Institute of Molecular Plant Sciences, Leiden University, Clusius
Laboratory, P.O. Box 9505, 2300 RA Leiden, The Netherlands
 |
INTRODUCTION |
During development of multicellular
organisms, most cells enter specific differentiation programs that
overtly change their structure and specialize them toward a distinct
function. Cell differentiation is preceded by the process of cell fate
commitment, in which the developmental potential of the cell becomes
restricted to a certain fate or subset of fates, but the explicit
demonstration and realization of that developmental pathway is not yet
apparent (Maclean and Hall, 1987
). Cell fate commitment has been
divided into the processes of specification and determination (Slack, 1991). However, the definitions of these processes are, by nature, operational, and knowledge of the molecular nature underlying the
effective functional properties of specified or determined cell states
in plants is still mostly lacking.
The vascular tissues of plants represent an attractive system
for the study of cell fate commitment. In fact, vascular tissues consist of several distinct cell types arranged in a network of continuous strands to form a system of exquisitely complex topography (Steeves and Sussex, 1989). Yet, all types of highly specialized vascular cells derive from an anatomically homogeneous population of
meristematic cells, the procambium, or provascular tissue. We have
previously developed assays with which different stages of procambial
cell fate commitment in rice (Oryza sativa) could be
distinguished (Scarpella et al., 2000). However, despite the detailed
studies available on different aspects of vascular development (for
review, see Sachs, 1981, 2000; Fukuda, 1996, 1997; Nelson and Dengler,
1997; Berleth et al., 2000; Aloni, 2001; Dengler, 2001; Dengler and
Kang, 2001; Kuriyama and Fukuda, 2001; Ye, 2002), the molecular
mechanisms underlying procambial cell fate commitment remain elusive
(e.g. Savidge, 2001). Nevertheless, it is possible that the plant
hormone auxin may be involved in this process, because a number of
studies have provided evidence for a role of this hormone in different
aspects of vascular tissue development (for review, see Sachs, 1981;
Berleth et al., 2000).
One key feature of auxin action is its translocation from cell to
cell in a polar fashion, a process that is referred to as polar auxin
transport (PAT; Goldsmith, 1977; Lomax et al., 1995; Muday and DeLong,
2001; Muday and Murphy, 2002). In the shoot, auxin moves
unidirectionally through the vascular tissues from the apex to the
base. In the root, two distinct polarities of PAT exist. Auxin moves
acropetally through the central vascular cylinder and basipetally
through the outer layers of root cells. Auxin entry into cells is
facilitated by an auxin influx carrier that is thought to be
encoded by AUX1 (Marchant et al., 1999) and possibly by
related genes (Parry et al., 2001). Auxin moves out of plant cells
through an efflux carrier apparatus that is sensitive to synthetic PAT
inhibitors (PATIs), such as 1-N-naphthylphthalamic acid (NPA), and requires the activity of at least two
polypeptides (Morris, 2000; Muday and DeLong, 2001; Muday and Murphy,
2002). The first is an integral membrane transporter thought to be
encoded by one of the members of the PIN gene family (Palme
and Gälweiler, 1999). Both the AUX1 and the PIN proteins show an
asymmetric localization in the plasma membrane that is consistent with
a role in controlling the polarity of auxin movement (Gälweiler
et al., 1998; Muller et al., 1998; Swarup et al., 2001; Friml et al.,
2002). The second protein component of the auxin efflux carrier
apparatus performs a regulatory function, and represents a
high-affinity binding site for PATIs such as NPA (Rubery, 1990).
Several studies indicate that this NPA-binding protein (NBP) is a
peripheral membrane protein associated with the cytosolic face of the
plasma membrane (Morris, 2000; Muday, 2000).
We have previously proposed a role for the auxin-inducible homeobox
gene Oshox1 of rice in the regulation of provascular cell fate specification (Scarpella et al., 2000). First, we have shown that
Oshox1 expression is switched on at a specific, critical stage of procambial development and that cells marked by
Oshox1 expression have been specified but not stably
determined toward vascular differentiation. We have subsequently
studied the effects of transgenic expression of Oshox1 under
control of the cauliflower mosaic virus 35S promoter on vascular
development in the root. The 35S promoter drives Oshox1
expression also in procambial cells close to the root tip, which in
wild type do not yet express Oshox1 and are still in an
uncommitted stage of development. The result of ectopic
Oshox1 expression was that both cell fate specification and
determination occurred closer to the root tip, as a consequence of
which a premature vascular differentiation was observed. Also in the
shoot, ectopic Oshox1 expression similarly shifted vascular differentiation closer to the apex. Importantly, the premature vascular
differentiation in the shoot and root appeared as an extremely subtle
phenotype detected only by careful anatomical inspection, and occurred
without any effect on morphology or growth characteristics. Therefore,
the 35S-Oshox1 plants are particularly useful to investigate the
mechanism underlying provascular cell fate commitment. A comparison of
auxin physiology between wild type and 35S-Oshox1 showed that
Oshox1 overexpression is associated with a decreased
sensitivity of PAT toward inhibition by NPA (Scarpella et al., 2000).
We have proposed that this might represent an aspect of procambial cell
fate commitment and that Oshox1 might contribute to this
process by stabilizing procambial cell fate toward endogenous modulations of PAT.
To test our hypothesis of Oshox1 function, we have examined
here the response of leaf vascular pattern formation under conditions of reduced PAT by inhibitors of three different classes in wild-type and 35S-Oshox1 seedlings, and found that both size and spacing of veins
are under the control of PAT in wild-type rice and that Oshox1 overexpression confers insensitivity to the
vascular-patterning defects evoked by the different PATIs. Furthermore,
here, we present evidence that direct effects of Oshox1
overexpression are to reduce the affinity with which the NBP binds NPA
and to enhance PAT and its sensitivity toward auxin. These results
suggest that Oshox1 might promote fate commitment in
procambial cells by increasing their auxin conductivity properties and
by simultaneously stabilizing this newly acquired state against
modulations of PAT by an endogenous NPA-like molecule.
| |
RESULTS |
Effects of Different Classes of PATIs on Wild-Type Rice Seedling
Development
It is possible to interfere with PAT through a number of
chemically heterogeneous compounds collectively referred to as
PATIs. Most of these PATIs belong to two groups, namely the
phytotropins, exemplified by NPA, and the morphactins, typified by
2-chloro-9-hydroxyfluorene-9-carboxylic acid (HFCA).
However, not all PATIs can be included in these two classes. The
most common example is represented by 2,3,5-triiodobenzoic acid (TIBA).
