–3006.
19. Kerstetter, R.A., Bollman, K., Taylor, R.A., Bomblies, K., and
Poethig, R.S. (2001). KANADI regulates organ polarity in Arabidopsis.
Nature 411, 706–709.
20. Kidner, C.A., and Martienssen, R.A. (2004). Spatially restricted
microRNA directs leaf polarity through ARGONAUTE1. Nature
428, 81–84.
21. Sussex, I.M. (1955). Morphogenesis in Solanum tuberosum I:
experimental investigation of leaf dorsiventrality and orientation
in the juvenile shoot. Phytomorphology 5, 286–300.
22. Reddy, G.V., Heisler, M.G., Ehrhardt, D.W., and Meyerowitz,
E.M. (2004). Real-time lineage analysis reveals oriented cell divisions
associated with morphogenesis at the shoot apex of
Arabidopsis thaliana. Development 131, 4225–4237.
23. Grandjean, O., Vernoux, T., Laufs, P., Belcram, K., Mizukami, Y.,
and Traas, J. (2004). In vivo analysis of cell division, cell growth,
and differentiation at the shoot apical meristem in arabidopsis.
Plant Cell 16, 74–87.
24. Berg, R.H. (2004). Evaluation of spectral imaging for plant cell
analysis. J. Microsc. 214, 174–181.
25. Vernoux, T., Kronenberger, J., Grandjean, O., Laufs, P., and
Traas, J. (2000). PIN-FORMED 1 regulates cell fate at the periphery
of the shoot apical meristem. Development 127, 5157–
5165.
26. Galweiler, L., Guan, C., Muller, A., Wisman, E., Mendgen, K.,
Yephremov, A., and Palme, K. (1998). Regulation of polar auxin
transport by AtPIN1 in Arabidopsis vascular tissue. Science
282, 2226–2230.
27. Smyth, D.R., Bowman, J.L., and Meyerowitz, E.M. (1990). Early
flower development in Arabidopsis. Plant Cell 2, 755–767.
28. Nagai, T., Ibata, K., Park, E.S., Kubota, M., Mikoshiba, K., and
Miyawaki, A. (2002). A variant of yellow fluorescent protein with
fast and efficient maturation for cell-biological applications.
Nat. Biotechnol. 20, 87–90.
29. Long, J., and Barton, M.K. (2000). Initiation of axillary and floral
meristems in Arabidopsis. Dev. Biol. 218, 341–353.
30. Furutani, M., Vernoux, T., Traas, J., Kato, T., Tasaka, M., and
Aida, M. (2004). PIN-FORMED1 and PINOID regulate boundary
formation and cotyledon development in Arabidopsis embryogenesis.
Development 131, 5021–5030.
31. Aida, M., Vernoux, T., Furutani, M., Traas, J., and Tasaka, M.
(2002). Roles of PIN-FORMED1 and MONOPTEROS in pattern
formation of the apical region of the Arabidopsis embryo. Development
129, 3965–3974.
32. Ishida, T., Aida, M., Takada, S., and Tasaka, M. (2000). Involvement
of CUP-SHAPED COTYLEDON genes in gynoecium and
ovule development in Arabidopsis thaliana. Plant Cell Physiol.
41, 60–67.
33. Prigge, M.J., Otsuga, D., Alonso, J.M., Ecker, J.R., Drews, G.N.,
and Clark, S.E. (2005). Class III homeodomain-leucine zipper
gene family members have overlapping, antagonistic, and distinct
roles in Arabidopsis development. Plant Cell 17, 61–76.
Auxin Transport and Gene Expression in Arabidopsis
1911
34. Zhong, R., and Ye, Z.H. (2001). Alteration of auxin polar transport
in the Arabidopsis ifl1 mutants. Plant Physiol. 126, 549–
563.
35. Talbert, P.B., Adler, H.T., Parks, D.W., and Comai, L. (1995). The
REVOLUTA gene is necessary for apical meristem development
and for limiting cell divisions in the leaves and stems of
Arabidopsis thaliana. Development 121, 2723–2735.
36. Weigel, D., Alvarez, J., Smyth, D.R., Yanofsky, M.F., and Meyerowitz,
E.M. (1992). Leafy controls floral meristem identity in
Arabidopsis. Cell 69, 843–859.
37. Reinhardt, D., Frenz, M., Mandel, T., and Kuhlemeier, C. (2003).
Microsurgical and laser ablation analysis of interactions between
the zones and layers of the tomato shoot apical meristem.
Development 130, 4073–4083.
38. Benjamins, R., Quint, A., Weijers, D., Hooykaas, P., and Offringa,
R. (2001). The PINOID protein kinase regulates organ
development in Arabidopsis by enhancing polar auxin transport.
Development 128, 4057–4067.
39. Christensen, S.K., Dagenais, N., Chory, J., and Weigel, D.
(2000). Regulation of auxin response by the protein kinase
PINOID. Cell 100, 469–478.