We have previously studied rice seedlings overexpressing
Oshox1, a proposed regulator of procambial cell fate
commitment, and reported that both the processes of PAT and
adventitious and lateral root development displayed reduced sensitivity
toward the inhibitory effects of NPA (Scarpella et al., 2000). To test whether this property could be extended to other classes of PATIs or to
other developmental processes, we decided to monitor the effects of
different concentrations of the PATIs NPA, HFCA, and TIBA on wild-type
and 35S-Oshox1 seedling development. Because the effects of these PATIs
have not previously been studied in rice, we will first describe their
influence on wild-type rice seedling development.
Germination of wild-type rice seedlings in the presence of PATIs
resulted in pronounced and reproducible effects on their development.
In all cases, the leaves of seedlings germinated on 5 or 10 µM of PATIs appeared shorter and narrower than the untreated ones. Furthermore, the leaves were frequently rolled and
displayed different types and degrees of perturbations of the
blade-sheath boundary, such as hypomorphism or complete absence of the
ligule and displacement or absence of the auricles (e.g. Fig.
1, B and C). In addition to common
effects of the different classes of PATIs, we could also recognize the
presence of seemingly class-specific phenotypes. Seedlings germinated
on 5 or 10 µM NPA showed dwarf and coiled shoots (Fig. 1,
E and F). Germination of wild-type seedlings on concentrations as low
as 1 µM of HFCA induced shoot waving and premature
development of axillary shoots (Fig. 1, G-I). The shoots of seedlings
germinated on TIBA did not show obvious deviation from the untreated
ones (Fig. 1, J-L). The roots of seedlings germinated on 1 or 5 µM NPA or HFCA were short and agravitropic (Fig. 1, D, E,
G, and H), whereas those of seedlings germinated on 10 µM
NPA or HFCA appeared stunted and swollen (Fig. 1, F and I).
Furthermore, both of these PATIs caused a reduction of the number of
adventitious and lateral roots (Fig. 2).
Roots of seedlings germinated on TIBA showed only a slight
agravitropism, and the effect on root elongation was less pronounced
(Fig. 1, J-L). Furthermore, TIBA inhibited lateral root formation, but
not adventitious root development (Fig. 2).
View larger version (27K):
[in this window]
[in a new window]
|
Figure 1.
Effects of PATIs on wild-type rice seedling
morphology. Seedlings germinated and grown for 2 weeks on medium
without PATIs (A) or supplemented with different concentrations of
PATIs (D-L). B and C, Detail of the fourth leaf of seedlings
germinated and grown for 4 weeks on medium without PATIs (B) or
supplemented with 5 µM TIBA (C). G and I, Arrowheads
point at axillary shoots. a, Auricle; l, ligule. Scale bars in A and D
through L = 5 mm; in B and C, scale bars = 1 mm.
|
|
View larger version (15K):
[in this window]
[in a new window]
|
Figure 2.
Effects of PATIs on root system complexity of
wild-type rice. A, Number of lateral roots bore by the seminal root. B,
Number of adventitious roots. C, Number of lateral roots bore by the
adventitious roots. The results represent the mean ± SE of two separate experiments each performed on a
population of 15  n 30 seedlings per
treatment. Difference between treated and untreated wild-type
populations was determined by one-way ANOVA followed by Dunnett's test
and was significant (P < 0.001) at all concentration
points, except in B, where the TIBA-treated population was not
significantly different from the untreated population.
|
|
Effects of PATIs on Vascular Pattern Formation in Wild-Type Rice
Leaves
PATIs have been reported to have a strong impact on leaf vascular
pattern formation in dicot plants (Mattsson et al., 1999; Sieburth,
1999). Therefore, we decided to investigate whether PATIs had a similar
influence on leaf vascular tissue organization in rice, a monocot
species. To address this question, the fourth leaf of untreated and
PATI-germinated wild-type seedlings was cleared and inspected with
dark-field illumination for the presence of vascular pattern
alterations. In rice, the fourth leaf is the first one that is
initiated postembryonically (Hoshikawa, 1993; E. Scarpella, S. Rueb,
and A.H. Meijer, unpublished data). We chose this leaf for our
analysis, because the position and width of procambial strands can only
be affected before the emergence of these strands during leaf
development, whereas the strands are insensitive to PATI-application
once they are anatomically distinguishable (Mattsson et al., 1999). The
fact that the fourth leaf of rice has not been initiated yet at the
moment of PATI application excludes that our treatments might have been
performed at a late and thus unresponsive stage of leaf development.
Wild-type rice leaves show the typical striate venation pattern, in
which longitudinal veins of three orders, the midvein and the large and
small veins, lie parallel along the proximo-distal axis of the leaf and
are connected transversely by minor (commissural) veins (Kaufman, 1959;
Fig. 3A). The distribution and
arrangement of these classes of veins follow a highly regular pattern,
which can be described by a series of venation pattern parameters, such as the distance between longitudinal or transverse veins and the number
of large veins and that of small veins in between two adjacent large
ones. Vein morphology in PATI-treated leaves deviated from that of the
untreated ones in that longitudinal veins appeared thicker (Fig. 3B).
This broadening of the vascular bundles often made it very difficult to
unambiguously distinguish between small and large veins in the leaf. As
a consequence, it became impossible to determine with certainty the
number of large veins and that of small veins in between two adjacent
large ones. However, PATI-treated rice leaves did show vascular
tissue-patterning defects, in that at rising concentrations of PATIs,
the distance between transverse veins decreased and that between
longitudinal veins increased (Fig. 3, A and B). All three classes of
PATIs showed qualitatively similar defects, but quantitative analysis
of the leaf venation pattern parameters was restricted to NPA-treated
leaves (Fig. 3, C and D).
View larger version (59K):
[in this window]
[in a new window]
|
Figure 3.
Effects of PATIs on the leaf vascular pattern of
wild-type rice. A and B, Detail of the vascular pattern of the fourth
leaf of seedlings germinated and grown for 4 weeks on medium without
PATIs (A) or containing 10 µM NPA (B), as viewed with
dark-field illumination after clearing. C and D, Effect of NPA
concentration on the distance between longitudinal veins (LV; C) and on
that between transverse commissural veins (CV; D). Arrows in A
exemplify how the distance between each adjacent pair of LVs or CVs was
measured in all experiments. The results represent the mean ± SE of two separate experiments each performed on a
population of 15 n 30 seedlings per
treatment. Difference between treated and untreated wild-type
populations was determined by one-way ANOVA followed by Dunnett's test
and was significant (P < 0.001) at all concentration
points. Scale bars = 200 µm.
|
|
Effects of PATIs on Root Development and Leaf Vascular Patterning
in 35S-Oshox1 Rice Seedlings
When germinated under the same conditions of PAT inhibition as for
wild type, 35S-Oshox1 seedlings did not show any appreciable deviation
from wild-type behavior as to shoot morphology, root elongation, and
gravitropism (data not shown). However, significant differences between
35S-Oshox1 and wild type were observed when the effects of PATIs on
root system complexity were evaluated. In all cases, adventitious as
well as lateral root development in 35S-Oshox1 seedlings appeared more
resistant to low concentrations of PATIs (Fig.
4). The venation pattern parameters of
the fourth leaf of wild-type and 35S-Oshox1 seedlings germinated in the
absence of PATIs were not significantly different (Fig.
5, C and D). Furthermore, when we
assessed the effect of PATIs on vein morphology in 35S-Oshox1 seedlings, we could observe, similarly as in wild type, hypertrophy of
longitudinal veins of all orders (Fig. 5, A and B). However, in
contrast to wild type, the distance between longitudinal or transverse
veins was not significantly altered in 35S-Oshox1 by germination on
PATIs. Again, all three classes of PATIs induced qualitatively similar
defects, but quantitative analysis of the leaf venation pattern
parameters was restricted to NPA-treated leaves (Fig. 5, C and
D).
View larger version (23K):
[in this window]
[in a new window]
|
Figure 4.
Effects of PATIs on root system complexity of
wild-type and 35S-Oshox1 rice. A, Number of lateral roots bore by the
seminal root. B, Number of adventitious roots. C, Number of lateral
roots bore by the adventitious roots. Three independent 35S-Oshox1
lines were used in all experiments and considered as different
genotypes in the subsequent statistical analysis. The results represent
the mean ± SE of two separate experiments each
performed on a population of 15 n 30 seedlings per genotype and per treatment. Difference between wild-type
and the three independent 35S-Oshox1 populations was determined by
two-way ANOVA followed by Bonferroni's test. No significant difference
in the behavior of the three different 35S-Oshox1 lines used was
detected. Untreated wild-type and 35S-Oshox1 populations were not
significantly different. Difference between PATI-treated wild-type and
35S-Oshox1 populations at the 1 µM
concentration point was significant at P < 0.001 in
all cases, except for TIBA treatment in B. At the 5 µM concentration point, difference was
significant only for NPA treatment in B and for TIBA treatment in A and
C.
|
|
View larger version (62K):
[in this window]
[in a new window]
|
Figure 5.
Effects of PATIs on the leaf vascular pattern of
wild-type and 35S-Oshox1 rice. A and B, Detail of the vascular pattern
of the fourth leaf of wild-type (A) and 35S-Oshox1 (B) seedlings
germinated and grown for 4 weeks on medium containing 5 µM NPA, as viewed with dark-field illumination after
clearing. C and D, Effect of NPA concentration on the distance between
longitudinal veins (LV; C) and on that between transverse commissural
veins (CV; D) in wild type and 35S-Oshox1. In all experiments, the
distance between each adjacent pair of LVs or CVs was measured as
illustrated in Figure 3A. Three independent 35S-Oshox1 lines were used
in all experiments and were considered as different genotypes in the
subsequent statistical analysis. Data represent the mean ± SE of two separate experiments each performed on a
population of 15 n 30 seedlings per genotype
and per treatment. Difference between treated wild-type and 35S-Oshox1
populations was determined by two-way ANOVA followed by Bonferroni's
test and was significant at 0.01 P < 0.05 at
the 1 µM concentration point and at 0.001 P < 0.01 at the 5 µM
concentration point in A, and significant at P < 0.001 at all concentration points in B. No significant difference in the
behavior of the three different 35S-Oshox1 lines used was detected.
Untreated wild-type and 35S-Oshox1 populations were not significantly
different. Scale bars = 200 µm.
|
|
NPA-Binding Constants in 35S-Oshox1 Rice Seedlings and
Suspension-Cultured Cells
PATIs of the phytotropin and morphactin classes have been reported
to interfere with PAT through the binding to a membrane-localized NBP
(Rubery, 1990). To determine whether the reduced NPA sensitivity displayed by 35S-Oshox1 seedlings was associated with changes in
NBP-binding constants, NPA binding to microsomal membranes from
wild-type and 35S-Oshox1 seedlings was assayed. Analysis of the
NPA-binding data revealed that the affinity of the NBP toward NPA,
described by the affinity constant Ka, was
reduced in 35S-Oshox1 seedlings, whereas the number of NPA-binding
sites (i.e. the amount of NBP), estimated by the
Bmax, was not significantly altered (Table
I). However, it was possible that the
reduced Ka might have been, at least
partly, a consequence of other aspects of the 35S-Oshox1 phenotype. In
fact, 35S-Oshox1 seedlings display a premature vascular differentiation
associated with reduced PAT capacity (Scarpella et al., 2000), both of
which features might affect the NBP-binding constants (Suttle, 1991;
Ruegger et al., 1997; Zhong and Ye, 2001). Therefore, to exclude any
influence of overt vascular differentiation in our studies, we decided
to study the effect of Oshox1 overexpression in a rice cell
suspension culture, where vascular differentiation does not occur (data
not shown). Overexpression of Oshox1 in a cell suspension
system had no appreciable effect on cell morphology, nor induced
vascular differentiation (data not shown), suggesting that
Oshox1 expression per se is not sufficient for vascular
differentiation to occur. Although not affecting cell morphology,
overexpression of Oshox1 did alter functional properties of
the suspension-cultured cells. Like in seedlings, Oshox1
overexpression in a cell suspension system resulted in a reduced
sensitivity toward NPA, as measured by the reduced amount of auxin that
35S-Oshox1 cells accumulate in the presence of NPA, when compared with
the wild type (Fig. 6B). Furthermore,
NPA-binding assays performed on microsomal membrane preparations from
wild-type and 35S-Oshox1 cells confirmed that, as in seedlings, the
Ka for NPA binding was reduced in
35S-Oshox1 cells, whereas the Bmax was not
significantly affected (Table I).
View this table:
[in this window]
[in a new window]
|
Table I.
Binding constants of the NBP in wild type and
35S-Oshox1
The affinity constant Ka corresponds to the
inverse of the dissociation constant Kd
(Ka = 1/Kd). Three
(seedling) or two (cell suspension) independent 35S-Oshox1 lines were
used in all experiments and were considered as different genotypes in
the subsequent statistical analysis. Data represent the mean ± SE of three experiments, each of which performed in
duplicate. Asterisks indicate the significance of the difference
between wild-type and 35S-Oshox1 populations as determined by two-way
ANOVA followed by Bonferroni's test. No significant difference in the
behavior of the two (cell suspension) or three (seedling) different
35S-Oshox1 lines used was detected. **, 0.001 P < 0.01.
|
|
View larger version (8K):
[in this window]
[in a new window]
|
Figure 6.