40. Friml, J., Yang, X., Michniewicz, M., Weijers, D., Quint, A., Tietz,
O., Benjamins, R., Ouwerkerk, P.B.F., Ljung, K., Sandberg, G.,
et al. (2004). A PINOID-dependent binary switch in apical-basal
PIN polar targeting directs auxin efflux. Science 306, 862–865.
41. Reinhardt, D., Frenz, M., Mandel, T., and Kuhlemeier, C. (2005).
Microsurgical and laser ablation analysis of leaf positioning
and dorsoventral patterning in tomato. Development 132, 15–
26.
42. Snow, M.S., and Snow, R. (1931). Experiments on phyllotaxis I.
The effect of isolating a primordium. Philos. Trans. R. Soc.
Lond. B Biol. Sci. 221, 1–43.
43. Aida, M., Ishida, T., and Tasaka, M. (1999). Shoot apical meristem
and cotyledon formation during Arabidopsis embryogenesis:
interaction among the CUP-SHAPED COTYLEDON and
SHOOT MERISTEMLESS genes. Development 126, 1563–1570.
44. Daimon, Y., Takabe, K., and Tasaka, M. (2003). The CUPSHAPED
COTYLEDON genes promote adventitious shoot formation
on calli. Plant Cell Physiol. 44, 113–121.
45. Baker, C.C., Sieber, P., Wellmer, F., and Meyerowitz, E.M.
(2005). The early extra petals1 mutant uncovers a role for
microRNA miR164c in regulating petal number in Arabidopsis.
Curr. Biol. 15, 303–315.
46. Mallory, A.C., Dugas, D.V., Bartel, D.P., and Bartel, B. (2004).
MicroRNA regulation of NAC-domain targets is required for
proper formation and separation of adjacent embryonic, vegetative,
and floral organs. Curr. Biol. 14, 1035–1046.
47. Watanabe, K., and Okada, K. (2003). Two discrete cis elements
control the abaxial side-specific expression of the FILAMENTOUS
FLOWER gene in Arabidopsis. Plant Cell 15, 2592–2602.
48. Sachs, T. (1981). The control of the patterned differentiation of
vascular tissues. Adv. Bot. Res. 9, 151–162.
49. Sieburth, L.E. (1999). Auxin is required for leaf vein pattern in
Arabidopsis. Plant Physiol. 121, 1179–1190.
50. Friml, J., Vieten, A., Sauer, M., Weijers, D., Schwarz, H., Hamann,
T., Offringa, R., and Jurgens, G. (2003). Efflux-dependent
auxin gradients establish the apical-basal axis of Arabidopsis.
Nature 426, 147–153.
51. Friml, J., Wisniewska, J., Benkova, E., Mendgen, K., and Palme,
K. (2002). Lateral relocation of auxin efflux regulator PIN3 mediates
tropism in Arabidopsis. Nature 415, 806–809.
52. Blilou, I., Xu, J., Wildwater, M., Willemsen, V., Paponov, I., Friml,
J., Heidstra, R., Aida, M., Palme, K., and Scheres, B. (2005).
The PIN auxin efflux facilitator network controls growth and
patterning in Arabidopsis roots. Nature 433, 39–44.
53. Friml, J., Benkova, E., Blilou, I., Wisniewska, J., Hamann, T.,
Ljung, K., Woody, S., Sandberg, G., Scheres, B., Jurgens, G.,
et al. (2002). AtPIN4 mediates sink-driven auxin gradients and
root patterning in Arabidopsis. Cell 108, 661–673.
54. Aida, M., Beis, D., Heidstra, R., Willemsen, V., Blilou, I., Galinha,
C., Nussaume, L., Noh, Y.S., Amasino, R., and Scheres, B.
(2004). The PLETHORA genes mediate patterning of the Arabidopsis
root stem cell niche. Cell 119, 109–120.
55. Gleave, A.P. (1992). A versatile binary vector system with a
T-DNA organisational structure conducive to efficient integration
of cloned DNA into the plant genome. Plant Mol. Biol.
20, 1203–1207.
56. Cutler, S.R., Ehrhardt, D.W., Griffitts, J.S., and Somerville, C.R.
(2000). Random GFP: cDNA fusions enable visualization of
subcellular structures in cells of Arabidopsis at a high frequency.
Proc. Natl. Acad. Sci. USA 97, 3718–3723.
57. Eshed, Y., Baum, S.F., Perea, J.V., and Bowman, J.L. (2001).
Establishment of polarity in lateral organs of plants. Curr. Biol.
11, 1251–1260.
58. Hajdukiewicz, P., Svab, Z., and Maliga, P. (1994). The small,
versatile pPZP family of Agrobacterium binary vectors for plant
transformation. Plant Mol. Biol. 25, 989–994.
59. Livak, K.J., and Schmittgen, T.D. (2001). Analysis of relative
gene expression data using real-time quantitative PCR and the
2(–Delta Delta C(T)) method. Methods 25, 402–408.