Auxin influx and efflux physiology in wild-type
and 35S-Oshox1 rice suspension-cultured cells. A, Auxin influx. B,
NPA-induced auxin accumulation. C, Auxin efflux. Two independent
35S-Oshox1 lines were used in all experiments and considered as
different genotypes in the subsequent statistical analysis. Data
represent the mean ± SE of six (A), five (B), or
three (C) separate experiments. Difference between treated wild-type
and 35S-Oshox1 populations was determined by two-way ANOVA followed by
Bonferroni's test and was significant at P < 0.001 in
B and C and not significant in A. No significant difference in the
behavior of the two different 35S-Oshox1 lines used was detected.
|
|
PAT Capacity in 35S-Oshox1 Rice Seedlings and Suspension-Cultured
Cells
Because the cell suspension system conveniently allows evaluating
the direct effects of Oshox1 by uncoupling them from any secondary overt phenotypical effect, we decided to investigate whether
other aspects of auxin transport physiology were altered by
Oshox1 overexpression in the rice cell suspension system. As in 35S-Oshox1 seedlings (Fig. 8A), no significant changes in auxin influx capacity were observed in 35S-Oshox1 cells, when compared with
wild type (Fig. 6A). We previously reported that in roots of 10-d-old
35S-Oshox1 seedlings, PAT capacity was reduced and vascular
differentiation occurred closer to the root tip (Scarpella et al.,
2000). However, it was not possible to discriminate between cause and
effect. Therefore, we decided to monitor auxin efflux capacity in
wild-type and 35S-Oshox1 cells, where vascular differentiation does not
occur. We unexpectedly observed a clear enhancement of auxin efflux in
35S-Oshox1 cells (Fig. 6C), suggesting that this is likely to be
a direct effect of Oshox1 and that the reduced PAT capacity
that we previously measured in roots of 10-d-old 35S-Oshox1 seedlings
is likely to be a consequence, rather than a cause, of the premature
vascular differentiation. To confirm this interpretation in planta, we
carefully monitored the course of vascular development in wild-type and
35S-Oshox1 roots during the first 10 d post-germination (dpg), to
determine when exactly the premature vascular differentiation induced
by Oshox1 overexpression could be first anatomically
detected. By means of confocal microscopy, we monitored the elongation
of the procambial precursor of the central late metaxylem element,
which is the first procambial cell in the root vascular cylinder that
shows overt anatomical signs of an ongoing vascular differentiation
process (Kawata et al., 1978; Scarpella et al., 2000). As inferred by
the elongation of this xylem procambial precursor, we could detect a
clear premature vascular differentiation in roots of 7-d-old 35S-Oshox1
seedlings (Fig. 7, C and D), whereas
roots of 3-d-old 35S-Oshox1 seedlings were anatomically
indistinguishable from wild type (Fig. 7, A and B). Because the 35S
promoter is equally active in 3- and 7-d-old root tip regions (data not
shown), we conclude that the premature vascular differentiation
observed in 10-d-old 35S-Oshox1 roots cannot be interpreted as a direct
consequence of the Oshox1 ectopic expression. To evaluate
the primary effect of Oshox1 on PAT capacity and its
sensitivity toward NPA inhibition, we therefore decided to measure
these physiological parameters in the anatomically indistinguishable
wild-type and 35S-Oshox1 3-d-old roots. Consistent with what observed
in the cell suspension system, 35S-Oshox1 3-d-old roots showed an
increased PAT capacity and a reduced sensitivity to the inhibitory
effects of NPA, compared with wild type (Fig. 8, B and C).
View larger version (107K):
[in this window]
[in a new window]
|
Figure 7.
Vascular differentiation in the root of wild-type
and 35S-Oshox1 seedlings 3 or 7 dpg. A through D, Detail of the
vascular cylinder in the region between 261 and 488 µm below the root
tip (root cap excluded) in a longitudinal confocal laser scanning
microscopic optical section. In all panels, the cell file of the
central late metaxylem precursor is positioned at the middle of the
section. Three independent 35S-Oshox1 lines were used for anatomical
inspection. Scale bar = 50 µm.
|
|
View larger version (16K):
[in this window]
[in a new window]
|
Figure 8.
Auxin influx and PAT physiology in wild-type and
35S-Oshox1 roots of 3-d-old seedlings. A, Auxin influx. B, Acropetal
PAT. C, NPA-induced auxin accumulation. D, Auxin-induced PAT increase.
Three independent 35S-Oshox1 lines were used in all experiments and
considered as different genotypes in the subsequent statistical
analysis. Data represent the mean ± SE of two (A and
D) or four (B and C) separate experiments each performed on 11 n 24 (A-C) or 19 n 24 (D) seedlings per genotype. Difference between treated wild-type and
35S-Oshox1 populations was determined by two-way ANOVA followed by
Bonferroni's test and was significant at P < 0.001 in
B through D and not significant in A. No significant difference in the
behavior of the three different 35S-Oshox1 lines used was
detected.
|
|
Sensitivity of PAT toward Auxin in 35S-Oshox1 Rice
Seedlings
PAT is autoregulated by the endogenous auxin level through its
influence on the auxin efflux carrier (Hertel and Flory, 1968; Rayle et
al., 1969; Goldsmith, 1977; Yoon and Kang, 1992). We have previously
shown that adventitious and lateral root development in 35S-Oshox1
seedlings displayed increased auxin sensitivity, compared with wild
type (Scarpella et al., 2000). To test whether Oshox1
overexpression led to changes in the autoregulatory properties of PAT,
we monitored the increase in PAT capacity that is induced by exogenous
auxin in the anatomically indistinguishable wild-type and 35S-Oshox1
3-d-old roots. The data reported in Figure 8D show that whereas in
wild-type, PAT was enhanced of approximately 40% by the presence of
auxin, in 35S-Oshox1, the increase in PAT induced by auxin was of
approximately 100%.
| |
DISCUSSION |
PAT Inhibition Evokes Specific Vascular Pattern Defects in
Wild-Type Rice Leaves
A large number of classical studies (for review, see Sachs, 1981)
together with more recent genetic and novel experimental evidence (for
review, see Berleth et al., 2000) have associated the plant hormone
auxin with vascular tissue development in dicot plants. However, the
existence of such a link in monocot plants has always been rather
speculative. Tsiantis et al. (1999) have recently shown that PAT
inhibition induced hypertrophic vascular bundles in maize leaves,
supporting the possible interplay between PAT and vascular development
in a monocot species. Confirming and extending these findings, we have
shown here that NPA, HFCA, and TIBA, PATIs of three different classes,
not only induce vein hypertrophy in rice leaves but also evoke highly
reproducible alterations in the pattern of vein spacing, including an
increased distance between longitudinal veins and a reduced distance
between transverse veins. These results indicate a role for PAT in
restricting vascular tissue differentiation to narrow regions and
regulating the distance between longitudinal and transverse veins.
Two recent studies have addressed the critical role of PAT in vascular
patterning in dicot plants (Mattsson et al., 1999; Sieburth, 1999).
Like we observed in rice, veins of PATI-treated dicot leaves also
appeared hypertrophic. However, in rice, alterations in vein spacing
were the only pattern defects, whereas in leaves of dicot seedlings,
PATIs also induced vein interruptions or bifurcations, anomalously
shaped veins, and regions of increased or ectopic vascularization along
the margins (Mattsson et al., 1999; Sieburth, 1999). The differences
between the response of dicot and rice leaf vascular patterns toward
PAT inhibition could have several explanations. The reticulate venation
patterns of dicot leaves and the striate patterns of monocot leaves are
fundamentally different, both as to their eventual appearance in mature
leaves and as to their ontogeny during leaf development (Nelson and
Dengler, 1997). Therefore, it is possible that inherent differences in
the mechanisms underlying vascular pattern formation in dicot and
monocot leaves may lead to different phenotypical outcomes when leaves
are elicited with similar treatments. Adding further to the relation
between auxin and vein spacing in monocot leaves, we have recently
identified an auxin-insensitive mutant of rice that displays opposite
effects on the spacing of veins as those evoked by PATI treatment (E. Scarpella, S. Rueb, and A.H. Meijer, unpublished data).
An alternative explanation for the differences between the responses of
leaf vascular pattern to PAT inhibition in dicots and in rice could be
searched for in the reduced auxin sensitivity displayed by monocot
leaves compared with dicot leaves (e.g. Wenicke et al., 1981; Schmidt
and Willmitzer, 1988). As a consequence, ectopic accumulation of auxin
attributable to treatment with PATI inhibitors would not induce similar
responses in monocot and dicot tissues because of fundamental
differences in auxin sensitivity. However, we do not favor this
hypothesis, because the reduced auxin sensitivity displayed by rice
leaves seems to be restricted to their mature stages. In fact, immature
rice leaves can be induced to callus formation by the very same auxin
treatments that prove to be ineffective on mature leaves (Wenicke et
al., 1981). In our study, embryos were directly germinated in the
presence of PATIs and our evaluation was based on analysis of a leaf
that is initiated completely postembryonically. Therefore, in our
experimental design PATIs are expected to be able to affect any process
in leaf development that is sensitive to inhibition of PAT.
In addition to vascular tissue development, PAT has been shown to be
essential for other aspects of dicot plant development, such as
adventitious shoot and root formation, lateral root development, root
elongation, and gravitropic response (e.g. Lomax et al., 1995). We have
confirmed here that in rice, these developmental processes also are
dependent on PAT. However, not all of these processes seem to be
equally sensitive to different classes of PATIs. Variable effects were
observed on adventitious shoot development, adventitious root
development, and shoot morphogenesis. On the other hand, lateral root
development, leaf proximo-distal polarity acquisition, and vascular
tissue pattern formation seem to be equally affected by all classes of
PATIs. This suggests that these latter developmental parameters,
because they are independent of the chemical nature of the PATI, might
represent more bona fide phenotypical indicators of specific PAT alterations.
Oshox1 Confers Resistance toward the Effects
of PAT Inhibition on Lateral and Adventitious Root Development and on
Leaf Vascular Pattern Formation
We previously reported that seedlings overexpressing the rice
homeobox gene Oshox1, a proposed regulator of procambial
cell fate commitment, display reduced sensitivity to the inhibitory effects of NPA on the acropetal component of PAT in the root and on
lateral root development (Scarpella et al., 2000), which itself specifically depends on acropetal PAT (Reed et al., 1998). Root elongation and gravitropic response, both dependent on basipetal PAT
(Rashotte et al., 2000), conversely displayed wild-type NPA sensitivity
in the 35S-Oshox1 seedlings. We have here extended our previous
observations showing that in 35S-Oshox1 seedlings lateral and
adventitious root development is less sensitive to inhibition by
different classes of PATIs. Furthermore, we have shown that in
35S-Oshox1 seedlings, leaf vascular pattern is insensitive to PATI
treatment. PAT-competent cells are localized in the vascular tissues of
the shoot, and the cells of the vascular cylinder of the root are those
responsible for the acropetal component of PAT (Lomax et al., 1995, and
refs. therein). Taken together, these observations suggest that
Oshox1 overexpression seems to confer resistance to PAT
toward inhibition only as to the specific component of it that takes
place in the vascular tissues.
Oshox1 Reduces the Affinity of the NBP toward
NPA
We have shown here for the first time, to our knowledge, that
alterations in the expression level of a gene, Oshox1, can
affect the affinity with which the NBP binds NPA in seedlings, without any effect on the abundance of the NBP itself. Genetic evidence from
the Arabidopsis mutants tir3 and ifl1 has
associated aberrant vascular tissue development and reduced PAT
capacity with reduced number of NPA-binding sites (Ruegger et al.,
1997; Zhong and Ye, 2001). Furthermore, physiological studies have also
provided evidence for an association among vascular tissue development,
PAT capacity, and NPA-binding activity (Suttle, 1991). To exclude that
the reduced NPA-binding affinity measured in 35S-Oshox1 seedlings was
not, at least partly, a consequence of the premature vascular
differentiation and associated decreased PAT capacity that we observed
in 10-d-old roots of these seedlings (Scarpella et al., 2000), we
analyzed the effect of Oshox1 overexpression in
suspension-cultured cells and found that also here, NPA-binding
affinity was reduced without any effect on the number of NPA-binding
sites. This suggests that the reduced affinity of the NBP toward NPA
displayed by 35S-Oshox1 seedlings and isolated cells is a direct effect
of the Oshox1 overexpression and not a consequence of the
altered vascular development and PAT capacity observed in seedlings.
The reduced affinity of the NBP toward NPA is the likely cause of the
PATI-resistance of Oshox1-overexpressing seedlings and suspension-cultured cells. However, the NBP has been reported to
specifically interact with PATIs of the phytotropin and morphactin classes but not with TIBA (Rubery, 1990; E. Scarpella, K.J.M. Boot, S. Rueb, and A.H. Meijer, unpublished data). The present model for the
action of TIBA on PAT inhibition predicts that this PATI binds directly
to the efflux carrier rather than to the NBP (Rubery, 1990). In view of
this, it is difficult to explain how a reduced affinity of the NBP
toward NPA may account for the increased resistance to TIBA that
the 35S-Oshox1 seedlings display. However, a change in the number of
NPA-binding sites as well as in their affinity to NPA has been
previously reported to increase resistance also to TIBA (Suttle,
1991).
Specific inhibition of PAT correlates with the high-affinity binding of
phytotropins and morphactins to the NBP (for review, see Rubery, 1990;
Lomax et al., 1995; Muday, 2000). However, additional lower affinity
NPA-binding activities have been found in Arabidopsis seedlings (Jensen
et al., 1998; Murphy and Taiz, 1999; Murphy et al., 2000). Several
proteins have been recently purified by NPA-affinity chromatography
from membrane fractions derived from Arabidopsis seedlings (Murphy et
al., 2000, 2002; Noh et al., 2001). These proteins fall into the lower
affinity category, and some of them show some similarities with
proteins involved in mammalian actin-dependent vesicular cycling (Muday
and Murphy, 2002). Two other recent studies strongly suggest interplay
between PATIs and actin-dependent vesicular cycling of the putative
auxin efflux carrier protein PIN1 in the regulation of PAT in
Arabidopsis (Geldner et al., 2001; Gil et al., 2001; for review, see
Muday and Murphy, 2002). Treatments with brefeldin A (BFA), an
inhibitor of intracellular vesicle movement, result in mislocalization
of the PIN1 protein, and high concentrations of PATIs prevent the BFA-induced PIN1 mislocalization or the recovery of PIN1 polar subcellular localization after BFA removal (Geldner et al., 2001). Oshox1 seems not to be an integrated component of such a
regulatory pathway, in that its overexpression does not alter the
wild-type sensitivity toward the inhibitory effects of BFA on lateral
and adventitious root development (data not shown). Furthermore, the differences in sensitivity to PAT inhibition that we observed between wild type and 35S-Oshox1 were evident at concentrations of NPA
that are not sufficient to prevent the BFA-induced mislocalization of
the PIN1 protein or the recovery of its localization after BFA removal.
Therefore, consistently with the measured
Ka values for NPA binding in wild-type and
35S-Oshox1, both falling in the high-affinity range, we conclude that
Oshox1 specifically interferes with the action of a
high-affinity NBP, the identity of which at present remains still unknown.
Oshox1 Enhances PAT and Its Sensitivity toward Auxin
We previously reported that 10-d-old 35S-Oshox1 roots showed
reduced PAT capacity associated with premature vascular
differentiation, but we were unable to discriminate between cause and
effect (Scarpella et al., 2000). To address this question, we have
studied here the effect of Oshox1 on auxin efflux in
suspension-cultured rice cells, where Oshox1 overexpression
does not lead to any overt morphological changes. In 35S-Oshox1 cells,
auxin efflux was unexpectedly enhanced compared with wild-type. To
solve the apparent contradiction with our previous report, we monitored
the course of vascular tissue differentiation in wild-type and
35S-Oshox1 roots and observed that the premature vascular
differentiation could be unambiguously observed in 7-d-old 35S-Oshox1
roots but not in 3-d-old roots. Direct PAT measurements on 3-d-old
roots confirmed the conclusion from the cell suspension experiments
that Oshox1 overexpression confers enhanced PAT capacity in
the absence of anatomical differences with wild type. Furthermore, as
already observed in older seedlings, PAT was more resistant toward NPA
inhibition in 3-d-old 35S-Oshox1 roots.
Flavonoids have long been proposed as endogenous negative regulators of
PAT (Jacobs and Rubery, 1988; Rubery, 1990), a hypothesis recently
substantiated by observations that the transparent testa4 (tt4) mutant of Arabidopsis, which is completely devoid of
flavonoids, displays enhanced PAT while retaining wild-type NPA-binding
constants (Brown et al., 2001). Because Oshox1
overexpression directly affects the affinity of the NBP toward NPA, it
is conceivable that this gene might reduce sensitivity of the NBP
toward an endogenous NPA-like molecule, which could be the primary
cause of PAT elevation in 35S-Oshox1 3-d-old roots. However, it still
remains to be explained how a transcription factor, such as
Oshox1, may modify the affinity of a receptor for a ligand
without directly affecting the abundance of such receptor. A tentative
explanation could be that Oshox1 regulates the transcription
of a protein that modulates the binding properties of the NBP.
Physiological and genetic evidence support a role for reversible
protein phosphorylation in the regulation of PAT and its sensitivity
toward NPA (for review, see Muday and DeLong, 2001; DeLong et al.,
2002). The transcriptional regulation of genes encoding phosphatase or
kinase activities might thus represent a possible mechanism through
which Oshox1 could induce changes in the affinity of the NBP
toward NPA.
We previously reported that Oshox1 overexpression confers
enhanced sensitivity to the effect of exogenous auxin on adventitious and lateral root development (Scarpella et al., 2000). It is possible that this hypersensitivity may be directly attributable to an increased
sensitivity of PAT toward auxin, because low concentrations of polarly
transported auxins have been reported to have a positive effect on PAT
(Hertel and Flory, 1968; Rayle et al., 1969; Goldsmith, 1977; Yoon and
Kang, 1992). In the 3-d-old 35S-Oshox1 roots, which are anatomically
indistinguishable from wild type, PAT shows increased sensitivity to
exogenous auxin. Therefore, Oshox1 seems to increase the
sensitivity of the PAT machinery toward auxin signals. It is possible
that this increased sensitivity of PAT toward auxin might represent one
of the primary causes for the increased PAT capacity of these roots.
Furthermore, it is conceivable that the same pathway that leads to the
increase in sensitivity of PAT toward auxin signals might be
interconnected at the biochemical level with that eventually resulting
in the reduction of the affinity of the NPB toward NPA or an endogenous PATI.
A Model for the Action of Oshox1 in Procambial Cell
Fate Commitment
One direct consequence of the altered expression of
Oshox1 seems to be the reinforcement of the auxin
conductivity properties of procambial cells, possibly accomplished
through an increased sensitivity of the PAT machinery toward auxin
signals. The possibility that changes in auxin conductivity may
underlie procambial cell fate commitment is consistent with the fact
that one of the earliest detectable aspects of vascular tissue
development is the enhanced PAT capacity of incipient vascular cells,
which is measurable before any overt anatomical sign of vascular
differentiation (Gersani and Sachs, 1984). An additional direct
consequence of Oshox1 overexpression is the decreased
sensitivity of PAT toward inhibition. This effect can be conceivably
ascribed to a reduced affinity of the NBP toward an endogenous NPA-like
molecule. The existence of an NPA-insensitive auxin efflux carrier was
previously postulated by Mattsson et al. (1999) to tentatively explain
the unresponsiveness of anatomically recognizable procambial strands to
PAT inhibition. Taken together, these observations prompt us to propose
here that Oshox1 would mediate the acquisition of a cell
state in procambium commitment that is associated with an increase in
auxin conductivity properties. Through its effect on the affinity with
which the NBP binds an endogenous PATI, Oshox1 would
simultaneously stabilize this newly acquired cell state by reducing the
sensitivity of the PAT machinery toward inhibition.
The importance of the proposed role of Oshox1 in stabilizing
a particular cell state acquired during procambial cell fate commitment
can be appreciated when considering the high degree of flexibility of
vascular tissues under both experimentally manipulated conditions and
during the undisturbed course of development (e.g. Sachs, 1975, 1981).
Reorientation of the course of vascular strands has been shown to be
preceded by changes in the polarity of auxin transport (Gersani and
Sachs, 1984). Because of the relative ease with which a previously
established polarity of auxin transport can be reoriented in response
to external stimuli or internal signals, it thus seems that once a
certain polarity of auxin transport has been established, this needs to
be stably maintained (at least until irreversible differentiation has
been achieved) to eventually originate a coherent and consistent
pattern of vascular tissues. The stable maintenance is particularly
important during procambium early development, which displays features
of high sensitivity to PAT inhibition (Mattsson et al., 1999).
From our hypothesis for the action of Oshox1 proposed above,
it follows that in 35S-Oshox1 seedlings, as a consequence of enhanced
PAT and reduced NPA sensitivity, procambial cells would enter a phase
of cell fate stabilization (commitment) earlier (i.e. closer to the
meristem) than in wild type. This would eventually result in a
premature vascular differentiation, which would consequently reduce the
PAT capacity of the tissues, as we previously explained (Scarpella et
al., 2000). However, altered Oshox1 expression per se is not
sufficient to result in premature or ectopic vascular differentiation,
as shown by the 35S-Oshox1 cell suspension system discussed above and
by analysis of 35S-Oshox1 embryos (E. Scarpella, S. Rueb, and A.H.
Meijer, unpublished data). The premature vascular differentiation in 35S-Oshox1 seedlings could therefore more likely be
attributable to the action of genes that are specifically expressed in
the seedling vasculature and not in suspension-cultured cells, and the
expression of which is not under the direct control of Oshox1 gene activity. These vascular
differentiation-specific genes, one example of which is the
TED3 gene of zinnia (Zinnia elegans; Demura and
Fukuda, 1994) are plausibly active only postembryonically, in that
vascular differentiation does not take place during wild-type embryogenesis.
| |
MATERIALS AND METHODS |
Plant Materials and Growth Conditions
Rice (Oryza sativa L. Japonica cv Taipei 309), in
which the 35S-Oshox1 transgene was introduced (Scarpella et al., 2000), was used as a wild-type control strain in all studies. As an additional control, a 35S-
-glucuronidase transgenic line (Scarpella et al., 2000) was used in all experiments to monitor the possible effects of
hygromycin, which was used to select the 35S-Oshox1 transgenics, on the
measured parameters. Because, in all cases, we could not find any
significant difference between wild-type and 35S--glucuronidase populations (data not shown), only the data from the comparison between
wild-type and 35S-Oshox1 populations are presented. We have previously
reported the characterization of nine independent 35S-Oshox1 transgenic
lines and shown that we could detect identical phenotypes in all of
them. Among these nine lines, we have selected two (cell suspension
studies) or three (all other analyses) of them for the detailed
characterization described here. All seeds were surface sterilized
(Rueb et al., 1994) and germinated in the dark at 28°C for 4 d
on described medium (Scarpella et al., 2000) supplemented with 75 mg
L
1 hygromycin in the case of transgenics and with or
without different aliquots of filter-sterilized aqueous stocks of NPA
(Supelco, Bellefonte, PA), HFCA (Riedel-De Haën, Seelze, Germany)
or TIBA (Sigma-Aldrich, St. Louis). Germinated seeds were subsequently grown at a 12-h-light/12-h-dark cycle at 28°C for 2 weeks, or 4 weeks
in the case of vascular pattern parameters measurements, after which
the responses to the different growth conditions were evaluated.
Containers were placed vertically under an angle of approximately 30°
during both the germination and growth phases. All suspension-cultured
cells were initiated simultaneously from scutellum-derived calli (Rueb
et al., 1994) and grown on AA medium (Muller and Grafe, 1978)
containing 4 mg L1 naphthalene-1-acetic acid (BDH, Poole,
Dorset, UK) and 2 mg L1 kinetin (Research Organics,
Cleveland) at 28°C in the dark on a gyratory shaker at 120 rpm. Stock
suspensions were subcultured every 7 d at an initial density of
approximately 30 mg mL1. For all experiments, 7-d-old
suspensions at an approximate density of 100 mg mL1 were used.
Microtechniques and Microscopy
Seedling morphology was monitored with a stereoscopic microscope
(MZ12, Leica, Wetzlar, Germany). Whole-mount cleared preparations of
the fourth leaf, which in rice is the first leaf that is
initiated postembryonically, were obtained by autoclaving dissected
samples in 80% (w/w) lactic acid for 20 min at 121°C. Samples
were mounted in fresh 80% (w/w) lactic acid and viewed with an
Axioplan 2 Imaging microscope (Zeiss, Welwyn Garden City, UK) using
dark-field optics settings. Roots of wild-type and 35S-Oshox1 etiolated
seedlings at 3 or 7 dpg were dissected and treated according to a
modified version of the procedure described by Braselton et al. (1996). Roots were fixed for 1 h in ethanol:acetic acid (3:1, v/v) and, after rehydration, carbohydrates were hydrolyzed by a 15-min treatment with 5 N HCl and followed by an incubation of 2.5 h
with Schiff's reagent (Sigma-Aldrich). After two washing steps with
cold water of 15 min each, samples were dehydrated, infiltrated with LR
White resin (London Resin, Theale, Berkshire, UK) and embedded directly on microscope slides. Samples were observed with a Zeiss Axioplan microscope equipped with a confocal laser scanning unit (MRC 1024 ES,
Bio-Rad, Hercules, CA) using a 488- and a 512-nm excitation line and a
585-nm LP barrier emission filter. Microscopic images were acquired
with a 3CCD digital photo camera (DKC-5000, Sony, Tokyo). All images
were processed using Adobe Photoshop 5.5 (Adobe Systems, Mountain View,
CA). Morphometric analysis of vascular pattern parameters was performed
on digital pictures using the ImageJ 1.21 software.
NPA-Binding Assays
Microsomal membranes were isolated from 10-d-old etiolated
wild-type and 35S-Oshox1 seedlings or from cells at 7 d after
subculturing. Cells were harvested by filtration under vacuum over a
30-mesh nylon cloth. Approximately 2 to 4 g of cells or whole
seedlings were ground in liquid nitrogen and resuspended in cold
microsome-isolation buffer (50 mM Tris-HCl, pH 8.0, 250 mM Suc, 0.1 mM MgCl2, and 1 mM EDTA) to which 10 mM ascorbic acid and 1 mM dithiothreitol were freshly added. Samples were
homogenized in a Potter S homogenizer (B. Braun, Melsungen, Germany)
with 10 strokes at 1,500 rpm, and the homogenate was filtered over four
layers of cheesecloth. The filtrate was centrifuged at 10,000 rpm for
20 min, and the supernatant was centrifuged at 35,000 rpm for 45 min.
The pellet was resuspended in NPA-binding buffer (10 mM
citrate-acetic acid buffer, pH 5.5, 250 mM Suc, and 5 mM MgCl2), homogenized as described above and directly used for protein quantification. The whole isolation procedure was performed at 4°C. Binding assays were performed in 1 mL
total volume of NPA-binding buffer with a final protein concentration
of 0.1 to 0.2 mg mL1. Microsomes were incubated on ice in
the dark for 1 h with 109 M
[2,3,4,5(n)-3H]NPA (58 Ci
mmol1; Moravek Biochemicals, Brea, CA) and unlabeled NPA
concentrations ranging from 1010 to 103
M. Samples were subsequently filtrated under vacuum over
GF/C glass microfiber filters (Whatman, Maidstone, Kent, UK)
with a 1225 sampling manifold (Millipore, Bedford, MA) apparatus.
Filters were washed with cold water and radioactivity was measured with a liquid scintillation counter (1214 Rackbeta, LKB-Wallac, Turku, Finland). Initial estimations for the
Ka and Bmax were
done on Scatchard plots (Scatchard, 1949) obtained from the
transformation of the displacement plots and used in the Ligand program
(Munson and Rodbard, 1980) to determine the precise binding constants.
Auxin Influx and Efflux Assays
For auxin influx experiments, contribution of
carrier-mediated indole-3-acetic acid (IAA) uptake
(IAAcarrier) was deduced by subtracting the accumulation of
[5(n)-3H]IAA ([3H]IAA; Amersham, Little
Chalfont, Buckinghamshire, UK) measured in the presence of
104 M unlabeled IAA (Sigma), i.e. the
non-saturable IAA uptake (IAAdiffusion), from that measured
in the absence of unlabeled IAA (total IAA uptake:
IAAtotal): IAAcarrier = IAAtotal 
IAAdiffusion. For
simplification, we assume that the non-saturable component of the IAA
uptake corresponds to the diffusion of IAA through the membrane,
although it probably also includes the [3H]IAA
accumulated in the small volume of medium entrapped inside the cell
pellet (Delbarre et al., 1996). Values were normalized to the total IAA
uptake: IAAinflux = (IAAcarrier/IAAtotal) × 100%. Auxin
influx experiments in roots of 3-dpg wild-type and 35S-Oshox1 etiolated
seedlings were performed essentially as described for the PAT
measurements (Scarpella et al., 2000). The most distal 2 cm of the
seminal roots were excised and the most basal part of them
(approximately 5 mm) was placed in agar blocks containing 107 M [3H]IAA, in the presence
or absence of 104 M unlabeled IAA, and
incubated in the dark at room temperature for 30 min. To prevent the
tissues from drying, they were overlaid with silicon oil. After
incubation, roots were cut into two segments: a basal segment (5 mm),
which was that enclosed in the agar block, and an apical one (15 mm).
Radioactivity in the basal segment was measured with a liquid
scintillation counter. To determine auxin influx in cells, these were
filtered over a 30-mesh nylon cloth, washed twice with cold water,
resuspended in the same volume of hormone-free AA medium supplemented
with 50 mM MES (pH 5.5), and incubated at 28°C for 20 min. [3H]IAA (108 M) was
subsequently added in the presence or absence of 104
M unlabeled IAA or 106 M NPA.
Duplicate samples (0.25 mL) were taken 15 min after the addition of
[3H]IAA, when equilibrium had been reached. Samples were
filtered over GF/C filters, and radioactivity in the cells was measured with a liquid scintillation counter. NPA-induced IAA accumulation was
expressed as the relative increase in total [3H]IAA
uptake in the presence of 106 M NPA at
equilibrium. For auxin efflux experiments, cells were harvested,
resuspended, and incubated in hormone-free AA medium supplemented with
50 mM MES (pH 5.5) as described above.
[3H]IAA (108 M) was
subsequently added, and cells were incubated at 28°C for 15 min.
Duplicate samples (0.5 mL each) were taken at this point (time 0) and
filtered over GF/C filters, and radioactivity in the cells was measured
with a liquid scintillation counter (IAAt = 0). The
remaining cells were filtered over GF/C filters and resuspended in 50 mM MES-buffered medium (pH 5.5) without
[3H]IAA. Duplicate samples were taken 3 min after the
transfer to this medium and filtered over GF/C filters, and
radioactivity in the cells was measured with a liquid scintillation
counter (IAAt = 3). Auxin efflux was expressed as the
percentage of radioactivity lost by the cells at the 3-min sampling
point with respect to the radioactivity present in the cells at time 0:
([IAAt = 0 IAAt = 3]/IAAt = 0) × 100%.
PAT Assays
Acropetal PAT and NPA-induced IAA accumulation were
measured in roots of 3-dpg wild-type and 35S-Oshox1 etiolated
seedlings, as described (Scarpella et al., 2000). IAA-induced
stimulation of PAT was measured by performing the PAT measurements in
the presence or absence of 106 M unlabeled
IAA and reducing the incubation time to 30 min and was expressed as the
relative increase in PAT measured in the presence of the unlabeled IAA.
We thank Raoul Latib for invaluable help in morphometric
analysis, Elly Schrijnemakers for plant care, Dolf Weijers and
René Benjamins for help with confocal pictures, and Peter Hock
for artwork. We are grateful to Thomas Berleth and Gloria Muday for critically reading the manuscript.
Received May 30, 2002; returned for revision June 27, 2002; accepted July 12, 2002.
Article, publication date, and citation information can be found at
www.plantphysiol.org/cgi/doi/10.1104/pp.009167.
TOP
ABSTRACT
INTRODUCTION